diff options
Diffstat (limited to 'drivers/misc')
163 files changed, 35196 insertions, 5122 deletions
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index b151b7c1bd5..ee9402324a2 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -54,6 +54,7 @@ config AD525X_DPOT_SPI config ATMEL_PWM tristate "Atmel AT32/AT91 PWM support" depends on HAVE_CLK + depends on AVR32 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91SAM9G45 help This option enables device driver support for the PWM channels on certain Atmel processors. Pulse Width Modulation is used for @@ -93,6 +94,14 @@ config ATMEL_TCB_CLKSRC_BLOCK TC can be used for other purposes, such as PWM generation and interval timing. +config DUMMY_IRQ + tristate "Dummy IRQ handler" + default n + ---help--- + This module accepts a single 'irq' parameter, which it should register for. + The sole purpose of this module is to help with debugging of systems on + which spurious IRQs would happen on disabled IRQ vector. + config IBM_ASM tristate "Device driver for IBM RSA service processor" depends on X86 && PCI && INPUT @@ -127,7 +136,7 @@ config PHANTOM config INTEL_MID_PTI tristate "Parallel Trace Interface for MIPI P1149.7 cJTAG standard" - depends on PCI + depends on PCI && TTY && (X86_INTEL_MID || COMPILE_TEST) default n help The PTI (Parallel Trace Interface) driver directs @@ -192,7 +201,7 @@ config ICS932S401 config ATMEL_SSC tristate "Device driver for Atmel SSC peripheral" - depends on AVR32 || ARCH_AT91 + depends on HAS_IOMEM && (AVR32 || ARCH_AT91 || COMPILE_TEST) ---help--- This option enables device driver support for Atmel Synchronized Serial Communication peripheral (SSC). @@ -227,7 +236,7 @@ config SGI_XP config CS5535_MFGPT tristate "CS5535/CS5536 Geode Multi-Function General Purpose Timer (MFGPT) support" - depends on PCI && X86 && MFD_CS5535 + depends on MFD_CS5535 default n help This driver provides access to MFGPT functionality for other @@ -292,8 +301,8 @@ config SGI_GRU_DEBUG depends on SGI_GRU default n ---help--- - This option enables addition debugging code for the SGI GRU driver. If - you are unsure, say N. + This option enables additional debugging code for the SGI GRU driver. + If you are unsure, say N. config APDS9802ALS tristate "Medfield Avago APDS9802 ALS Sensor module" @@ -373,19 +382,6 @@ config HMC6352 This driver provides support for the Honeywell HMC6352 compass, providing configuration and heading data via sysfs. -config EP93XX_PWM - tristate "EP93xx PWM support" - depends on ARCH_EP93XX - help - This option enables device driver support for the PWM channels - on the Cirrus EP93xx processors. The EP9307 chip only has one - PWM channel all the others have two, the second channel is an - alternate function of the EGPIO14 pin. A sysfs interface is - provided to control the PWM channels. - - To compile this driver as a module, choose M here: the module will - be called ep93xx_pwm. - config DS1682 tristate "Dallas DS1682 Total Elapsed Time Recorder with Alarm" depends on I2C @@ -398,7 +394,7 @@ config DS1682 config SPEAR13XX_PCIE_GADGET bool "PCIe gadget support for SPEAr13XX platform" - depends on ARCH_SPEAR13XX + depends on ARCH_SPEAR13XX && BROKEN default n help This option enables gadget support for PCIe controller. If @@ -418,7 +414,7 @@ config TI_DAC7512 config VMWARE_BALLOON tristate "VMware Balloon Driver" - depends on X86 + depends on X86 && HYPERVISOR_GUEST help This is VMware physical memory management driver which acts like a "balloon" that can be inflated to reclaim physical pages @@ -472,7 +468,8 @@ config BMP085_SPI config PCH_PHUB tristate "Intel EG20T PCH/LAPIS Semicon IOH(ML7213/ML7223/ML7831) PHUB" - depends on PCI + select GENERIC_NET_UTILS + depends on PCI && (X86_32 || COMPILE_TEST) help This driver is for PCH(Platform controller Hub) PHUB(Packet Hub) of Intel Topcliff which is an IOH(Input/Output Hub) for x86 embedded @@ -499,6 +496,35 @@ config USB_SWITCH_FSA9480 stereo and mono audio, video, microphone and UART data to use a common connector port. +config LATTICE_ECP3_CONFIG + tristate "Lattice ECP3 FPGA bitstream configuration via SPI" + depends on SPI && SYSFS + select FW_LOADER + default n + help + This option enables support for bitstream configuration (programming + or loading) of the Lattice ECP3 FPGA family via SPI. + + If unsure, say N. + +config SRAM + bool "Generic on-chip SRAM driver" + depends on HAS_IOMEM + select GENERIC_ALLOCATOR + help + This driver allows you to declare a memory region to be managed by + the genalloc API. It is supposed to be used for small on-chip SRAM + areas found on many SoCs. + +config VEXPRESS_SYSCFG + bool "Versatile Express System Configuration driver" + depends on VEXPRESS_CONFIG + default y + help + ARM Ltd. Versatile Express uses specialised platform configuration + bus. System Configuration interface is one of the possible means + of generating transactions on this bus. + source "drivers/misc/c2port/Kconfig" source "drivers/misc/eeprom/Kconfig" source "drivers/misc/cb710/Kconfig" @@ -507,4 +533,8 @@ source "drivers/misc/lis3lv02d/Kconfig" source "drivers/misc/carma/Kconfig" source "drivers/misc/altera-stapl/Kconfig" source "drivers/misc/mei/Kconfig" +source "drivers/misc/vmw_vmci/Kconfig" +source "drivers/misc/mic/Kconfig" +source "drivers/misc/genwqe/Kconfig" +source "drivers/misc/echo/Kconfig" endmenu diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile index 2129377c0de..d59ce1261b3 100644 --- a/drivers/misc/Makefile +++ b/drivers/misc/Makefile @@ -13,6 +13,7 @@ obj-$(CONFIG_ATMEL_TCLIB) += atmel_tclib.o obj-$(CONFIG_BMP085) += bmp085.o obj-$(CONFIG_BMP085_I2C) += bmp085-i2c.o obj-$(CONFIG_BMP085_SPI) += bmp085-spi.o +obj-$(CONFIG_DUMMY_IRQ) += dummy-irq.o obj-$(CONFIG_ICS932S401) += ics932s401.o obj-$(CONFIG_LKDTM) += lkdtm.o obj-$(CONFIG_TIFM_CORE) += tifm_core.o @@ -32,7 +33,6 @@ obj-$(CONFIG_APDS9802ALS) += apds9802als.o obj-$(CONFIG_ISL29003) += isl29003.o obj-$(CONFIG_ISL29020) += isl29020.o obj-$(CONFIG_SENSORS_TSL2550) += tsl2550.o -obj-$(CONFIG_EP93XX_PWM) += ep93xx_pwm.o obj-$(CONFIG_DS1682) += ds1682.o obj-$(CONFIG_TI_DAC7512) += ti_dac7512.o obj-$(CONFIG_C2PORT) += c2port/ @@ -49,3 +49,10 @@ obj-y += carma/ obj-$(CONFIG_USB_SWITCH_FSA9480) += fsa9480.o obj-$(CONFIG_ALTERA_STAPL) +=altera-stapl/ obj-$(CONFIG_INTEL_MEI) += mei/ +obj-$(CONFIG_VMWARE_VMCI) += vmw_vmci/ +obj-$(CONFIG_LATTICE_ECP3_CONFIG) += lattice-ecp3-config.o +obj-$(CONFIG_SRAM) += sram.o +obj-y += mic/ +obj-$(CONFIG_GENWQE) += genwqe/ +obj-$(CONFIG_ECHO) += echo/ +obj-$(CONFIG_VEXPRESS_SYSCFG) += vexpress-syscfg.o diff --git a/drivers/misc/ad525x_dpot-i2c.c b/drivers/misc/ad525x_dpot-i2c.c index 820826270b6..705b881e186 100644 --- a/drivers/misc/ad525x_dpot-i2c.c +++ b/drivers/misc/ad525x_dpot-i2c.c @@ -51,7 +51,7 @@ static const struct ad_dpot_bus_ops bops = { .write_r8d16 = write_r8d16, }; -static int __devinit ad_dpot_i2c_probe(struct i2c_client *client, +static int ad_dpot_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct ad_dpot_bus_data bdata = { @@ -68,7 +68,7 @@ static int __devinit ad_dpot_i2c_probe(struct i2c_client *client, return ad_dpot_probe(&client->dev, &bdata, id->driver_data, id->name); } -static int __devexit ad_dpot_i2c_remove(struct i2c_client *client) +static int ad_dpot_i2c_remove(struct i2c_client *client) { return ad_dpot_remove(&client->dev); } @@ -109,7 +109,7 @@ static struct i2c_driver ad_dpot_i2c_driver = { .owner = THIS_MODULE, }, .probe = ad_dpot_i2c_probe, - .remove = __devexit_p(ad_dpot_i2c_remove), + .remove = ad_dpot_i2c_remove, .id_table = ad_dpot_id, }; diff --git a/drivers/misc/ad525x_dpot-spi.c b/drivers/misc/ad525x_dpot-spi.c index f62317540d0..9da04ede04f 100644 --- a/drivers/misc/ad525x_dpot-spi.c +++ b/drivers/misc/ad525x_dpot-spi.c @@ -75,7 +75,7 @@ static const struct ad_dpot_bus_ops bops = { .write_r8d8 = write16, .write_r8d16 = write24, }; -static int __devinit ad_dpot_spi_probe(struct spi_device *spi) +static int ad_dpot_spi_probe(struct spi_device *spi) { struct ad_dpot_bus_data bdata = { .client = spi, @@ -87,7 +87,7 @@ static int __devinit ad_dpot_spi_probe(struct spi_device *spi) spi_get_device_id(spi)->name); } -static int __devexit ad_dpot_spi_remove(struct spi_device *spi) +static int ad_dpot_spi_remove(struct spi_device *spi) { return ad_dpot_remove(&spi->dev); } @@ -131,7 +131,7 @@ static struct spi_driver ad_dpot_spi_driver = { .owner = THIS_MODULE, }, .probe = ad_dpot_spi_probe, - .remove = __devexit_p(ad_dpot_spi_remove), + .remove = ad_dpot_spi_remove, .id_table = ad_dpot_spi_id, }; diff --git a/drivers/misc/ad525x_dpot.c b/drivers/misc/ad525x_dpot.c index 6938f1be664..a43053daad0 100644 --- a/drivers/misc/ad525x_dpot.c +++ b/drivers/misc/ad525x_dpot.c @@ -72,7 +72,6 @@ #include <linux/module.h> #include <linux/device.h> #include <linux/kernel.h> -#include <linux/init.h> #include <linux/delay.h> #include <linux/slab.h> @@ -470,7 +469,7 @@ static ssize_t sysfs_set_reg(struct device *dev, !test_bit(DPOT_RDAC_MASK & reg, data->otp_en_mask)) return -EPERM; - err = strict_strtoul(buf, 10, &value); + err = kstrtoul(buf, 10, &value); if (err) return err; @@ -641,7 +640,7 @@ static const struct attribute_group ad525x_group_commands = { .attrs = ad525x_attributes_commands, }; -__devinit int ad_dpot_add_files(struct device *dev, +static int ad_dpot_add_files(struct device *dev, unsigned features, unsigned rdac) { int err = sysfs_create_file(&dev->kobj, @@ -666,7 +665,7 @@ __devinit int ad_dpot_add_files(struct device *dev, return err; } -inline void ad_dpot_remove_files(struct device *dev, +static inline void ad_dpot_remove_files(struct device *dev, unsigned features, unsigned rdac) { sysfs_remove_file(&dev->kobj, @@ -685,7 +684,7 @@ inline void ad_dpot_remove_files(struct device *dev, } } -int __devinit ad_dpot_probe(struct device *dev, +int ad_dpot_probe(struct device *dev, struct ad_dpot_bus_data *bdata, unsigned long devid, const char *name) { diff --git a/drivers/misc/apds9802als.c b/drivers/misc/apds9802als.c index 0314773f6db..c6cc3dc8ae1 100644 --- a/drivers/misc/apds9802als.c +++ b/drivers/misc/apds9802als.c @@ -22,7 +22,6 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <linux/slab.h> #include <linux/i2c.h> #include <linux/err.h> @@ -68,7 +67,7 @@ static int als_wait_for_data_ready(struct device *dev) ret = i2c_smbus_read_byte_data(client, 0x86); } while (!(ret & 0x80) && retry--); - if (!retry) { + if (retry < 0) { dev_warn(dev, "timeout waiting for data ready\n"); return -ETIMEDOUT; } @@ -126,8 +125,9 @@ static ssize_t als_sensing_range_store(struct device *dev, int ret_val; unsigned long val; - if (strict_strtoul(buf, 10, &val)) - return -EINVAL; + ret_val = kstrtoul(buf, 10, &val); + if (ret_val) + return ret_val; if (val < 4096) val = 1; @@ -254,7 +254,7 @@ als_error1: return res; } -static int __devexit apds9802als_remove(struct i2c_client *client) +static int apds9802als_remove(struct i2c_client *client) { struct als_data *data = i2c_get_clientdata(client); @@ -272,19 +272,8 @@ static int __devexit apds9802als_remove(struct i2c_client *client) } #ifdef CONFIG_PM -static int apds9802als_suspend(struct i2c_client *client, pm_message_t mesg) -{ - als_set_power_state(client, false); - return 0; -} -static int apds9802als_resume(struct i2c_client *client) -{ - als_set_default_config(client); - return 0; -} - -static int apds9802als_runtime_suspend(struct device *dev) +static int apds9802als_suspend(struct device *dev) { struct i2c_client *client = to_i2c_client(dev); @@ -292,7 +281,7 @@ static int apds9802als_runtime_suspend(struct device *dev) return 0; } -static int apds9802als_runtime_resume(struct device *dev) +static int apds9802als_resume(struct device *dev) { struct i2c_client *client = to_i2c_client(dev); @@ -300,16 +289,12 @@ static int apds9802als_runtime_resume(struct device *dev) return 0; } -static const struct dev_pm_ops apds9802als_pm_ops = { - .runtime_suspend = apds9802als_runtime_suspend, - .runtime_resume = apds9802als_runtime_resume, -}; +static UNIVERSAL_DEV_PM_OPS(apds9802als_pm_ops, apds9802als_suspend, + apds9802als_resume, NULL); #define APDS9802ALS_PM_OPS (&apds9802als_pm_ops) #else /* CONFIG_PM */ -#define apds9802als_suspend NULL -#define apds9802als_resume NULL #define APDS9802ALS_PM_OPS NULL #endif /* CONFIG_PM */ @@ -326,9 +311,7 @@ static struct i2c_driver apds9802als_driver = { .pm = APDS9802ALS_PM_OPS, }, .probe = apds9802als_probe, - .remove = __devexit_p(apds9802als_remove), - .suspend = apds9802als_suspend, - .resume = apds9802als_resume, + .remove = apds9802als_remove, .id_table = apds9802als_id, }; diff --git a/drivers/misc/apds990x.c b/drivers/misc/apds990x.c index ee74244aa03..868a30a1b41 100644 --- a/drivers/misc/apds990x.c +++ b/drivers/misc/apds990x.c @@ -696,12 +696,11 @@ static ssize_t apds990x_lux_calib_store(struct device *dev, { struct apds990x_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; - - if (chip->lux_calib > APDS_RANGE) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; chip->lux_calib = value; @@ -762,8 +761,9 @@ static ssize_t apds990x_rate_store(struct device *dev, unsigned long value; int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; mutex_lock(&chip->mutex); ret = apds990x_set_arate(chip, value); @@ -816,9 +816,11 @@ static ssize_t apds990x_prox_enable_store(struct device *dev, { struct apds990x_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; mutex_lock(&chip->mutex); @@ -895,11 +897,12 @@ static ssize_t apds990x_lux_thresh_below_show(struct device *dev, static ssize_t apds990x_set_lux_thresh(struct apds990x_chip *chip, u32 *target, const char *buf) { - int ret = 0; unsigned long thresh; + int ret; - if (strict_strtoul(buf, 0, &thresh)) - return -EINVAL; + ret = kstrtoul(buf, 0, &thresh); + if (ret) + return ret; if (thresh > APDS_RANGE) return -EINVAL; @@ -960,9 +963,11 @@ static ssize_t apds990x_prox_threshold_store(struct device *dev, { struct apds990x_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; if ((value > APDS_RANGE) || (value == 0) || (value < APDS_PROX_HYSTERESIS)) @@ -993,9 +998,12 @@ static ssize_t apds990x_power_state_store(struct device *dev, { struct apds990x_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; + + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; if (value) { pm_runtime_get_sync(dev); mutex_lock(&chip->mutex); @@ -1047,7 +1055,7 @@ static struct attribute_group apds990x_attribute_group[] = { {.attrs = sysfs_attrs_ctrl }, }; -static int __devinit apds990x_probe(struct i2c_client *client, +static int apds990x_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct apds990x_chip *chip; @@ -1181,7 +1189,7 @@ fail1: return err; } -static int __devexit apds990x_remove(struct i2c_client *client) +static int apds990x_remove(struct i2c_client *client) { struct apds990x_chip *chip = i2c_get_clientdata(client); @@ -1204,7 +1212,7 @@ static int __devexit apds990x_remove(struct i2c_client *client) return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP static int apds990x_suspend(struct device *dev) { struct i2c_client *client = container_of(dev, struct i2c_client, dev); @@ -1227,10 +1235,6 @@ static int apds990x_resume(struct device *dev) return 0; } -#else -#define apds990x_suspend NULL -#define apds990x_resume NULL -#define apds990x_shutdown NULL #endif #ifdef CONFIG_PM_RUNTIME @@ -1275,7 +1279,7 @@ static struct i2c_driver apds990x_driver = { .pm = &apds990x_pm_ops, }, .probe = apds990x_probe, - .remove = __devexit_p(apds990x_remove), + .remove = apds990x_remove, .id_table = apds990x_id, }; diff --git a/drivers/misc/arm-charlcd.c b/drivers/misc/arm-charlcd.c index fe8616a8d28..c72e96b523e 100644 --- a/drivers/misc/arm-charlcd.c +++ b/drivers/misc/arm-charlcd.c @@ -11,6 +11,7 @@ #include <linux/module.h> #include <linux/interrupt.h> #include <linux/platform_device.h> +#include <linux/of.h> #include <linux/completion.h> #include <linux/delay.h> #include <linux/io.h> @@ -291,13 +292,13 @@ static int __init charlcd_probe(struct platform_device *pdev) lcd->virtbase = ioremap(lcd->phybase, lcd->physize); if (!lcd->virtbase) { ret = -ENOMEM; - goto out_no_remap; + goto out_no_memregion; } lcd->irq = platform_get_irq(pdev, 0); /* If no IRQ is supplied, we'll survive without it */ if (lcd->irq >= 0) { - if (request_irq(lcd->irq, charlcd_interrupt, IRQF_DISABLED, + if (request_irq(lcd->irq, charlcd_interrupt, 0, DRIVERNAME, lcd)) { ret = -EIO; goto out_no_irq; @@ -320,8 +321,6 @@ static int __init charlcd_probe(struct platform_device *pdev) out_no_irq: iounmap(lcd->virtbase); -out_no_remap: - platform_set_drvdata(pdev, NULL); out_no_memregion: release_mem_region(lcd->phybase, SZ_4K); out_no_resource: @@ -337,7 +336,6 @@ static int __exit charlcd_remove(struct platform_device *pdev) free_irq(lcd->irq, lcd); iounmap(lcd->virtbase); release_mem_region(lcd->phybase, lcd->physize); - platform_set_drvdata(pdev, NULL); kfree(lcd); } @@ -369,27 +367,22 @@ static const struct dev_pm_ops charlcd_pm_ops = { .resume = charlcd_resume, }; +static const struct of_device_id charlcd_match[] = { + { .compatible = "arm,versatile-lcd", }, + {} +}; + static struct platform_driver charlcd_driver = { .driver = { .name = DRIVERNAME, .owner = THIS_MODULE, .pm = &charlcd_pm_ops, + .of_match_table = of_match_ptr(charlcd_match), }, .remove = __exit_p(charlcd_remove), }; -static int __init charlcd_init(void) -{ - return platform_driver_probe(&charlcd_driver, charlcd_probe); -} - -static void __exit charlcd_exit(void) -{ - platform_driver_unregister(&charlcd_driver); -} - -module_init(charlcd_init); -module_exit(charlcd_exit); +module_platform_driver_probe(charlcd_driver, charlcd_probe); MODULE_AUTHOR("Linus Walleij <triad@df.lth.se>"); MODULE_DESCRIPTION("ARM Character LCD Driver"); diff --git a/drivers/misc/atmel-ssc.c b/drivers/misc/atmel-ssc.c index 5bb18778107..22de1372764 100644 --- a/drivers/misc/atmel-ssc.c +++ b/drivers/misc/atmel-ssc.c @@ -18,6 +18,8 @@ #include <linux/slab.h> #include <linux/module.h> +#include <linux/of.h> + /* Serialize access to ssc_list and user count */ static DEFINE_SPINLOCK(user_lock); static LIST_HEAD(ssc_list); @@ -29,7 +31,13 @@ struct ssc_device *ssc_request(unsigned int ssc_num) spin_lock(&user_lock); list_for_each_entry(ssc, &ssc_list, list) { - if (ssc->pdev->id == ssc_num) { + if (ssc->pdev->dev.of_node) { + if (of_alias_get_id(ssc->pdev->dev.of_node, "ssc") + == ssc_num) { + ssc_valid = 1; + break; + } + } else if (ssc->pdev->id == ssc_num) { ssc_valid = 1; break; } @@ -49,7 +57,7 @@ struct ssc_device *ssc_request(unsigned int ssc_num) ssc->user++; spin_unlock(&user_lock); - clk_enable(ssc->clk); + clk_prepare_enable(ssc->clk); return ssc; } @@ -57,63 +65,120 @@ EXPORT_SYMBOL(ssc_request); void ssc_free(struct ssc_device *ssc) { + bool disable_clk = true; + spin_lock(&user_lock); - if (ssc->user) { + if (ssc->user) ssc->user--; - clk_disable(ssc->clk); - } else { + else { + disable_clk = false; dev_dbg(&ssc->pdev->dev, "device already free\n"); } spin_unlock(&user_lock); + + if (disable_clk) + clk_disable_unprepare(ssc->clk); } EXPORT_SYMBOL(ssc_free); -static int __init ssc_probe(struct platform_device *pdev) +static struct atmel_ssc_platform_data at91rm9200_config = { + .use_dma = 0, +}; + +static struct atmel_ssc_platform_data at91sam9g45_config = { + .use_dma = 1, +}; + +static const struct platform_device_id atmel_ssc_devtypes[] = { + { + .name = "at91rm9200_ssc", + .driver_data = (unsigned long) &at91rm9200_config, + }, { + .name = "at91sam9g45_ssc", + .driver_data = (unsigned long) &at91sam9g45_config, + }, { + /* sentinel */ + } +}; + +#ifdef CONFIG_OF +static const struct of_device_id atmel_ssc_dt_ids[] = { + { + .compatible = "atmel,at91rm9200-ssc", + .data = &at91rm9200_config, + }, { + .compatible = "atmel,at91sam9g45-ssc", + .data = &at91sam9g45_config, + }, { + /* sentinel */ + } +}; +MODULE_DEVICE_TABLE(of, atmel_ssc_dt_ids); +#endif + +static inline const struct atmel_ssc_platform_data * __init + atmel_ssc_get_driver_data(struct platform_device *pdev) +{ + if (pdev->dev.of_node) { + const struct of_device_id *match; + match = of_match_node(atmel_ssc_dt_ids, pdev->dev.of_node); + if (match == NULL) + return NULL; + return match->data; + } + + return (struct atmel_ssc_platform_data *) + platform_get_device_id(pdev)->driver_data; +} + +static int ssc_probe(struct platform_device *pdev) { - int retval = 0; struct resource *regs; struct ssc_device *ssc; + const struct atmel_ssc_platform_data *plat_dat; - ssc = kzalloc(sizeof(struct ssc_device), GFP_KERNEL); + ssc = devm_kzalloc(&pdev->dev, sizeof(struct ssc_device), GFP_KERNEL); if (!ssc) { dev_dbg(&pdev->dev, "out of memory\n"); - retval = -ENOMEM; - goto out; + return -ENOMEM; } - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!regs) { - dev_dbg(&pdev->dev, "no mmio resource defined\n"); - retval = -ENXIO; - goto out_free; + ssc->pdev = pdev; + + plat_dat = atmel_ssc_get_driver_data(pdev); + if (!plat_dat) + return -ENODEV; + ssc->pdata = (struct atmel_ssc_platform_data *)plat_dat; + + if (pdev->dev.of_node) { + struct device_node *np = pdev->dev.of_node; + ssc->clk_from_rk_pin = + of_property_read_bool(np, "atmel,clk-from-rk-pin"); } - ssc->clk = clk_get(&pdev->dev, "pclk"); + regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); + ssc->regs = devm_ioremap_resource(&pdev->dev, regs); + if (IS_ERR(ssc->regs)) + return PTR_ERR(ssc->regs); + + ssc->phybase = regs->start; + + ssc->clk = devm_clk_get(&pdev->dev, "pclk"); if (IS_ERR(ssc->clk)) { dev_dbg(&pdev->dev, "no pclk clock defined\n"); - retval = -ENXIO; - goto out_free; - } - - ssc->pdev = pdev; - ssc->regs = ioremap(regs->start, resource_size(regs)); - if (!ssc->regs) { - dev_dbg(&pdev->dev, "ioremap failed\n"); - retval = -EINVAL; - goto out_clk; + return -ENXIO; } /* disable all interrupts */ - clk_enable(ssc->clk); - ssc_writel(ssc->regs, IDR, ~0UL); + clk_prepare_enable(ssc->clk); + ssc_writel(ssc->regs, IDR, -1); ssc_readl(ssc->regs, SR); - clk_disable(ssc->clk); + clk_disable_unprepare(ssc->clk); ssc->irq = platform_get_irq(pdev, 0); if (!ssc->irq) { dev_dbg(&pdev->dev, "could not get irq\n"); - retval = -ENXIO; - goto out_unmap; + return -ENXIO; } spin_lock(&user_lock); @@ -125,51 +190,31 @@ static int __init ssc_probe(struct platform_device *pdev) dev_info(&pdev->dev, "Atmel SSC device at 0x%p (irq %d)\n", ssc->regs, ssc->irq); - goto out; - -out_unmap: - iounmap(ssc->regs); -out_clk: - clk_put(ssc->clk); -out_free: - kfree(ssc); -out: - return retval; + return 0; } -static int __devexit ssc_remove(struct platform_device *pdev) +static int ssc_remove(struct platform_device *pdev) { struct ssc_device *ssc = platform_get_drvdata(pdev); spin_lock(&user_lock); - iounmap(ssc->regs); - clk_put(ssc->clk); list_del(&ssc->list); - kfree(ssc); spin_unlock(&user_lock); return 0; } static struct platform_driver ssc_driver = { - .remove = __devexit_p(ssc_remove), .driver = { .name = "ssc", .owner = THIS_MODULE, + .of_match_table = of_match_ptr(atmel_ssc_dt_ids), }, + .id_table = atmel_ssc_devtypes, + .probe = ssc_probe, + .remove = ssc_remove, }; - -static int __init ssc_init(void) -{ - return platform_driver_probe(&ssc_driver, ssc_probe); -} -module_init(ssc_init); - -static void __exit ssc_exit(void) -{ - platform_driver_unregister(&ssc_driver); -} -module_exit(ssc_exit); +module_platform_driver(ssc_driver); MODULE_AUTHOR("Hans-Christian Egtvedt <hcegtvedt@atmel.com>"); MODULE_DESCRIPTION("SSC driver for Atmel AVR32 and AT91"); diff --git a/drivers/misc/atmel_pwm.c b/drivers/misc/atmel_pwm.c index 28f5aaa19d4..a6dc56e1bc5 100644 --- a/drivers/misc/atmel_pwm.c +++ b/drivers/misc/atmel_pwm.c @@ -90,8 +90,10 @@ int pwm_channel_alloc(int index, struct pwm_channel *ch) unsigned long flags; int status = 0; - /* insist on PWM init, with this signal pinned out */ - if (!pwm || !(pwm->mask & 1 << index)) + if (!pwm) + return -EPROBE_DEFER; + + if (!(pwm->mask & 1 << index)) return -ENODEV; if (index < 0 || index >= PWM_NCHAN || !ch) @@ -393,17 +395,7 @@ static struct platform_driver atmel_pwm_driver = { */ }; -static int __init pwm_init(void) -{ - return platform_driver_probe(&atmel_pwm_driver, pwm_probe); -} -module_init(pwm_init); - -static void __exit pwm_exit(void) -{ - platform_driver_unregister(&atmel_pwm_driver); -} -module_exit(pwm_exit); +module_platform_driver_probe(atmel_pwm_driver, pwm_probe); MODULE_DESCRIPTION("Driver for AT32/AT91 PWM module"); MODULE_LICENSE("GPL"); diff --git a/drivers/misc/bh1770glc.c b/drivers/misc/bh1770glc.c index 3d56ae7ef8d..99a04686e45 100644 --- a/drivers/misc/bh1770glc.c +++ b/drivers/misc/bh1770glc.c @@ -651,8 +651,9 @@ static ssize_t bh1770_power_state_store(struct device *dev, unsigned long value; ssize_t ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; mutex_lock(&chip->mutex); if (value) { @@ -726,9 +727,11 @@ static ssize_t bh1770_prox_enable_store(struct device *dev, { struct bh1770_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; mutex_lock(&chip->mutex); /* Assume no proximity. Sensor will tell real state soon */ @@ -824,9 +827,11 @@ static ssize_t bh1770_set_prox_rate_above(struct device *dev, { struct bh1770_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; mutex_lock(&chip->mutex); chip->prox_rate_threshold = bh1770_prox_rate_validate(value); @@ -840,9 +845,11 @@ static ssize_t bh1770_set_prox_rate_below(struct device *dev, { struct bh1770_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; mutex_lock(&chip->mutex); chip->prox_rate = bh1770_prox_rate_validate(value); @@ -865,8 +872,10 @@ static ssize_t bh1770_set_prox_thres(struct device *dev, unsigned long value; int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; + if (value > BH1770_PROX_RANGE) return -EINVAL; @@ -893,9 +902,11 @@ static ssize_t bh1770_prox_persistence_store(struct device *dev, { struct bh1770_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; if (value > BH1770_PROX_MAX_PERSISTENCE) return -EINVAL; @@ -918,9 +929,11 @@ static ssize_t bh1770_prox_abs_thres_store(struct device *dev, { struct bh1770_chip *chip = dev_get_drvdata(dev); unsigned long value; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; if (value > BH1770_PROX_RANGE) return -EINVAL; @@ -963,9 +976,11 @@ static ssize_t bh1770_lux_calib_store(struct device *dev, unsigned long value; u32 old_calib; u32 new_corr; + int ret; - if (strict_strtoul(buf, 0, &value)) - return -EINVAL; + ret = kstrtoul(buf, 0, &value); + if (ret) + return ret; mutex_lock(&chip->mutex); old_calib = chip->lux_calib; @@ -1012,8 +1027,9 @@ static ssize_t bh1770_set_lux_rate(struct device *dev, unsigned long rate_hz; int ret, i; - if (strict_strtoul(buf, 0, &rate_hz)) - return -EINVAL; + ret = kstrtoul(buf, 0, &rate_hz); + if (ret) + return ret; for (i = 0; i < ARRAY_SIZE(lux_rates_hz) - 1; i++) if (rate_hz >= lux_rates_hz[i]) @@ -1047,11 +1063,12 @@ static ssize_t bh1770_get_lux_thresh_below(struct device *dev, static ssize_t bh1770_set_lux_thresh(struct bh1770_chip *chip, u16 *target, const char *buf) { - int ret = 0; unsigned long thresh; + int ret; - if (strict_strtoul(buf, 0, &thresh)) - return -EINVAL; + ret = kstrtoul(buf, 0, &thresh); + if (ret) + return ret; if (thresh > BH1770_LUX_RANGE) return -EINVAL; @@ -1162,7 +1179,7 @@ static struct attribute_group bh1770_attribute_group = { .attrs = sysfs_attrs }; -static int __devinit bh1770_probe(struct i2c_client *client, +static int bh1770_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct bh1770_chip *chip; @@ -1285,7 +1302,7 @@ fail1: return err; } -static int __devexit bh1770_remove(struct i2c_client *client) +static int bh1770_remove(struct i2c_client *client) { struct bh1770_chip *chip = i2c_get_clientdata(client); @@ -1310,7 +1327,7 @@ static int __devexit bh1770_remove(struct i2c_client *client) return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP static int bh1770_suspend(struct device *dev) { struct i2c_client *client = container_of(dev, struct i2c_client, dev); @@ -1346,11 +1363,6 @@ static int bh1770_resume(struct device *dev) } return ret; } - -#else -#define bh1770_suspend NULL -#define bh1770_shutdown NULL -#define bh1770_resume NULL #endif #ifdef CONFIG_PM_RUNTIME @@ -1395,7 +1407,7 @@ static struct i2c_driver bh1770_driver = { .pm = &bh1770_pm_ops, }, .probe = bh1770_probe, - .remove = __devexit_p(bh1770_remove), + .remove = bh1770_remove, .id_table = bh1770_id, }; diff --git a/drivers/misc/bh1780gli.c b/drivers/misc/bh1780gli.c index f1f9877f3fd..48ea33d15a7 100644 --- a/drivers/misc/bh1780gli.c +++ b/drivers/misc/bh1780gli.c @@ -23,6 +23,7 @@ #include <linux/platform_device.h> #include <linux/delay.h> #include <linux/module.h> +#include <linux/of.h> #define BH1780_REG_CONTROL 0x80 #define BH1780_REG_PARTID 0x8A @@ -107,7 +108,7 @@ static ssize_t bh1780_store_power_state(struct device *dev, unsigned long val; int error; - error = strict_strtoul(buf, 0, &val); + error = kstrtoul(buf, 0, &val); if (error) return error; @@ -144,7 +145,7 @@ static const struct attribute_group bh1780_attr_group = { .attrs = bh1780_attributes, }; -static int __devinit bh1780_probe(struct i2c_client *client, +static int bh1780_probe(struct i2c_client *client, const struct i2c_device_id *id) { int ret; @@ -185,7 +186,7 @@ err_op_failed: return ret; } -static int __devexit bh1780_remove(struct i2c_client *client) +static int bh1780_remove(struct i2c_client *client) { struct bh1780_data *ddata; @@ -196,7 +197,7 @@ static int __devexit bh1780_remove(struct i2c_client *client) return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP static int bh1780_suspend(struct device *dev) { struct bh1780_data *ddata; @@ -235,24 +236,32 @@ static int bh1780_resume(struct device *dev) return 0; } +#endif /* CONFIG_PM_SLEEP */ + static SIMPLE_DEV_PM_OPS(bh1780_pm, bh1780_suspend, bh1780_resume); -#define BH1780_PMOPS (&bh1780_pm) -#else -#define BH1780_PMOPS NULL -#endif /* CONFIG_PM */ static const struct i2c_device_id bh1780_id[] = { { "bh1780", 0 }, { }, }; +#ifdef CONFIG_OF +static const struct of_device_id of_bh1780_match[] = { + { .compatible = "rohm,bh1780gli", }, + {}, +}; + +MODULE_DEVICE_TABLE(of, of_bh1780_match); +#endif + static struct i2c_driver bh1780_driver = { .probe = bh1780_probe, - .remove = __devexit_p(bh1780_remove), + .remove = bh1780_remove, .id_table = bh1780_id, .driver = { .name = "bh1780", - .pm = BH1780_PMOPS, + .pm = &bh1780_pm, + .of_match_table = of_match_ptr(of_bh1780_match), }, }; diff --git a/drivers/misc/bmp085-i2c.c b/drivers/misc/bmp085-i2c.c index a4f33c995ea..a7c16295b81 100644 --- a/drivers/misc/bmp085-i2c.c +++ b/drivers/misc/bmp085-i2c.c @@ -36,7 +36,7 @@ static int bmp085_i2c_detect(struct i2c_client *client, return bmp085_detect(&client->dev); } -static int __devinit bmp085_i2c_probe(struct i2c_client *client, +static int bmp085_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { int err; @@ -49,7 +49,7 @@ static int __devinit bmp085_i2c_probe(struct i2c_client *client, return err; } - return bmp085_probe(&client->dev, regmap); + return bmp085_probe(&client->dev, regmap, client->irq); } static int bmp085_i2c_remove(struct i2c_client *client) @@ -71,7 +71,7 @@ static struct i2c_driver bmp085_i2c_driver = { }, .id_table = bmp085_id, .probe = bmp085_i2c_probe, - .remove = __devexit_p(bmp085_i2c_remove), + .remove = bmp085_i2c_remove, .detect = bmp085_i2c_detect, .address_list = normal_i2c diff --git a/drivers/misc/bmp085-spi.c b/drivers/misc/bmp085-spi.c index 5e982af9973..864ecac3237 100644 --- a/drivers/misc/bmp085-spi.c +++ b/drivers/misc/bmp085-spi.c @@ -22,7 +22,7 @@ #include <linux/err.h> #include "bmp085.h" -static int __devinit bmp085_spi_probe(struct spi_device *client) +static int bmp085_spi_probe(struct spi_device *client) { int err; struct regmap *regmap; @@ -41,7 +41,7 @@ static int __devinit bmp085_spi_probe(struct spi_device *client) return err; } - return bmp085_probe(&client->dev, regmap); + return bmp085_probe(&client->dev, regmap, client->irq); } static int bmp085_spi_remove(struct spi_device *client) @@ -70,7 +70,7 @@ static struct spi_driver bmp085_spi_driver = { }, .id_table = bmp085_id, .probe = bmp085_spi_probe, - .remove = __devexit_p(bmp085_spi_remove) + .remove = bmp085_spi_remove }; module_spi_driver(bmp085_spi_driver); diff --git a/drivers/misc/bmp085.c b/drivers/misc/bmp085.c index 62e418293b7..9b313f7810f 100644 --- a/drivers/misc/bmp085.c +++ b/drivers/misc/bmp085.c @@ -47,11 +47,12 @@ #include <linux/module.h> #include <linux/device.h> -#include <linux/init.h> #include <linux/slab.h> -#include <linux/delay.h> #include <linux/of.h> #include "bmp085.h" +#include <linux/interrupt.h> +#include <linux/completion.h> +#include <linux/gpio.h> #define BMP085_CHIP_ID 0x55 #define BMP085_CALIBRATION_DATA_START 0xAA @@ -84,8 +85,19 @@ struct bmp085_data { unsigned long last_temp_measurement; u8 chip_id; s32 b6; /* calculated temperature correction coefficient */ + int irq; + struct completion done; }; +static irqreturn_t bmp085_eoc_isr(int irq, void *devid) +{ + struct bmp085_data *data = devid; + + complete(&data->done); + + return IRQ_HANDLED; +} + static s32 bmp085_read_calibration_data(struct bmp085_data *data) { u16 tmp[BMP085_CALIBRATION_DATA_LENGTH]; @@ -116,6 +128,9 @@ static s32 bmp085_update_raw_temperature(struct bmp085_data *data) s32 status; mutex_lock(&data->lock); + + init_completion(&data->done); + status = regmap_write(data->regmap, BMP085_CTRL_REG, BMP085_TEMP_MEASUREMENT); if (status < 0) { @@ -123,7 +138,8 @@ static s32 bmp085_update_raw_temperature(struct bmp085_data *data) "Error while requesting temperature measurement.\n"); goto exit; } - msleep(BMP085_TEMP_CONVERSION_TIME); + wait_for_completion_timeout(&data->done, 1 + msecs_to_jiffies( + BMP085_TEMP_CONVERSION_TIME)); status = regmap_bulk_read(data->regmap, BMP085_CONVERSION_REGISTER_MSB, &tmp, sizeof(tmp)); @@ -147,6 +163,9 @@ static s32 bmp085_update_raw_pressure(struct bmp085_data *data) s32 status; mutex_lock(&data->lock); + + init_completion(&data->done); + status = regmap_write(data->regmap, BMP085_CTRL_REG, BMP085_PRESSURE_MEASUREMENT + (data->oversampling_setting << 6)); @@ -157,8 +176,8 @@ static s32 bmp085_update_raw_pressure(struct bmp085_data *data) } /* wait for the end of conversion */ - msleep(2+(3 << data->oversampling_setting)); - + wait_for_completion_timeout(&data->done, 1 + msecs_to_jiffies( + 2+(3 << data->oversampling_setting))); /* copy data into a u32 (4 bytes), but skip the first byte. */ status = regmap_bulk_read(data->regmap, BMP085_CONVERSION_REGISTER_MSB, ((u8 *)&tmp)+1, 3); @@ -374,7 +393,7 @@ int bmp085_detect(struct device *dev) } EXPORT_SYMBOL_GPL(bmp085_detect); -static void __init bmp085_get_of_properties(struct bmp085_data *data) +static void bmp085_get_of_properties(struct bmp085_data *data) { #ifdef CONFIG_OF struct device_node *np = data->dev->of_node; @@ -420,7 +439,7 @@ struct regmap_config bmp085_regmap_config = { }; EXPORT_SYMBOL_GPL(bmp085_regmap_config); -__devinit int bmp085_probe(struct device *dev, struct regmap *regmap) +int bmp085_probe(struct device *dev, struct regmap *regmap, int irq) { struct bmp085_data *data; int err = 0; @@ -434,6 +453,15 @@ __devinit int bmp085_probe(struct device *dev, struct regmap *regmap) dev_set_drvdata(dev, data); data->dev = dev; data->regmap = regmap; + data->irq = irq; + + if (data->irq > 0) { + err = devm_request_irq(dev, data->irq, bmp085_eoc_isr, + IRQF_TRIGGER_RISING, "bmp085", + data); + if (err < 0) + goto exit_free; + } /* Initialize the BMP085 chip */ err = bmp085_init_client(data); diff --git a/drivers/misc/bmp085.h b/drivers/misc/bmp085.h index 2b8f615bca9..8b8e3b1f5ca 100644 --- a/drivers/misc/bmp085.h +++ b/drivers/misc/bmp085.h @@ -26,7 +26,7 @@ extern struct regmap_config bmp085_regmap_config; -int bmp085_probe(struct device *dev, struct regmap *regmap); +int bmp085_probe(struct device *dev, struct regmap *regmap, int irq); int bmp085_remove(struct device *dev); int bmp085_detect(struct device *dev); diff --git a/drivers/misc/c2port/core.c b/drivers/misc/c2port/core.c index f428d86bfc1..464419b3644 100644 --- a/drivers/misc/c2port/core.c +++ b/drivers/misc/c2port/core.c @@ -311,6 +311,7 @@ static ssize_t c2port_show_name(struct device *dev, return sprintf(buf, "%s\n", c2dev->name); } +static DEVICE_ATTR(name, 0444, c2port_show_name, NULL); static ssize_t c2port_show_flash_blocks_num(struct device *dev, struct device_attribute *attr, char *buf) @@ -320,6 +321,7 @@ static ssize_t c2port_show_flash_blocks_num(struct device *dev, return sprintf(buf, "%d\n", ops->blocks_num); } +static DEVICE_ATTR(flash_blocks_num, 0444, c2port_show_flash_blocks_num, NULL); static ssize_t c2port_show_flash_block_size(struct device *dev, struct device_attribute *attr, char *buf) @@ -329,6 +331,7 @@ static ssize_t c2port_show_flash_block_size(struct device *dev, return sprintf(buf, "%d\n", ops->block_size); } +static DEVICE_ATTR(flash_block_size, 0444, c2port_show_flash_block_size, NULL); static ssize_t c2port_show_flash_size(struct device *dev, struct device_attribute *attr, char *buf) @@ -338,18 +341,18 @@ static ssize_t c2port_show_flash_size(struct device *dev, return sprintf(buf, "%d\n", ops->blocks_num * ops->block_size); } +static DEVICE_ATTR(flash_size, 0444, c2port_show_flash_size, NULL); -static ssize_t c2port_show_access(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t access_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct c2port_device *c2dev = dev_get_drvdata(dev); return sprintf(buf, "%d\n", c2dev->access); } -static ssize_t c2port_store_access(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +static ssize_t access_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) { struct c2port_device *c2dev = dev_get_drvdata(dev); struct c2port_ops *ops = c2dev->ops; @@ -375,6 +378,7 @@ static ssize_t c2port_store_access(struct device *dev, return count; } +static DEVICE_ATTR_RW(access); static ssize_t c2port_store_reset(struct device *dev, struct device_attribute *attr, @@ -395,6 +399,7 @@ static ssize_t c2port_store_reset(struct device *dev, return count; } +static DEVICE_ATTR(reset, 0200, NULL, c2port_store_reset); static ssize_t __c2port_show_dev_id(struct c2port_device *dev, char *buf) { @@ -431,6 +436,7 @@ static ssize_t c2port_show_dev_id(struct device *dev, return ret; } +static DEVICE_ATTR(dev_id, 0444, c2port_show_dev_id, NULL); static ssize_t __c2port_show_rev_id(struct c2port_device *dev, char *buf) { @@ -467,6 +473,7 @@ static ssize_t c2port_show_rev_id(struct device *dev, return ret; } +static DEVICE_ATTR(rev_id, 0444, c2port_show_rev_id, NULL); static ssize_t c2port_show_flash_access(struct device *dev, struct device_attribute *attr, char *buf) @@ -536,6 +543,8 @@ static ssize_t c2port_store_flash_access(struct device *dev, return count; } +static DEVICE_ATTR(flash_access, 0644, c2port_show_flash_access, + c2port_store_flash_access); static ssize_t __c2port_write_flash_erase(struct c2port_device *dev) { @@ -616,6 +625,7 @@ static ssize_t c2port_store_flash_erase(struct device *dev, return count; } +static DEVICE_ATTR(flash_erase, 0200, NULL, c2port_store_flash_erase); static ssize_t __c2port_read_flash_data(struct c2port_device *dev, char *buffer, loff_t offset, size_t count) @@ -846,35 +856,40 @@ static ssize_t c2port_write_flash_data(struct file *filp, struct kobject *kobj, return ret; } +/* size is computed at run-time */ +static BIN_ATTR(flash_data, 0644, c2port_read_flash_data, + c2port_write_flash_data, 0); /* * Class attributes */ +static struct attribute *c2port_attrs[] = { + &dev_attr_name.attr, + &dev_attr_flash_blocks_num.attr, + &dev_attr_flash_block_size.attr, + &dev_attr_flash_size.attr, + &dev_attr_access.attr, + &dev_attr_reset.attr, + &dev_attr_dev_id.attr, + &dev_attr_rev_id.attr, + &dev_attr_flash_access.attr, + &dev_attr_flash_erase.attr, + NULL, +}; -static struct device_attribute c2port_attrs[] = { - __ATTR(name, 0444, c2port_show_name, NULL), - __ATTR(flash_blocks_num, 0444, c2port_show_flash_blocks_num, NULL), - __ATTR(flash_block_size, 0444, c2port_show_flash_block_size, NULL), - __ATTR(flash_size, 0444, c2port_show_flash_size, NULL), - __ATTR(access, 0644, c2port_show_access, c2port_store_access), - __ATTR(reset, 0200, NULL, c2port_store_reset), - __ATTR(dev_id, 0444, c2port_show_dev_id, NULL), - __ATTR(rev_id, 0444, c2port_show_rev_id, NULL), - - __ATTR(flash_access, 0644, c2port_show_flash_access, - c2port_store_flash_access), - __ATTR(flash_erase, 0200, NULL, c2port_store_flash_erase), - __ATTR_NULL, +static struct bin_attribute *c2port_bin_attrs[] = { + &bin_attr_flash_data, + NULL, }; -static struct bin_attribute c2port_bin_attrs = { - .attr = { - .name = "flash_data", - .mode = 0644 - }, - .read = c2port_read_flash_data, - .write = c2port_write_flash_data, - /* .size is computed at run-time */ +static const struct attribute_group c2port_group = { + .attrs = c2port_attrs, + .bin_attrs = c2port_bin_attrs, +}; + +static const struct attribute_group *c2port_groups[] = { + &c2port_group, + NULL, }; /* @@ -885,7 +900,7 @@ struct c2port_device *c2port_device_register(char *name, struct c2port_ops *ops, void *devdata) { struct c2port_device *c2dev; - int id, ret; + int ret; if (unlikely(!ops) || unlikely(!ops->access) || \ unlikely(!ops->c2d_dir) || unlikely(!ops->c2ck_set) || \ @@ -897,22 +912,20 @@ struct c2port_device *c2port_device_register(char *name, if (unlikely(!c2dev)) return ERR_PTR(-ENOMEM); - ret = idr_pre_get(&c2port_idr, GFP_KERNEL); - if (!ret) { - ret = -ENOMEM; - goto error_idr_get_new; - } - + idr_preload(GFP_KERNEL); spin_lock_irq(&c2port_idr_lock); - ret = idr_get_new(&c2port_idr, c2dev, &id); + ret = idr_alloc(&c2port_idr, c2dev, 0, 0, GFP_NOWAIT); spin_unlock_irq(&c2port_idr_lock); + idr_preload_end(); if (ret < 0) - goto error_idr_get_new; - c2dev->id = id; + goto error_idr_alloc; + c2dev->id = ret; + + bin_attr_flash_data.size = ops->blocks_num * ops->block_size; c2dev->dev = device_create(c2port_class, NULL, 0, c2dev, - "c2port%d", id); + "c2port%d", c2dev->id); if (unlikely(IS_ERR(c2dev->dev))) { ret = PTR_ERR(c2dev->dev); goto error_device_create; @@ -923,12 +936,6 @@ struct c2port_device *c2port_device_register(char *name, c2dev->ops = ops; mutex_init(&c2dev->mutex); - /* Create binary file */ - c2port_bin_attrs.size = ops->blocks_num * ops->block_size; - ret = device_create_bin_file(c2dev->dev, &c2port_bin_attrs); - if (unlikely(ret)) - goto error_device_create_bin_file; - /* By default C2 port access is off */ c2dev->access = c2dev->flash_access = 0; ops->access(c2dev, 0); @@ -941,15 +948,12 @@ struct c2port_device *c2port_device_register(char *name, return c2dev; -error_device_create_bin_file: - device_destroy(c2port_class, 0); - error_device_create: spin_lock_irq(&c2port_idr_lock); - idr_remove(&c2port_idr, id); + idr_remove(&c2port_idr, c2dev->id); spin_unlock_irq(&c2port_idr_lock); -error_idr_get_new: +error_idr_alloc: kfree(c2dev); return ERR_PTR(ret); @@ -963,7 +967,6 @@ void c2port_device_unregister(struct c2port_device *c2dev) dev_info(c2dev->dev, "C2 port %s removed\n", c2dev->name); - device_remove_bin_file(c2dev->dev, &c2port_bin_attrs); spin_lock_irq(&c2port_idr_lock); idr_remove(&c2port_idr, c2dev->id); spin_unlock_irq(&c2port_idr_lock); @@ -988,7 +991,7 @@ static int __init c2port_init(void) printk(KERN_ERR "c2port: failed to allocate class\n"); return PTR_ERR(c2port_class); } - c2port_class->dev_attrs = c2port_attrs; + c2port_class->dev_groups = c2port_groups; return 0; } diff --git a/drivers/misc/carma/carma-fpga-program.c b/drivers/misc/carma/carma-fpga-program.c index eaddfe9db14..7be89832db1 100644 --- a/drivers/misc/carma/carma-fpga-program.c +++ b/drivers/misc/carma/carma-fpga-program.c @@ -10,6 +10,8 @@ */ #include <linux/dma-mapping.h> +#include <linux/of_address.h> +#include <linux/of_irq.h> #include <linux/of_platform.h> #include <linux/completion.h> #include <linux/miscdevice.h> @@ -546,7 +548,7 @@ static noinline int fpga_program_dma(struct fpga_dev *priv) goto out_dma_unmap; } - dma_async_memcpy_issue_pending(chan); + dma_async_issue_pending(chan); /* Set the total byte count */ fpga_set_byte_count(priv->regs, priv->bytes); @@ -830,8 +832,9 @@ static ssize_t penable_store(struct device *dev, struct device_attribute *attr, unsigned long val; int ret; - if (strict_strtoul(buf, 0, &val)) - return -EINVAL; + ret = kstrtoul(buf, 0, &val); + if (ret) + return ret; if (val) { ret = fpga_enable_power_supplies(priv); @@ -859,8 +862,9 @@ static ssize_t program_store(struct device *dev, struct device_attribute *attr, unsigned long val; int ret; - if (strict_strtoul(buf, 0, &val)) - return -EINVAL; + ret = kstrtoul(buf, 0, &val); + if (ret) + return ret; /* We can't have an image writer and be programming simultaneously */ if (mutex_lock_interruptible(&priv->lock)) @@ -919,7 +923,7 @@ static bool dma_filter(struct dma_chan *chan, void *data) static int fpga_of_remove(struct platform_device *op) { - struct fpga_dev *priv = dev_get_drvdata(&op->dev); + struct fpga_dev *priv = platform_get_drvdata(op); struct device *this_device = priv->miscdev.this_device; sysfs_remove_group(&this_device->kobj, &fpga_attr_group); @@ -969,7 +973,7 @@ static int fpga_of_probe(struct platform_device *op) kref_init(&priv->ref); - dev_set_drvdata(&op->dev, priv); + platform_set_drvdata(op, priv); priv->dev = &op->dev; mutex_init(&priv->lock); init_completion(&priv->completion); diff --git a/drivers/misc/carma/carma-fpga.c b/drivers/misc/carma/carma-fpga.c index 8835eabb3b8..14d90eae605 100644 --- a/drivers/misc/carma/carma-fpga.c +++ b/drivers/misc/carma/carma-fpga.c @@ -88,6 +88,8 @@ * interrupt source to the GPIO pin. Tada, we hid the interrupt. :) */ +#include <linux/of_address.h> +#include <linux/of_irq.h> #include <linux/of_platform.h> #include <linux/dma-mapping.h> #include <linux/miscdevice.h> @@ -99,7 +101,6 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/poll.h> -#include <linux/init.h> #include <linux/slab.h> #include <linux/kref.h> #include <linux/io.h> @@ -631,6 +632,7 @@ static int data_submit_dma(struct fpga_device *priv, struct data_buf *buf) struct dma_async_tx_descriptor *tx; dma_cookie_t cookie; dma_addr_t dst, src; + unsigned long dma_flags = 0; dst_sg = buf->vb.sglist; dst_nents = buf->vb.sglen; @@ -666,7 +668,7 @@ static int data_submit_dma(struct fpga_device *priv, struct data_buf *buf) src = SYS_FPGA_BLOCK; tx = chan->device->device_prep_dma_memcpy(chan, dst, src, REG_BLOCK_SIZE, - 0); + dma_flags); if (!tx) { dev_err(priv->dev, "unable to prep SYS-FPGA DMA\n"); return -ENOMEM; @@ -749,7 +751,7 @@ static irqreturn_t data_irq(int irq, void *dev_id) submitted = true; /* Start the DMA Engine */ - dma_async_memcpy_issue_pending(priv->chan); + dma_async_issue_pending(priv->chan); out: /* If no DMA was submitted, re-enable interrupts */ @@ -1000,10 +1002,10 @@ static ssize_t data_en_set(struct device *dev, struct device_attribute *attr, unsigned long enable; int ret; - ret = strict_strtoul(buf, 0, &enable); + ret = kstrtoul(buf, 0, &enable); if (ret) { dev_err(priv->dev, "unable to parse enable input\n"); - return -EINVAL; + return ret; } /* protect against concurrent enable/disable */ @@ -1294,7 +1296,7 @@ static int data_of_probe(struct platform_device *op) goto out_return; } - dev_set_drvdata(&op->dev, priv); + platform_set_drvdata(op, priv); priv->dev = &op->dev; kref_init(&priv->ref); mutex_init(&priv->mutex); @@ -1398,7 +1400,7 @@ out_return: static int data_of_remove(struct platform_device *op) { - struct fpga_device *priv = dev_get_drvdata(&op->dev); + struct fpga_device *priv = platform_get_drvdata(op); struct device *this_device = priv->miscdev.this_device; /* remove all sysfs files, now the device cannot be re-enabled */ diff --git a/drivers/misc/cb710/core.c b/drivers/misc/cb710/core.c index 9d5eed75466..fb397e7d1cc 100644 --- a/drivers/misc/cb710/core.c +++ b/drivers/misc/cb710/core.c @@ -30,7 +30,7 @@ void cb710_pci_update_config_reg(struct pci_dev *pdev, EXPORT_SYMBOL_GPL(cb710_pci_update_config_reg); /* Some magic writes based on Windows driver init code */ -static int __devinit cb710_pci_configure(struct pci_dev *pdev) +static int cb710_pci_configure(struct pci_dev *pdev) { unsigned int devfn = PCI_DEVFN(PCI_SLOT(pdev->devfn), 0); struct pci_dev *pdev0; @@ -96,7 +96,7 @@ static void cb710_release_slot(struct device *dev) #endif } -static int __devinit cb710_register_slot(struct cb710_chip *chip, +static int cb710_register_slot(struct cb710_chip *chip, unsigned slot_mask, unsigned io_offset, const char *name) { int nr = chip->slots; @@ -176,7 +176,7 @@ static int cb710_suspend(struct pci_dev *pdev, pm_message_t state) { struct cb710_chip *chip = pci_get_drvdata(pdev); - free_irq(pdev->irq, chip); + devm_free_irq(&pdev->dev, pdev->irq, chip); pci_save_state(pdev); pci_disable_device(pdev); if (state.event & PM_EVENT_SLEEP) @@ -201,7 +201,7 @@ static int cb710_resume(struct pci_dev *pdev) #endif /* CONFIG_PM */ -static int __devinit cb710_probe(struct pci_dev *pdev, +static int cb710_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { struct cb710_chip *chip; @@ -305,7 +305,7 @@ unreg_mmc: return err; } -static void __devexit cb710_remove_one(struct pci_dev *pdev) +static void cb710_remove_one(struct pci_dev *pdev) { struct cb710_chip *chip = pci_get_drvdata(pdev); unsigned long flags; @@ -332,7 +332,7 @@ static struct pci_driver cb710_driver = { .name = KBUILD_MODNAME, .id_table = cb710_pci_tbl, .probe = cb710_probe, - .remove = __devexit_p(cb710_remove_one), + .remove = cb710_remove_one, #ifdef CONFIG_PM .suspend = cb710_suspend, .resume = cb710_resume, diff --git a/drivers/misc/cs5535-mfgpt.c b/drivers/misc/cs5535-mfgpt.c index f505a40a8f4..effd8c6b2b9 100644 --- a/drivers/misc/cs5535-mfgpt.c +++ b/drivers/misc/cs5535-mfgpt.c @@ -24,8 +24,11 @@ static int mfgpt_reset_timers; module_param_named(mfgptfix, mfgpt_reset_timers, int, 0644); -MODULE_PARM_DESC(mfgptfix, "Reset the MFGPT timers during init; " - "required by some broken BIOSes (ie, TinyBIOS < 0.99)."); +MODULE_PARM_DESC(mfgptfix, "Try to reset the MFGPT timers during init; " + "required by some broken BIOSes (ie, TinyBIOS < 0.99) or kexec " + "(1 = reset the MFGPT using an undocumented bit, " + "2 = perform a soft reset by unconfiguring all timers); " + "use what works best for you."); struct cs5535_mfgpt_timer { struct cs5535_mfgpt_chip *chip; @@ -246,7 +249,7 @@ EXPORT_SYMBOL_GPL(cs5535_mfgpt_write); * Jordan tells me that he and Mitch once played w/ it, but it's unclear * what the results of that were (and they experienced some instability). */ -static void __devinit reset_all_timers(void) +static void reset_all_timers(void) { uint32_t val, dummy; @@ -256,13 +259,35 @@ static void __devinit reset_all_timers(void) } /* + * This is another sledgehammer to reset all MFGPT timers. + * Instead of using the undocumented bit method it clears + * IRQ, NMI and RESET settings. + */ +static void soft_reset(void) +{ + int i; + struct cs5535_mfgpt_timer t; + + for (i = 0; i < MFGPT_MAX_TIMERS; i++) { + t.nr = i; + + cs5535_mfgpt_toggle_event(&t, MFGPT_CMP1, MFGPT_EVENT_RESET, 0); + cs5535_mfgpt_toggle_event(&t, MFGPT_CMP2, MFGPT_EVENT_RESET, 0); + cs5535_mfgpt_toggle_event(&t, MFGPT_CMP1, MFGPT_EVENT_NMI, 0); + cs5535_mfgpt_toggle_event(&t, MFGPT_CMP2, MFGPT_EVENT_NMI, 0); + cs5535_mfgpt_toggle_event(&t, MFGPT_CMP1, MFGPT_EVENT_IRQ, 0); + cs5535_mfgpt_toggle_event(&t, MFGPT_CMP2, MFGPT_EVENT_IRQ, 0); + } +} + +/* * Check whether any MFGPTs are available for the kernel to use. In most * cases, firmware that uses AMD's VSA code will claim all timers during * bootup; we certainly don't want to take them if they're already in use. * In other cases (such as with VSAless OpenFirmware), the system firmware * leaves timers available for us to use. */ -static int __devinit scan_timers(struct cs5535_mfgpt_chip *mfgpt) +static int scan_timers(struct cs5535_mfgpt_chip *mfgpt) { struct cs5535_mfgpt_timer timer = { .chip = mfgpt }; unsigned long flags; @@ -271,15 +296,17 @@ static int __devinit scan_timers(struct cs5535_mfgpt_chip *mfgpt) int i; /* bios workaround */ - if (mfgpt_reset_timers) + if (mfgpt_reset_timers == 1) reset_all_timers(); + else if (mfgpt_reset_timers == 2) + soft_reset(); /* just to be safe, protect this section w/ lock */ spin_lock_irqsave(&mfgpt->lock, flags); for (i = 0; i < MFGPT_MAX_TIMERS; i++) { timer.nr = i; val = cs5535_mfgpt_read(&timer, MFGPT_REG_SETUP); - if (!(val & MFGPT_SETUP_SETUP)) { + if (!(val & MFGPT_SETUP_SETUP) || mfgpt_reset_timers == 2) { __set_bit(i, mfgpt->avail); timers++; } @@ -289,11 +316,17 @@ static int __devinit scan_timers(struct cs5535_mfgpt_chip *mfgpt) return timers; } -static int __devinit cs5535_mfgpt_probe(struct platform_device *pdev) +static int cs5535_mfgpt_probe(struct platform_device *pdev) { struct resource *res; int err = -EIO, t; + if (mfgpt_reset_timers < 0 || mfgpt_reset_timers > 2) { + dev_err(&pdev->dev, "Bad mfgpt_reset_timers value: %i\n", + mfgpt_reset_timers); + goto done; + } + /* There are two ways to get the MFGPT base address; one is by * fetching it from MSR_LBAR_MFGPT, the other is by reading the * PCI BAR info. The latter method is easier (especially across diff --git a/drivers/misc/ds1682.c b/drivers/misc/ds1682.c index 154b02e5094..b909fb30232 100644 --- a/drivers/misc/ds1682.c +++ b/drivers/misc/ds1682.c @@ -32,7 +32,6 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <linux/i2c.h> #include <linux/string.h> #include <linux/list.h> @@ -86,7 +85,6 @@ static ssize_t ds1682_store(struct device *dev, struct device_attribute *attr, { struct sensor_device_attribute_2 *sattr = to_sensor_dev_attr_2(attr); struct i2c_client *client = to_i2c_client(dev); - char *endp; u64 val; __le32 val_le; int rc; @@ -94,8 +92,8 @@ static ssize_t ds1682_store(struct device *dev, struct device_attribute *attr, dev_dbg(dev, "ds1682_store() called on %s\n", attr->attr.name); /* Decode input */ - val = simple_strtoull(buf, &endp, 0); - if (buf == endp) { + rc = kstrtoull(buf, 0, &val); + if (rc < 0) { dev_dbg(dev, "input string not a number\n"); return -EINVAL; } diff --git a/drivers/misc/dummy-irq.c b/drivers/misc/dummy-irq.c new file mode 100644 index 00000000000..4d0db15df11 --- /dev/null +++ b/drivers/misc/dummy-irq.c @@ -0,0 +1,63 @@ +/* + * Dummy IRQ handler driver. + * + * This module only registers itself as a handler that is specified to it + * by the 'irq' parameter. + * + * The sole purpose of this module is to help with debugging of systems on + * which spurious IRQs would happen on disabled IRQ vector. + * + * Copyright (C) 2013 Jiri Kosina + */ + +/* + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + */ +#include <linux/module.h> +#include <linux/irq.h> +#include <linux/interrupt.h> + +static int irq = -1; + +static irqreturn_t dummy_interrupt(int irq, void *dev_id) +{ + static int count = 0; + + if (count == 0) { + printk(KERN_INFO "dummy-irq: interrupt occurred on IRQ %d\n", + irq); + count++; + } + + return IRQ_NONE; +} + +static int __init dummy_irq_init(void) +{ + if (irq < 0) { + printk(KERN_ERR "dummy-irq: no IRQ given. Use irq=N\n"); + return -EIO; + } + if (request_irq(irq, &dummy_interrupt, IRQF_SHARED, "dummy_irq", &irq)) { + printk(KERN_ERR "dummy-irq: cannot register IRQ %d\n", irq); + return -EIO; + } + printk(KERN_INFO "dummy-irq: registered for IRQ %d\n", irq); + return 0; +} + +static void __exit dummy_irq_exit(void) +{ + printk(KERN_INFO "dummy-irq unloaded\n"); + free_irq(irq, &irq); +} + +module_init(dummy_irq_init); +module_exit(dummy_irq_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Jiri Kosina"); +module_param(irq, uint, 0444); +MODULE_PARM_DESC(irq, "The IRQ to register for"); diff --git a/drivers/misc/echo/Kconfig b/drivers/misc/echo/Kconfig new file mode 100644 index 00000000000..f1d41ea9cd4 --- /dev/null +++ b/drivers/misc/echo/Kconfig @@ -0,0 +1,9 @@ +config ECHO + tristate "Line Echo Canceller support" + default n + ---help--- + This driver provides line echo cancelling support for mISDN and + Zaptel drivers. + + To compile this driver as a module, choose M here. The module + will be called echo. diff --git a/drivers/misc/echo/Makefile b/drivers/misc/echo/Makefile new file mode 100644 index 00000000000..7d4caac12a8 --- /dev/null +++ b/drivers/misc/echo/Makefile @@ -0,0 +1 @@ +obj-$(CONFIG_ECHO) += echo.o diff --git a/drivers/misc/echo/echo.c b/drivers/misc/echo/echo.c new file mode 100644 index 00000000000..9597e9523ca --- /dev/null +++ b/drivers/misc/echo/echo.c @@ -0,0 +1,674 @@ +/* + * SpanDSP - a series of DSP components for telephony + * + * echo.c - A line echo canceller. This code is being developed + * against and partially complies with G168. + * + * Written by Steve Underwood <steveu@coppice.org> + * and David Rowe <david_at_rowetel_dot_com> + * + * Copyright (C) 2001, 2003 Steve Underwood, 2007 David Rowe + * + * Based on a bit from here, a bit from there, eye of toad, ear of + * bat, 15 years of failed attempts by David and a few fried brain + * cells. + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +/*! \file */ + +/* Implementation Notes + David Rowe + April 2007 + + This code started life as Steve's NLMS algorithm with a tap + rotation algorithm to handle divergence during double talk. I + added a Geigel Double Talk Detector (DTD) [2] and performed some + G168 tests. However I had trouble meeting the G168 requirements, + especially for double talk - there were always cases where my DTD + failed, for example where near end speech was under the 6dB + threshold required for declaring double talk. + + So I tried a two path algorithm [1], which has so far given better + results. The original tap rotation/Geigel algorithm is available + in SVN http://svn.rowetel.com/software/oslec/tags/before_16bit. + It's probably possible to make it work if some one wants to put some + serious work into it. + + At present no special treatment is provided for tones, which + generally cause NLMS algorithms to diverge. Initial runs of a + subset of the G168 tests for tones (e.g ./echo_test 6) show the + current algorithm is passing OK, which is kind of surprising. The + full set of tests needs to be performed to confirm this result. + + One other interesting change is that I have managed to get the NLMS + code to work with 16 bit coefficients, rather than the original 32 + bit coefficents. This reduces the MIPs and storage required. + I evaulated the 16 bit port using g168_tests.sh and listening tests + on 4 real-world samples. + + I also attempted the implementation of a block based NLMS update + [2] but although this passes g168_tests.sh it didn't converge well + on the real-world samples. I have no idea why, perhaps a scaling + problem. The block based code is also available in SVN + http://svn.rowetel.com/software/oslec/tags/before_16bit. If this + code can be debugged, it will lead to further reduction in MIPS, as + the block update code maps nicely onto DSP instruction sets (it's a + dot product) compared to the current sample-by-sample update. + + Steve also has some nice notes on echo cancellers in echo.h + + References: + + [1] Ochiai, Areseki, and Ogihara, "Echo Canceller with Two Echo + Path Models", IEEE Transactions on communications, COM-25, + No. 6, June + 1977. + http://www.rowetel.com/images/echo/dual_path_paper.pdf + + [2] The classic, very useful paper that tells you how to + actually build a real world echo canceller: + Messerschmitt, Hedberg, Cole, Haoui, Winship, "Digital Voice + Echo Canceller with a TMS320020, + http://www.rowetel.com/images/echo/spra129.pdf + + [3] I have written a series of blog posts on this work, here is + Part 1: http://www.rowetel.com/blog/?p=18 + + [4] The source code http://svn.rowetel.com/software/oslec/ + + [5] A nice reference on LMS filters: + http://en.wikipedia.org/wiki/Least_mean_squares_filter + + Credits: + + Thanks to Steve Underwood, Jean-Marc Valin, and Ramakrishnan + Muthukrishnan for their suggestions and email discussions. Thanks + also to those people who collected echo samples for me such as + Mark, Pawel, and Pavel. +*/ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/slab.h> + +#include "echo.h" + +#define MIN_TX_POWER_FOR_ADAPTION 64 +#define MIN_RX_POWER_FOR_ADAPTION 64 +#define DTD_HANGOVER 600 /* 600 samples, or 75ms */ +#define DC_LOG2BETA 3 /* log2() of DC filter Beta */ + +/* adapting coeffs using the traditional stochastic descent (N)LMS algorithm */ + +#ifdef __bfin__ +static inline void lms_adapt_bg(struct oslec_state *ec, int clean, int shift) +{ + int i; + int offset1; + int offset2; + int factor; + int exp; + int16_t *phist; + int n; + + if (shift > 0) + factor = clean << shift; + else + factor = clean >> -shift; + + /* Update the FIR taps */ + + offset2 = ec->curr_pos; + offset1 = ec->taps - offset2; + phist = &ec->fir_state_bg.history[offset2]; + + /* st: and en: help us locate the assembler in echo.s */ + + /* asm("st:"); */ + n = ec->taps; + for (i = 0; i < n; i++) { + exp = *phist++ * factor; + ec->fir_taps16[1][i] += (int16_t) ((exp + (1 << 14)) >> 15); + } + /* asm("en:"); */ + + /* Note the asm for the inner loop above generated by Blackfin gcc + 4.1.1 is pretty good (note even parallel instructions used): + + R0 = W [P0++] (X); + R0 *= R2; + R0 = R0 + R3 (NS) || + R1 = W [P1] (X) || + nop; + R0 >>>= 15; + R0 = R0 + R1; + W [P1++] = R0; + + A block based update algorithm would be much faster but the + above can't be improved on much. Every instruction saved in + the loop above is 2 MIPs/ch! The for loop above is where the + Blackfin spends most of it's time - about 17 MIPs/ch measured + with speedtest.c with 256 taps (32ms). Write-back and + Write-through cache gave about the same performance. + */ +} + +/* + IDEAS for further optimisation of lms_adapt_bg(): + + 1/ The rounding is quite costly. Could we keep as 32 bit coeffs + then make filter pluck the MS 16-bits of the coeffs when filtering? + However this would lower potential optimisation of filter, as I + think the dual-MAC architecture requires packed 16 bit coeffs. + + 2/ Block based update would be more efficient, as per comments above, + could use dual MAC architecture. + + 3/ Look for same sample Blackfin LMS code, see if we can get dual-MAC + packing. + + 4/ Execute the whole e/c in a block of say 20ms rather than sample + by sample. Processing a few samples every ms is inefficient. +*/ + +#else +static inline void lms_adapt_bg(struct oslec_state *ec, int clean, int shift) +{ + int i; + + int offset1; + int offset2; + int factor; + int exp; + + if (shift > 0) + factor = clean << shift; + else + factor = clean >> -shift; + + /* Update the FIR taps */ + + offset2 = ec->curr_pos; + offset1 = ec->taps - offset2; + + for (i = ec->taps - 1; i >= offset1; i--) { + exp = (ec->fir_state_bg.history[i - offset1] * factor); + ec->fir_taps16[1][i] += (int16_t) ((exp + (1 << 14)) >> 15); + } + for (; i >= 0; i--) { + exp = (ec->fir_state_bg.history[i + offset2] * factor); + ec->fir_taps16[1][i] += (int16_t) ((exp + (1 << 14)) >> 15); + } +} +#endif + +static inline int top_bit(unsigned int bits) +{ + if (bits == 0) + return -1; + else + return (int)fls((int32_t) bits) - 1; +} + +struct oslec_state *oslec_create(int len, int adaption_mode) +{ + struct oslec_state *ec; + int i; + const int16_t *history; + + ec = kzalloc(sizeof(*ec), GFP_KERNEL); + if (!ec) + return NULL; + + ec->taps = len; + ec->log2taps = top_bit(len); + ec->curr_pos = ec->taps - 1; + + ec->fir_taps16[0] = + kcalloc(ec->taps, sizeof(int16_t), GFP_KERNEL); + if (!ec->fir_taps16[0]) + goto error_oom_0; + + ec->fir_taps16[1] = + kcalloc(ec->taps, sizeof(int16_t), GFP_KERNEL); + if (!ec->fir_taps16[1]) + goto error_oom_1; + + history = fir16_create(&ec->fir_state, ec->fir_taps16[0], ec->taps); + if (!history) + goto error_state; + history = fir16_create(&ec->fir_state_bg, ec->fir_taps16[1], ec->taps); + if (!history) + goto error_state_bg; + + for (i = 0; i < 5; i++) + ec->xvtx[i] = ec->yvtx[i] = ec->xvrx[i] = ec->yvrx[i] = 0; + + ec->cng_level = 1000; + oslec_adaption_mode(ec, adaption_mode); + + ec->snapshot = kcalloc(ec->taps, sizeof(int16_t), GFP_KERNEL); + if (!ec->snapshot) + goto error_snap; + + ec->cond_met = 0; + ec->pstates = 0; + ec->ltxacc = ec->lrxacc = ec->lcleanacc = ec->lclean_bgacc = 0; + ec->ltx = ec->lrx = ec->lclean = ec->lclean_bg = 0; + ec->tx_1 = ec->tx_2 = ec->rx_1 = ec->rx_2 = 0; + ec->lbgn = ec->lbgn_acc = 0; + ec->lbgn_upper = 200; + ec->lbgn_upper_acc = ec->lbgn_upper << 13; + + return ec; + +error_snap: + fir16_free(&ec->fir_state_bg); +error_state_bg: + fir16_free(&ec->fir_state); +error_state: + kfree(ec->fir_taps16[1]); +error_oom_1: + kfree(ec->fir_taps16[0]); +error_oom_0: + kfree(ec); + return NULL; +} +EXPORT_SYMBOL_GPL(oslec_create); + +void oslec_free(struct oslec_state *ec) +{ + int i; + + fir16_free(&ec->fir_state); + fir16_free(&ec->fir_state_bg); + for (i = 0; i < 2; i++) + kfree(ec->fir_taps16[i]); + kfree(ec->snapshot); + kfree(ec); +} +EXPORT_SYMBOL_GPL(oslec_free); + +void oslec_adaption_mode(struct oslec_state *ec, int adaption_mode) +{ + ec->adaption_mode = adaption_mode; +} +EXPORT_SYMBOL_GPL(oslec_adaption_mode); + +void oslec_flush(struct oslec_state *ec) +{ + int i; + + ec->ltxacc = ec->lrxacc = ec->lcleanacc = ec->lclean_bgacc = 0; + ec->ltx = ec->lrx = ec->lclean = ec->lclean_bg = 0; + ec->tx_1 = ec->tx_2 = ec->rx_1 = ec->rx_2 = 0; + + ec->lbgn = ec->lbgn_acc = 0; + ec->lbgn_upper = 200; + ec->lbgn_upper_acc = ec->lbgn_upper << 13; + + ec->nonupdate_dwell = 0; + + fir16_flush(&ec->fir_state); + fir16_flush(&ec->fir_state_bg); + ec->fir_state.curr_pos = ec->taps - 1; + ec->fir_state_bg.curr_pos = ec->taps - 1; + for (i = 0; i < 2; i++) + memset(ec->fir_taps16[i], 0, ec->taps * sizeof(int16_t)); + + ec->curr_pos = ec->taps - 1; + ec->pstates = 0; +} +EXPORT_SYMBOL_GPL(oslec_flush); + +void oslec_snapshot(struct oslec_state *ec) +{ + memcpy(ec->snapshot, ec->fir_taps16[0], ec->taps * sizeof(int16_t)); +} +EXPORT_SYMBOL_GPL(oslec_snapshot); + +/* Dual Path Echo Canceller */ + +int16_t oslec_update(struct oslec_state *ec, int16_t tx, int16_t rx) +{ + int32_t echo_value; + int clean_bg; + int tmp; + int tmp1; + + /* + * Input scaling was found be required to prevent problems when tx + * starts clipping. Another possible way to handle this would be the + * filter coefficent scaling. + */ + + ec->tx = tx; + ec->rx = rx; + tx >>= 1; + rx >>= 1; + + /* + * Filter DC, 3dB point is 160Hz (I think), note 32 bit precision + * required otherwise values do not track down to 0. Zero at DC, Pole + * at (1-Beta) on real axis. Some chip sets (like Si labs) don't + * need this, but something like a $10 X100P card does. Any DC really + * slows down convergence. + * + * Note: removes some low frequency from the signal, this reduces the + * speech quality when listening to samples through headphones but may + * not be obvious through a telephone handset. + * + * Note that the 3dB frequency in radians is approx Beta, e.g. for Beta + * = 2^(-3) = 0.125, 3dB freq is 0.125 rads = 159Hz. + */ + + if (ec->adaption_mode & ECHO_CAN_USE_RX_HPF) { + tmp = rx << 15; + + /* + * Make sure the gain of the HPF is 1.0. This can still + * saturate a little under impulse conditions, and it might + * roll to 32768 and need clipping on sustained peak level + * signals. However, the scale of such clipping is small, and + * the error due to any saturation should not markedly affect + * the downstream processing. + */ + tmp -= (tmp >> 4); + + ec->rx_1 += -(ec->rx_1 >> DC_LOG2BETA) + tmp - ec->rx_2; + + /* + * hard limit filter to prevent clipping. Note that at this + * stage rx should be limited to +/- 16383 due to right shift + * above + */ + tmp1 = ec->rx_1 >> 15; + if (tmp1 > 16383) + tmp1 = 16383; + if (tmp1 < -16383) + tmp1 = -16383; + rx = tmp1; + ec->rx_2 = tmp; + } + + /* Block average of power in the filter states. Used for + adaption power calculation. */ + + { + int new, old; + + /* efficient "out with the old and in with the new" algorithm so + we don't have to recalculate over the whole block of + samples. */ + new = (int)tx * (int)tx; + old = (int)ec->fir_state.history[ec->fir_state.curr_pos] * + (int)ec->fir_state.history[ec->fir_state.curr_pos]; + ec->pstates += + ((new - old) + (1 << (ec->log2taps - 1))) >> ec->log2taps; + if (ec->pstates < 0) + ec->pstates = 0; + } + + /* Calculate short term average levels using simple single pole IIRs */ + + ec->ltxacc += abs(tx) - ec->ltx; + ec->ltx = (ec->ltxacc + (1 << 4)) >> 5; + ec->lrxacc += abs(rx) - ec->lrx; + ec->lrx = (ec->lrxacc + (1 << 4)) >> 5; + + /* Foreground filter */ + + ec->fir_state.coeffs = ec->fir_taps16[0]; + echo_value = fir16(&ec->fir_state, tx); + ec->clean = rx - echo_value; + ec->lcleanacc += abs(ec->clean) - ec->lclean; + ec->lclean = (ec->lcleanacc + (1 << 4)) >> 5; + + /* Background filter */ + + echo_value = fir16(&ec->fir_state_bg, tx); + clean_bg = rx - echo_value; + ec->lclean_bgacc += abs(clean_bg) - ec->lclean_bg; + ec->lclean_bg = (ec->lclean_bgacc + (1 << 4)) >> 5; + + /* Background Filter adaption */ + + /* Almost always adap bg filter, just simple DT and energy + detection to minimise adaption in cases of strong double talk. + However this is not critical for the dual path algorithm. + */ + ec->factor = 0; + ec->shift = 0; + if ((ec->nonupdate_dwell == 0)) { + int p, logp, shift; + + /* Determine: + + f = Beta * clean_bg_rx/P ------ (1) + + where P is the total power in the filter states. + + The Boffins have shown that if we obey (1) we converge + quickly and avoid instability. + + The correct factor f must be in Q30, as this is the fixed + point format required by the lms_adapt_bg() function, + therefore the scaled version of (1) is: + + (2^30) * f = (2^30) * Beta * clean_bg_rx/P + factor = (2^30) * Beta * clean_bg_rx/P ----- (2) + + We have chosen Beta = 0.25 by experiment, so: + + factor = (2^30) * (2^-2) * clean_bg_rx/P + + (30 - 2 - log2(P)) + factor = clean_bg_rx 2 ----- (3) + + To avoid a divide we approximate log2(P) as top_bit(P), + which returns the position of the highest non-zero bit in + P. This approximation introduces an error as large as a + factor of 2, but the algorithm seems to handle it OK. + + Come to think of it a divide may not be a big deal on a + modern DSP, so its probably worth checking out the cycles + for a divide versus a top_bit() implementation. + */ + + p = MIN_TX_POWER_FOR_ADAPTION + ec->pstates; + logp = top_bit(p) + ec->log2taps; + shift = 30 - 2 - logp; + ec->shift = shift; + + lms_adapt_bg(ec, clean_bg, shift); + } + + /* very simple DTD to make sure we dont try and adapt with strong + near end speech */ + + ec->adapt = 0; + if ((ec->lrx > MIN_RX_POWER_FOR_ADAPTION) && (ec->lrx > ec->ltx)) + ec->nonupdate_dwell = DTD_HANGOVER; + if (ec->nonupdate_dwell) + ec->nonupdate_dwell--; + + /* Transfer logic */ + + /* These conditions are from the dual path paper [1], I messed with + them a bit to improve performance. */ + + if ((ec->adaption_mode & ECHO_CAN_USE_ADAPTION) && + (ec->nonupdate_dwell == 0) && + /* (ec->Lclean_bg < 0.875*ec->Lclean) */ + (8 * ec->lclean_bg < 7 * ec->lclean) && + /* (ec->Lclean_bg < 0.125*ec->Ltx) */ + (8 * ec->lclean_bg < ec->ltx)) { + if (ec->cond_met == 6) { + /* + * BG filter has had better results for 6 consecutive + * samples + */ + ec->adapt = 1; + memcpy(ec->fir_taps16[0], ec->fir_taps16[1], + ec->taps * sizeof(int16_t)); + } else + ec->cond_met++; + } else + ec->cond_met = 0; + + /* Non-Linear Processing */ + + ec->clean_nlp = ec->clean; + if (ec->adaption_mode & ECHO_CAN_USE_NLP) { + /* + * Non-linear processor - a fancy way to say "zap small + * signals, to avoid residual echo due to (uLaw/ALaw) + * non-linearity in the channel.". + */ + + if ((16 * ec->lclean < ec->ltx)) { + /* + * Our e/c has improved echo by at least 24 dB (each + * factor of 2 is 6dB, so 2*2*2*2=16 is the same as + * 6+6+6+6=24dB) + */ + if (ec->adaption_mode & ECHO_CAN_USE_CNG) { + ec->cng_level = ec->lbgn; + + /* + * Very elementary comfort noise generation. + * Just random numbers rolled off very vaguely + * Hoth-like. DR: This noise doesn't sound + * quite right to me - I suspect there are some + * overflow issues in the filtering as it's too + * "crackly". + * TODO: debug this, maybe just play noise at + * high level or look at spectrum. + */ + + ec->cng_rndnum = + 1664525U * ec->cng_rndnum + 1013904223U; + ec->cng_filter = + ((ec->cng_rndnum & 0xFFFF) - 32768 + + 5 * ec->cng_filter) >> 3; + ec->clean_nlp = + (ec->cng_filter * ec->cng_level * 8) >> 14; + + } else if (ec->adaption_mode & ECHO_CAN_USE_CLIP) { + /* This sounds much better than CNG */ + if (ec->clean_nlp > ec->lbgn) + ec->clean_nlp = ec->lbgn; + if (ec->clean_nlp < -ec->lbgn) + ec->clean_nlp = -ec->lbgn; + } else { + /* + * just mute the residual, doesn't sound very + * good, used mainly in G168 tests + */ + ec->clean_nlp = 0; + } + } else { + /* + * Background noise estimator. I tried a few + * algorithms here without much luck. This very simple + * one seems to work best, we just average the level + * using a slow (1 sec time const) filter if the + * current level is less than a (experimentally + * derived) constant. This means we dont include high + * level signals like near end speech. When combined + * with CNG or especially CLIP seems to work OK. + */ + if (ec->lclean < 40) { + ec->lbgn_acc += abs(ec->clean) - ec->lbgn; + ec->lbgn = (ec->lbgn_acc + (1 << 11)) >> 12; + } + } + } + + /* Roll around the taps buffer */ + if (ec->curr_pos <= 0) + ec->curr_pos = ec->taps; + ec->curr_pos--; + + if (ec->adaption_mode & ECHO_CAN_DISABLE) + ec->clean_nlp = rx; + + /* Output scaled back up again to match input scaling */ + + return (int16_t) ec->clean_nlp << 1; +} +EXPORT_SYMBOL_GPL(oslec_update); + +/* This function is separated from the echo canceller is it is usually called + as part of the tx process. See rx HP (DC blocking) filter above, it's + the same design. + + Some soft phones send speech signals with a lot of low frequency + energy, e.g. down to 20Hz. This can make the hybrid non-linear + which causes the echo canceller to fall over. This filter can help + by removing any low frequency before it gets to the tx port of the + hybrid. + + It can also help by removing and DC in the tx signal. DC is bad + for LMS algorithms. + + This is one of the classic DC removal filters, adjusted to provide + sufficient bass rolloff to meet the above requirement to protect hybrids + from things that upset them. The difference between successive samples + produces a lousy HPF, and then a suitably placed pole flattens things out. + The final result is a nicely rolled off bass end. The filtering is + implemented with extended fractional precision, which noise shapes things, + giving very clean DC removal. +*/ + +int16_t oslec_hpf_tx(struct oslec_state *ec, int16_t tx) +{ + int tmp; + int tmp1; + + if (ec->adaption_mode & ECHO_CAN_USE_TX_HPF) { + tmp = tx << 15; + + /* + * Make sure the gain of the HPF is 1.0. The first can still + * saturate a little under impulse conditions, and it might + * roll to 32768 and need clipping on sustained peak level + * signals. However, the scale of such clipping is small, and + * the error due to any saturation should not markedly affect + * the downstream processing. + */ + tmp -= (tmp >> 4); + + ec->tx_1 += -(ec->tx_1 >> DC_LOG2BETA) + tmp - ec->tx_2; + tmp1 = ec->tx_1 >> 15; + if (tmp1 > 32767) + tmp1 = 32767; + if (tmp1 < -32767) + tmp1 = -32767; + tx = tmp1; + ec->tx_2 = tmp; + } + + return tx; +} +EXPORT_SYMBOL_GPL(oslec_hpf_tx); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("David Rowe"); +MODULE_DESCRIPTION("Open Source Line Echo Canceller"); +MODULE_VERSION("0.3.0"); diff --git a/drivers/misc/echo/echo.h b/drivers/misc/echo/echo.h new file mode 100644 index 00000000000..9b08c63e636 --- /dev/null +++ b/drivers/misc/echo/echo.h @@ -0,0 +1,187 @@ +/* + * SpanDSP - a series of DSP components for telephony + * + * echo.c - A line echo canceller. This code is being developed + * against and partially complies with G168. + * + * Written by Steve Underwood <steveu@coppice.org> + * and David Rowe <david_at_rowetel_dot_com> + * + * Copyright (C) 2001 Steve Underwood and 2007 David Rowe + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#ifndef __ECHO_H +#define __ECHO_H + +/* +Line echo cancellation for voice + +What does it do? + +This module aims to provide G.168-2002 compliant echo cancellation, to remove +electrical echoes (e.g. from 2-4 wire hybrids) from voice calls. + +How does it work? + +The heart of the echo cancellor is FIR filter. This is adapted to match the +echo impulse response of the telephone line. It must be long enough to +adequately cover the duration of that impulse response. The signal transmitted +to the telephone line is passed through the FIR filter. Once the FIR is +properly adapted, the resulting output is an estimate of the echo signal +received from the line. This is subtracted from the received signal. The result +is an estimate of the signal which originated at the far end of the line, free +from echos of our own transmitted signal. + +The least mean squares (LMS) algorithm is attributed to Widrow and Hoff, and +was introduced in 1960. It is the commonest form of filter adaption used in +things like modem line equalisers and line echo cancellers. There it works very +well. However, it only works well for signals of constant amplitude. It works +very poorly for things like speech echo cancellation, where the signal level +varies widely. This is quite easy to fix. If the signal level is normalised - +similar to applying AGC - LMS can work as well for a signal of varying +amplitude as it does for a modem signal. This normalised least mean squares +(NLMS) algorithm is the commonest one used for speech echo cancellation. Many +other algorithms exist - e.g. RLS (essentially the same as Kalman filtering), +FAP, etc. Some perform significantly better than NLMS. However, factors such +as computational complexity and patents favour the use of NLMS. + +A simple refinement to NLMS can improve its performance with speech. NLMS tends +to adapt best to the strongest parts of a signal. If the signal is white noise, +the NLMS algorithm works very well. However, speech has more low frequency than +high frequency content. Pre-whitening (i.e. filtering the signal to flatten its +spectrum) the echo signal improves the adapt rate for speech, and ensures the +final residual signal is not heavily biased towards high frequencies. A very +low complexity filter is adequate for this, so pre-whitening adds little to the +compute requirements of the echo canceller. + +An FIR filter adapted using pre-whitened NLMS performs well, provided certain +conditions are met: + + - The transmitted signal has poor self-correlation. + - There is no signal being generated within the environment being + cancelled. + +The difficulty is that neither of these can be guaranteed. + +If the adaption is performed while transmitting noise (or something fairly +noise like, such as voice) the adaption works very well. If the adaption is +performed while transmitting something highly correlative (typically narrow +band energy such as signalling tones or DTMF), the adaption can go seriously +wrong. The reason is there is only one solution for the adaption on a near +random signal - the impulse response of the line. For a repetitive signal, +there are any number of solutions which converge the adaption, and nothing +guides the adaption to choose the generalised one. Allowing an untrained +canceller to converge on this kind of narrowband energy probably a good thing, +since at least it cancels the tones. Allowing a well converged canceller to +continue converging on such energy is just a way to ruin its generalised +adaption. A narrowband detector is needed, so adapation can be suspended at +appropriate times. + +The adaption process is based on trying to eliminate the received signal. When +there is any signal from within the environment being cancelled it may upset +the adaption process. Similarly, if the signal we are transmitting is small, +noise may dominate and disturb the adaption process. If we can ensure that the +adaption is only performed when we are transmitting a significant signal level, +and the environment is not, things will be OK. Clearly, it is easy to tell when +we are sending a significant signal. Telling, if the environment is generating +a significant signal, and doing it with sufficient speed that the adaption will +not have diverged too much more we stop it, is a little harder. + +The key problem in detecting when the environment is sourcing significant +energy is that we must do this very quickly. Given a reasonably long sample of +the received signal, there are a number of strategies which may be used to +assess whether that signal contains a strong far end component. However, by the +time that assessment is complete the far end signal will have already caused +major mis-convergence in the adaption process. An assessment algorithm is +needed which produces a fairly accurate result from a very short burst of far +end energy. + +How do I use it? + +The echo cancellor processes both the transmit and receive streams sample by +sample. The processing function is not declared inline. Unfortunately, +cancellation requires many operations per sample, so the call overhead is only +a minor burden. +*/ + +#include "fir.h" +#include "oslec.h" + +/* + G.168 echo canceller descriptor. This defines the working state for a line + echo canceller. +*/ +struct oslec_state { + int16_t tx; + int16_t rx; + int16_t clean; + int16_t clean_nlp; + + int nonupdate_dwell; + int curr_pos; + int taps; + int log2taps; + int adaption_mode; + + int cond_met; + int32_t pstates; + int16_t adapt; + int32_t factor; + int16_t shift; + + /* Average levels and averaging filter states */ + int ltxacc; + int lrxacc; + int lcleanacc; + int lclean_bgacc; + int ltx; + int lrx; + int lclean; + int lclean_bg; + int lbgn; + int lbgn_acc; + int lbgn_upper; + int lbgn_upper_acc; + + /* foreground and background filter states */ + struct fir16_state_t fir_state; + struct fir16_state_t fir_state_bg; + int16_t *fir_taps16[2]; + + /* DC blocking filter states */ + int tx_1; + int tx_2; + int rx_1; + int rx_2; + + /* optional High Pass Filter states */ + int32_t xvtx[5]; + int32_t yvtx[5]; + int32_t xvrx[5]; + int32_t yvrx[5]; + + /* Parameters for the optional Hoth noise generator */ + int cng_level; + int cng_rndnum; + int cng_filter; + + /* snapshot sample of coeffs used for development */ + int16_t *snapshot; +}; + +#endif /* __ECHO_H */ diff --git a/drivers/misc/echo/fir.h b/drivers/misc/echo/fir.h new file mode 100644 index 00000000000..7b9fabf1fea --- /dev/null +++ b/drivers/misc/echo/fir.h @@ -0,0 +1,216 @@ +/* + * SpanDSP - a series of DSP components for telephony + * + * fir.h - General telephony FIR routines + * + * Written by Steve Underwood <steveu@coppice.org> + * + * Copyright (C) 2002 Steve Underwood + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#if !defined(_FIR_H_) +#define _FIR_H_ + +/* + Blackfin NOTES & IDEAS: + + A simple dot product function is used to implement the filter. This performs + just one MAC/cycle which is inefficient but was easy to implement as a first + pass. The current Blackfin code also uses an unrolled form of the filter + history to avoid 0 length hardware loop issues. This is wasteful of + memory. + + Ideas for improvement: + + 1/ Rewrite filter for dual MAC inner loop. The issue here is handling + history sample offsets that are 16 bit aligned - the dual MAC needs + 32 bit aligmnent. There are some good examples in libbfdsp. + + 2/ Use the hardware circular buffer facility tohalve memory usage. + + 3/ Consider using internal memory. + + Using less memory might also improve speed as cache misses will be + reduced. A drop in MIPs and memory approaching 50% should be + possible. + + The foreground and background filters currenlty use a total of + about 10 MIPs/ch as measured with speedtest.c on a 256 TAP echo + can. +*/ + +/* + * 16 bit integer FIR descriptor. This defines the working state for a single + * instance of an FIR filter using 16 bit integer coefficients. + */ +struct fir16_state_t { + int taps; + int curr_pos; + const int16_t *coeffs; + int16_t *history; +}; + +/* + * 32 bit integer FIR descriptor. This defines the working state for a single + * instance of an FIR filter using 32 bit integer coefficients, and filtering + * 16 bit integer data. + */ +struct fir32_state_t { + int taps; + int curr_pos; + const int32_t *coeffs; + int16_t *history; +}; + +/* + * Floating point FIR descriptor. This defines the working state for a single + * instance of an FIR filter using floating point coefficients and data. + */ +struct fir_float_state_t { + int taps; + int curr_pos; + const float *coeffs; + float *history; +}; + +static inline const int16_t *fir16_create(struct fir16_state_t *fir, + const int16_t *coeffs, int taps) +{ + fir->taps = taps; + fir->curr_pos = taps - 1; + fir->coeffs = coeffs; +#if defined(__bfin__) + fir->history = kcalloc(2 * taps, sizeof(int16_t), GFP_KERNEL); +#else + fir->history = kcalloc(taps, sizeof(int16_t), GFP_KERNEL); +#endif + return fir->history; +} + +static inline void fir16_flush(struct fir16_state_t *fir) +{ +#if defined(__bfin__) + memset(fir->history, 0, 2 * fir->taps * sizeof(int16_t)); +#else + memset(fir->history, 0, fir->taps * sizeof(int16_t)); +#endif +} + +static inline void fir16_free(struct fir16_state_t *fir) +{ + kfree(fir->history); +} + +#ifdef __bfin__ +static inline int32_t dot_asm(short *x, short *y, int len) +{ + int dot; + + len--; + + __asm__("I0 = %1;\n\t" + "I1 = %2;\n\t" + "A0 = 0;\n\t" + "R0.L = W[I0++] || R1.L = W[I1++];\n\t" + "LOOP dot%= LC0 = %3;\n\t" + "LOOP_BEGIN dot%=;\n\t" + "A0 += R0.L * R1.L (IS) || R0.L = W[I0++] || R1.L = W[I1++];\n\t" + "LOOP_END dot%=;\n\t" + "A0 += R0.L*R1.L (IS);\n\t" + "R0 = A0;\n\t" + "%0 = R0;\n\t" + : "=&d"(dot) + : "a"(x), "a"(y), "a"(len) + : "I0", "I1", "A1", "A0", "R0", "R1" + ); + + return dot; +} +#endif + +static inline int16_t fir16(struct fir16_state_t *fir, int16_t sample) +{ + int32_t y; +#if defined(__bfin__) + fir->history[fir->curr_pos] = sample; + fir->history[fir->curr_pos + fir->taps] = sample; + y = dot_asm((int16_t *) fir->coeffs, &fir->history[fir->curr_pos], + fir->taps); +#else + int i; + int offset1; + int offset2; + + fir->history[fir->curr_pos] = sample; + + offset2 = fir->curr_pos; + offset1 = fir->taps - offset2; + y = 0; + for (i = fir->taps - 1; i >= offset1; i--) + y += fir->coeffs[i] * fir->history[i - offset1]; + for (; i >= 0; i--) + y += fir->coeffs[i] * fir->history[i + offset2]; +#endif + if (fir->curr_pos <= 0) + fir->curr_pos = fir->taps; + fir->curr_pos--; + return (int16_t) (y >> 15); +} + +static inline const int16_t *fir32_create(struct fir32_state_t *fir, + const int32_t *coeffs, int taps) +{ + fir->taps = taps; + fir->curr_pos = taps - 1; + fir->coeffs = coeffs; + fir->history = kcalloc(taps, sizeof(int16_t), GFP_KERNEL); + return fir->history; +} + +static inline void fir32_flush(struct fir32_state_t *fir) +{ + memset(fir->history, 0, fir->taps * sizeof(int16_t)); +} + +static inline void fir32_free(struct fir32_state_t *fir) +{ + kfree(fir->history); +} + +static inline int16_t fir32(struct fir32_state_t *fir, int16_t sample) +{ + int i; + int32_t y; + int offset1; + int offset2; + + fir->history[fir->curr_pos] = sample; + offset2 = fir->curr_pos; + offset1 = fir->taps - offset2; + y = 0; + for (i = fir->taps - 1; i >= offset1; i--) + y += fir->coeffs[i] * fir->history[i - offset1]; + for (; i >= 0; i--) + y += fir->coeffs[i] * fir->history[i + offset2]; + if (fir->curr_pos <= 0) + fir->curr_pos = fir->taps; + fir->curr_pos--; + return (int16_t) (y >> 15); +} + +#endif diff --git a/drivers/misc/echo/oslec.h b/drivers/misc/echo/oslec.h new file mode 100644 index 00000000000..f4175360ce2 --- /dev/null +++ b/drivers/misc/echo/oslec.h @@ -0,0 +1,94 @@ +/* + * OSLEC - A line echo canceller. This code is being developed + * against and partially complies with G168. Using code from SpanDSP + * + * Written by Steve Underwood <steveu@coppice.org> + * and David Rowe <david_at_rowetel_dot_com> + * + * Copyright (C) 2001 Steve Underwood and 2007-2008 David Rowe + * + * All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + * + */ + +#ifndef __OSLEC_H +#define __OSLEC_H + +/* Mask bits for the adaption mode */ +#define ECHO_CAN_USE_ADAPTION 0x01 +#define ECHO_CAN_USE_NLP 0x02 +#define ECHO_CAN_USE_CNG 0x04 +#define ECHO_CAN_USE_CLIP 0x08 +#define ECHO_CAN_USE_TX_HPF 0x10 +#define ECHO_CAN_USE_RX_HPF 0x20 +#define ECHO_CAN_DISABLE 0x40 + +/** + * oslec_state: G.168 echo canceller descriptor. + * + * This defines the working state for a line echo canceller. + */ +struct oslec_state; + +/** + * oslec_create - Create a voice echo canceller context. + * @len: The length of the canceller, in samples. + * @return: The new canceller context, or NULL if the canceller could not be + * created. + */ +struct oslec_state *oslec_create(int len, int adaption_mode); + +/** + * oslec_free - Free a voice echo canceller context. + * @ec: The echo canceller context. + */ +void oslec_free(struct oslec_state *ec); + +/** + * oslec_flush - Flush (reinitialise) a voice echo canceller context. + * @ec: The echo canceller context. + */ +void oslec_flush(struct oslec_state *ec); + +/** + * oslec_adaption_mode - set the adaption mode of a voice echo canceller context. + * @ec The echo canceller context. + * @adaption_mode: The mode. + */ +void oslec_adaption_mode(struct oslec_state *ec, int adaption_mode); + +void oslec_snapshot(struct oslec_state *ec); + +/** + * oslec_update: Process a sample through a voice echo canceller. + * @ec: The echo canceller context. + * @tx: The transmitted audio sample. + * @rx: The received audio sample. + * + * The return value is the clean (echo cancelled) received sample. + */ +int16_t oslec_update(struct oslec_state *ec, int16_t tx, int16_t rx); + +/** + * oslec_hpf_tx: Process to high pass filter the tx signal. + * @ec: The echo canceller context. + * @tx: The transmitted auio sample. + * + * The return value is the HP filtered transmit sample, send this to your D/A. + */ +int16_t oslec_hpf_tx(struct oslec_state *ec, int16_t tx); + +#endif /* __OSLEC_H */ diff --git a/drivers/misc/eeprom/Kconfig b/drivers/misc/eeprom/Kconfig index c9e695ea7c9..9536852fd4c 100644 --- a/drivers/misc/eeprom/Kconfig +++ b/drivers/misc/eeprom/Kconfig @@ -1,13 +1,14 @@ menu "EEPROM support" config EEPROM_AT24 - tristate "I2C EEPROMs from most vendors" + tristate "I2C EEPROMs / RAMs / ROMs from most vendors" depends on I2C && SYSFS help - Enable this driver to get read/write support to most I2C EEPROMs, - after you configure the driver to know about each EEPROM on - your target board. Use these generic chip names, instead of - vendor-specific ones like at24c64 or 24lc02: + Enable this driver to get read/write support to most I2C EEPROMs + and compatible devices like FRAMs, SRAMs, ROMs etc. After you + configure the driver to know about each chip on your target + board. Use these generic chip names, instead of vendor-specific + ones like at24c64, 24lc02 or fm24c04: 24c00, 24c01, 24c02, spd (readonly 24c02), 24c04, 24c08, 24c16, 24c32, 24c64, 24c128, 24c256, 24c512, 24c1024 @@ -95,4 +96,17 @@ config EEPROM_DIGSY_MTC_CFG If unsure, say N. +config EEPROM_SUNXI_SID + tristate "Allwinner sunxi security ID support" + depends on ARCH_SUNXI && SYSFS + help + This is a driver for the 'security ID' available on various Allwinner + devices. + + Due to the potential risks involved with changing e-fuses, + this driver is read-only. + + This driver can also be built as a module. If so, the module + will be called sunxi_sid. + endmenu diff --git a/drivers/misc/eeprom/Makefile b/drivers/misc/eeprom/Makefile index fc1e81d2926..9507aec95e9 100644 --- a/drivers/misc/eeprom/Makefile +++ b/drivers/misc/eeprom/Makefile @@ -4,4 +4,5 @@ obj-$(CONFIG_EEPROM_LEGACY) += eeprom.o obj-$(CONFIG_EEPROM_MAX6875) += max6875.o obj-$(CONFIG_EEPROM_93CX6) += eeprom_93cx6.o obj-$(CONFIG_EEPROM_93XX46) += eeprom_93xx46.o +obj-$(CONFIG_EEPROM_SUNXI_SID) += sunxi_sid.o obj-$(CONFIG_EEPROM_DIGSY_MTC_CFG) += digsy_mtc_eeprom.o diff --git a/drivers/misc/eeprom/at24.c b/drivers/misc/eeprom/at24.c index ab1ad41786d..d87f77f790d 100644 --- a/drivers/misc/eeprom/at24.c +++ b/drivers/misc/eeprom/at24.c @@ -22,7 +22,7 @@ #include <linux/jiffies.h> #include <linux/of.h> #include <linux/i2c.h> -#include <linux/i2c/at24.h> +#include <linux/platform_data/at24.h> /* * I2C EEPROMs from most vendors are inexpensive and mostly interchangeable. @@ -428,6 +428,9 @@ static ssize_t at24_bin_write(struct file *filp, struct kobject *kobj, { struct at24_data *at24; + if (unlikely(off >= attr->size)) + return -EFBIG; + at24 = dev_get_drvdata(container_of(kobj, struct device, kobj)); return at24_write(at24, buf, off, count); } @@ -492,10 +495,9 @@ static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id) if (client->dev.platform_data) { chip = *(struct at24_platform_data *)client->dev.platform_data; } else { - if (!id->driver_data) { - err = -ENODEV; - goto err_out; - } + if (!id->driver_data) + return -ENODEV; + magic = id->driver_data; chip.byte_len = BIT(magic & AT24_BITMASK(AT24_SIZE_BYTELEN)); magic >>= AT24_SIZE_BYTELEN; @@ -519,8 +521,7 @@ static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id) "byte_len looks suspicious (no power of 2)!\n"); if (!chip.page_size) { dev_err(&client->dev, "page_size must not be 0!\n"); - err = -EINVAL; - goto err_out; + return -EINVAL; } if (!is_power_of_2(chip.page_size)) dev_warn(&client->dev, @@ -528,10 +529,9 @@ static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id) /* Use I2C operations unless we're stuck with SMBus extensions. */ if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) { - if (chip.flags & AT24_FLAG_ADDR16) { - err = -EPFNOSUPPORT; - goto err_out; - } + if (chip.flags & AT24_FLAG_ADDR16) + return -EPFNOSUPPORT; + if (i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_READ_I2C_BLOCK)) { use_smbus = I2C_SMBUS_I2C_BLOCK_DATA; @@ -542,8 +542,7 @@ static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id) I2C_FUNC_SMBUS_READ_BYTE_DATA)) { use_smbus = I2C_SMBUS_BYTE_DATA; } else { - err = -EPFNOSUPPORT; - goto err_out; + return -EPFNOSUPPORT; } } @@ -553,12 +552,10 @@ static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id) num_addresses = DIV_ROUND_UP(chip.byte_len, (chip.flags & AT24_FLAG_ADDR16) ? 65536 : 256); - at24 = kzalloc(sizeof(struct at24_data) + + at24 = devm_kzalloc(&client->dev, sizeof(struct at24_data) + num_addresses * sizeof(struct i2c_client *), GFP_KERNEL); - if (!at24) { - err = -ENOMEM; - goto err_out; - } + if (!at24) + return -ENOMEM; mutex_init(&at24->lock); at24->use_smbus = use_smbus; @@ -596,11 +593,10 @@ static int at24_probe(struct i2c_client *client, const struct i2c_device_id *id) at24->write_max = write_max; /* buffer (data + address at the beginning) */ - at24->writebuf = kmalloc(write_max + 2, GFP_KERNEL); - if (!at24->writebuf) { - err = -ENOMEM; - goto err_struct; - } + at24->writebuf = devm_kzalloc(&client->dev, + write_max + 2, GFP_KERNEL); + if (!at24->writebuf) + return -ENOMEM; } else { dev_warn(&client->dev, "cannot write due to controller restrictions."); @@ -648,15 +644,10 @@ err_clients: if (at24->client[i]) i2c_unregister_device(at24->client[i]); - kfree(at24->writebuf); -err_struct: - kfree(at24); -err_out: - dev_dbg(&client->dev, "probe error %d\n", err); return err; } -static int __devexit at24_remove(struct i2c_client *client) +static int at24_remove(struct i2c_client *client) { struct at24_data *at24; int i; @@ -667,8 +658,6 @@ static int __devexit at24_remove(struct i2c_client *client) for (i = 1; i < at24->num_addresses; i++) i2c_unregister_device(at24->client[i]); - kfree(at24->writebuf); - kfree(at24); return 0; } @@ -680,7 +669,7 @@ static struct i2c_driver at24_driver = { .owner = THIS_MODULE, }, .probe = at24_probe, - .remove = __devexit_p(at24_remove), + .remove = at24_remove, .id_table = at24_ids, }; diff --git a/drivers/misc/eeprom/at25.c b/drivers/misc/eeprom/at25.c index 4ed93dd5411..634f72929e1 100644 --- a/drivers/misc/eeprom/at25.c +++ b/drivers/misc/eeprom/at25.c @@ -10,7 +10,6 @@ */ #include <linux/kernel.h> -#include <linux/init.h> #include <linux/module.h> #include <linux/slab.h> #include <linux/delay.h> @@ -371,11 +370,10 @@ static int at25_probe(struct spi_device *spi) if (np) { err = at25_np_to_chip(&spi->dev, np, &chip); if (err) - goto fail; + return err; } else { dev_err(&spi->dev, "Error: no chip description\n"); - err = -ENODEV; - goto fail; + return -ENODEV; } } else chip = *(struct spi_eeprom *)spi->dev.platform_data; @@ -389,8 +387,7 @@ static int at25_probe(struct spi_device *spi) addrlen = 3; else { dev_dbg(&spi->dev, "unsupported address type\n"); - err = -EINVAL; - goto fail; + return -EINVAL; } /* Ping the chip ... the status register is pretty portable, @@ -400,19 +397,17 @@ static int at25_probe(struct spi_device *spi) sr = spi_w8r8(spi, AT25_RDSR); if (sr < 0 || sr & AT25_SR_nRDY) { dev_dbg(&spi->dev, "rdsr --> %d (%02x)\n", sr, sr); - err = -ENXIO; - goto fail; + return -ENXIO; } - if (!(at25 = kzalloc(sizeof *at25, GFP_KERNEL))) { - err = -ENOMEM; - goto fail; - } + at25 = devm_kzalloc(&spi->dev, sizeof(struct at25_data), GFP_KERNEL); + if (!at25) + return -ENOMEM; mutex_init(&at25->lock); at25->chip = chip; at25->spi = spi_dev_get(spi); - dev_set_drvdata(&spi->dev, at25); + spi_set_drvdata(spi, at25); at25->addrlen = addrlen; /* Export the EEPROM bytes through sysfs, since that's convenient. @@ -439,7 +434,7 @@ static int at25_probe(struct spi_device *spi) err = sysfs_create_bin_file(&spi->dev.kobj, &at25->bin); if (err) - goto fail; + return err; if (chip.setup) chip.setup(&at25->mem, chip.context); @@ -453,31 +448,33 @@ static int at25_probe(struct spi_device *spi) (chip.flags & EE_READONLY) ? " (readonly)" : "", at25->chip.page_size); return 0; -fail: - dev_dbg(&spi->dev, "probe err %d\n", err); - kfree(at25); - return err; } -static int __devexit at25_remove(struct spi_device *spi) +static int at25_remove(struct spi_device *spi) { struct at25_data *at25; - at25 = dev_get_drvdata(&spi->dev); + at25 = spi_get_drvdata(spi); sysfs_remove_bin_file(&spi->dev.kobj, &at25->bin); - kfree(at25); return 0; } /*-------------------------------------------------------------------------*/ +static const struct of_device_id at25_of_match[] = { + { .compatible = "atmel,at25", }, + { } +}; +MODULE_DEVICE_TABLE(of, at25_of_match); + static struct spi_driver at25_driver = { .driver = { .name = "at25", .owner = THIS_MODULE, + .of_match_table = at25_of_match, }, .probe = at25_probe, - .remove = __devexit_p(at25_remove), + .remove = at25_remove, }; module_spi_driver(at25_driver); diff --git a/drivers/misc/eeprom/eeprom.c b/drivers/misc/eeprom/eeprom.c index c169e07654c..33f8673d23a 100644 --- a/drivers/misc/eeprom/eeprom.c +++ b/drivers/misc/eeprom/eeprom.c @@ -3,7 +3,7 @@ * Philip Edelbrock <phil@netroedge.com> * Copyright (C) 2003 Greg Kroah-Hartman <greg@kroah.com> * Copyright (C) 2003 IBM Corp. - * Copyright (C) 2004 Jean Delvare <khali@linux-fr.org> + * Copyright (C) 2004 Jean Delvare <jdelvare@suse.de> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -17,7 +17,6 @@ */ #include <linux/kernel.h> -#include <linux/init.h> #include <linux/module.h> #include <linux/slab.h> #include <linux/jiffies.h> diff --git a/drivers/misc/eeprom/eeprom_93xx46.c b/drivers/misc/eeprom/eeprom_93xx46.c index ce3fe3633dd..9ebeacdb8ec 100644 --- a/drivers/misc/eeprom/eeprom_93xx46.c +++ b/drivers/misc/eeprom/eeprom_93xx46.c @@ -11,7 +11,6 @@ #include <linux/delay.h> #include <linux/device.h> #include <linux/kernel.h> -#include <linux/init.h> #include <linux/module.h> #include <linux/mutex.h> #include <linux/slab.h> @@ -202,7 +201,7 @@ eeprom_93xx46_bin_write(struct file *filp, struct kobject *kobj, edev = dev_get_drvdata(dev); if (unlikely(off >= edev->bin.size)) - return 0; + return -EFBIG; if ((off + count) > edev->bin.size) count = edev->bin.size - off; if (unlikely(!count)) @@ -309,7 +308,7 @@ static ssize_t eeprom_93xx46_store_erase(struct device *dev, } static DEVICE_ATTR(erase, S_IWUSR, NULL, eeprom_93xx46_store_erase); -static int __devinit eeprom_93xx46_probe(struct spi_device *spi) +static int eeprom_93xx46_probe(struct spi_device *spi) { struct eeprom_93xx46_platform_data *pd; struct eeprom_93xx46_dev *edev; @@ -363,22 +362,21 @@ static int __devinit eeprom_93xx46_probe(struct spi_device *spi) dev_err(&spi->dev, "can't create erase interface\n"); } - dev_set_drvdata(&spi->dev, edev); + spi_set_drvdata(spi, edev); return 0; fail: kfree(edev); return err; } -static int __devexit eeprom_93xx46_remove(struct spi_device *spi) +static int eeprom_93xx46_remove(struct spi_device *spi) { - struct eeprom_93xx46_dev *edev = dev_get_drvdata(&spi->dev); + struct eeprom_93xx46_dev *edev = spi_get_drvdata(spi); if (!(edev->pdata->flags & EE_READONLY)) device_remove_file(&spi->dev, &dev_attr_erase); sysfs_remove_bin_file(&spi->dev.kobj, &edev->bin); - dev_set_drvdata(&spi->dev, NULL); kfree(edev); return 0; } @@ -389,7 +387,7 @@ static struct spi_driver eeprom_93xx46_driver = { .owner = THIS_MODULE, }, .probe = eeprom_93xx46_probe, - .remove = __devexit_p(eeprom_93xx46_remove), + .remove = eeprom_93xx46_remove, }; module_spi_driver(eeprom_93xx46_driver); diff --git a/drivers/misc/eeprom/max6875.c b/drivers/misc/eeprom/max6875.c index e36157d5d3a..580ff9df552 100644 --- a/drivers/misc/eeprom/max6875.c +++ b/drivers/misc/eeprom/max6875.c @@ -27,7 +27,6 @@ */ #include <linux/kernel.h> -#include <linux/init.h> #include <linux/module.h> #include <linux/slab.h> #include <linux/i2c.h> diff --git a/drivers/misc/eeprom/sunxi_sid.c b/drivers/misc/eeprom/sunxi_sid.c new file mode 100644 index 00000000000..3f2b625b203 --- /dev/null +++ b/drivers/misc/eeprom/sunxi_sid.c @@ -0,0 +1,157 @@ +/* + * Copyright (c) 2013 Oliver Schinagl <oliver@schinagl.nl> + * http://www.linux-sunxi.org + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * This driver exposes the Allwinner security ID, efuses exported in byte- + * sized chunks. + */ + +#include <linux/compiler.h> +#include <linux/device.h> +#include <linux/err.h> +#include <linux/export.h> +#include <linux/fs.h> +#include <linux/io.h> +#include <linux/kernel.h> +#include <linux/kobject.h> +#include <linux/module.h> +#include <linux/of_device.h> +#include <linux/platform_device.h> +#include <linux/random.h> +#include <linux/slab.h> +#include <linux/stat.h> +#include <linux/sysfs.h> +#include <linux/types.h> + +#define DRV_NAME "sunxi-sid" + +struct sunxi_sid_data { + void __iomem *reg_base; + unsigned int keysize; +}; + +/* We read the entire key, due to a 32 bit read alignment requirement. Since we + * want to return the requested byte, this results in somewhat slower code and + * uses 4 times more reads as needed but keeps code simpler. Since the SID is + * only very rarely probed, this is not really an issue. + */ +static u8 sunxi_sid_read_byte(const struct sunxi_sid_data *sid_data, + const unsigned int offset) +{ + u32 sid_key; + + if (offset >= sid_data->keysize) + return 0; + + sid_key = ioread32be(sid_data->reg_base + round_down(offset, 4)); + sid_key >>= (offset % 4) * 8; + + return sid_key; /* Only return the last byte */ +} + +static ssize_t sid_read(struct file *fd, struct kobject *kobj, + struct bin_attribute *attr, char *buf, + loff_t pos, size_t size) +{ + struct platform_device *pdev; + struct sunxi_sid_data *sid_data; + int i; + + pdev = to_platform_device(kobj_to_dev(kobj)); + sid_data = platform_get_drvdata(pdev); + + if (pos < 0 || pos >= sid_data->keysize) + return 0; + if (size > sid_data->keysize - pos) + size = sid_data->keysize - pos; + + for (i = 0; i < size; i++) + buf[i] = sunxi_sid_read_byte(sid_data, pos + i); + + return i; +} + +static struct bin_attribute sid_bin_attr = { + .attr = { .name = "eeprom", .mode = S_IRUGO, }, + .read = sid_read, +}; + +static int sunxi_sid_remove(struct platform_device *pdev) +{ + device_remove_bin_file(&pdev->dev, &sid_bin_attr); + dev_dbg(&pdev->dev, "driver unloaded\n"); + + return 0; +} + +static const struct of_device_id sunxi_sid_of_match[] = { + { .compatible = "allwinner,sun4i-a10-sid", .data = (void *)16}, + { .compatible = "allwinner,sun7i-a20-sid", .data = (void *)512}, + {/* sentinel */}, +}; +MODULE_DEVICE_TABLE(of, sunxi_sid_of_match); + +static int sunxi_sid_probe(struct platform_device *pdev) +{ + struct sunxi_sid_data *sid_data; + struct resource *res; + const struct of_device_id *of_dev_id; + u8 *entropy; + unsigned int i; + + sid_data = devm_kzalloc(&pdev->dev, sizeof(struct sunxi_sid_data), + GFP_KERNEL); + if (!sid_data) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + sid_data->reg_base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(sid_data->reg_base)) + return PTR_ERR(sid_data->reg_base); + + of_dev_id = of_match_device(sunxi_sid_of_match, &pdev->dev); + if (!of_dev_id) + return -ENODEV; + sid_data->keysize = (int)of_dev_id->data; + + platform_set_drvdata(pdev, sid_data); + + sid_bin_attr.size = sid_data->keysize; + if (device_create_bin_file(&pdev->dev, &sid_bin_attr)) + return -ENODEV; + + entropy = kzalloc(sizeof(u8) * sid_data->keysize, GFP_KERNEL); + for (i = 0; i < sid_data->keysize; i++) + entropy[i] = sunxi_sid_read_byte(sid_data, i); + add_device_randomness(entropy, sid_data->keysize); + kfree(entropy); + + dev_dbg(&pdev->dev, "loaded\n"); + + return 0; +} + +static struct platform_driver sunxi_sid_driver = { + .probe = sunxi_sid_probe, + .remove = sunxi_sid_remove, + .driver = { + .name = DRV_NAME, + .owner = THIS_MODULE, + .of_match_table = sunxi_sid_of_match, + }, +}; +module_platform_driver(sunxi_sid_driver); + +MODULE_AUTHOR("Oliver Schinagl <oliver@schinagl.nl>"); +MODULE_DESCRIPTION("Allwinner sunxi security id driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/misc/enclosure.c b/drivers/misc/enclosure.c index 00e5fcac8fd..2cf2bbc0b92 100644 --- a/drivers/misc/enclosure.c +++ b/drivers/misc/enclosure.c @@ -198,6 +198,13 @@ static void enclosure_remove_links(struct enclosure_component *cdev) { char name[ENCLOSURE_NAME_SIZE]; + /* + * In odd circumstances, like multipath devices, something else may + * already have removed the links, so check for this condition first. + */ + if (!cdev->dev->kobj.sd) + return; + enclosure_link_name(cdev, name); sysfs_remove_link(&cdev->dev->kobj, name); sysfs_remove_link(&cdev->cdev.kobj, "device"); @@ -239,7 +246,7 @@ static void enclosure_component_release(struct device *dev) put_device(dev->parent); } -static const struct attribute_group *enclosure_groups[]; +static const struct attribute_group *enclosure_component_groups[]; /** * enclosure_component_register - add a particular component to an enclosure @@ -282,7 +289,7 @@ enclosure_component_register(struct enclosure_device *edev, dev_set_name(cdev, "%u", number); cdev->release = enclosure_component_release; - cdev->groups = enclosure_groups; + cdev->groups = enclosure_component_groups; err = device_register(cdev); if (err) { @@ -365,25 +372,26 @@ EXPORT_SYMBOL_GPL(enclosure_remove_device); * sysfs pieces below */ -static ssize_t enclosure_show_components(struct device *cdev, - struct device_attribute *attr, - char *buf) +static ssize_t components_show(struct device *cdev, + struct device_attribute *attr, char *buf) { struct enclosure_device *edev = to_enclosure_device(cdev); return snprintf(buf, 40, "%d\n", edev->components); } +static DEVICE_ATTR_RO(components); -static struct device_attribute enclosure_attrs[] = { - __ATTR(components, S_IRUGO, enclosure_show_components, NULL), - __ATTR_NULL +static struct attribute *enclosure_class_attrs[] = { + &dev_attr_components.attr, + NULL, }; +ATTRIBUTE_GROUPS(enclosure_class); static struct class enclosure_class = { .name = "enclosure", .owner = THIS_MODULE, .dev_release = enclosure_release, - .dev_attrs = enclosure_attrs, + .dev_groups = enclosure_class_groups, }; static const char *const enclosure_status [] = { @@ -536,15 +544,7 @@ static struct attribute *enclosure_component_attrs[] = { &dev_attr_type.attr, NULL }; - -static struct attribute_group enclosure_group = { - .attrs = enclosure_component_attrs, -}; - -static const struct attribute_group *enclosure_groups[] = { - &enclosure_group, - NULL -}; +ATTRIBUTE_GROUPS(enclosure_component); static int __init enclosure_init(void) { diff --git a/drivers/misc/ep93xx_pwm.c b/drivers/misc/ep93xx_pwm.c deleted file mode 100644 index 16d7179e2f9..00000000000 --- a/drivers/misc/ep93xx_pwm.c +++ /dev/null @@ -1,385 +0,0 @@ -/* - * Simple PWM driver for EP93XX - * - * (c) Copyright 2009 Matthieu Crapet <mcrapet@gmail.com> - * (c) Copyright 2009 H Hartley Sweeten <hsweeten@visionengravers.com> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * EP9307 has only one channel: - * - PWMOUT - * - * EP9301/02/12/15 have two channels: - * - PWMOUT - * - PWMOUT1 (alternate function for EGPIO14) - */ - -#include <linux/module.h> -#include <linux/platform_device.h> -#include <linux/slab.h> -#include <linux/clk.h> -#include <linux/err.h> -#include <linux/io.h> - -#include <mach/platform.h> - -#define EP93XX_PWMx_TERM_COUNT 0x00 -#define EP93XX_PWMx_DUTY_CYCLE 0x04 -#define EP93XX_PWMx_ENABLE 0x08 -#define EP93XX_PWMx_INVERT 0x0C - -#define EP93XX_PWM_MAX_COUNT 0xFFFF - -struct ep93xx_pwm { - void __iomem *mmio_base; - struct clk *clk; - u32 duty_percent; -}; - -static inline void ep93xx_pwm_writel(struct ep93xx_pwm *pwm, - unsigned int val, unsigned int off) -{ - __raw_writel(val, pwm->mmio_base + off); -} - -static inline unsigned int ep93xx_pwm_readl(struct ep93xx_pwm *pwm, - unsigned int off) -{ - return __raw_readl(pwm->mmio_base + off); -} - -static inline void ep93xx_pwm_write_tc(struct ep93xx_pwm *pwm, u16 value) -{ - ep93xx_pwm_writel(pwm, value, EP93XX_PWMx_TERM_COUNT); -} - -static inline u16 ep93xx_pwm_read_tc(struct ep93xx_pwm *pwm) -{ - return ep93xx_pwm_readl(pwm, EP93XX_PWMx_TERM_COUNT); -} - -static inline void ep93xx_pwm_write_dc(struct ep93xx_pwm *pwm, u16 value) -{ - ep93xx_pwm_writel(pwm, value, EP93XX_PWMx_DUTY_CYCLE); -} - -static inline void ep93xx_pwm_enable(struct ep93xx_pwm *pwm) -{ - ep93xx_pwm_writel(pwm, 0x1, EP93XX_PWMx_ENABLE); -} - -static inline void ep93xx_pwm_disable(struct ep93xx_pwm *pwm) -{ - ep93xx_pwm_writel(pwm, 0x0, EP93XX_PWMx_ENABLE); -} - -static inline int ep93xx_pwm_is_enabled(struct ep93xx_pwm *pwm) -{ - return ep93xx_pwm_readl(pwm, EP93XX_PWMx_ENABLE) & 0x1; -} - -static inline void ep93xx_pwm_invert(struct ep93xx_pwm *pwm) -{ - ep93xx_pwm_writel(pwm, 0x1, EP93XX_PWMx_INVERT); -} - -static inline void ep93xx_pwm_normal(struct ep93xx_pwm *pwm) -{ - ep93xx_pwm_writel(pwm, 0x0, EP93XX_PWMx_INVERT); -} - -static inline int ep93xx_pwm_is_inverted(struct ep93xx_pwm *pwm) -{ - return ep93xx_pwm_readl(pwm, EP93XX_PWMx_INVERT) & 0x1; -} - -/* - * /sys/devices/platform/ep93xx-pwm.N - * /min_freq read-only minimum pwm output frequency - * /max_req read-only maximum pwm output frequency - * /freq read-write pwm output frequency (0 = disable output) - * /duty_percent read-write pwm duty cycle percent (1..99) - * /invert read-write invert pwm output - */ - -static ssize_t ep93xx_pwm_get_min_freq(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - unsigned long rate = clk_get_rate(pwm->clk); - - return sprintf(buf, "%ld\n", rate / (EP93XX_PWM_MAX_COUNT + 1)); -} - -static ssize_t ep93xx_pwm_get_max_freq(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - unsigned long rate = clk_get_rate(pwm->clk); - - return sprintf(buf, "%ld\n", rate / 2); -} - -static ssize_t ep93xx_pwm_get_freq(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - - if (ep93xx_pwm_is_enabled(pwm)) { - unsigned long rate = clk_get_rate(pwm->clk); - u16 term = ep93xx_pwm_read_tc(pwm); - - return sprintf(buf, "%ld\n", rate / (term + 1)); - } else { - return sprintf(buf, "disabled\n"); - } -} - -static ssize_t ep93xx_pwm_set_freq(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - long val; - int err; - - err = strict_strtol(buf, 10, &val); - if (err) - return -EINVAL; - - if (val == 0) { - ep93xx_pwm_disable(pwm); - } else if (val <= (clk_get_rate(pwm->clk) / 2)) { - u32 term, duty; - - val = (clk_get_rate(pwm->clk) / val) - 1; - if (val > EP93XX_PWM_MAX_COUNT) - val = EP93XX_PWM_MAX_COUNT; - if (val < 1) - val = 1; - - term = ep93xx_pwm_read_tc(pwm); - duty = ((val + 1) * pwm->duty_percent / 100) - 1; - - /* If pwm is running, order is important */ - if (val > term) { - ep93xx_pwm_write_tc(pwm, val); - ep93xx_pwm_write_dc(pwm, duty); - } else { - ep93xx_pwm_write_dc(pwm, duty); - ep93xx_pwm_write_tc(pwm, val); - } - - if (!ep93xx_pwm_is_enabled(pwm)) - ep93xx_pwm_enable(pwm); - } else { - return -EINVAL; - } - - return count; -} - -static ssize_t ep93xx_pwm_get_duty_percent(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - - return sprintf(buf, "%d\n", pwm->duty_percent); -} - -static ssize_t ep93xx_pwm_set_duty_percent(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - long val; - int err; - - err = strict_strtol(buf, 10, &val); - if (err) - return -EINVAL; - - if (val > 0 && val < 100) { - u32 term = ep93xx_pwm_read_tc(pwm); - ep93xx_pwm_write_dc(pwm, ((term + 1) * val / 100) - 1); - pwm->duty_percent = val; - return count; - } - - return -EINVAL; -} - -static ssize_t ep93xx_pwm_get_invert(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - - return sprintf(buf, "%d\n", ep93xx_pwm_is_inverted(pwm)); -} - -static ssize_t ep93xx_pwm_set_invert(struct device *dev, - struct device_attribute *attr, const char *buf, size_t count) -{ - struct platform_device *pdev = to_platform_device(dev); - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - long val; - int err; - - err = strict_strtol(buf, 10, &val); - if (err) - return -EINVAL; - - if (val == 0) - ep93xx_pwm_normal(pwm); - else if (val == 1) - ep93xx_pwm_invert(pwm); - else - return -EINVAL; - - return count; -} - -static DEVICE_ATTR(min_freq, S_IRUGO, ep93xx_pwm_get_min_freq, NULL); -static DEVICE_ATTR(max_freq, S_IRUGO, ep93xx_pwm_get_max_freq, NULL); -static DEVICE_ATTR(freq, S_IWUSR | S_IRUGO, - ep93xx_pwm_get_freq, ep93xx_pwm_set_freq); -static DEVICE_ATTR(duty_percent, S_IWUSR | S_IRUGO, - ep93xx_pwm_get_duty_percent, ep93xx_pwm_set_duty_percent); -static DEVICE_ATTR(invert, S_IWUSR | S_IRUGO, - ep93xx_pwm_get_invert, ep93xx_pwm_set_invert); - -static struct attribute *ep93xx_pwm_attrs[] = { - &dev_attr_min_freq.attr, - &dev_attr_max_freq.attr, - &dev_attr_freq.attr, - &dev_attr_duty_percent.attr, - &dev_attr_invert.attr, - NULL -}; - -static const struct attribute_group ep93xx_pwm_sysfs_files = { - .attrs = ep93xx_pwm_attrs, -}; - -static int __init ep93xx_pwm_probe(struct platform_device *pdev) -{ - struct ep93xx_pwm *pwm; - struct resource *res; - int err; - - err = ep93xx_pwm_acquire_gpio(pdev); - if (err) - return err; - - pwm = kzalloc(sizeof(struct ep93xx_pwm), GFP_KERNEL); - if (!pwm) { - err = -ENOMEM; - goto fail_no_mem; - } - - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (res == NULL) { - err = -ENXIO; - goto fail_no_mem_resource; - } - - res = request_mem_region(res->start, resource_size(res), pdev->name); - if (res == NULL) { - err = -EBUSY; - goto fail_no_mem_resource; - } - - pwm->mmio_base = ioremap(res->start, resource_size(res)); - if (pwm->mmio_base == NULL) { - err = -ENXIO; - goto fail_no_ioremap; - } - - err = sysfs_create_group(&pdev->dev.kobj, &ep93xx_pwm_sysfs_files); - if (err) - goto fail_no_sysfs; - - pwm->clk = clk_get(&pdev->dev, "pwm_clk"); - if (IS_ERR(pwm->clk)) { - err = PTR_ERR(pwm->clk); - goto fail_no_clk; - } - - pwm->duty_percent = 50; - - platform_set_drvdata(pdev, pwm); - - /* disable pwm at startup. Avoids zero value. */ - ep93xx_pwm_disable(pwm); - ep93xx_pwm_write_tc(pwm, EP93XX_PWM_MAX_COUNT); - ep93xx_pwm_write_dc(pwm, EP93XX_PWM_MAX_COUNT / 2); - - clk_enable(pwm->clk); - - return 0; - -fail_no_clk: - sysfs_remove_group(&pdev->dev.kobj, &ep93xx_pwm_sysfs_files); -fail_no_sysfs: - iounmap(pwm->mmio_base); -fail_no_ioremap: - release_mem_region(res->start, resource_size(res)); -fail_no_mem_resource: - kfree(pwm); -fail_no_mem: - ep93xx_pwm_release_gpio(pdev); - return err; -} - -static int __exit ep93xx_pwm_remove(struct platform_device *pdev) -{ - struct ep93xx_pwm *pwm = platform_get_drvdata(pdev); - struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - - ep93xx_pwm_disable(pwm); - clk_disable(pwm->clk); - clk_put(pwm->clk); - platform_set_drvdata(pdev, NULL); - sysfs_remove_group(&pdev->dev.kobj, &ep93xx_pwm_sysfs_files); - iounmap(pwm->mmio_base); - release_mem_region(res->start, resource_size(res)); - kfree(pwm); - ep93xx_pwm_release_gpio(pdev); - - return 0; -} - -static struct platform_driver ep93xx_pwm_driver = { - .driver = { - .name = "ep93xx-pwm", - .owner = THIS_MODULE, - }, - .remove = __exit_p(ep93xx_pwm_remove), -}; - -static int __init ep93xx_pwm_init(void) -{ - return platform_driver_probe(&ep93xx_pwm_driver, ep93xx_pwm_probe); -} - -static void __exit ep93xx_pwm_exit(void) -{ - platform_driver_unregister(&ep93xx_pwm_driver); -} - -module_init(ep93xx_pwm_init); -module_exit(ep93xx_pwm_exit); - -MODULE_AUTHOR("Matthieu Crapet <mcrapet@gmail.com>, " - "H Hartley Sweeten <hsweeten@visionengravers.com>"); -MODULE_DESCRIPTION("EP93xx PWM driver"); -MODULE_LICENSE("GPL"); -MODULE_ALIAS("platform:ep93xx-pwm"); diff --git a/drivers/misc/fsa9480.c b/drivers/misc/fsa9480.c index ac96c3a4034..71d2793b372 100644 --- a/drivers/misc/fsa9480.c +++ b/drivers/misc/fsa9480.c @@ -396,7 +396,7 @@ static int fsa9480_irq_init(struct fsa9480_usbsw *usbsw) IRQF_TRIGGER_FALLING | IRQF_ONESHOT, "fsa9480 micro USB", usbsw); if (ret) { - dev_err(&client->dev, "failed to reqeust IRQ\n"); + dev_err(&client->dev, "failed to request IRQ\n"); return ret; } @@ -407,7 +407,7 @@ static int fsa9480_irq_init(struct fsa9480_usbsw *usbsw) return 0; } -static int __devinit fsa9480_probe(struct i2c_client *client, +static int fsa9480_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent); @@ -462,7 +462,7 @@ fail1: return ret; } -static int __devexit fsa9480_remove(struct i2c_client *client) +static int fsa9480_remove(struct i2c_client *client) { struct fsa9480_usbsw *usbsw = i2c_get_clientdata(client); if (client->irq) @@ -474,10 +474,11 @@ static int __devexit fsa9480_remove(struct i2c_client *client) return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP -static int fsa9480_suspend(struct i2c_client *client, pm_message_t state) +static int fsa9480_suspend(struct device *dev) { + struct i2c_client *client = to_i2c_client(dev); struct fsa9480_usbsw *usbsw = i2c_get_clientdata(client); struct fsa9480_platform_data *pdata = usbsw->pdata; @@ -490,8 +491,9 @@ static int fsa9480_suspend(struct i2c_client *client, pm_message_t state) return 0; } -static int fsa9480_resume(struct i2c_client *client) +static int fsa9480_resume(struct device *dev) { + struct i2c_client *client = to_i2c_client(dev); struct fsa9480_usbsw *usbsw = i2c_get_clientdata(client); int dev1, dev2; @@ -515,12 +517,14 @@ static int fsa9480_resume(struct i2c_client *client) return 0; } +static SIMPLE_DEV_PM_OPS(fsa9480_pm_ops, fsa9480_suspend, fsa9480_resume); +#define FSA9480_PM_OPS (&fsa9480_pm_ops) + #else -#define fsa9480_suspend NULL -#define fsa9480_resume NULL +#define FSA9480_PM_OPS NULL -#endif /* CONFIG_PM */ +#endif /* CONFIG_PM_SLEEP */ static const struct i2c_device_id fsa9480_id[] = { {"fsa9480", 0}, @@ -531,11 +535,10 @@ MODULE_DEVICE_TABLE(i2c, fsa9480_id); static struct i2c_driver fsa9480_i2c_driver = { .driver = { .name = "fsa9480", + .pm = FSA9480_PM_OPS, }, .probe = fsa9480_probe, - .remove = __devexit_p(fsa9480_remove), - .resume = fsa9480_resume, - .suspend = fsa9480_suspend, + .remove = fsa9480_remove, .id_table = fsa9480_id, }; diff --git a/drivers/misc/genwqe/Kconfig b/drivers/misc/genwqe/Kconfig new file mode 100644 index 00000000000..6069d8cd79d --- /dev/null +++ b/drivers/misc/genwqe/Kconfig @@ -0,0 +1,13 @@ +# +# IBM Accelerator Family 'GenWQE' +# + +menuconfig GENWQE + tristate "GenWQE PCIe Accelerator" + depends on PCI && 64BIT + select CRC_ITU_T + default n + help + Enables PCIe card driver for IBM GenWQE accelerators. + The user-space interface is described in + include/linux/genwqe/genwqe_card.h. diff --git a/drivers/misc/genwqe/Makefile b/drivers/misc/genwqe/Makefile new file mode 100644 index 00000000000..98a2b4f0b18 --- /dev/null +++ b/drivers/misc/genwqe/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for GenWQE driver +# + +obj-$(CONFIG_GENWQE) := genwqe_card.o +genwqe_card-objs := card_base.o card_dev.o card_ddcb.o card_sysfs.o \ + card_debugfs.o card_utils.o diff --git a/drivers/misc/genwqe/card_base.c b/drivers/misc/genwqe/card_base.c new file mode 100644 index 00000000000..74d51c9bb85 --- /dev/null +++ b/drivers/misc/genwqe/card_base.c @@ -0,0 +1,1205 @@ +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +/* + * Module initialization and PCIe setup. Card health monitoring and + * recovery functionality. Character device creation and deletion are + * controlled from here. + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/err.h> +#include <linux/aer.h> +#include <linux/string.h> +#include <linux/sched.h> +#include <linux/wait.h> +#include <linux/delay.h> +#include <linux/dma-mapping.h> +#include <linux/module.h> +#include <linux/notifier.h> +#include <linux/device.h> +#include <linux/log2.h> +#include <linux/genwqe/genwqe_card.h> + +#include "card_base.h" +#include "card_ddcb.h" + +MODULE_AUTHOR("Frank Haverkamp <haver@linux.vnet.ibm.com>"); +MODULE_AUTHOR("Michael Ruettger <michael@ibmra.de>"); +MODULE_AUTHOR("Joerg-Stephan Vogt <jsvogt@de.ibm.com>"); +MODULE_AUTHOR("Michal Jung <mijung@de.ibm.com>"); + +MODULE_DESCRIPTION("GenWQE Card"); +MODULE_VERSION(DRV_VERS_STRING); +MODULE_LICENSE("GPL"); + +static char genwqe_driver_name[] = GENWQE_DEVNAME; +static struct class *class_genwqe; +static struct dentry *debugfs_genwqe; +static struct genwqe_dev *genwqe_devices[GENWQE_CARD_NO_MAX]; + +/* PCI structure for identifying device by PCI vendor and device ID */ +static DEFINE_PCI_DEVICE_TABLE(genwqe_device_table) = { + { .vendor = PCI_VENDOR_ID_IBM, + .device = PCI_DEVICE_GENWQE, + .subvendor = PCI_SUBVENDOR_ID_IBM, + .subdevice = PCI_SUBSYSTEM_ID_GENWQE5, + .class = (PCI_CLASSCODE_GENWQE5 << 8), + .class_mask = ~0, + .driver_data = 0 }, + + /* Initial SR-IOV bring-up image */ + { .vendor = PCI_VENDOR_ID_IBM, + .device = PCI_DEVICE_GENWQE, + .subvendor = PCI_SUBVENDOR_ID_IBM_SRIOV, + .subdevice = PCI_SUBSYSTEM_ID_GENWQE5_SRIOV, + .class = (PCI_CLASSCODE_GENWQE5_SRIOV << 8), + .class_mask = ~0, + .driver_data = 0 }, + + { .vendor = PCI_VENDOR_ID_IBM, /* VF Vendor ID */ + .device = 0x0000, /* VF Device ID */ + .subvendor = PCI_SUBVENDOR_ID_IBM_SRIOV, + .subdevice = PCI_SUBSYSTEM_ID_GENWQE5_SRIOV, + .class = (PCI_CLASSCODE_GENWQE5_SRIOV << 8), + .class_mask = ~0, + .driver_data = 0 }, + + /* Fixed up image */ + { .vendor = PCI_VENDOR_ID_IBM, + .device = PCI_DEVICE_GENWQE, + .subvendor = PCI_SUBVENDOR_ID_IBM_SRIOV, + .subdevice = PCI_SUBSYSTEM_ID_GENWQE5, + .class = (PCI_CLASSCODE_GENWQE5_SRIOV << 8), + .class_mask = ~0, + .driver_data = 0 }, + + { .vendor = PCI_VENDOR_ID_IBM, /* VF Vendor ID */ + .device = 0x0000, /* VF Device ID */ + .subvendor = PCI_SUBVENDOR_ID_IBM_SRIOV, + .subdevice = PCI_SUBSYSTEM_ID_GENWQE5, + .class = (PCI_CLASSCODE_GENWQE5_SRIOV << 8), + .class_mask = ~0, + .driver_data = 0 }, + + /* Even one more ... */ + { .vendor = PCI_VENDOR_ID_IBM, + .device = PCI_DEVICE_GENWQE, + .subvendor = PCI_SUBVENDOR_ID_IBM, + .subdevice = PCI_SUBSYSTEM_ID_GENWQE5_NEW, + .class = (PCI_CLASSCODE_GENWQE5 << 8), + .class_mask = ~0, + .driver_data = 0 }, + + { 0, } /* 0 terminated list. */ +}; + +MODULE_DEVICE_TABLE(pci, genwqe_device_table); + +/** + * genwqe_dev_alloc() - Create and prepare a new card descriptor + * + * Return: Pointer to card descriptor, or ERR_PTR(err) on error + */ +static struct genwqe_dev *genwqe_dev_alloc(void) +{ + unsigned int i = 0, j; + struct genwqe_dev *cd; + + for (i = 0; i < GENWQE_CARD_NO_MAX; i++) { + if (genwqe_devices[i] == NULL) + break; + } + if (i >= GENWQE_CARD_NO_MAX) + return ERR_PTR(-ENODEV); + + cd = kzalloc(sizeof(struct genwqe_dev), GFP_KERNEL); + if (!cd) + return ERR_PTR(-ENOMEM); + + cd->card_idx = i; + cd->class_genwqe = class_genwqe; + cd->debugfs_genwqe = debugfs_genwqe; + + init_waitqueue_head(&cd->queue_waitq); + + spin_lock_init(&cd->file_lock); + INIT_LIST_HEAD(&cd->file_list); + + cd->card_state = GENWQE_CARD_UNUSED; + spin_lock_init(&cd->print_lock); + + cd->ddcb_software_timeout = genwqe_ddcb_software_timeout; + cd->kill_timeout = genwqe_kill_timeout; + + for (j = 0; j < GENWQE_MAX_VFS; j++) + cd->vf_jobtimeout_msec[j] = genwqe_vf_jobtimeout_msec; + + genwqe_devices[i] = cd; + return cd; +} + +static void genwqe_dev_free(struct genwqe_dev *cd) +{ + if (!cd) + return; + + genwqe_devices[cd->card_idx] = NULL; + kfree(cd); +} + +/** + * genwqe_bus_reset() - Card recovery + * + * pci_reset_function() will recover the device and ensure that the + * registers are accessible again when it completes with success. If + * not, the card will stay dead and registers will be unaccessible + * still. + */ +static int genwqe_bus_reset(struct genwqe_dev *cd) +{ + int bars, rc = 0; + struct pci_dev *pci_dev = cd->pci_dev; + void __iomem *mmio; + + if (cd->err_inject & GENWQE_INJECT_BUS_RESET_FAILURE) + return -EIO; + + mmio = cd->mmio; + cd->mmio = NULL; + pci_iounmap(pci_dev, mmio); + + bars = pci_select_bars(pci_dev, IORESOURCE_MEM); + pci_release_selected_regions(pci_dev, bars); + + /* + * Firmware/BIOS might change memory mapping during bus reset. + * Settings like enable bus-mastering, ... are backuped and + * restored by the pci_reset_function(). + */ + dev_dbg(&pci_dev->dev, "[%s] pci_reset function ...\n", __func__); + rc = pci_reset_function(pci_dev); + if (rc) { + dev_err(&pci_dev->dev, + "[%s] err: failed reset func (rc %d)\n", __func__, rc); + return rc; + } + dev_dbg(&pci_dev->dev, "[%s] done with rc=%d\n", __func__, rc); + + /* + * Here is the right spot to clear the register read + * failure. pci_bus_reset() does this job in real systems. + */ + cd->err_inject &= ~(GENWQE_INJECT_HARDWARE_FAILURE | + GENWQE_INJECT_GFIR_FATAL | + GENWQE_INJECT_GFIR_INFO); + + rc = pci_request_selected_regions(pci_dev, bars, genwqe_driver_name); + if (rc) { + dev_err(&pci_dev->dev, + "[%s] err: request bars failed (%d)\n", __func__, rc); + return -EIO; + } + + cd->mmio = pci_iomap(pci_dev, 0, 0); + if (cd->mmio == NULL) { + dev_err(&pci_dev->dev, + "[%s] err: mapping BAR0 failed\n", __func__); + return -ENOMEM; + } + return 0; +} + +/* + * Hardware circumvention section. Certain bitstreams in our test-lab + * had different kinds of problems. Here is where we adjust those + * bitstreams to function will with this version of our device driver. + * + * Thise circumventions are applied to the physical function only. + * The magical numbers below are identifying development/manufacturing + * versions of the bitstream used on the card. + * + * Turn off error reporting for old/manufacturing images. + */ + +bool genwqe_need_err_masking(struct genwqe_dev *cd) +{ + return (cd->slu_unitcfg & 0xFFFF0ull) < 0x32170ull; +} + +static void genwqe_tweak_hardware(struct genwqe_dev *cd) +{ + struct pci_dev *pci_dev = cd->pci_dev; + + /* Mask FIRs for development images */ + if (((cd->slu_unitcfg & 0xFFFF0ull) >= 0x32000ull) && + ((cd->slu_unitcfg & 0xFFFF0ull) <= 0x33250ull)) { + dev_warn(&pci_dev->dev, + "FIRs masked due to bitstream %016llx.%016llx\n", + cd->slu_unitcfg, cd->app_unitcfg); + + __genwqe_writeq(cd, IO_APP_SEC_LEM_DEBUG_OVR, + 0xFFFFFFFFFFFFFFFFull); + + __genwqe_writeq(cd, IO_APP_ERR_ACT_MASK, + 0x0000000000000000ull); + } +} + +/** + * genwqe_recovery_on_fatal_gfir_required() - Version depended actions + * + * Bitstreams older than 2013-02-17 have a bug where fatal GFIRs must + * be ignored. This is e.g. true for the bitstream we gave to the card + * manufacturer, but also for some old bitstreams we released to our + * test-lab. + */ +int genwqe_recovery_on_fatal_gfir_required(struct genwqe_dev *cd) +{ + return (cd->slu_unitcfg & 0xFFFF0ull) >= 0x32170ull; +} + +int genwqe_flash_readback_fails(struct genwqe_dev *cd) +{ + return (cd->slu_unitcfg & 0xFFFF0ull) < 0x32170ull; +} + +/** + * genwqe_T_psec() - Calculate PF/VF timeout register content + * + * Note: From a design perspective it turned out to be a bad idea to + * use codes here to specifiy the frequency/speed values. An old + * driver cannot understand new codes and is therefore always a + * problem. Better is to measure out the value or put the + * speed/frequency directly into a register which is always a valid + * value for old as well as for new software. + */ +/* T = 1/f */ +static int genwqe_T_psec(struct genwqe_dev *cd) +{ + u16 speed; /* 1/f -> 250, 200, 166, 175 */ + static const int T[] = { 4000, 5000, 6000, 5714 }; + + speed = (u16)((cd->slu_unitcfg >> 28) & 0x0full); + if (speed >= ARRAY_SIZE(T)) + return -1; /* illegal value */ + + return T[speed]; +} + +/** + * genwqe_setup_pf_jtimer() - Setup PF hardware timeouts for DDCB execution + * + * Do this _after_ card_reset() is called. Otherwise the values will + * vanish. The settings need to be done when the queues are inactive. + * + * The max. timeout value is 2^(10+x) * T (6ns for 166MHz) * 15/16. + * The min. timeout value is 2^(10+x) * T (6ns for 166MHz) * 14/16. + */ +static bool genwqe_setup_pf_jtimer(struct genwqe_dev *cd) +{ + u32 T = genwqe_T_psec(cd); + u64 x; + + if (genwqe_pf_jobtimeout_msec == 0) + return false; + + /* PF: large value needed, flash update 2sec per block */ + x = ilog2(genwqe_pf_jobtimeout_msec * + 16000000000uL/(T * 15)) - 10; + + genwqe_write_vreg(cd, IO_SLC_VF_APPJOB_TIMEOUT, + 0xff00 | (x & 0xff), 0); + return true; +} + +/** + * genwqe_setup_vf_jtimer() - Setup VF hardware timeouts for DDCB execution + */ +static bool genwqe_setup_vf_jtimer(struct genwqe_dev *cd) +{ + struct pci_dev *pci_dev = cd->pci_dev; + unsigned int vf; + u32 T = genwqe_T_psec(cd); + u64 x; + + for (vf = 0; vf < pci_sriov_get_totalvfs(pci_dev); vf++) { + + if (cd->vf_jobtimeout_msec[vf] == 0) + continue; + + x = ilog2(cd->vf_jobtimeout_msec[vf] * + 16000000000uL/(T * 15)) - 10; + + genwqe_write_vreg(cd, IO_SLC_VF_APPJOB_TIMEOUT, + 0xff00 | (x & 0xff), vf + 1); + } + return true; +} + +static int genwqe_ffdc_buffs_alloc(struct genwqe_dev *cd) +{ + unsigned int type, e = 0; + + for (type = 0; type < GENWQE_DBG_UNITS; type++) { + switch (type) { + case GENWQE_DBG_UNIT0: + e = genwqe_ffdc_buff_size(cd, 0); + break; + case GENWQE_DBG_UNIT1: + e = genwqe_ffdc_buff_size(cd, 1); + break; + case GENWQE_DBG_UNIT2: + e = genwqe_ffdc_buff_size(cd, 2); + break; + case GENWQE_DBG_REGS: + e = GENWQE_FFDC_REGS; + break; + } + + /* currently support only the debug units mentioned here */ + cd->ffdc[type].entries = e; + cd->ffdc[type].regs = kmalloc(e * sizeof(struct genwqe_reg), + GFP_KERNEL); + /* + * regs == NULL is ok, the using code treats this as no regs, + * Printing warning is ok in this case. + */ + } + return 0; +} + +static void genwqe_ffdc_buffs_free(struct genwqe_dev *cd) +{ + unsigned int type; + + for (type = 0; type < GENWQE_DBG_UNITS; type++) { + kfree(cd->ffdc[type].regs); + cd->ffdc[type].regs = NULL; + } +} + +static int genwqe_read_ids(struct genwqe_dev *cd) +{ + int err = 0; + int slu_id; + struct pci_dev *pci_dev = cd->pci_dev; + + cd->slu_unitcfg = __genwqe_readq(cd, IO_SLU_UNITCFG); + if (cd->slu_unitcfg == IO_ILLEGAL_VALUE) { + dev_err(&pci_dev->dev, + "err: SLUID=%016llx\n", cd->slu_unitcfg); + err = -EIO; + goto out_err; + } + + slu_id = genwqe_get_slu_id(cd); + if (slu_id < GENWQE_SLU_ARCH_REQ || slu_id == 0xff) { + dev_err(&pci_dev->dev, + "err: incompatible SLU Architecture %u\n", slu_id); + err = -ENOENT; + goto out_err; + } + + cd->app_unitcfg = __genwqe_readq(cd, IO_APP_UNITCFG); + if (cd->app_unitcfg == IO_ILLEGAL_VALUE) { + dev_err(&pci_dev->dev, + "err: APPID=%016llx\n", cd->app_unitcfg); + err = -EIO; + goto out_err; + } + genwqe_read_app_id(cd, cd->app_name, sizeof(cd->app_name)); + + /* + * Is access to all registers possible? If we are a VF the + * answer is obvious. If we run fully virtualized, we need to + * check if we can access all registers. If we do not have + * full access we will cause an UR and some informational FIRs + * in the PF, but that should not harm. + */ + if (pci_dev->is_virtfn) + cd->is_privileged = 0; + else + cd->is_privileged = (__genwqe_readq(cd, IO_SLU_BITSTREAM) + != IO_ILLEGAL_VALUE); + + out_err: + return err; +} + +static int genwqe_start(struct genwqe_dev *cd) +{ + int err; + struct pci_dev *pci_dev = cd->pci_dev; + + err = genwqe_read_ids(cd); + if (err) + return err; + + if (genwqe_is_privileged(cd)) { + /* do this after the tweaks. alloc fail is acceptable */ + genwqe_ffdc_buffs_alloc(cd); + genwqe_stop_traps(cd); + + /* Collect registers e.g. FIRs, UNITIDs, traces ... */ + genwqe_read_ffdc_regs(cd, cd->ffdc[GENWQE_DBG_REGS].regs, + cd->ffdc[GENWQE_DBG_REGS].entries, 0); + + genwqe_ffdc_buff_read(cd, GENWQE_DBG_UNIT0, + cd->ffdc[GENWQE_DBG_UNIT0].regs, + cd->ffdc[GENWQE_DBG_UNIT0].entries); + + genwqe_ffdc_buff_read(cd, GENWQE_DBG_UNIT1, + cd->ffdc[GENWQE_DBG_UNIT1].regs, + cd->ffdc[GENWQE_DBG_UNIT1].entries); + + genwqe_ffdc_buff_read(cd, GENWQE_DBG_UNIT2, + cd->ffdc[GENWQE_DBG_UNIT2].regs, + cd->ffdc[GENWQE_DBG_UNIT2].entries); + + genwqe_start_traps(cd); + + if (cd->card_state == GENWQE_CARD_FATAL_ERROR) { + dev_warn(&pci_dev->dev, + "[%s] chip reload/recovery!\n", __func__); + + /* + * Stealth Mode: Reload chip on either hot + * reset or PERST. + */ + cd->softreset = 0x7Cull; + __genwqe_writeq(cd, IO_SLC_CFGREG_SOFTRESET, + cd->softreset); + + err = genwqe_bus_reset(cd); + if (err != 0) { + dev_err(&pci_dev->dev, + "[%s] err: bus reset failed!\n", + __func__); + goto out; + } + + /* + * Re-read the IDs because + * it could happen that the bitstream load + * failed! + */ + err = genwqe_read_ids(cd); + if (err) + goto out; + } + } + + err = genwqe_setup_service_layer(cd); /* does a reset to the card */ + if (err != 0) { + dev_err(&pci_dev->dev, + "[%s] err: could not setup servicelayer!\n", __func__); + err = -ENODEV; + goto out; + } + + if (genwqe_is_privileged(cd)) { /* code is running _after_ reset */ + genwqe_tweak_hardware(cd); + + genwqe_setup_pf_jtimer(cd); + genwqe_setup_vf_jtimer(cd); + } + + err = genwqe_device_create(cd); + if (err < 0) { + dev_err(&pci_dev->dev, + "err: chdev init failed! (err=%d)\n", err); + goto out_release_service_layer; + } + return 0; + + out_release_service_layer: + genwqe_release_service_layer(cd); + out: + if (genwqe_is_privileged(cd)) + genwqe_ffdc_buffs_free(cd); + return -EIO; +} + +/** + * genwqe_stop() - Stop card operation + * + * Recovery notes: + * As long as genwqe_thread runs we might access registers during + * error data capture. Same is with the genwqe_health_thread. + * When genwqe_bus_reset() fails this function might called two times: + * first by the genwqe_health_thread() and later by genwqe_remove() to + * unbind the device. We must be able to survive that. + * + * This function must be robust enough to be called twice. + */ +static int genwqe_stop(struct genwqe_dev *cd) +{ + genwqe_finish_queue(cd); /* no register access */ + genwqe_device_remove(cd); /* device removed, procs killed */ + genwqe_release_service_layer(cd); /* here genwqe_thread is stopped */ + + if (genwqe_is_privileged(cd)) { + pci_disable_sriov(cd->pci_dev); /* access pci config space */ + genwqe_ffdc_buffs_free(cd); + } + + return 0; +} + +/** + * genwqe_recover_card() - Try to recover the card if it is possible + * + * If fatal_err is set no register access is possible anymore. It is + * likely that genwqe_start fails in that situation. Proper error + * handling is required in this case. + * + * genwqe_bus_reset() will cause the pci code to call genwqe_remove() + * and later genwqe_probe() for all virtual functions. + */ +static int genwqe_recover_card(struct genwqe_dev *cd, int fatal_err) +{ + int rc; + struct pci_dev *pci_dev = cd->pci_dev; + + genwqe_stop(cd); + + /* + * Make sure chip is not reloaded to maintain FFDC. Write SLU + * Reset Register, CPLDReset field to 0. + */ + if (!fatal_err) { + cd->softreset = 0x70ull; + __genwqe_writeq(cd, IO_SLC_CFGREG_SOFTRESET, cd->softreset); + } + + rc = genwqe_bus_reset(cd); + if (rc != 0) { + dev_err(&pci_dev->dev, + "[%s] err: card recovery impossible!\n", __func__); + return rc; + } + + rc = genwqe_start(cd); + if (rc < 0) { + dev_err(&pci_dev->dev, + "[%s] err: failed to launch device!\n", __func__); + return rc; + } + return 0; +} + +static int genwqe_health_check_cond(struct genwqe_dev *cd, u64 *gfir) +{ + *gfir = __genwqe_readq(cd, IO_SLC_CFGREG_GFIR); + return (*gfir & GFIR_ERR_TRIGGER) && + genwqe_recovery_on_fatal_gfir_required(cd); +} + +/** + * genwqe_fir_checking() - Check the fault isolation registers of the card + * + * If this code works ok, can be tried out with help of the genwqe_poke tool: + * sudo ./tools/genwqe_poke 0x8 0xfefefefefef + * + * Now the relevant FIRs/sFIRs should be printed out and the driver should + * invoke recovery (devices are removed and readded). + */ +static u64 genwqe_fir_checking(struct genwqe_dev *cd) +{ + int j, iterations = 0; + u64 mask, fir, fec, uid, gfir, gfir_masked, sfir, sfec; + u32 fir_addr, fir_clr_addr, fec_addr, sfir_addr, sfec_addr; + struct pci_dev *pci_dev = cd->pci_dev; + + healthMonitor: + iterations++; + if (iterations > 16) { + dev_err(&pci_dev->dev, "* exit looping after %d times\n", + iterations); + goto fatal_error; + } + + gfir = __genwqe_readq(cd, IO_SLC_CFGREG_GFIR); + if (gfir != 0x0) + dev_err(&pci_dev->dev, "* 0x%08x 0x%016llx\n", + IO_SLC_CFGREG_GFIR, gfir); + if (gfir == IO_ILLEGAL_VALUE) + goto fatal_error; + + /* + * Avoid printing when to GFIR bit is on prevents contignous + * printout e.g. for the following bug: + * FIR set without a 2ndary FIR/FIR cannot be cleared + * Comment out the following if to get the prints: + */ + if (gfir == 0) + return 0; + + gfir_masked = gfir & GFIR_ERR_TRIGGER; /* fatal errors */ + + for (uid = 0; uid < GENWQE_MAX_UNITS; uid++) { /* 0..2 in zEDC */ + + /* read the primary FIR (pfir) */ + fir_addr = (uid << 24) + 0x08; + fir = __genwqe_readq(cd, fir_addr); + if (fir == 0x0) + continue; /* no error in this unit */ + + dev_err(&pci_dev->dev, "* 0x%08x 0x%016llx\n", fir_addr, fir); + if (fir == IO_ILLEGAL_VALUE) + goto fatal_error; + + /* read primary FEC */ + fec_addr = (uid << 24) + 0x18; + fec = __genwqe_readq(cd, fec_addr); + + dev_err(&pci_dev->dev, "* 0x%08x 0x%016llx\n", fec_addr, fec); + if (fec == IO_ILLEGAL_VALUE) + goto fatal_error; + + for (j = 0, mask = 1ULL; j < 64; j++, mask <<= 1) { + + /* secondary fir empty, skip it */ + if ((fir & mask) == 0x0) + continue; + + sfir_addr = (uid << 24) + 0x100 + 0x08 * j; + sfir = __genwqe_readq(cd, sfir_addr); + + if (sfir == IO_ILLEGAL_VALUE) + goto fatal_error; + dev_err(&pci_dev->dev, + "* 0x%08x 0x%016llx\n", sfir_addr, sfir); + + sfec_addr = (uid << 24) + 0x300 + 0x08 * j; + sfec = __genwqe_readq(cd, sfec_addr); + + if (sfec == IO_ILLEGAL_VALUE) + goto fatal_error; + dev_err(&pci_dev->dev, + "* 0x%08x 0x%016llx\n", sfec_addr, sfec); + + gfir = __genwqe_readq(cd, IO_SLC_CFGREG_GFIR); + if (gfir == IO_ILLEGAL_VALUE) + goto fatal_error; + + /* gfir turned on during routine! get out and + start over. */ + if ((gfir_masked == 0x0) && + (gfir & GFIR_ERR_TRIGGER)) { + goto healthMonitor; + } + + /* do not clear if we entered with a fatal gfir */ + if (gfir_masked == 0x0) { + + /* NEW clear by mask the logged bits */ + sfir_addr = (uid << 24) + 0x100 + 0x08 * j; + __genwqe_writeq(cd, sfir_addr, sfir); + + dev_dbg(&pci_dev->dev, + "[HM] Clearing 2ndary FIR 0x%08x " + "with 0x%016llx\n", sfir_addr, sfir); + + /* + * note, these cannot be error-Firs + * since gfir_masked is 0 after sfir + * was read. Also, it is safe to do + * this write if sfir=0. Still need to + * clear the primary. This just means + * there is no secondary FIR. + */ + + /* clear by mask the logged bit. */ + fir_clr_addr = (uid << 24) + 0x10; + __genwqe_writeq(cd, fir_clr_addr, mask); + + dev_dbg(&pci_dev->dev, + "[HM] Clearing primary FIR 0x%08x " + "with 0x%016llx\n", fir_clr_addr, + mask); + } + } + } + gfir = __genwqe_readq(cd, IO_SLC_CFGREG_GFIR); + if (gfir == IO_ILLEGAL_VALUE) + goto fatal_error; + + if ((gfir_masked == 0x0) && (gfir & GFIR_ERR_TRIGGER)) { + /* + * Check once more that it didn't go on after all the + * FIRS were cleared. + */ + dev_dbg(&pci_dev->dev, "ACK! Another FIR! Recursing %d!\n", + iterations); + goto healthMonitor; + } + return gfir_masked; + + fatal_error: + return IO_ILLEGAL_VALUE; +} + +/** + * genwqe_health_thread() - Health checking thread + * + * This thread is only started for the PF of the card. + * + * This thread monitors the health of the card. A critical situation + * is when we read registers which contain -1 (IO_ILLEGAL_VALUE). In + * this case we need to be recovered from outside. Writing to + * registers will very likely not work either. + * + * This thread must only exit if kthread_should_stop() becomes true. + * + * Condition for the health-thread to trigger: + * a) when a kthread_stop() request comes in or + * b) a critical GFIR occured + * + * Informational GFIRs are checked and potentially printed in + * health_check_interval seconds. + */ +static int genwqe_health_thread(void *data) +{ + int rc, should_stop = 0; + struct genwqe_dev *cd = data; + struct pci_dev *pci_dev = cd->pci_dev; + u64 gfir, gfir_masked, slu_unitcfg, app_unitcfg; + + while (!kthread_should_stop()) { + rc = wait_event_interruptible_timeout(cd->health_waitq, + (genwqe_health_check_cond(cd, &gfir) || + (should_stop = kthread_should_stop())), + genwqe_health_check_interval * HZ); + + if (should_stop) + break; + + if (gfir == IO_ILLEGAL_VALUE) { + dev_err(&pci_dev->dev, + "[%s] GFIR=%016llx\n", __func__, gfir); + goto fatal_error; + } + + slu_unitcfg = __genwqe_readq(cd, IO_SLU_UNITCFG); + if (slu_unitcfg == IO_ILLEGAL_VALUE) { + dev_err(&pci_dev->dev, + "[%s] SLU_UNITCFG=%016llx\n", + __func__, slu_unitcfg); + goto fatal_error; + } + + app_unitcfg = __genwqe_readq(cd, IO_APP_UNITCFG); + if (app_unitcfg == IO_ILLEGAL_VALUE) { + dev_err(&pci_dev->dev, + "[%s] APP_UNITCFG=%016llx\n", + __func__, app_unitcfg); + goto fatal_error; + } + + gfir = __genwqe_readq(cd, IO_SLC_CFGREG_GFIR); + if (gfir == IO_ILLEGAL_VALUE) { + dev_err(&pci_dev->dev, + "[%s] %s: GFIR=%016llx\n", __func__, + (gfir & GFIR_ERR_TRIGGER) ? "err" : "info", + gfir); + goto fatal_error; + } + + gfir_masked = genwqe_fir_checking(cd); + if (gfir_masked == IO_ILLEGAL_VALUE) + goto fatal_error; + + /* + * GFIR ErrorTrigger bits set => reset the card! + * Never do this for old/manufacturing images! + */ + if ((gfir_masked) && !cd->skip_recovery && + genwqe_recovery_on_fatal_gfir_required(cd)) { + + cd->card_state = GENWQE_CARD_FATAL_ERROR; + + rc = genwqe_recover_card(cd, 0); + if (rc < 0) { + /* FIXME Card is unusable and needs unbind! */ + goto fatal_error; + } + } + + cd->last_gfir = gfir; + cond_resched(); + } + + return 0; + + fatal_error: + dev_err(&pci_dev->dev, + "[%s] card unusable. Please trigger unbind!\n", __func__); + + /* Bring down logical devices to inform user space via udev remove. */ + cd->card_state = GENWQE_CARD_FATAL_ERROR; + genwqe_stop(cd); + + /* genwqe_bus_reset failed(). Now wait for genwqe_remove(). */ + while (!kthread_should_stop()) + cond_resched(); + + return -EIO; +} + +static int genwqe_health_check_start(struct genwqe_dev *cd) +{ + int rc; + + if (genwqe_health_check_interval <= 0) + return 0; /* valid for disabling the service */ + + /* moved before request_irq() */ + /* init_waitqueue_head(&cd->health_waitq); */ + + cd->health_thread = kthread_run(genwqe_health_thread, cd, + GENWQE_DEVNAME "%d_health", + cd->card_idx); + if (IS_ERR(cd->health_thread)) { + rc = PTR_ERR(cd->health_thread); + cd->health_thread = NULL; + return rc; + } + return 0; +} + +static int genwqe_health_thread_running(struct genwqe_dev *cd) +{ + return cd->health_thread != NULL; +} + +static int genwqe_health_check_stop(struct genwqe_dev *cd) +{ + int rc; + + if (!genwqe_health_thread_running(cd)) + return -EIO; + + rc = kthread_stop(cd->health_thread); + cd->health_thread = NULL; + return 0; +} + +/** + * genwqe_pci_setup() - Allocate PCIe related resources for our card + */ +static int genwqe_pci_setup(struct genwqe_dev *cd) +{ + int err, bars; + struct pci_dev *pci_dev = cd->pci_dev; + + bars = pci_select_bars(pci_dev, IORESOURCE_MEM); + err = pci_enable_device_mem(pci_dev); + if (err) { + dev_err(&pci_dev->dev, + "err: failed to enable pci memory (err=%d)\n", err); + goto err_out; + } + + /* Reserve PCI I/O and memory resources */ + err = pci_request_selected_regions(pci_dev, bars, genwqe_driver_name); + if (err) { + dev_err(&pci_dev->dev, + "[%s] err: request bars failed (%d)\n", __func__, err); + err = -EIO; + goto err_disable_device; + } + + /* check for 64-bit DMA address supported (DAC) */ + if (!pci_set_dma_mask(pci_dev, DMA_BIT_MASK(64))) { + err = pci_set_consistent_dma_mask(pci_dev, DMA_BIT_MASK(64)); + if (err) { + dev_err(&pci_dev->dev, + "err: DMA64 consistent mask error\n"); + err = -EIO; + goto out_release_resources; + } + /* check for 32-bit DMA address supported (SAC) */ + } else if (!pci_set_dma_mask(pci_dev, DMA_BIT_MASK(32))) { + err = pci_set_consistent_dma_mask(pci_dev, DMA_BIT_MASK(32)); + if (err) { + dev_err(&pci_dev->dev, + "err: DMA32 consistent mask error\n"); + err = -EIO; + goto out_release_resources; + } + } else { + dev_err(&pci_dev->dev, + "err: neither DMA32 nor DMA64 supported\n"); + err = -EIO; + goto out_release_resources; + } + + pci_set_master(pci_dev); + pci_enable_pcie_error_reporting(pci_dev); + + /* request complete BAR-0 space (length = 0) */ + cd->mmio_len = pci_resource_len(pci_dev, 0); + cd->mmio = pci_iomap(pci_dev, 0, 0); + if (cd->mmio == NULL) { + dev_err(&pci_dev->dev, + "[%s] err: mapping BAR0 failed\n", __func__); + err = -ENOMEM; + goto out_release_resources; + } + + cd->num_vfs = pci_sriov_get_totalvfs(pci_dev); + + err = genwqe_read_ids(cd); + if (err) + goto out_iounmap; + + return 0; + + out_iounmap: + pci_iounmap(pci_dev, cd->mmio); + out_release_resources: + pci_release_selected_regions(pci_dev, bars); + err_disable_device: + pci_disable_device(pci_dev); + err_out: + return err; +} + +/** + * genwqe_pci_remove() - Free PCIe related resources for our card + */ +static void genwqe_pci_remove(struct genwqe_dev *cd) +{ + int bars; + struct pci_dev *pci_dev = cd->pci_dev; + + if (cd->mmio) + pci_iounmap(pci_dev, cd->mmio); + + bars = pci_select_bars(pci_dev, IORESOURCE_MEM); + pci_release_selected_regions(pci_dev, bars); + pci_disable_device(pci_dev); +} + +/** + * genwqe_probe() - Device initialization + * @pdev: PCI device information struct + * + * Callable for multiple cards. This function is called on bind. + * + * Return: 0 if succeeded, < 0 when failed + */ +static int genwqe_probe(struct pci_dev *pci_dev, + const struct pci_device_id *id) +{ + int err; + struct genwqe_dev *cd; + + genwqe_init_crc32(); + + cd = genwqe_dev_alloc(); + if (IS_ERR(cd)) { + dev_err(&pci_dev->dev, "err: could not alloc mem (err=%d)!\n", + (int)PTR_ERR(cd)); + return PTR_ERR(cd); + } + + dev_set_drvdata(&pci_dev->dev, cd); + cd->pci_dev = pci_dev; + + err = genwqe_pci_setup(cd); + if (err < 0) { + dev_err(&pci_dev->dev, + "err: problems with PCI setup (err=%d)\n", err); + goto out_free_dev; + } + + err = genwqe_start(cd); + if (err < 0) { + dev_err(&pci_dev->dev, + "err: cannot start card services! (err=%d)\n", err); + goto out_pci_remove; + } + + if (genwqe_is_privileged(cd)) { + err = genwqe_health_check_start(cd); + if (err < 0) { + dev_err(&pci_dev->dev, + "err: cannot start health checking! " + "(err=%d)\n", err); + goto out_stop_services; + } + } + return 0; + + out_stop_services: + genwqe_stop(cd); + out_pci_remove: + genwqe_pci_remove(cd); + out_free_dev: + genwqe_dev_free(cd); + return err; +} + +/** + * genwqe_remove() - Called when device is removed (hot-plugable) + * + * Or when driver is unloaded respecitively when unbind is done. + */ +static void genwqe_remove(struct pci_dev *pci_dev) +{ + struct genwqe_dev *cd = dev_get_drvdata(&pci_dev->dev); + + genwqe_health_check_stop(cd); + + /* + * genwqe_stop() must survive if it is called twice + * sequentially. This happens when the health thread calls it + * and fails on genwqe_bus_reset(). + */ + genwqe_stop(cd); + genwqe_pci_remove(cd); + genwqe_dev_free(cd); +} + +/* + * genwqe_err_error_detected() - Error detection callback + * + * This callback is called by the PCI subsystem whenever a PCI bus + * error is detected. + */ +static pci_ers_result_t genwqe_err_error_detected(struct pci_dev *pci_dev, + enum pci_channel_state state) +{ + struct genwqe_dev *cd; + + dev_err(&pci_dev->dev, "[%s] state=%d\n", __func__, state); + + if (pci_dev == NULL) + return PCI_ERS_RESULT_NEED_RESET; + + cd = dev_get_drvdata(&pci_dev->dev); + if (cd == NULL) + return PCI_ERS_RESULT_NEED_RESET; + + switch (state) { + case pci_channel_io_normal: + return PCI_ERS_RESULT_CAN_RECOVER; + case pci_channel_io_frozen: + return PCI_ERS_RESULT_NEED_RESET; + case pci_channel_io_perm_failure: + return PCI_ERS_RESULT_DISCONNECT; + } + + return PCI_ERS_RESULT_NEED_RESET; +} + +static pci_ers_result_t genwqe_err_result_none(struct pci_dev *dev) +{ + return PCI_ERS_RESULT_NONE; +} + +static void genwqe_err_resume(struct pci_dev *dev) +{ +} + +static int genwqe_sriov_configure(struct pci_dev *dev, int numvfs) +{ + struct genwqe_dev *cd = dev_get_drvdata(&dev->dev); + + if (numvfs > 0) { + genwqe_setup_vf_jtimer(cd); + pci_enable_sriov(dev, numvfs); + return numvfs; + } + if (numvfs == 0) { + pci_disable_sriov(dev); + return 0; + } + return 0; +} + +static struct pci_error_handlers genwqe_err_handler = { + .error_detected = genwqe_err_error_detected, + .mmio_enabled = genwqe_err_result_none, + .link_reset = genwqe_err_result_none, + .slot_reset = genwqe_err_result_none, + .resume = genwqe_err_resume, +}; + +static struct pci_driver genwqe_driver = { + .name = genwqe_driver_name, + .id_table = genwqe_device_table, + .probe = genwqe_probe, + .remove = genwqe_remove, + .sriov_configure = genwqe_sriov_configure, + .err_handler = &genwqe_err_handler, +}; + +/** + * genwqe_init_module() - Driver registration and initialization + */ +static int __init genwqe_init_module(void) +{ + int rc; + + class_genwqe = class_create(THIS_MODULE, GENWQE_DEVNAME); + if (IS_ERR(class_genwqe)) { + pr_err("[%s] create class failed\n", __func__); + return -ENOMEM; + } + + debugfs_genwqe = debugfs_create_dir(GENWQE_DEVNAME, NULL); + if (!debugfs_genwqe) { + rc = -ENOMEM; + goto err_out; + } + + rc = pci_register_driver(&genwqe_driver); + if (rc != 0) { + pr_err("[%s] pci_reg_driver (rc=%d)\n", __func__, rc); + goto err_out0; + } + + return rc; + + err_out0: + debugfs_remove(debugfs_genwqe); + err_out: + class_destroy(class_genwqe); + return rc; +} + +/** + * genwqe_exit_module() - Driver exit + */ +static void __exit genwqe_exit_module(void) +{ + pci_unregister_driver(&genwqe_driver); + debugfs_remove(debugfs_genwqe); + class_destroy(class_genwqe); +} + +module_init(genwqe_init_module); +module_exit(genwqe_exit_module); diff --git a/drivers/misc/genwqe/card_base.h b/drivers/misc/genwqe/card_base.h new file mode 100644 index 00000000000..0e608a28860 --- /dev/null +++ b/drivers/misc/genwqe/card_base.h @@ -0,0 +1,577 @@ +#ifndef __CARD_BASE_H__ +#define __CARD_BASE_H__ + +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +/* + * Interfaces within the GenWQE module. Defines genwqe_card and + * ddcb_queue as well as ddcb_requ. + */ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/cdev.h> +#include <linux/stringify.h> +#include <linux/pci.h> +#include <linux/semaphore.h> +#include <linux/uaccess.h> +#include <linux/io.h> +#include <linux/version.h> +#include <linux/debugfs.h> +#include <linux/slab.h> + +#include <linux/genwqe/genwqe_card.h> +#include "genwqe_driver.h" + +#define GENWQE_MSI_IRQS 4 /* Just one supported, no MSIx */ +#define GENWQE_FLAG_MSI_ENABLED (1 << 0) + +#define GENWQE_MAX_VFS 15 /* maximum 15 VFs are possible */ +#define GENWQE_MAX_FUNCS 16 /* 1 PF and 15 VFs */ +#define GENWQE_CARD_NO_MAX (16 * GENWQE_MAX_FUNCS) + +/* Compile parameters, some of them appear in debugfs for later adjustment */ +#define genwqe_ddcb_max 32 /* DDCBs on the work-queue */ +#define genwqe_polling_enabled 0 /* in case of irqs not working */ +#define genwqe_ddcb_software_timeout 10 /* timeout per DDCB in seconds */ +#define genwqe_kill_timeout 8 /* time until process gets killed */ +#define genwqe_vf_jobtimeout_msec 250 /* 250 msec */ +#define genwqe_pf_jobtimeout_msec 8000 /* 8 sec should be ok */ +#define genwqe_health_check_interval 4 /* <= 0: disabled */ + +/* Sysfs attribute groups used when we create the genwqe device */ +extern const struct attribute_group *genwqe_attribute_groups[]; + +/* + * Config space for Genwqe5 A7: + * 00:[14 10 4b 04]40 00 10 00[00 00 00 12]00 00 00 00 + * 10: 0c 00 00 f0 07 3c 00 00 00 00 00 00 00 00 00 00 + * 20: 00 00 00 00 00 00 00 00 00 00 00 00[14 10 4b 04] + * 30: 00 00 00 00 50 00 00 00 00 00 00 00 00 00 00 00 + */ +#define PCI_DEVICE_GENWQE 0x044b /* Genwqe DeviceID */ + +#define PCI_SUBSYSTEM_ID_GENWQE5 0x035f /* Genwqe A5 Subsystem-ID */ +#define PCI_SUBSYSTEM_ID_GENWQE5_NEW 0x044b /* Genwqe A5 Subsystem-ID */ +#define PCI_CLASSCODE_GENWQE5 0x1200 /* UNKNOWN */ + +#define PCI_SUBVENDOR_ID_IBM_SRIOV 0x0000 +#define PCI_SUBSYSTEM_ID_GENWQE5_SRIOV 0x0000 /* Genwqe A5 Subsystem-ID */ +#define PCI_CLASSCODE_GENWQE5_SRIOV 0x1200 /* UNKNOWN */ + +#define GENWQE_SLU_ARCH_REQ 2 /* Required SLU architecture level */ + +/** + * struct genwqe_reg - Genwqe data dump functionality + */ +struct genwqe_reg { + u32 addr; + u32 idx; + u64 val; +}; + +/* + * enum genwqe_dbg_type - Specify chip unit to dump/debug + */ +enum genwqe_dbg_type { + GENWQE_DBG_UNIT0 = 0, /* captured before prev errs cleared */ + GENWQE_DBG_UNIT1 = 1, + GENWQE_DBG_UNIT2 = 2, + GENWQE_DBG_UNIT3 = 3, + GENWQE_DBG_UNIT4 = 4, + GENWQE_DBG_UNIT5 = 5, + GENWQE_DBG_UNIT6 = 6, + GENWQE_DBG_UNIT7 = 7, + GENWQE_DBG_REGS = 8, + GENWQE_DBG_DMA = 9, + GENWQE_DBG_UNITS = 10, /* max number of possible debug units */ +}; + +/* Software error injection to simulate card failures */ +#define GENWQE_INJECT_HARDWARE_FAILURE 0x00000001 /* injects -1 reg reads */ +#define GENWQE_INJECT_BUS_RESET_FAILURE 0x00000002 /* pci_bus_reset fail */ +#define GENWQE_INJECT_GFIR_FATAL 0x00000004 /* GFIR = 0x0000ffff */ +#define GENWQE_INJECT_GFIR_INFO 0x00000008 /* GFIR = 0xffff0000 */ + +/* + * Genwqe card description and management data. + * + * Error-handling in case of card malfunction + * ------------------------------------------ + * + * If the card is detected to be defective the outside environment + * will cause the PCI layer to call deinit (the cleanup function for + * probe). This is the same effect like doing a unbind/bind operation + * on the card. + * + * The genwqe card driver implements a health checking thread which + * verifies the card function. If this detects a problem the cards + * device is being shutdown and restarted again, along with a reset of + * the card and queue. + * + * All functions accessing the card device return either -EIO or -ENODEV + * code to indicate the malfunction to the user. The user has to close + * the file descriptor and open a new one, once the card becomes + * available again. + * + * If the open file descriptor is setup to receive SIGIO, the signal is + * genereated for the application which has to provide a handler to + * react on it. If the application does not close the open + * file descriptor a SIGKILL is send to enforce freeing the cards + * resources. + * + * I did not find a different way to prevent kernel problems due to + * reference counters for the cards character devices getting out of + * sync. The character device deallocation does not block, even if + * there is still an open file descriptor pending. If this pending + * descriptor is closed, the data structures used by the character + * device is reinstantiated, which will lead to the reference counter + * dropping below the allowed values. + * + * Card recovery + * ------------- + * + * To test the internal driver recovery the following command can be used: + * sudo sh -c 'echo 0xfffff > /sys/class/genwqe/genwqe0_card/err_inject' + */ + + +/** + * struct dma_mapping_type - Mapping type definition + * + * To avoid memcpying data arround we use user memory directly. To do + * this we need to pin/swap-in the memory and request a DMA address + * for it. + */ +enum dma_mapping_type { + GENWQE_MAPPING_RAW = 0, /* contignous memory buffer */ + GENWQE_MAPPING_SGL_TEMP, /* sglist dynamically used */ + GENWQE_MAPPING_SGL_PINNED, /* sglist used with pinning */ +}; + +/** + * struct dma_mapping - Information about memory mappings done by the driver + */ +struct dma_mapping { + enum dma_mapping_type type; + + void *u_vaddr; /* user-space vaddr/non-aligned */ + void *k_vaddr; /* kernel-space vaddr/non-aligned */ + dma_addr_t dma_addr; /* physical DMA address */ + + struct page **page_list; /* list of pages used by user buff */ + dma_addr_t *dma_list; /* list of dma addresses per page */ + unsigned int nr_pages; /* number of pages */ + unsigned int size; /* size in bytes */ + + struct list_head card_list; /* list of usr_maps for card */ + struct list_head pin_list; /* list of pinned memory for dev */ +}; + +static inline void genwqe_mapping_init(struct dma_mapping *m, + enum dma_mapping_type type) +{ + memset(m, 0, sizeof(*m)); + m->type = type; +} + +/** + * struct ddcb_queue - DDCB queue data + * @ddcb_max: Number of DDCBs on the queue + * @ddcb_next: Next free DDCB + * @ddcb_act: Next DDCB supposed to finish + * @ddcb_seq: Sequence number of last DDCB + * @ddcbs_in_flight: Currently enqueued DDCBs + * @ddcbs_completed: Number of already completed DDCBs + * @busy: Number of -EBUSY returns + * @ddcb_daddr: DMA address of first DDCB in the queue + * @ddcb_vaddr: Kernel virtual address of first DDCB in the queue + * @ddcb_req: Associated requests (one per DDCB) + * @ddcb_waitqs: Associated wait queues (one per DDCB) + * @ddcb_lock: Lock to protect queuing operations + * @ddcb_waitq: Wait on next DDCB finishing + */ + +struct ddcb_queue { + int ddcb_max; /* amount of DDCBs */ + int ddcb_next; /* next available DDCB num */ + int ddcb_act; /* DDCB to be processed */ + u16 ddcb_seq; /* slc seq num */ + unsigned int ddcbs_in_flight; /* number of ddcbs in processing */ + unsigned int ddcbs_completed; + unsigned int ddcbs_max_in_flight; + unsigned int busy; /* how many times -EBUSY? */ + + dma_addr_t ddcb_daddr; /* DMA address */ + struct ddcb *ddcb_vaddr; /* kernel virtual addr for DDCBs */ + struct ddcb_requ **ddcb_req; /* ddcb processing parameter */ + wait_queue_head_t *ddcb_waitqs; /* waitqueue per ddcb */ + + spinlock_t ddcb_lock; /* exclusive access to queue */ + wait_queue_head_t ddcb_waitq; /* wait for ddcb processing */ + + /* registers or the respective queue to be used */ + u32 IO_QUEUE_CONFIG; + u32 IO_QUEUE_STATUS; + u32 IO_QUEUE_SEGMENT; + u32 IO_QUEUE_INITSQN; + u32 IO_QUEUE_WRAP; + u32 IO_QUEUE_OFFSET; + u32 IO_QUEUE_WTIME; + u32 IO_QUEUE_ERRCNTS; + u32 IO_QUEUE_LRW; +}; + +/* + * GFIR, SLU_UNITCFG, APP_UNITCFG + * 8 Units with FIR/FEC + 64 * 2ndary FIRS/FEC. + */ +#define GENWQE_FFDC_REGS (3 + (8 * (2 + 2 * 64))) + +struct genwqe_ffdc { + unsigned int entries; + struct genwqe_reg *regs; +}; + +/** + * struct genwqe_dev - GenWQE device information + * @card_state: Card operation state, see above + * @ffdc: First Failure Data Capture buffers for each unit + * @card_thread: Working thread to operate the DDCB queue + * @card_waitq: Wait queue used in card_thread + * @queue: DDCB queue + * @health_thread: Card monitoring thread (only for PFs) + * @health_waitq: Wait queue used in health_thread + * @pci_dev: Associated PCI device (function) + * @mmio: Base address of 64-bit register space + * @mmio_len: Length of register area + * @file_lock: Lock to protect access to file_list + * @file_list: List of all processes with open GenWQE file descriptors + * + * This struct contains all information needed to communicate with a + * GenWQE card. It is initialized when a GenWQE device is found and + * destroyed when it goes away. It holds data to maintain the queue as + * well as data needed to feed the user interfaces. + */ +struct genwqe_dev { + enum genwqe_card_state card_state; + spinlock_t print_lock; + + int card_idx; /* card index 0..CARD_NO_MAX-1 */ + u64 flags; /* general flags */ + + /* FFDC data gathering */ + struct genwqe_ffdc ffdc[GENWQE_DBG_UNITS]; + + /* DDCB workqueue */ + struct task_struct *card_thread; + wait_queue_head_t queue_waitq; + struct ddcb_queue queue; /* genwqe DDCB queue */ + unsigned int irqs_processed; + + /* Card health checking thread */ + struct task_struct *health_thread; + wait_queue_head_t health_waitq; + + /* char device */ + dev_t devnum_genwqe; /* major/minor num card */ + struct class *class_genwqe; /* reference to class object */ + struct device *dev; /* for device creation */ + struct cdev cdev_genwqe; /* char device for card */ + + struct dentry *debugfs_root; /* debugfs card root directory */ + struct dentry *debugfs_genwqe; /* debugfs driver root directory */ + + /* pci resources */ + struct pci_dev *pci_dev; /* PCI device */ + void __iomem *mmio; /* BAR-0 MMIO start */ + unsigned long mmio_len; + u16 num_vfs; + u32 vf_jobtimeout_msec[GENWQE_MAX_VFS]; + int is_privileged; /* access to all regs possible */ + + /* config regs which we need often */ + u64 slu_unitcfg; + u64 app_unitcfg; + u64 softreset; + u64 err_inject; + u64 last_gfir; + char app_name[5]; + + spinlock_t file_lock; /* lock for open files */ + struct list_head file_list; /* list of open files */ + + /* debugfs parameters */ + int ddcb_software_timeout; /* wait until DDCB times out */ + int skip_recovery; /* circumvention if recovery fails */ + int kill_timeout; /* wait after sending SIGKILL */ +}; + +/** + * enum genwqe_requ_state - State of a DDCB execution request + */ +enum genwqe_requ_state { + GENWQE_REQU_NEW = 0, + GENWQE_REQU_ENQUEUED = 1, + GENWQE_REQU_TAPPED = 2, + GENWQE_REQU_FINISHED = 3, + GENWQE_REQU_STATE_MAX, +}; + +/** + * struct genwqe_sgl - Scatter gather list describing user-space memory + * @sgl: scatter gather list needs to be 128 byte aligned + * @sgl_dma_addr: dma address of sgl + * @sgl_size: size of area used for sgl + * @user_addr: user-space address of memory area + * @user_size: size of user-space memory area + * @page: buffer for partial pages if needed + * @page_dma_addr: dma address partial pages + */ +struct genwqe_sgl { + dma_addr_t sgl_dma_addr; + struct sg_entry *sgl; + size_t sgl_size; /* size of sgl */ + + void __user *user_addr; /* user-space base-address */ + size_t user_size; /* size of memory area */ + + unsigned long nr_pages; + unsigned long fpage_offs; + size_t fpage_size; + size_t lpage_size; + + void *fpage; + dma_addr_t fpage_dma_addr; + + void *lpage; + dma_addr_t lpage_dma_addr; +}; + +int genwqe_alloc_sync_sgl(struct genwqe_dev *cd, struct genwqe_sgl *sgl, + void __user *user_addr, size_t user_size); + +int genwqe_setup_sgl(struct genwqe_dev *cd, struct genwqe_sgl *sgl, + dma_addr_t *dma_list); + +int genwqe_free_sync_sgl(struct genwqe_dev *cd, struct genwqe_sgl *sgl); + +/** + * struct ddcb_requ - Kernel internal representation of the DDCB request + * @cmd: User space representation of the DDCB execution request + */ +struct ddcb_requ { + /* kernel specific content */ + enum genwqe_requ_state req_state; /* request status */ + int num; /* ddcb_no for this request */ + struct ddcb_queue *queue; /* associated queue */ + + struct dma_mapping dma_mappings[DDCB_FIXUPS]; + struct genwqe_sgl sgls[DDCB_FIXUPS]; + + /* kernel/user shared content */ + struct genwqe_ddcb_cmd cmd; /* ddcb_no for this request */ + struct genwqe_debug_data debug_data; +}; + +/** + * struct genwqe_file - Information for open GenWQE devices + */ +struct genwqe_file { + struct genwqe_dev *cd; + struct genwqe_driver *client; + struct file *filp; + + struct fasync_struct *async_queue; + struct task_struct *owner; + struct list_head list; /* entry in list of open files */ + + spinlock_t map_lock; /* lock for dma_mappings */ + struct list_head map_list; /* list of dma_mappings */ + + spinlock_t pin_lock; /* lock for pinned memory */ + struct list_head pin_list; /* list of pinned memory */ +}; + +int genwqe_setup_service_layer(struct genwqe_dev *cd); /* for PF only */ +int genwqe_finish_queue(struct genwqe_dev *cd); +int genwqe_release_service_layer(struct genwqe_dev *cd); + +/** + * genwqe_get_slu_id() - Read Service Layer Unit Id + * Return: 0x00: Development code + * 0x01: SLC1 (old) + * 0x02: SLC2 (sept2012) + * 0x03: SLC2 (feb2013, generic driver) + */ +static inline int genwqe_get_slu_id(struct genwqe_dev *cd) +{ + return (int)((cd->slu_unitcfg >> 32) & 0xff); +} + +int genwqe_ddcbs_in_flight(struct genwqe_dev *cd); + +u8 genwqe_card_type(struct genwqe_dev *cd); +int genwqe_card_reset(struct genwqe_dev *cd); +int genwqe_set_interrupt_capability(struct genwqe_dev *cd, int count); +void genwqe_reset_interrupt_capability(struct genwqe_dev *cd); + +int genwqe_device_create(struct genwqe_dev *cd); +int genwqe_device_remove(struct genwqe_dev *cd); + +/* debugfs */ +int genwqe_init_debugfs(struct genwqe_dev *cd); +void genqwe_exit_debugfs(struct genwqe_dev *cd); + +int genwqe_read_softreset(struct genwqe_dev *cd); + +/* Hardware Circumventions */ +int genwqe_recovery_on_fatal_gfir_required(struct genwqe_dev *cd); +int genwqe_flash_readback_fails(struct genwqe_dev *cd); + +/** + * genwqe_write_vreg() - Write register in VF window + * @cd: genwqe device + * @reg: register address + * @val: value to write + * @func: 0: PF, 1: VF0, ..., 15: VF14 + */ +int genwqe_write_vreg(struct genwqe_dev *cd, u32 reg, u64 val, int func); + +/** + * genwqe_read_vreg() - Read register in VF window + * @cd: genwqe device + * @reg: register address + * @func: 0: PF, 1: VF0, ..., 15: VF14 + * + * Return: content of the register + */ +u64 genwqe_read_vreg(struct genwqe_dev *cd, u32 reg, int func); + +/* FFDC Buffer Management */ +int genwqe_ffdc_buff_size(struct genwqe_dev *cd, int unit_id); +int genwqe_ffdc_buff_read(struct genwqe_dev *cd, int unit_id, + struct genwqe_reg *regs, unsigned int max_regs); +int genwqe_read_ffdc_regs(struct genwqe_dev *cd, struct genwqe_reg *regs, + unsigned int max_regs, int all); +int genwqe_ffdc_dump_dma(struct genwqe_dev *cd, + struct genwqe_reg *regs, unsigned int max_regs); + +int genwqe_init_debug_data(struct genwqe_dev *cd, + struct genwqe_debug_data *d); + +void genwqe_init_crc32(void); +int genwqe_read_app_id(struct genwqe_dev *cd, char *app_name, int len); + +/* Memory allocation/deallocation; dma address handling */ +int genwqe_user_vmap(struct genwqe_dev *cd, struct dma_mapping *m, + void *uaddr, unsigned long size, + struct ddcb_requ *req); + +int genwqe_user_vunmap(struct genwqe_dev *cd, struct dma_mapping *m, + struct ddcb_requ *req); + +static inline bool dma_mapping_used(struct dma_mapping *m) +{ + if (!m) + return 0; + return m->size != 0; +} + +/** + * __genwqe_execute_ddcb() - Execute DDCB request with addr translation + * + * This function will do the address translation changes to the DDCBs + * according to the definitions required by the ATS field. It looks up + * the memory allocation buffer or does vmap/vunmap for the respective + * user-space buffers, inclusive page pinning and scatter gather list + * buildup and teardown. + */ +int __genwqe_execute_ddcb(struct genwqe_dev *cd, + struct genwqe_ddcb_cmd *cmd); + +/** + * __genwqe_execute_raw_ddcb() - Execute DDCB request without addr translation + * + * This version will not do address translation or any modifcation of + * the DDCB data. It is used e.g. for the MoveFlash DDCB which is + * entirely prepared by the driver itself. That means the appropriate + * DMA addresses are already in the DDCB and do not need any + * modification. + */ +int __genwqe_execute_raw_ddcb(struct genwqe_dev *cd, + struct genwqe_ddcb_cmd *cmd); + +int __genwqe_enqueue_ddcb(struct genwqe_dev *cd, struct ddcb_requ *req); +int __genwqe_wait_ddcb(struct genwqe_dev *cd, struct ddcb_requ *req); +int __genwqe_purge_ddcb(struct genwqe_dev *cd, struct ddcb_requ *req); + +/* register access */ +int __genwqe_writeq(struct genwqe_dev *cd, u64 byte_offs, u64 val); +u64 __genwqe_readq(struct genwqe_dev *cd, u64 byte_offs); +int __genwqe_writel(struct genwqe_dev *cd, u64 byte_offs, u32 val); +u32 __genwqe_readl(struct genwqe_dev *cd, u64 byte_offs); + +void *__genwqe_alloc_consistent(struct genwqe_dev *cd, size_t size, + dma_addr_t *dma_handle); +void __genwqe_free_consistent(struct genwqe_dev *cd, size_t size, + void *vaddr, dma_addr_t dma_handle); + +/* Base clock frequency in MHz */ +int genwqe_base_clock_frequency(struct genwqe_dev *cd); + +/* Before FFDC is captured the traps should be stopped. */ +void genwqe_stop_traps(struct genwqe_dev *cd); +void genwqe_start_traps(struct genwqe_dev *cd); + +/* Hardware circumvention */ +bool genwqe_need_err_masking(struct genwqe_dev *cd); + +/** + * genwqe_is_privileged() - Determine operation mode for PCI function + * + * On Intel with SRIOV support we see: + * PF: is_physfn = 1 is_virtfn = 0 + * VF: is_physfn = 0 is_virtfn = 1 + * + * On Systems with no SRIOV support _and_ virtualized systems we get: + * is_physfn = 0 is_virtfn = 0 + * + * Other vendors have individual pci device ids to distinguish between + * virtual function drivers and physical function drivers. GenWQE + * unfortunately has just on pci device id for both, VFs and PF. + * + * The following code is used to distinguish if the card is running in + * privileged mode, either as true PF or in a virtualized system with + * full register access e.g. currently on PowerPC. + * + * if (pci_dev->is_virtfn) + * cd->is_privileged = 0; + * else + * cd->is_privileged = (__genwqe_readq(cd, IO_SLU_BITSTREAM) + * != IO_ILLEGAL_VALUE); + */ +static inline int genwqe_is_privileged(struct genwqe_dev *cd) +{ + return cd->is_privileged; +} + +#endif /* __CARD_BASE_H__ */ diff --git a/drivers/misc/genwqe/card_ddcb.c b/drivers/misc/genwqe/card_ddcb.c new file mode 100644 index 00000000000..c8046db2d5a --- /dev/null +++ b/drivers/misc/genwqe/card_ddcb.c @@ -0,0 +1,1380 @@ +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +/* + * Device Driver Control Block (DDCB) queue support. Definition of + * interrupt handlers for queue support as well as triggering the + * health monitor code in case of problems. The current hardware uses + * an MSI interrupt which is shared between error handling and + * functional code. + */ + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/wait.h> +#include <linux/pci.h> +#include <linux/string.h> +#include <linux/dma-mapping.h> +#include <linux/delay.h> +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/crc-itu-t.h> + +#include "card_base.h" +#include "card_ddcb.h" + +/* + * N: next DDCB, this is where the next DDCB will be put. + * A: active DDCB, this is where the code will look for the next completion. + * x: DDCB is enqueued, we are waiting for its completion. + + * Situation (1): Empty queue + * +---+---+---+---+---+---+---+---+ + * | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | + * | | | | | | | | | + * +---+---+---+---+---+---+---+---+ + * A/N + * enqueued_ddcbs = A - N = 2 - 2 = 0 + * + * Situation (2): Wrapped, N > A + * +---+---+---+---+---+---+---+---+ + * | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | + * | | | x | x | | | | | + * +---+---+---+---+---+---+---+---+ + * A N + * enqueued_ddcbs = N - A = 4 - 2 = 2 + * + * Situation (3): Queue wrapped, A > N + * +---+---+---+---+---+---+---+---+ + * | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | + * | x | x | | | x | x | x | x | + * +---+---+---+---+---+---+---+---+ + * N A + * enqueued_ddcbs = queue_max - (A - N) = 8 - (4 - 2) = 6 + * + * Situation (4a): Queue full N > A + * +---+---+---+---+---+---+---+---+ + * | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | + * | x | x | x | x | x | x | x | | + * +---+---+---+---+---+---+---+---+ + * A N + * + * enqueued_ddcbs = N - A = 7 - 0 = 7 + * + * Situation (4a): Queue full A > N + * +---+---+---+---+---+---+---+---+ + * | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | + * | x | x | x | | x | x | x | x | + * +---+---+---+---+---+---+---+---+ + * N A + * enqueued_ddcbs = queue_max - (A - N) = 8 - (4 - 3) = 7 + */ + +static int queue_empty(struct ddcb_queue *queue) +{ + return queue->ddcb_next == queue->ddcb_act; +} + +static int queue_enqueued_ddcbs(struct ddcb_queue *queue) +{ + if (queue->ddcb_next >= queue->ddcb_act) + return queue->ddcb_next - queue->ddcb_act; + + return queue->ddcb_max - (queue->ddcb_act - queue->ddcb_next); +} + +static int queue_free_ddcbs(struct ddcb_queue *queue) +{ + int free_ddcbs = queue->ddcb_max - queue_enqueued_ddcbs(queue) - 1; + + if (WARN_ON_ONCE(free_ddcbs < 0)) { /* must never ever happen! */ + return 0; + } + return free_ddcbs; +} + +/* + * Use of the PRIV field in the DDCB for queue debugging: + * + * (1) Trying to get rid of a DDCB which saw a timeout: + * pddcb->priv[6] = 0xcc; # cleared + * + * (2) Append a DDCB via NEXT bit: + * pddcb->priv[7] = 0xaa; # appended + * + * (3) DDCB needed tapping: + * pddcb->priv[7] = 0xbb; # tapped + * + * (4) DDCB marked as correctly finished: + * pddcb->priv[6] = 0xff; # finished + */ + +static inline void ddcb_mark_tapped(struct ddcb *pddcb) +{ + pddcb->priv[7] = 0xbb; /* tapped */ +} + +static inline void ddcb_mark_appended(struct ddcb *pddcb) +{ + pddcb->priv[7] = 0xaa; /* appended */ +} + +static inline void ddcb_mark_cleared(struct ddcb *pddcb) +{ + pddcb->priv[6] = 0xcc; /* cleared */ +} + +static inline void ddcb_mark_finished(struct ddcb *pddcb) +{ + pddcb->priv[6] = 0xff; /* finished */ +} + +static inline void ddcb_mark_unused(struct ddcb *pddcb) +{ + pddcb->priv_64 = cpu_to_be64(0); /* not tapped */ +} + +/** + * genwqe_crc16() - Generate 16-bit crc as required for DDCBs + * @buff: pointer to data buffer + * @len: length of data for calculation + * @init: initial crc (0xffff at start) + * + * Polynomial = x^16 + x^12 + x^5 + 1 (0x1021) + * Example: 4 bytes 0x01 0x02 0x03 0x04 with init = 0xffff + * should result in a crc16 of 0x89c3 + * + * Return: crc16 checksum in big endian format ! + */ +static inline u16 genwqe_crc16(const u8 *buff, size_t len, u16 init) +{ + return crc_itu_t(init, buff, len); +} + +static void print_ddcb_info(struct genwqe_dev *cd, struct ddcb_queue *queue) +{ + int i; + struct ddcb *pddcb; + unsigned long flags; + struct pci_dev *pci_dev = cd->pci_dev; + + spin_lock_irqsave(&cd->print_lock, flags); + + dev_info(&pci_dev->dev, + "DDCB list for card #%d (ddcb_act=%d / ddcb_next=%d):\n", + cd->card_idx, queue->ddcb_act, queue->ddcb_next); + + pddcb = queue->ddcb_vaddr; + for (i = 0; i < queue->ddcb_max; i++) { + dev_err(&pci_dev->dev, + " %c %-3d: RETC=%03x SEQ=%04x " + "HSI=%02X SHI=%02x PRIV=%06llx CMD=%03x\n", + i == queue->ddcb_act ? '>' : ' ', + i, + be16_to_cpu(pddcb->retc_16), + be16_to_cpu(pddcb->seqnum_16), + pddcb->hsi, + pddcb->shi, + be64_to_cpu(pddcb->priv_64), + pddcb->cmd); + pddcb++; + } + spin_unlock_irqrestore(&cd->print_lock, flags); +} + +struct genwqe_ddcb_cmd *ddcb_requ_alloc(void) +{ + struct ddcb_requ *req; + + req = kzalloc(sizeof(*req), GFP_ATOMIC); + if (!req) + return NULL; + + return &req->cmd; +} + +void ddcb_requ_free(struct genwqe_ddcb_cmd *cmd) +{ + struct ddcb_requ *req = container_of(cmd, struct ddcb_requ, cmd); + kfree(req); +} + +static inline enum genwqe_requ_state ddcb_requ_get_state(struct ddcb_requ *req) +{ + return req->req_state; +} + +static inline void ddcb_requ_set_state(struct ddcb_requ *req, + enum genwqe_requ_state new_state) +{ + req->req_state = new_state; +} + +static inline int ddcb_requ_collect_debug_data(struct ddcb_requ *req) +{ + return req->cmd.ddata_addr != 0x0; +} + +/** + * ddcb_requ_finished() - Returns the hardware state of the associated DDCB + * @cd: pointer to genwqe device descriptor + * @req: DDCB work request + * + * Status of ddcb_requ mirrors this hardware state, but is copied in + * the ddcb_requ on interrupt/polling function. The lowlevel code + * should check the hardware state directly, the higher level code + * should check the copy. + * + * This function will also return true if the state of the queue is + * not GENWQE_CARD_USED. This enables us to purge all DDCBs in the + * shutdown case. + */ +static int ddcb_requ_finished(struct genwqe_dev *cd, struct ddcb_requ *req) +{ + return (ddcb_requ_get_state(req) == GENWQE_REQU_FINISHED) || + (cd->card_state != GENWQE_CARD_USED); +} + +/** + * enqueue_ddcb() - Enqueue a DDCB + * @cd: pointer to genwqe device descriptor + * @queue: queue this operation should be done on + * @ddcb_no: pointer to ddcb number being tapped + * + * Start execution of DDCB by tapping or append to queue via NEXT + * bit. This is done by an atomic 'compare and swap' instruction and + * checking SHI and HSI of the previous DDCB. + * + * This function must only be called with ddcb_lock held. + * + * Return: 1 if new DDCB is appended to previous + * 2 if DDCB queue is tapped via register/simulation + */ +#define RET_DDCB_APPENDED 1 +#define RET_DDCB_TAPPED 2 + +static int enqueue_ddcb(struct genwqe_dev *cd, struct ddcb_queue *queue, + struct ddcb *pddcb, int ddcb_no) +{ + unsigned int try; + int prev_no; + struct ddcb *prev_ddcb; + __be32 old, new, icrc_hsi_shi; + u64 num; + + /* + * For performance checks a Dispatch Timestamp can be put into + * DDCB It is supposed to use the SLU's free running counter, + * but this requires PCIe cycles. + */ + ddcb_mark_unused(pddcb); + + /* check previous DDCB if already fetched */ + prev_no = (ddcb_no == 0) ? queue->ddcb_max - 1 : ddcb_no - 1; + prev_ddcb = &queue->ddcb_vaddr[prev_no]; + + /* + * It might have happened that the HSI.FETCHED bit is + * set. Retry in this case. Therefore I expect maximum 2 times + * trying. + */ + ddcb_mark_appended(pddcb); + for (try = 0; try < 2; try++) { + old = prev_ddcb->icrc_hsi_shi_32; /* read SHI/HSI in BE32 */ + + /* try to append via NEXT bit if prev DDCB is not completed */ + if ((old & DDCB_COMPLETED_BE32) != 0x00000000) + break; + + new = (old | DDCB_NEXT_BE32); + + wmb(); + icrc_hsi_shi = cmpxchg(&prev_ddcb->icrc_hsi_shi_32, old, new); + + if (icrc_hsi_shi == old) + return RET_DDCB_APPENDED; /* appended to queue */ + } + + /* Queue must be re-started by updating QUEUE_OFFSET */ + ddcb_mark_tapped(pddcb); + num = (u64)ddcb_no << 8; + + wmb(); + __genwqe_writeq(cd, queue->IO_QUEUE_OFFSET, num); /* start queue */ + + return RET_DDCB_TAPPED; +} + +/** + * copy_ddcb_results() - Copy output state from real DDCB to request + * + * Copy DDCB ASV to request struct. There is no endian + * conversion made, since data structure in ASV is still + * unknown here. + * + * This is needed by: + * - genwqe_purge_ddcb() + * - genwqe_check_ddcb_queue() + */ +static void copy_ddcb_results(struct ddcb_requ *req, int ddcb_no) +{ + struct ddcb_queue *queue = req->queue; + struct ddcb *pddcb = &queue->ddcb_vaddr[req->num]; + + memcpy(&req->cmd.asv[0], &pddcb->asv[0], DDCB_ASV_LENGTH); + + /* copy status flags of the variant part */ + req->cmd.vcrc = be16_to_cpu(pddcb->vcrc_16); + req->cmd.deque_ts = be64_to_cpu(pddcb->deque_ts_64); + req->cmd.cmplt_ts = be64_to_cpu(pddcb->cmplt_ts_64); + + req->cmd.attn = be16_to_cpu(pddcb->attn_16); + req->cmd.progress = be32_to_cpu(pddcb->progress_32); + req->cmd.retc = be16_to_cpu(pddcb->retc_16); + + if (ddcb_requ_collect_debug_data(req)) { + int prev_no = (ddcb_no == 0) ? + queue->ddcb_max - 1 : ddcb_no - 1; + struct ddcb *prev_pddcb = &queue->ddcb_vaddr[prev_no]; + + memcpy(&req->debug_data.ddcb_finished, pddcb, + sizeof(req->debug_data.ddcb_finished)); + memcpy(&req->debug_data.ddcb_prev, prev_pddcb, + sizeof(req->debug_data.ddcb_prev)); + } +} + +/** + * genwqe_check_ddcb_queue() - Checks DDCB queue for completed work equests. + * @cd: pointer to genwqe device descriptor + * + * Return: Number of DDCBs which were finished + */ +static int genwqe_check_ddcb_queue(struct genwqe_dev *cd, + struct ddcb_queue *queue) +{ + unsigned long flags; + int ddcbs_finished = 0; + struct pci_dev *pci_dev = cd->pci_dev; + + spin_lock_irqsave(&queue->ddcb_lock, flags); + + /* FIXME avoid soft locking CPU */ + while (!queue_empty(queue) && (ddcbs_finished < queue->ddcb_max)) { + + struct ddcb *pddcb; + struct ddcb_requ *req; + u16 vcrc, vcrc_16, retc_16; + + pddcb = &queue->ddcb_vaddr[queue->ddcb_act]; + + if ((pddcb->icrc_hsi_shi_32 & DDCB_COMPLETED_BE32) == + 0x00000000) + goto go_home; /* not completed, continue waiting */ + + /* Note: DDCB could be purged */ + + req = queue->ddcb_req[queue->ddcb_act]; + if (req == NULL) { + /* this occurs if DDCB is purged, not an error */ + /* Move active DDCB further; Nothing to do anymore. */ + goto pick_next_one; + } + + /* + * HSI=0x44 (fetched and completed), but RETC is + * 0x101, or even worse 0x000. + * + * In case of seeing the queue in inconsistent state + * we read the errcnts and the queue status to provide + * a trigger for our PCIe analyzer stop capturing. + */ + retc_16 = be16_to_cpu(pddcb->retc_16); + if ((pddcb->hsi == 0x44) && (retc_16 <= 0x101)) { + u64 errcnts, status; + u64 ddcb_offs = (u64)pddcb - (u64)queue->ddcb_vaddr; + + errcnts = __genwqe_readq(cd, queue->IO_QUEUE_ERRCNTS); + status = __genwqe_readq(cd, queue->IO_QUEUE_STATUS); + + dev_err(&pci_dev->dev, + "[%s] SEQN=%04x HSI=%02x RETC=%03x " + " Q_ERRCNTS=%016llx Q_STATUS=%016llx\n" + " DDCB_DMA_ADDR=%016llx\n", + __func__, be16_to_cpu(pddcb->seqnum_16), + pddcb->hsi, retc_16, errcnts, status, + queue->ddcb_daddr + ddcb_offs); + } + + copy_ddcb_results(req, queue->ddcb_act); + queue->ddcb_req[queue->ddcb_act] = NULL; /* take from queue */ + + dev_dbg(&pci_dev->dev, "FINISHED DDCB#%d\n", req->num); + genwqe_hexdump(pci_dev, pddcb, sizeof(*pddcb)); + + ddcb_mark_finished(pddcb); + + /* calculate CRC_16 to see if VCRC is correct */ + vcrc = genwqe_crc16(pddcb->asv, + VCRC_LENGTH(req->cmd.asv_length), + 0xffff); + vcrc_16 = be16_to_cpu(pddcb->vcrc_16); + if (vcrc != vcrc_16) { + printk_ratelimited(KERN_ERR + "%s %s: err: wrong VCRC pre=%02x vcrc_len=%d " + "bytes vcrc_data=%04x is not vcrc_card=%04x\n", + GENWQE_DEVNAME, dev_name(&pci_dev->dev), + pddcb->pre, VCRC_LENGTH(req->cmd.asv_length), + vcrc, vcrc_16); + } + + ddcb_requ_set_state(req, GENWQE_REQU_FINISHED); + queue->ddcbs_completed++; + queue->ddcbs_in_flight--; + + /* wake up process waiting for this DDCB */ + wake_up_interruptible(&queue->ddcb_waitqs[queue->ddcb_act]); + +pick_next_one: + queue->ddcb_act = (queue->ddcb_act + 1) % queue->ddcb_max; + ddcbs_finished++; + } + + go_home: + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + return ddcbs_finished; +} + +/** + * __genwqe_wait_ddcb(): Waits until DDCB is completed + * @cd: pointer to genwqe device descriptor + * @req: pointer to requsted DDCB parameters + * + * The Service Layer will update the RETC in DDCB when processing is + * pending or done. + * + * Return: > 0 remaining jiffies, DDCB completed + * -ETIMEDOUT when timeout + * -ERESTARTSYS when ^C + * -EINVAL when unknown error condition + * + * When an error is returned the called needs to ensure that + * purge_ddcb() is being called to get the &req removed from the + * queue. + */ +int __genwqe_wait_ddcb(struct genwqe_dev *cd, struct ddcb_requ *req) +{ + int rc; + unsigned int ddcb_no; + struct ddcb_queue *queue; + struct pci_dev *pci_dev = cd->pci_dev; + + if (req == NULL) + return -EINVAL; + + queue = req->queue; + if (queue == NULL) + return -EINVAL; + + ddcb_no = req->num; + if (ddcb_no >= queue->ddcb_max) + return -EINVAL; + + rc = wait_event_interruptible_timeout(queue->ddcb_waitqs[ddcb_no], + ddcb_requ_finished(cd, req), + genwqe_ddcb_software_timeout * HZ); + + /* + * We need to distinguish 3 cases here: + * 1. rc == 0 timeout occured + * 2. rc == -ERESTARTSYS signal received + * 3. rc > 0 remaining jiffies condition is true + */ + if (rc == 0) { + struct ddcb_queue *queue = req->queue; + struct ddcb *pddcb; + + /* + * Timeout may be caused by long task switching time. + * When timeout happens, check if the request has + * meanwhile completed. + */ + genwqe_check_ddcb_queue(cd, req->queue); + if (ddcb_requ_finished(cd, req)) + return rc; + + dev_err(&pci_dev->dev, + "[%s] err: DDCB#%d timeout rc=%d state=%d req @ %p\n", + __func__, req->num, rc, ddcb_requ_get_state(req), + req); + dev_err(&pci_dev->dev, + "[%s] IO_QUEUE_STATUS=0x%016llx\n", __func__, + __genwqe_readq(cd, queue->IO_QUEUE_STATUS)); + + pddcb = &queue->ddcb_vaddr[req->num]; + genwqe_hexdump(pci_dev, pddcb, sizeof(*pddcb)); + + print_ddcb_info(cd, req->queue); + return -ETIMEDOUT; + + } else if (rc == -ERESTARTSYS) { + return rc; + /* + * EINTR: Stops the application + * ERESTARTSYS: Restartable systemcall; called again + */ + + } else if (rc < 0) { + dev_err(&pci_dev->dev, + "[%s] err: DDCB#%d unknown result (rc=%d) %d!\n", + __func__, req->num, rc, ddcb_requ_get_state(req)); + return -EINVAL; + } + + /* Severe error occured. Driver is forced to stop operation */ + if (cd->card_state != GENWQE_CARD_USED) { + dev_err(&pci_dev->dev, + "[%s] err: DDCB#%d forced to stop (rc=%d)\n", + __func__, req->num, rc); + return -EIO; + } + return rc; +} + +/** + * get_next_ddcb() - Get next available DDCB + * @cd: pointer to genwqe device descriptor + * + * DDCB's content is completely cleared but presets for PRE and + * SEQNUM. This function must only be called when ddcb_lock is held. + * + * Return: NULL if no empty DDCB available otherwise ptr to next DDCB. + */ +static struct ddcb *get_next_ddcb(struct genwqe_dev *cd, + struct ddcb_queue *queue, + int *num) +{ + u64 *pu64; + struct ddcb *pddcb; + + if (queue_free_ddcbs(queue) == 0) /* queue is full */ + return NULL; + + /* find new ddcb */ + pddcb = &queue->ddcb_vaddr[queue->ddcb_next]; + + /* if it is not completed, we are not allowed to use it */ + /* barrier(); */ + if ((pddcb->icrc_hsi_shi_32 & DDCB_COMPLETED_BE32) == 0x00000000) + return NULL; + + *num = queue->ddcb_next; /* internal DDCB number */ + queue->ddcb_next = (queue->ddcb_next + 1) % queue->ddcb_max; + + /* clear important DDCB fields */ + pu64 = (u64 *)pddcb; + pu64[0] = 0ULL; /* offs 0x00 (ICRC,HSI,SHI,...) */ + pu64[1] = 0ULL; /* offs 0x01 (ACFUNC,CMD...) */ + + /* destroy previous results in ASV */ + pu64[0x80/8] = 0ULL; /* offs 0x80 (ASV + 0) */ + pu64[0x88/8] = 0ULL; /* offs 0x88 (ASV + 0x08) */ + pu64[0x90/8] = 0ULL; /* offs 0x90 (ASV + 0x10) */ + pu64[0x98/8] = 0ULL; /* offs 0x98 (ASV + 0x18) */ + pu64[0xd0/8] = 0ULL; /* offs 0xd0 (RETC,ATTN...) */ + + pddcb->pre = DDCB_PRESET_PRE; /* 128 */ + pddcb->seqnum_16 = cpu_to_be16(queue->ddcb_seq++); + return pddcb; +} + +/** + * __genwqe_purge_ddcb() - Remove a DDCB from the workqueue + * @cd: genwqe device descriptor + * @req: DDCB request + * + * This will fail when the request was already FETCHED. In this case + * we need to wait until it is finished. Else the DDCB can be + * reused. This function also ensures that the request data structure + * is removed from ddcb_req[]. + * + * Do not forget to call this function when genwqe_wait_ddcb() fails, + * such that the request gets really removed from ddcb_req[]. + * + * Return: 0 success + */ +int __genwqe_purge_ddcb(struct genwqe_dev *cd, struct ddcb_requ *req) +{ + struct ddcb *pddcb = NULL; + unsigned int t; + unsigned long flags; + struct ddcb_queue *queue = req->queue; + struct pci_dev *pci_dev = cd->pci_dev; + u64 queue_status; + __be32 icrc_hsi_shi = 0x0000; + __be32 old, new; + + /* unsigned long flags; */ + if (genwqe_ddcb_software_timeout <= 0) { + dev_err(&pci_dev->dev, + "[%s] err: software timeout is not set!\n", __func__); + return -EFAULT; + } + + pddcb = &queue->ddcb_vaddr[req->num]; + + for (t = 0; t < genwqe_ddcb_software_timeout * 10; t++) { + + spin_lock_irqsave(&queue->ddcb_lock, flags); + + /* Check if req was meanwhile finished */ + if (ddcb_requ_get_state(req) == GENWQE_REQU_FINISHED) + goto go_home; + + /* try to set PURGE bit if FETCHED/COMPLETED are not set */ + old = pddcb->icrc_hsi_shi_32; /* read SHI/HSI in BE32 */ + if ((old & DDCB_FETCHED_BE32) == 0x00000000) { + + new = (old | DDCB_PURGE_BE32); + icrc_hsi_shi = cmpxchg(&pddcb->icrc_hsi_shi_32, + old, new); + if (icrc_hsi_shi == old) + goto finish_ddcb; + } + + /* normal finish with HSI bit */ + barrier(); + icrc_hsi_shi = pddcb->icrc_hsi_shi_32; + if (icrc_hsi_shi & DDCB_COMPLETED_BE32) + goto finish_ddcb; + + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + + /* + * Here the check_ddcb() function will most likely + * discover this DDCB to be finished some point in + * time. It will mark the req finished and free it up + * in the list. + */ + + copy_ddcb_results(req, req->num); /* for the failing case */ + msleep(100); /* sleep for 1/10 second and try again */ + continue; + +finish_ddcb: + copy_ddcb_results(req, req->num); + ddcb_requ_set_state(req, GENWQE_REQU_FINISHED); + queue->ddcbs_in_flight--; + queue->ddcb_req[req->num] = NULL; /* delete from array */ + ddcb_mark_cleared(pddcb); + + /* Move active DDCB further; Nothing to do here anymore. */ + + /* + * We need to ensure that there is at least one free + * DDCB in the queue. To do that, we must update + * ddcb_act only if the COMPLETED bit is set for the + * DDCB we are working on else we treat that DDCB even + * if we PURGED it as occupied (hardware is supposed + * to set the COMPLETED bit yet!). + */ + icrc_hsi_shi = pddcb->icrc_hsi_shi_32; + if ((icrc_hsi_shi & DDCB_COMPLETED_BE32) && + (queue->ddcb_act == req->num)) { + queue->ddcb_act = ((queue->ddcb_act + 1) % + queue->ddcb_max); + } +go_home: + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + return 0; + } + + /* + * If the card is dead and the queue is forced to stop, we + * might see this in the queue status register. + */ + queue_status = __genwqe_readq(cd, queue->IO_QUEUE_STATUS); + + dev_dbg(&pci_dev->dev, "UN/FINISHED DDCB#%d\n", req->num); + genwqe_hexdump(pci_dev, pddcb, sizeof(*pddcb)); + + dev_err(&pci_dev->dev, + "[%s] err: DDCB#%d not purged and not completed " + "after %d seconds QSTAT=%016llx!!\n", + __func__, req->num, genwqe_ddcb_software_timeout, + queue_status); + + print_ddcb_info(cd, req->queue); + + return -EFAULT; +} + +int genwqe_init_debug_data(struct genwqe_dev *cd, struct genwqe_debug_data *d) +{ + int len; + struct pci_dev *pci_dev = cd->pci_dev; + + if (d == NULL) { + dev_err(&pci_dev->dev, + "[%s] err: invalid memory for debug data!\n", + __func__); + return -EFAULT; + } + + len = sizeof(d->driver_version); + snprintf(d->driver_version, len, "%s", DRV_VERS_STRING); + d->slu_unitcfg = cd->slu_unitcfg; + d->app_unitcfg = cd->app_unitcfg; + return 0; +} + +/** + * __genwqe_enqueue_ddcb() - Enqueue a DDCB + * @cd: pointer to genwqe device descriptor + * @req: pointer to DDCB execution request + * + * Return: 0 if enqueuing succeeded + * -EIO if card is unusable/PCIe problems + * -EBUSY if enqueuing failed + */ +int __genwqe_enqueue_ddcb(struct genwqe_dev *cd, struct ddcb_requ *req) +{ + struct ddcb *pddcb; + unsigned long flags; + struct ddcb_queue *queue; + struct pci_dev *pci_dev = cd->pci_dev; + u16 icrc; + + if (cd->card_state != GENWQE_CARD_USED) { + printk_ratelimited(KERN_ERR + "%s %s: [%s] Card is unusable/PCIe problem Req#%d\n", + GENWQE_DEVNAME, dev_name(&pci_dev->dev), + __func__, req->num); + return -EIO; + } + + queue = req->queue = &cd->queue; + + /* FIXME circumvention to improve performance when no irq is + * there. + */ + if (genwqe_polling_enabled) + genwqe_check_ddcb_queue(cd, queue); + + /* + * It must be ensured to process all DDCBs in successive + * order. Use a lock here in order to prevent nested DDCB + * enqueuing. + */ + spin_lock_irqsave(&queue->ddcb_lock, flags); + + pddcb = get_next_ddcb(cd, queue, &req->num); /* get ptr and num */ + if (pddcb == NULL) { + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + queue->busy++; + return -EBUSY; + } + + if (queue->ddcb_req[req->num] != NULL) { + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + + dev_err(&pci_dev->dev, + "[%s] picked DDCB %d with req=%p still in use!!\n", + __func__, req->num, req); + return -EFAULT; + } + ddcb_requ_set_state(req, GENWQE_REQU_ENQUEUED); + queue->ddcb_req[req->num] = req; + + pddcb->cmdopts_16 = cpu_to_be16(req->cmd.cmdopts); + pddcb->cmd = req->cmd.cmd; + pddcb->acfunc = req->cmd.acfunc; /* functional unit */ + + /* + * We know that we can get retc 0x104 with CRC error, do not + * stop the queue in those cases for this command. XDIR = 1 + * does not work for old SLU versions. + * + * Last bitstream with the old XDIR behavior had SLU_ID + * 0x34199. + */ + if ((cd->slu_unitcfg & 0xFFFF0ull) > 0x34199ull) + pddcb->xdir = 0x1; + else + pddcb->xdir = 0x0; + + + pddcb->psp = (((req->cmd.asiv_length / 8) << 4) | + ((req->cmd.asv_length / 8))); + pddcb->disp_ts_64 = cpu_to_be64(req->cmd.disp_ts); + + /* + * If copying the whole DDCB_ASIV_LENGTH is impacting + * performance we need to change it to + * req->cmd.asiv_length. But simulation benefits from some + * non-architectured bits behind the architectured content. + * + * How much data is copied depends on the availability of the + * ATS field, which was introduced late. If the ATS field is + * supported ASIV is 8 bytes shorter than it used to be. Since + * the ATS field is copied too, the code should do exactly + * what it did before, but I wanted to make copying of the ATS + * field very explicit. + */ + if (genwqe_get_slu_id(cd) <= 0x2) { + memcpy(&pddcb->__asiv[0], /* destination */ + &req->cmd.__asiv[0], /* source */ + DDCB_ASIV_LENGTH); /* req->cmd.asiv_length */ + } else { + pddcb->n.ats_64 = cpu_to_be64(req->cmd.ats); + memcpy(&pddcb->n.asiv[0], /* destination */ + &req->cmd.asiv[0], /* source */ + DDCB_ASIV_LENGTH_ATS); /* req->cmd.asiv_length */ + } + + pddcb->icrc_hsi_shi_32 = cpu_to_be32(0x00000000); /* for crc */ + + /* + * Calculate CRC_16 for corresponding range PSP(7:4). Include + * empty 4 bytes prior to the data. + */ + icrc = genwqe_crc16((const u8 *)pddcb, + ICRC_LENGTH(req->cmd.asiv_length), 0xffff); + pddcb->icrc_hsi_shi_32 = cpu_to_be32((u32)icrc << 16); + + /* enable DDCB completion irq */ + if (!genwqe_polling_enabled) + pddcb->icrc_hsi_shi_32 |= DDCB_INTR_BE32; + + dev_dbg(&pci_dev->dev, "INPUT DDCB#%d\n", req->num); + genwqe_hexdump(pci_dev, pddcb, sizeof(*pddcb)); + + if (ddcb_requ_collect_debug_data(req)) { + /* use the kernel copy of debug data. copying back to + user buffer happens later */ + + genwqe_init_debug_data(cd, &req->debug_data); + memcpy(&req->debug_data.ddcb_before, pddcb, + sizeof(req->debug_data.ddcb_before)); + } + + enqueue_ddcb(cd, queue, pddcb, req->num); + queue->ddcbs_in_flight++; + + if (queue->ddcbs_in_flight > queue->ddcbs_max_in_flight) + queue->ddcbs_max_in_flight = queue->ddcbs_in_flight; + + ddcb_requ_set_state(req, GENWQE_REQU_TAPPED); + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + wake_up_interruptible(&cd->queue_waitq); + + return 0; +} + +/** + * __genwqe_execute_raw_ddcb() - Setup and execute DDCB + * @cd: pointer to genwqe device descriptor + * @req: user provided DDCB request + */ +int __genwqe_execute_raw_ddcb(struct genwqe_dev *cd, + struct genwqe_ddcb_cmd *cmd) +{ + int rc = 0; + struct pci_dev *pci_dev = cd->pci_dev; + struct ddcb_requ *req = container_of(cmd, struct ddcb_requ, cmd); + + if (cmd->asiv_length > DDCB_ASIV_LENGTH) { + dev_err(&pci_dev->dev, "[%s] err: wrong asiv_length of %d\n", + __func__, cmd->asiv_length); + return -EINVAL; + } + if (cmd->asv_length > DDCB_ASV_LENGTH) { + dev_err(&pci_dev->dev, "[%s] err: wrong asv_length of %d\n", + __func__, cmd->asiv_length); + return -EINVAL; + } + rc = __genwqe_enqueue_ddcb(cd, req); + if (rc != 0) + return rc; + + rc = __genwqe_wait_ddcb(cd, req); + if (rc < 0) /* error or signal interrupt */ + goto err_exit; + + if (ddcb_requ_collect_debug_data(req)) { + if (copy_to_user((struct genwqe_debug_data __user *) + (unsigned long)cmd->ddata_addr, + &req->debug_data, + sizeof(struct genwqe_debug_data))) + return -EFAULT; + } + + /* + * Higher values than 0x102 indicate completion with faults, + * lower values than 0x102 indicate processing faults. Note + * that DDCB might have been purged. E.g. Cntl+C. + */ + if (cmd->retc != DDCB_RETC_COMPLETE) { + /* This might happen e.g. flash read, and needs to be + handled by the upper layer code. */ + rc = -EBADMSG; /* not processed/error retc */ + } + + return rc; + + err_exit: + __genwqe_purge_ddcb(cd, req); + + if (ddcb_requ_collect_debug_data(req)) { + if (copy_to_user((struct genwqe_debug_data __user *) + (unsigned long)cmd->ddata_addr, + &req->debug_data, + sizeof(struct genwqe_debug_data))) + return -EFAULT; + } + return rc; +} + +/** + * genwqe_next_ddcb_ready() - Figure out if the next DDCB is already finished + * + * We use this as condition for our wait-queue code. + */ +static int genwqe_next_ddcb_ready(struct genwqe_dev *cd) +{ + unsigned long flags; + struct ddcb *pddcb; + struct ddcb_queue *queue = &cd->queue; + + spin_lock_irqsave(&queue->ddcb_lock, flags); + + if (queue_empty(queue)) { /* emtpy queue */ + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + return 0; + } + + pddcb = &queue->ddcb_vaddr[queue->ddcb_act]; + if (pddcb->icrc_hsi_shi_32 & DDCB_COMPLETED_BE32) { /* ddcb ready */ + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + return 1; + } + + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + return 0; +} + +/** + * genwqe_ddcbs_in_flight() - Check how many DDCBs are in flight + * + * Keep track on the number of DDCBs which ware currently in the + * queue. This is needed for statistics as well as conditon if we want + * to wait or better do polling in case of no interrupts available. + */ +int genwqe_ddcbs_in_flight(struct genwqe_dev *cd) +{ + unsigned long flags; + int ddcbs_in_flight = 0; + struct ddcb_queue *queue = &cd->queue; + + spin_lock_irqsave(&queue->ddcb_lock, flags); + ddcbs_in_flight += queue->ddcbs_in_flight; + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + + return ddcbs_in_flight; +} + +static int setup_ddcb_queue(struct genwqe_dev *cd, struct ddcb_queue *queue) +{ + int rc, i; + struct ddcb *pddcb; + u64 val64; + unsigned int queue_size; + struct pci_dev *pci_dev = cd->pci_dev; + + if (genwqe_ddcb_max < 2) + return -EINVAL; + + queue_size = roundup(genwqe_ddcb_max * sizeof(struct ddcb), PAGE_SIZE); + + queue->ddcbs_in_flight = 0; /* statistics */ + queue->ddcbs_max_in_flight = 0; + queue->ddcbs_completed = 0; + queue->busy = 0; + + queue->ddcb_seq = 0x100; /* start sequence number */ + queue->ddcb_max = genwqe_ddcb_max; /* module parameter */ + queue->ddcb_vaddr = __genwqe_alloc_consistent(cd, queue_size, + &queue->ddcb_daddr); + if (queue->ddcb_vaddr == NULL) { + dev_err(&pci_dev->dev, + "[%s] **err: could not allocate DDCB **\n", __func__); + return -ENOMEM; + } + memset(queue->ddcb_vaddr, 0, queue_size); + + queue->ddcb_req = kzalloc(sizeof(struct ddcb_requ *) * + queue->ddcb_max, GFP_KERNEL); + if (!queue->ddcb_req) { + rc = -ENOMEM; + goto free_ddcbs; + } + + queue->ddcb_waitqs = kzalloc(sizeof(wait_queue_head_t) * + queue->ddcb_max, GFP_KERNEL); + if (!queue->ddcb_waitqs) { + rc = -ENOMEM; + goto free_requs; + } + + for (i = 0; i < queue->ddcb_max; i++) { + pddcb = &queue->ddcb_vaddr[i]; /* DDCBs */ + pddcb->icrc_hsi_shi_32 = DDCB_COMPLETED_BE32; + pddcb->retc_16 = cpu_to_be16(0xfff); + + queue->ddcb_req[i] = NULL; /* requests */ + init_waitqueue_head(&queue->ddcb_waitqs[i]); /* waitqueues */ + } + + queue->ddcb_act = 0; + queue->ddcb_next = 0; /* queue is empty */ + + spin_lock_init(&queue->ddcb_lock); + init_waitqueue_head(&queue->ddcb_waitq); + + val64 = ((u64)(queue->ddcb_max - 1) << 8); /* lastptr */ + __genwqe_writeq(cd, queue->IO_QUEUE_CONFIG, 0x07); /* iCRC/vCRC */ + __genwqe_writeq(cd, queue->IO_QUEUE_SEGMENT, queue->ddcb_daddr); + __genwqe_writeq(cd, queue->IO_QUEUE_INITSQN, queue->ddcb_seq); + __genwqe_writeq(cd, queue->IO_QUEUE_WRAP, val64); + return 0; + + free_requs: + kfree(queue->ddcb_req); + queue->ddcb_req = NULL; + free_ddcbs: + __genwqe_free_consistent(cd, queue_size, queue->ddcb_vaddr, + queue->ddcb_daddr); + queue->ddcb_vaddr = NULL; + queue->ddcb_daddr = 0ull; + return -ENODEV; + +} + +static int ddcb_queue_initialized(struct ddcb_queue *queue) +{ + return queue->ddcb_vaddr != NULL; +} + +static void free_ddcb_queue(struct genwqe_dev *cd, struct ddcb_queue *queue) +{ + unsigned int queue_size; + + queue_size = roundup(queue->ddcb_max * sizeof(struct ddcb), PAGE_SIZE); + + kfree(queue->ddcb_req); + queue->ddcb_req = NULL; + + if (queue->ddcb_vaddr) { + __genwqe_free_consistent(cd, queue_size, queue->ddcb_vaddr, + queue->ddcb_daddr); + queue->ddcb_vaddr = NULL; + queue->ddcb_daddr = 0ull; + } +} + +static irqreturn_t genwqe_pf_isr(int irq, void *dev_id) +{ + u64 gfir; + struct genwqe_dev *cd = (struct genwqe_dev *)dev_id; + struct pci_dev *pci_dev = cd->pci_dev; + + /* + * In case of fatal FIR error the queue is stopped, such that + * we can safely check it without risking anything. + */ + cd->irqs_processed++; + wake_up_interruptible(&cd->queue_waitq); + + /* + * Checking for errors before kicking the queue might be + * safer, but slower for the good-case ... See above. + */ + gfir = __genwqe_readq(cd, IO_SLC_CFGREG_GFIR); + if ((gfir & GFIR_ERR_TRIGGER) != 0x0) { + + wake_up_interruptible(&cd->health_waitq); + + /* + * By default GFIRs causes recovery actions. This + * count is just for debug when recovery is masked. + */ + printk_ratelimited(KERN_ERR + "%s %s: [%s] GFIR=%016llx\n", + GENWQE_DEVNAME, dev_name(&pci_dev->dev), + __func__, gfir); + } + + return IRQ_HANDLED; +} + +static irqreturn_t genwqe_vf_isr(int irq, void *dev_id) +{ + struct genwqe_dev *cd = (struct genwqe_dev *)dev_id; + + cd->irqs_processed++; + wake_up_interruptible(&cd->queue_waitq); + + return IRQ_HANDLED; +} + +/** + * genwqe_card_thread() - Work thread for the DDCB queue + * + * The idea is to check if there are DDCBs in processing. If there are + * some finished DDCBs, we process them and wakeup the + * requestors. Otherwise we give other processes time using + * cond_resched(). + */ +static int genwqe_card_thread(void *data) +{ + int should_stop = 0, rc = 0; + struct genwqe_dev *cd = (struct genwqe_dev *)data; + + while (!kthread_should_stop()) { + + genwqe_check_ddcb_queue(cd, &cd->queue); + + if (genwqe_polling_enabled) { + rc = wait_event_interruptible_timeout( + cd->queue_waitq, + genwqe_ddcbs_in_flight(cd) || + (should_stop = kthread_should_stop()), 1); + } else { + rc = wait_event_interruptible_timeout( + cd->queue_waitq, + genwqe_next_ddcb_ready(cd) || + (should_stop = kthread_should_stop()), HZ); + } + if (should_stop) + break; + + /* + * Avoid soft lockups on heavy loads; we do not want + * to disable our interrupts. + */ + cond_resched(); + } + return 0; +} + +/** + * genwqe_setup_service_layer() - Setup DDCB queue + * @cd: pointer to genwqe device descriptor + * + * Allocate DDCBs. Configure Service Layer Controller (SLC). + * + * Return: 0 success + */ +int genwqe_setup_service_layer(struct genwqe_dev *cd) +{ + int rc; + struct ddcb_queue *queue; + struct pci_dev *pci_dev = cd->pci_dev; + + if (genwqe_is_privileged(cd)) { + rc = genwqe_card_reset(cd); + if (rc < 0) { + dev_err(&pci_dev->dev, + "[%s] err: reset failed.\n", __func__); + return rc; + } + genwqe_read_softreset(cd); + } + + queue = &cd->queue; + queue->IO_QUEUE_CONFIG = IO_SLC_QUEUE_CONFIG; + queue->IO_QUEUE_STATUS = IO_SLC_QUEUE_STATUS; + queue->IO_QUEUE_SEGMENT = IO_SLC_QUEUE_SEGMENT; + queue->IO_QUEUE_INITSQN = IO_SLC_QUEUE_INITSQN; + queue->IO_QUEUE_OFFSET = IO_SLC_QUEUE_OFFSET; + queue->IO_QUEUE_WRAP = IO_SLC_QUEUE_WRAP; + queue->IO_QUEUE_WTIME = IO_SLC_QUEUE_WTIME; + queue->IO_QUEUE_ERRCNTS = IO_SLC_QUEUE_ERRCNTS; + queue->IO_QUEUE_LRW = IO_SLC_QUEUE_LRW; + + rc = setup_ddcb_queue(cd, queue); + if (rc != 0) { + rc = -ENODEV; + goto err_out; + } + + init_waitqueue_head(&cd->queue_waitq); + cd->card_thread = kthread_run(genwqe_card_thread, cd, + GENWQE_DEVNAME "%d_thread", + cd->card_idx); + if (IS_ERR(cd->card_thread)) { + rc = PTR_ERR(cd->card_thread); + cd->card_thread = NULL; + goto stop_free_queue; + } + + rc = genwqe_set_interrupt_capability(cd, GENWQE_MSI_IRQS); + if (rc > 0) + rc = genwqe_set_interrupt_capability(cd, rc); + if (rc != 0) { + rc = -ENODEV; + goto stop_kthread; + } + + /* + * We must have all wait-queues initialized when we enable the + * interrupts. Otherwise we might crash if we get an early + * irq. + */ + init_waitqueue_head(&cd->health_waitq); + + if (genwqe_is_privileged(cd)) { + rc = request_irq(pci_dev->irq, genwqe_pf_isr, IRQF_SHARED, + GENWQE_DEVNAME, cd); + } else { + rc = request_irq(pci_dev->irq, genwqe_vf_isr, IRQF_SHARED, + GENWQE_DEVNAME, cd); + } + if (rc < 0) { + dev_err(&pci_dev->dev, "irq %d not free.\n", pci_dev->irq); + goto stop_irq_cap; + } + + cd->card_state = GENWQE_CARD_USED; + return 0; + + stop_irq_cap: + genwqe_reset_interrupt_capability(cd); + stop_kthread: + kthread_stop(cd->card_thread); + cd->card_thread = NULL; + stop_free_queue: + free_ddcb_queue(cd, queue); + err_out: + return rc; +} + +/** + * queue_wake_up_all() - Handles fatal error case + * + * The PCI device got unusable and we have to stop all pending + * requests as fast as we can. The code after this must purge the + * DDCBs in question and ensure that all mappings are freed. + */ +static int queue_wake_up_all(struct genwqe_dev *cd) +{ + unsigned int i; + unsigned long flags; + struct ddcb_queue *queue = &cd->queue; + + spin_lock_irqsave(&queue->ddcb_lock, flags); + + for (i = 0; i < queue->ddcb_max; i++) + wake_up_interruptible(&queue->ddcb_waitqs[queue->ddcb_act]); + + spin_unlock_irqrestore(&queue->ddcb_lock, flags); + + return 0; +} + +/** + * genwqe_finish_queue() - Remove any genwqe devices and user-interfaces + * + * Relies on the pre-condition that there are no users of the card + * device anymore e.g. with open file-descriptors. + * + * This function must be robust enough to be called twice. + */ +int genwqe_finish_queue(struct genwqe_dev *cd) +{ + int i, rc = 0, in_flight; + int waitmax = genwqe_ddcb_software_timeout; + struct pci_dev *pci_dev = cd->pci_dev; + struct ddcb_queue *queue = &cd->queue; + + if (!ddcb_queue_initialized(queue)) + return 0; + + /* Do not wipe out the error state. */ + if (cd->card_state == GENWQE_CARD_USED) + cd->card_state = GENWQE_CARD_UNUSED; + + /* Wake up all requests in the DDCB queue such that they + should be removed nicely. */ + queue_wake_up_all(cd); + + /* We must wait to get rid of the DDCBs in flight */ + for (i = 0; i < waitmax; i++) { + in_flight = genwqe_ddcbs_in_flight(cd); + + if (in_flight == 0) + break; + + dev_dbg(&pci_dev->dev, + " DEBUG [%d/%d] waiting for queue to get empty: " + "%d requests!\n", i, waitmax, in_flight); + + /* + * Severe severe error situation: The card itself has + * 16 DDCB queues, each queue has e.g. 32 entries, + * each DDBC has a hardware timeout of currently 250 + * msec but the PFs have a hardware timeout of 8 sec + * ... so I take something large. + */ + msleep(1000); + } + if (i == waitmax) { + dev_err(&pci_dev->dev, " [%s] err: queue is not empty!!\n", + __func__); + rc = -EIO; + } + return rc; +} + +/** + * genwqe_release_service_layer() - Shutdown DDCB queue + * @cd: genwqe device descriptor + * + * This function must be robust enough to be called twice. + */ +int genwqe_release_service_layer(struct genwqe_dev *cd) +{ + struct pci_dev *pci_dev = cd->pci_dev; + + if (!ddcb_queue_initialized(&cd->queue)) + return 1; + + free_irq(pci_dev->irq, cd); + genwqe_reset_interrupt_capability(cd); + + if (cd->card_thread != NULL) { + kthread_stop(cd->card_thread); + cd->card_thread = NULL; + } + + free_ddcb_queue(cd, &cd->queue); + return 0; +} diff --git a/drivers/misc/genwqe/card_ddcb.h b/drivers/misc/genwqe/card_ddcb.h new file mode 100644 index 00000000000..c4f26720753 --- /dev/null +++ b/drivers/misc/genwqe/card_ddcb.h @@ -0,0 +1,188 @@ +#ifndef __CARD_DDCB_H__ +#define __CARD_DDCB_H__ + +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2, or (at your option) + * any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/types.h> +#include <asm/byteorder.h> + +#include "genwqe_driver.h" +#include "card_base.h" + +/** + * struct ddcb - Device Driver Control Block DDCB + * @hsi: Hardware software interlock + * @shi: Software hardware interlock. Hsi and shi are used to interlock + * software and hardware activities. We are using a compare and + * swap operation to ensure that there are no races when + * activating new DDCBs on the queue, or when we need to + * purge a DDCB from a running queue. + * @acfunc: Accelerator function addresses a unit within the chip + * @cmd: Command to work on + * @cmdopts_16: Options for the command + * @asiv: Input data + * @asv: Output data + * + * The DDCB data format is big endian. Multiple consequtive DDBCs form + * a DDCB queue. + */ +#define ASIV_LENGTH 104 /* Old specification without ATS field */ +#define ASIV_LENGTH_ATS 96 /* New specification with ATS field */ +#define ASV_LENGTH 64 + +struct ddcb { + union { + __be32 icrc_hsi_shi_32; /* iCRC, Hardware/SW interlock */ + struct { + __be16 icrc_16; + u8 hsi; + u8 shi; + }; + }; + u8 pre; /* Preamble */ + u8 xdir; /* Execution Directives */ + __be16 seqnum_16; /* Sequence Number */ + + u8 acfunc; /* Accelerator Function.. */ + u8 cmd; /* Command. */ + __be16 cmdopts_16; /* Command Options */ + u8 sur; /* Status Update Rate */ + u8 psp; /* Protection Section Pointer */ + __be16 rsvd_0e_16; /* Reserved invariant */ + + __be64 fwiv_64; /* Firmware Invariant. */ + + union { + struct { + __be64 ats_64; /* Address Translation Spec */ + u8 asiv[ASIV_LENGTH_ATS]; /* New ASIV */ + } n; + u8 __asiv[ASIV_LENGTH]; /* obsolete */ + }; + u8 asv[ASV_LENGTH]; /* Appl Spec Variant */ + + __be16 rsvd_c0_16; /* Reserved Variant */ + __be16 vcrc_16; /* Variant CRC */ + __be32 rsvd_32; /* Reserved unprotected */ + + __be64 deque_ts_64; /* Deque Time Stamp. */ + + __be16 retc_16; /* Return Code */ + __be16 attn_16; /* Attention/Extended Error Codes */ + __be32 progress_32; /* Progress indicator. */ + + __be64 cmplt_ts_64; /* Completion Time Stamp. */ + + /* The following layout matches the new service layer format */ + __be32 ibdc_32; /* Inbound Data Count (* 256) */ + __be32 obdc_32; /* Outbound Data Count (* 256) */ + + __be64 rsvd_SLH_64; /* Reserved for hardware */ + union { /* private data for driver */ + u8 priv[8]; + __be64 priv_64; + }; + __be64 disp_ts_64; /* Dispatch TimeStamp */ +} __attribute__((__packed__)); + +/* CRC polynomials for DDCB */ +#define CRC16_POLYNOMIAL 0x1021 + +/* + * SHI: Software to Hardware Interlock + * This 1 byte field is written by software to interlock the + * movement of one queue entry to another with the hardware in the + * chip. + */ +#define DDCB_SHI_INTR 0x04 /* Bit 2 */ +#define DDCB_SHI_PURGE 0x02 /* Bit 1 */ +#define DDCB_SHI_NEXT 0x01 /* Bit 0 */ + +/* + * HSI: Hardware to Software interlock + * This 1 byte field is written by hardware to interlock the movement + * of one queue entry to another with the software in the chip. + */ +#define DDCB_HSI_COMPLETED 0x40 /* Bit 6 */ +#define DDCB_HSI_FETCHED 0x04 /* Bit 2 */ + +/* + * Accessing HSI/SHI is done 32-bit wide + * Normally 16-bit access would work too, but on some platforms the + * 16 compare and swap operation is not supported. Therefore + * switching to 32-bit such that those platforms will work too. + * + * iCRC HSI/SHI + */ +#define DDCB_INTR_BE32 cpu_to_be32(0x00000004) +#define DDCB_PURGE_BE32 cpu_to_be32(0x00000002) +#define DDCB_NEXT_BE32 cpu_to_be32(0x00000001) +#define DDCB_COMPLETED_BE32 cpu_to_be32(0x00004000) +#define DDCB_FETCHED_BE32 cpu_to_be32(0x00000400) + +/* Definitions of DDCB presets */ +#define DDCB_PRESET_PRE 0x80 +#define ICRC_LENGTH(n) ((n) + 8 + 8 + 8) /* used ASIV + hdr fields */ +#define VCRC_LENGTH(n) ((n)) /* used ASV */ + +/* + * Genwqe Scatter Gather list + * Each element has up to 8 entries. + * The chaining element is element 0 cause of prefetching needs. + */ + +/* + * 0b0110 Chained descriptor. The descriptor is describing the next + * descriptor list. + */ +#define SG_CHAINED (0x6) + +/* + * 0b0010 First entry of a descriptor list. Start from a Buffer-Empty + * condition. + */ +#define SG_DATA (0x2) + +/* + * 0b0000 Early terminator. This is the last entry on the list + * irregardless of the length indicated. + */ +#define SG_END_LIST (0x0) + +/** + * struct sglist - Scatter gather list + * @target_addr: Either a dma addr of memory to work on or a + * dma addr or a subsequent sglist block. + * @len: Length of the data block. + * @flags: See above. + * + * Depending on the command the GenWQE card can use a scatter gather + * list to describe the memory it works on. Always 8 sg_entry's form + * a block. + */ +struct sg_entry { + __be64 target_addr; + __be32 len; + __be32 flags; +}; + +#endif /* __CARD_DDCB_H__ */ diff --git a/drivers/misc/genwqe/card_debugfs.c b/drivers/misc/genwqe/card_debugfs.c new file mode 100644 index 00000000000..0a33ade6410 --- /dev/null +++ b/drivers/misc/genwqe/card_debugfs.c @@ -0,0 +1,499 @@ +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +/* + * Debugfs interfaces for the GenWQE card. Help to debug potential + * problems. Dump internal chip state for debugging and failure + * determination. + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/debugfs.h> +#include <linux/seq_file.h> +#include <linux/uaccess.h> + +#include "card_base.h" +#include "card_ddcb.h" + +#define GENWQE_DEBUGFS_RO(_name, _showfn) \ + static int genwqe_debugfs_##_name##_open(struct inode *inode, \ + struct file *file) \ + { \ + return single_open(file, _showfn, inode->i_private); \ + } \ + static const struct file_operations genwqe_##_name##_fops = { \ + .open = genwqe_debugfs_##_name##_open, \ + .read = seq_read, \ + .llseek = seq_lseek, \ + .release = single_release, \ + } + +static void dbg_uidn_show(struct seq_file *s, struct genwqe_reg *regs, + int entries) +{ + unsigned int i; + u32 v_hi, v_lo; + + for (i = 0; i < entries; i++) { + v_hi = (regs[i].val >> 32) & 0xffffffff; + v_lo = (regs[i].val) & 0xffffffff; + + seq_printf(s, " 0x%08x 0x%08x 0x%08x 0x%08x EXT_ERR_REC\n", + regs[i].addr, regs[i].idx, v_hi, v_lo); + } +} + +static int curr_dbg_uidn_show(struct seq_file *s, void *unused, int uid) +{ + struct genwqe_dev *cd = s->private; + int entries; + struct genwqe_reg *regs; + + entries = genwqe_ffdc_buff_size(cd, uid); + if (entries < 0) + return -EINVAL; + + if (entries == 0) + return 0; + + regs = kcalloc(entries, sizeof(*regs), GFP_KERNEL); + if (regs == NULL) + return -ENOMEM; + + genwqe_stop_traps(cd); /* halt the traps while dumping data */ + genwqe_ffdc_buff_read(cd, uid, regs, entries); + genwqe_start_traps(cd); + + dbg_uidn_show(s, regs, entries); + kfree(regs); + return 0; +} + +static int genwqe_curr_dbg_uid0_show(struct seq_file *s, void *unused) +{ + return curr_dbg_uidn_show(s, unused, 0); +} + +GENWQE_DEBUGFS_RO(curr_dbg_uid0, genwqe_curr_dbg_uid0_show); + +static int genwqe_curr_dbg_uid1_show(struct seq_file *s, void *unused) +{ + return curr_dbg_uidn_show(s, unused, 1); +} + +GENWQE_DEBUGFS_RO(curr_dbg_uid1, genwqe_curr_dbg_uid1_show); + +static int genwqe_curr_dbg_uid2_show(struct seq_file *s, void *unused) +{ + return curr_dbg_uidn_show(s, unused, 2); +} + +GENWQE_DEBUGFS_RO(curr_dbg_uid2, genwqe_curr_dbg_uid2_show); + +static int prev_dbg_uidn_show(struct seq_file *s, void *unused, int uid) +{ + struct genwqe_dev *cd = s->private; + + dbg_uidn_show(s, cd->ffdc[uid].regs, cd->ffdc[uid].entries); + return 0; +} + +static int genwqe_prev_dbg_uid0_show(struct seq_file *s, void *unused) +{ + return prev_dbg_uidn_show(s, unused, 0); +} + +GENWQE_DEBUGFS_RO(prev_dbg_uid0, genwqe_prev_dbg_uid0_show); + +static int genwqe_prev_dbg_uid1_show(struct seq_file *s, void *unused) +{ + return prev_dbg_uidn_show(s, unused, 1); +} + +GENWQE_DEBUGFS_RO(prev_dbg_uid1, genwqe_prev_dbg_uid1_show); + +static int genwqe_prev_dbg_uid2_show(struct seq_file *s, void *unused) +{ + return prev_dbg_uidn_show(s, unused, 2); +} + +GENWQE_DEBUGFS_RO(prev_dbg_uid2, genwqe_prev_dbg_uid2_show); + +static int genwqe_curr_regs_show(struct seq_file *s, void *unused) +{ + struct genwqe_dev *cd = s->private; + unsigned int i; + struct genwqe_reg *regs; + + regs = kcalloc(GENWQE_FFDC_REGS, sizeof(*regs), GFP_KERNEL); + if (regs == NULL) + return -ENOMEM; + + genwqe_stop_traps(cd); + genwqe_read_ffdc_regs(cd, regs, GENWQE_FFDC_REGS, 1); + genwqe_start_traps(cd); + + for (i = 0; i < GENWQE_FFDC_REGS; i++) { + if (regs[i].addr == 0xffffffff) + break; /* invalid entries */ + + if (regs[i].val == 0x0ull) + continue; /* do not print 0x0 FIRs */ + + seq_printf(s, " 0x%08x 0x%016llx\n", + regs[i].addr, regs[i].val); + } + return 0; +} + +GENWQE_DEBUGFS_RO(curr_regs, genwqe_curr_regs_show); + +static int genwqe_prev_regs_show(struct seq_file *s, void *unused) +{ + struct genwqe_dev *cd = s->private; + unsigned int i; + struct genwqe_reg *regs = cd->ffdc[GENWQE_DBG_REGS].regs; + + if (regs == NULL) + return -EINVAL; + + for (i = 0; i < GENWQE_FFDC_REGS; i++) { + if (regs[i].addr == 0xffffffff) + break; /* invalid entries */ + + if (regs[i].val == 0x0ull) + continue; /* do not print 0x0 FIRs */ + + seq_printf(s, " 0x%08x 0x%016llx\n", + regs[i].addr, regs[i].val); + } + return 0; +} + +GENWQE_DEBUGFS_RO(prev_regs, genwqe_prev_regs_show); + +static int genwqe_jtimer_show(struct seq_file *s, void *unused) +{ + struct genwqe_dev *cd = s->private; + unsigned int vf_num; + u64 jtimer; + + jtimer = genwqe_read_vreg(cd, IO_SLC_VF_APPJOB_TIMEOUT, 0); + seq_printf(s, " PF 0x%016llx %d msec\n", jtimer, + genwqe_pf_jobtimeout_msec); + + for (vf_num = 0; vf_num < cd->num_vfs; vf_num++) { + jtimer = genwqe_read_vreg(cd, IO_SLC_VF_APPJOB_TIMEOUT, + vf_num + 1); + seq_printf(s, " VF%-2d 0x%016llx %d msec\n", vf_num, jtimer, + cd->vf_jobtimeout_msec[vf_num]); + } + return 0; +} + +GENWQE_DEBUGFS_RO(jtimer, genwqe_jtimer_show); + +static int genwqe_queue_working_time_show(struct seq_file *s, void *unused) +{ + struct genwqe_dev *cd = s->private; + unsigned int vf_num; + u64 t; + + t = genwqe_read_vreg(cd, IO_SLC_VF_QUEUE_WTIME, 0); + seq_printf(s, " PF 0x%016llx\n", t); + + for (vf_num = 0; vf_num < cd->num_vfs; vf_num++) { + t = genwqe_read_vreg(cd, IO_SLC_VF_QUEUE_WTIME, vf_num + 1); + seq_printf(s, " VF%-2d 0x%016llx\n", vf_num, t); + } + return 0; +} + +GENWQE_DEBUGFS_RO(queue_working_time, genwqe_queue_working_time_show); + +static int genwqe_ddcb_info_show(struct seq_file *s, void *unused) +{ + struct genwqe_dev *cd = s->private; + unsigned int i; + struct ddcb_queue *queue; + struct ddcb *pddcb; + + queue = &cd->queue; + seq_puts(s, "DDCB QUEUE:\n"); + seq_printf(s, " ddcb_max: %d\n" + " ddcb_daddr: %016llx - %016llx\n" + " ddcb_vaddr: %016llx\n" + " ddcbs_in_flight: %u\n" + " ddcbs_max_in_flight: %u\n" + " ddcbs_completed: %u\n" + " busy: %u\n" + " irqs_processed: %u\n", + queue->ddcb_max, (long long)queue->ddcb_daddr, + (long long)queue->ddcb_daddr + + (queue->ddcb_max * DDCB_LENGTH), + (long long)queue->ddcb_vaddr, queue->ddcbs_in_flight, + queue->ddcbs_max_in_flight, queue->ddcbs_completed, + queue->busy, cd->irqs_processed); + + /* Hardware State */ + seq_printf(s, " 0x%08x 0x%016llx IO_QUEUE_CONFIG\n" + " 0x%08x 0x%016llx IO_QUEUE_STATUS\n" + " 0x%08x 0x%016llx IO_QUEUE_SEGMENT\n" + " 0x%08x 0x%016llx IO_QUEUE_INITSQN\n" + " 0x%08x 0x%016llx IO_QUEUE_WRAP\n" + " 0x%08x 0x%016llx IO_QUEUE_OFFSET\n" + " 0x%08x 0x%016llx IO_QUEUE_WTIME\n" + " 0x%08x 0x%016llx IO_QUEUE_ERRCNTS\n" + " 0x%08x 0x%016llx IO_QUEUE_LRW\n", + queue->IO_QUEUE_CONFIG, + __genwqe_readq(cd, queue->IO_QUEUE_CONFIG), + queue->IO_QUEUE_STATUS, + __genwqe_readq(cd, queue->IO_QUEUE_STATUS), + queue->IO_QUEUE_SEGMENT, + __genwqe_readq(cd, queue->IO_QUEUE_SEGMENT), + queue->IO_QUEUE_INITSQN, + __genwqe_readq(cd, queue->IO_QUEUE_INITSQN), + queue->IO_QUEUE_WRAP, + __genwqe_readq(cd, queue->IO_QUEUE_WRAP), + queue->IO_QUEUE_OFFSET, + __genwqe_readq(cd, queue->IO_QUEUE_OFFSET), + queue->IO_QUEUE_WTIME, + __genwqe_readq(cd, queue->IO_QUEUE_WTIME), + queue->IO_QUEUE_ERRCNTS, + __genwqe_readq(cd, queue->IO_QUEUE_ERRCNTS), + queue->IO_QUEUE_LRW, + __genwqe_readq(cd, queue->IO_QUEUE_LRW)); + + seq_printf(s, "DDCB list (ddcb_act=%d/ddcb_next=%d):\n", + queue->ddcb_act, queue->ddcb_next); + + pddcb = queue->ddcb_vaddr; + for (i = 0; i < queue->ddcb_max; i++) { + seq_printf(s, " %-3d: RETC=%03x SEQ=%04x HSI/SHI=%02x/%02x ", + i, be16_to_cpu(pddcb->retc_16), + be16_to_cpu(pddcb->seqnum_16), + pddcb->hsi, pddcb->shi); + seq_printf(s, "PRIV=%06llx CMD=%02x\n", + be64_to_cpu(pddcb->priv_64), pddcb->cmd); + pddcb++; + } + return 0; +} + +GENWQE_DEBUGFS_RO(ddcb_info, genwqe_ddcb_info_show); + +static int genwqe_info_show(struct seq_file *s, void *unused) +{ + struct genwqe_dev *cd = s->private; + u16 val16, type; + u64 app_id, slu_id, bitstream = -1; + struct pci_dev *pci_dev = cd->pci_dev; + + slu_id = __genwqe_readq(cd, IO_SLU_UNITCFG); + app_id = __genwqe_readq(cd, IO_APP_UNITCFG); + + if (genwqe_is_privileged(cd)) + bitstream = __genwqe_readq(cd, IO_SLU_BITSTREAM); + + val16 = (u16)(slu_id & 0x0fLLU); + type = (u16)((slu_id >> 20) & 0xffLLU); + + seq_printf(s, "%s driver version: %s\n" + " Device Name/Type: %s %s CardIdx: %d\n" + " SLU/APP Config : 0x%016llx/0x%016llx\n" + " Build Date : %u/%x/%u\n" + " Base Clock : %u MHz\n" + " Arch/SVN Release: %u/%llx\n" + " Bitstream : %llx\n", + GENWQE_DEVNAME, DRV_VERS_STRING, dev_name(&pci_dev->dev), + genwqe_is_privileged(cd) ? + "Physical" : "Virtual or no SR-IOV", + cd->card_idx, slu_id, app_id, + (u16)((slu_id >> 12) & 0x0fLLU), /* month */ + (u16)((slu_id >> 4) & 0xffLLU), /* day */ + (u16)((slu_id >> 16) & 0x0fLLU) + 2010, /* year */ + genwqe_base_clock_frequency(cd), + (u16)((slu_id >> 32) & 0xffLLU), slu_id >> 40, + bitstream); + + return 0; +} + +GENWQE_DEBUGFS_RO(info, genwqe_info_show); + +int genwqe_init_debugfs(struct genwqe_dev *cd) +{ + struct dentry *root; + struct dentry *file; + int ret; + char card_name[64]; + char name[64]; + unsigned int i; + + sprintf(card_name, "%s%d_card", GENWQE_DEVNAME, cd->card_idx); + + root = debugfs_create_dir(card_name, cd->debugfs_genwqe); + if (!root) { + ret = -ENOMEM; + goto err0; + } + + /* non privileged interfaces are done here */ + file = debugfs_create_file("ddcb_info", S_IRUGO, root, cd, + &genwqe_ddcb_info_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("info", S_IRUGO, root, cd, + &genwqe_info_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_x64("err_inject", 0666, root, &cd->err_inject); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_u32("ddcb_software_timeout", 0666, root, + &cd->ddcb_software_timeout); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_u32("kill_timeout", 0666, root, + &cd->kill_timeout); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + /* privileged interfaces follow here */ + if (!genwqe_is_privileged(cd)) { + cd->debugfs_root = root; + return 0; + } + + file = debugfs_create_file("curr_regs", S_IRUGO, root, cd, + &genwqe_curr_regs_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("curr_dbg_uid0", S_IRUGO, root, cd, + &genwqe_curr_dbg_uid0_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("curr_dbg_uid1", S_IRUGO, root, cd, + &genwqe_curr_dbg_uid1_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("curr_dbg_uid2", S_IRUGO, root, cd, + &genwqe_curr_dbg_uid2_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("prev_regs", S_IRUGO, root, cd, + &genwqe_prev_regs_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("prev_dbg_uid0", S_IRUGO, root, cd, + &genwqe_prev_dbg_uid0_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("prev_dbg_uid1", S_IRUGO, root, cd, + &genwqe_prev_dbg_uid1_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("prev_dbg_uid2", S_IRUGO, root, cd, + &genwqe_prev_dbg_uid2_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + for (i = 0; i < GENWQE_MAX_VFS; i++) { + sprintf(name, "vf%u_jobtimeout_msec", i); + + file = debugfs_create_u32(name, 0666, root, + &cd->vf_jobtimeout_msec[i]); + if (!file) { + ret = -ENOMEM; + goto err1; + } + } + + file = debugfs_create_file("jobtimer", S_IRUGO, root, cd, + &genwqe_jtimer_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_file("queue_working_time", S_IRUGO, root, cd, + &genwqe_queue_working_time_fops); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + file = debugfs_create_u32("skip_recovery", 0666, root, + &cd->skip_recovery); + if (!file) { + ret = -ENOMEM; + goto err1; + } + + cd->debugfs_root = root; + return 0; +err1: + debugfs_remove_recursive(root); +err0: + return ret; +} + +void genqwe_exit_debugfs(struct genwqe_dev *cd) +{ + debugfs_remove_recursive(cd->debugfs_root); +} diff --git a/drivers/misc/genwqe/card_dev.c b/drivers/misc/genwqe/card_dev.c new file mode 100644 index 00000000000..1d2f163a190 --- /dev/null +++ b/drivers/misc/genwqe/card_dev.c @@ -0,0 +1,1403 @@ +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +/* + * Character device representation of the GenWQE device. This allows + * user-space applications to communicate with the card. + */ + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/string.h> +#include <linux/fs.h> +#include <linux/sched.h> +#include <linux/wait.h> +#include <linux/delay.h> +#include <linux/atomic.h> + +#include "card_base.h" +#include "card_ddcb.h" + +static int genwqe_open_files(struct genwqe_dev *cd) +{ + int rc; + unsigned long flags; + + spin_lock_irqsave(&cd->file_lock, flags); + rc = list_empty(&cd->file_list); + spin_unlock_irqrestore(&cd->file_lock, flags); + return !rc; +} + +static void genwqe_add_file(struct genwqe_dev *cd, struct genwqe_file *cfile) +{ + unsigned long flags; + + cfile->owner = current; + spin_lock_irqsave(&cd->file_lock, flags); + list_add(&cfile->list, &cd->file_list); + spin_unlock_irqrestore(&cd->file_lock, flags); +} + +static int genwqe_del_file(struct genwqe_dev *cd, struct genwqe_file *cfile) +{ + unsigned long flags; + + spin_lock_irqsave(&cd->file_lock, flags); + list_del(&cfile->list); + spin_unlock_irqrestore(&cd->file_lock, flags); + + return 0; +} + +static void genwqe_add_pin(struct genwqe_file *cfile, struct dma_mapping *m) +{ + unsigned long flags; + + spin_lock_irqsave(&cfile->pin_lock, flags); + list_add(&m->pin_list, &cfile->pin_list); + spin_unlock_irqrestore(&cfile->pin_lock, flags); +} + +static int genwqe_del_pin(struct genwqe_file *cfile, struct dma_mapping *m) +{ + unsigned long flags; + + spin_lock_irqsave(&cfile->pin_lock, flags); + list_del(&m->pin_list); + spin_unlock_irqrestore(&cfile->pin_lock, flags); + + return 0; +} + +/** + * genwqe_search_pin() - Search for the mapping for a userspace address + * @cfile: Descriptor of opened file + * @u_addr: User virtual address + * @size: Size of buffer + * @dma_addr: DMA address to be updated + * + * Return: Pointer to the corresponding mapping NULL if not found + */ +static struct dma_mapping *genwqe_search_pin(struct genwqe_file *cfile, + unsigned long u_addr, + unsigned int size, + void **virt_addr) +{ + unsigned long flags; + struct dma_mapping *m; + + spin_lock_irqsave(&cfile->pin_lock, flags); + + list_for_each_entry(m, &cfile->pin_list, pin_list) { + if ((((u64)m->u_vaddr) <= (u_addr)) && + (((u64)m->u_vaddr + m->size) >= (u_addr + size))) { + + if (virt_addr) + *virt_addr = m->k_vaddr + + (u_addr - (u64)m->u_vaddr); + + spin_unlock_irqrestore(&cfile->pin_lock, flags); + return m; + } + } + spin_unlock_irqrestore(&cfile->pin_lock, flags); + return NULL; +} + +static void __genwqe_add_mapping(struct genwqe_file *cfile, + struct dma_mapping *dma_map) +{ + unsigned long flags; + + spin_lock_irqsave(&cfile->map_lock, flags); + list_add(&dma_map->card_list, &cfile->map_list); + spin_unlock_irqrestore(&cfile->map_lock, flags); +} + +static void __genwqe_del_mapping(struct genwqe_file *cfile, + struct dma_mapping *dma_map) +{ + unsigned long flags; + + spin_lock_irqsave(&cfile->map_lock, flags); + list_del(&dma_map->card_list); + spin_unlock_irqrestore(&cfile->map_lock, flags); +} + + +/** + * __genwqe_search_mapping() - Search for the mapping for a userspace address + * @cfile: descriptor of opened file + * @u_addr: user virtual address + * @size: size of buffer + * @dma_addr: DMA address to be updated + * Return: Pointer to the corresponding mapping NULL if not found + */ +static struct dma_mapping *__genwqe_search_mapping(struct genwqe_file *cfile, + unsigned long u_addr, + unsigned int size, + dma_addr_t *dma_addr, + void **virt_addr) +{ + unsigned long flags; + struct dma_mapping *m; + struct pci_dev *pci_dev = cfile->cd->pci_dev; + + spin_lock_irqsave(&cfile->map_lock, flags); + list_for_each_entry(m, &cfile->map_list, card_list) { + + if ((((u64)m->u_vaddr) <= (u_addr)) && + (((u64)m->u_vaddr + m->size) >= (u_addr + size))) { + + /* match found: current is as expected and + addr is in range */ + if (dma_addr) + *dma_addr = m->dma_addr + + (u_addr - (u64)m->u_vaddr); + + if (virt_addr) + *virt_addr = m->k_vaddr + + (u_addr - (u64)m->u_vaddr); + + spin_unlock_irqrestore(&cfile->map_lock, flags); + return m; + } + } + spin_unlock_irqrestore(&cfile->map_lock, flags); + + dev_err(&pci_dev->dev, + "[%s] Entry not found: u_addr=%lx, size=%x\n", + __func__, u_addr, size); + + return NULL; +} + +static void genwqe_remove_mappings(struct genwqe_file *cfile) +{ + int i = 0; + struct list_head *node, *next; + struct dma_mapping *dma_map; + struct genwqe_dev *cd = cfile->cd; + struct pci_dev *pci_dev = cfile->cd->pci_dev; + + list_for_each_safe(node, next, &cfile->map_list) { + dma_map = list_entry(node, struct dma_mapping, card_list); + + list_del_init(&dma_map->card_list); + + /* + * This is really a bug, because those things should + * have been already tidied up. + * + * GENWQE_MAPPING_RAW should have been removed via mmunmap(). + * GENWQE_MAPPING_SGL_TEMP should be removed by tidy up code. + */ + dev_err(&pci_dev->dev, + "[%s] %d. cleanup mapping: u_vaddr=%p " + "u_kaddr=%016lx dma_addr=%lx\n", __func__, i++, + dma_map->u_vaddr, (unsigned long)dma_map->k_vaddr, + (unsigned long)dma_map->dma_addr); + + if (dma_map->type == GENWQE_MAPPING_RAW) { + /* we allocated this dynamically */ + __genwqe_free_consistent(cd, dma_map->size, + dma_map->k_vaddr, + dma_map->dma_addr); + kfree(dma_map); + } else if (dma_map->type == GENWQE_MAPPING_SGL_TEMP) { + /* we use dma_map statically from the request */ + genwqe_user_vunmap(cd, dma_map, NULL); + } + } +} + +static void genwqe_remove_pinnings(struct genwqe_file *cfile) +{ + struct list_head *node, *next; + struct dma_mapping *dma_map; + struct genwqe_dev *cd = cfile->cd; + + list_for_each_safe(node, next, &cfile->pin_list) { + dma_map = list_entry(node, struct dma_mapping, pin_list); + + /* + * This is not a bug, because a killed processed might + * not call the unpin ioctl, which is supposed to free + * the resources. + * + * Pinnings are dymically allocated and need to be + * deleted. + */ + list_del_init(&dma_map->pin_list); + genwqe_user_vunmap(cd, dma_map, NULL); + kfree(dma_map); + } +} + +/** + * genwqe_kill_fasync() - Send signal to all processes with open GenWQE files + * + * E.g. genwqe_send_signal(cd, SIGIO); + */ +static int genwqe_kill_fasync(struct genwqe_dev *cd, int sig) +{ + unsigned int files = 0; + unsigned long flags; + struct genwqe_file *cfile; + + spin_lock_irqsave(&cd->file_lock, flags); + list_for_each_entry(cfile, &cd->file_list, list) { + if (cfile->async_queue) + kill_fasync(&cfile->async_queue, sig, POLL_HUP); + files++; + } + spin_unlock_irqrestore(&cd->file_lock, flags); + return files; +} + +static int genwqe_force_sig(struct genwqe_dev *cd, int sig) +{ + unsigned int files = 0; + unsigned long flags; + struct genwqe_file *cfile; + + spin_lock_irqsave(&cd->file_lock, flags); + list_for_each_entry(cfile, &cd->file_list, list) { + force_sig(sig, cfile->owner); + files++; + } + spin_unlock_irqrestore(&cd->file_lock, flags); + return files; +} + +/** + * genwqe_open() - file open + * @inode: file system information + * @filp: file handle + * + * This function is executed whenever an application calls + * open("/dev/genwqe",..). + * + * Return: 0 if successful or <0 if errors + */ +static int genwqe_open(struct inode *inode, struct file *filp) +{ + struct genwqe_dev *cd; + struct genwqe_file *cfile; + struct pci_dev *pci_dev; + + cfile = kzalloc(sizeof(*cfile), GFP_KERNEL); + if (cfile == NULL) + return -ENOMEM; + + cd = container_of(inode->i_cdev, struct genwqe_dev, cdev_genwqe); + pci_dev = cd->pci_dev; + cfile->cd = cd; + cfile->filp = filp; + cfile->client = NULL; + + spin_lock_init(&cfile->map_lock); /* list of raw memory allocations */ + INIT_LIST_HEAD(&cfile->map_list); + + spin_lock_init(&cfile->pin_lock); /* list of user pinned memory */ + INIT_LIST_HEAD(&cfile->pin_list); + + filp->private_data = cfile; + + genwqe_add_file(cd, cfile); + return 0; +} + +/** + * genwqe_fasync() - Setup process to receive SIGIO. + * @fd: file descriptor + * @filp: file handle + * @mode: file mode + * + * Sending a signal is working as following: + * + * if (cdev->async_queue) + * kill_fasync(&cdev->async_queue, SIGIO, POLL_IN); + * + * Some devices also implement asynchronous notification to indicate + * when the device can be written; in this case, of course, + * kill_fasync must be called with a mode of POLL_OUT. + */ +static int genwqe_fasync(int fd, struct file *filp, int mode) +{ + struct genwqe_file *cdev = (struct genwqe_file *)filp->private_data; + return fasync_helper(fd, filp, mode, &cdev->async_queue); +} + + +/** + * genwqe_release() - file close + * @inode: file system information + * @filp: file handle + * + * This function is executed whenever an application calls 'close(fd_genwqe)' + * + * Return: always 0 + */ +static int genwqe_release(struct inode *inode, struct file *filp) +{ + struct genwqe_file *cfile = (struct genwqe_file *)filp->private_data; + struct genwqe_dev *cd = cfile->cd; + + /* there must be no entries in these lists! */ + genwqe_remove_mappings(cfile); + genwqe_remove_pinnings(cfile); + + /* remove this filp from the asynchronously notified filp's */ + genwqe_fasync(-1, filp, 0); + + /* + * For this to work we must not release cd when this cfile is + * not yet released, otherwise the list entry is invalid, + * because the list itself gets reinstantiated! + */ + genwqe_del_file(cd, cfile); + kfree(cfile); + return 0; +} + +static void genwqe_vma_open(struct vm_area_struct *vma) +{ + /* nothing ... */ +} + +/** + * genwqe_vma_close() - Called each time when vma is unmapped + * + * Free memory which got allocated by GenWQE mmap(). + */ +static void genwqe_vma_close(struct vm_area_struct *vma) +{ + unsigned long vsize = vma->vm_end - vma->vm_start; + struct inode *inode = vma->vm_file->f_dentry->d_inode; + struct dma_mapping *dma_map; + struct genwqe_dev *cd = container_of(inode->i_cdev, struct genwqe_dev, + cdev_genwqe); + struct pci_dev *pci_dev = cd->pci_dev; + dma_addr_t d_addr = 0; + struct genwqe_file *cfile = vma->vm_private_data; + + dma_map = __genwqe_search_mapping(cfile, vma->vm_start, vsize, + &d_addr, NULL); + if (dma_map == NULL) { + dev_err(&pci_dev->dev, + " [%s] err: mapping not found: v=%lx, p=%lx s=%lx\n", + __func__, vma->vm_start, vma->vm_pgoff << PAGE_SHIFT, + vsize); + return; + } + __genwqe_del_mapping(cfile, dma_map); + __genwqe_free_consistent(cd, dma_map->size, dma_map->k_vaddr, + dma_map->dma_addr); + kfree(dma_map); +} + +static struct vm_operations_struct genwqe_vma_ops = { + .open = genwqe_vma_open, + .close = genwqe_vma_close, +}; + +/** + * genwqe_mmap() - Provide contignous buffers to userspace + * + * We use mmap() to allocate contignous buffers used for DMA + * transfers. After the buffer is allocated we remap it to user-space + * and remember a reference to our dma_mapping data structure, where + * we store the associated DMA address and allocated size. + * + * When we receive a DDCB execution request with the ATS bits set to + * plain buffer, we lookup our dma_mapping list to find the + * corresponding DMA address for the associated user-space address. + */ +static int genwqe_mmap(struct file *filp, struct vm_area_struct *vma) +{ + int rc; + unsigned long pfn, vsize = vma->vm_end - vma->vm_start; + struct genwqe_file *cfile = (struct genwqe_file *)filp->private_data; + struct genwqe_dev *cd = cfile->cd; + struct dma_mapping *dma_map; + + if (vsize == 0) + return -EINVAL; + + if (get_order(vsize) > MAX_ORDER) + return -ENOMEM; + + dma_map = kzalloc(sizeof(struct dma_mapping), GFP_ATOMIC); + if (dma_map == NULL) + return -ENOMEM; + + genwqe_mapping_init(dma_map, GENWQE_MAPPING_RAW); + dma_map->u_vaddr = (void *)vma->vm_start; + dma_map->size = vsize; + dma_map->nr_pages = DIV_ROUND_UP(vsize, PAGE_SIZE); + dma_map->k_vaddr = __genwqe_alloc_consistent(cd, vsize, + &dma_map->dma_addr); + if (dma_map->k_vaddr == NULL) { + rc = -ENOMEM; + goto free_dma_map; + } + + if (capable(CAP_SYS_ADMIN) && (vsize > sizeof(dma_addr_t))) + *(dma_addr_t *)dma_map->k_vaddr = dma_map->dma_addr; + + pfn = virt_to_phys(dma_map->k_vaddr) >> PAGE_SHIFT; + rc = remap_pfn_range(vma, + vma->vm_start, + pfn, + vsize, + vma->vm_page_prot); + if (rc != 0) { + rc = -EFAULT; + goto free_dma_mem; + } + + vma->vm_private_data = cfile; + vma->vm_ops = &genwqe_vma_ops; + __genwqe_add_mapping(cfile, dma_map); + + return 0; + + free_dma_mem: + __genwqe_free_consistent(cd, dma_map->size, + dma_map->k_vaddr, + dma_map->dma_addr); + free_dma_map: + kfree(dma_map); + return rc; +} + +/** + * do_flash_update() - Excute flash update (write image or CVPD) + * @cd: genwqe device + * @load: details about image load + * + * Return: 0 if successful + */ + +#define FLASH_BLOCK 0x40000 /* we use 256k blocks */ + +static int do_flash_update(struct genwqe_file *cfile, + struct genwqe_bitstream *load) +{ + int rc = 0; + int blocks_to_flash; + dma_addr_t dma_addr; + u64 flash = 0; + size_t tocopy = 0; + u8 __user *buf; + u8 *xbuf; + u32 crc; + u8 cmdopts; + struct genwqe_dev *cd = cfile->cd; + struct pci_dev *pci_dev = cd->pci_dev; + + if ((load->size & 0x3) != 0) + return -EINVAL; + + if (((unsigned long)(load->data_addr) & ~PAGE_MASK) != 0) + return -EINVAL; + + /* FIXME Bits have changed for new service layer! */ + switch ((char)load->partition) { + case '0': + cmdopts = 0x14; + break; /* download/erase_first/part_0 */ + case '1': + cmdopts = 0x1C; + break; /* download/erase_first/part_1 */ + case 'v': + cmdopts = 0x0C; + break; /* download/erase_first/vpd */ + default: + return -EINVAL; + } + + buf = (u8 __user *)load->data_addr; + xbuf = __genwqe_alloc_consistent(cd, FLASH_BLOCK, &dma_addr); + if (xbuf == NULL) + return -ENOMEM; + + blocks_to_flash = load->size / FLASH_BLOCK; + while (load->size) { + struct genwqe_ddcb_cmd *req; + + /* + * We must be 4 byte aligned. Buffer must be 0 appened + * to have defined values when calculating CRC. + */ + tocopy = min_t(size_t, load->size, FLASH_BLOCK); + + rc = copy_from_user(xbuf, buf, tocopy); + if (rc) { + rc = -EFAULT; + goto free_buffer; + } + crc = genwqe_crc32(xbuf, tocopy, 0xffffffff); + + dev_dbg(&pci_dev->dev, + "[%s] DMA: %lx CRC: %08x SZ: %ld %d\n", + __func__, (unsigned long)dma_addr, crc, tocopy, + blocks_to_flash); + + /* prepare DDCB for SLU process */ + req = ddcb_requ_alloc(); + if (req == NULL) { + rc = -ENOMEM; + goto free_buffer; + } + + req->cmd = SLCMD_MOVE_FLASH; + req->cmdopts = cmdopts; + + /* prepare invariant values */ + if (genwqe_get_slu_id(cd) <= 0x2) { + *(__be64 *)&req->__asiv[0] = cpu_to_be64(dma_addr); + *(__be64 *)&req->__asiv[8] = cpu_to_be64(tocopy); + *(__be64 *)&req->__asiv[16] = cpu_to_be64(flash); + *(__be32 *)&req->__asiv[24] = cpu_to_be32(0); + req->__asiv[24] = load->uid; + *(__be32 *)&req->__asiv[28] = cpu_to_be32(crc); + + /* for simulation only */ + *(__be64 *)&req->__asiv[88] = cpu_to_be64(load->slu_id); + *(__be64 *)&req->__asiv[96] = cpu_to_be64(load->app_id); + req->asiv_length = 32; /* bytes included in crc calc */ + } else { /* setup DDCB for ATS architecture */ + *(__be64 *)&req->asiv[0] = cpu_to_be64(dma_addr); + *(__be32 *)&req->asiv[8] = cpu_to_be32(tocopy); + *(__be32 *)&req->asiv[12] = cpu_to_be32(0); /* resvd */ + *(__be64 *)&req->asiv[16] = cpu_to_be64(flash); + *(__be32 *)&req->asiv[24] = cpu_to_be32(load->uid<<24); + *(__be32 *)&req->asiv[28] = cpu_to_be32(crc); + + /* for simulation only */ + *(__be64 *)&req->asiv[80] = cpu_to_be64(load->slu_id); + *(__be64 *)&req->asiv[88] = cpu_to_be64(load->app_id); + + /* Rd only */ + req->ats = 0x4ULL << 44; + req->asiv_length = 40; /* bytes included in crc calc */ + } + req->asv_length = 8; + + /* For Genwqe5 we get back the calculated CRC */ + *(u64 *)&req->asv[0] = 0ULL; /* 0x80 */ + + rc = __genwqe_execute_raw_ddcb(cd, req); + + load->retc = req->retc; + load->attn = req->attn; + load->progress = req->progress; + + if (rc < 0) { + ddcb_requ_free(req); + goto free_buffer; + } + + if (req->retc != DDCB_RETC_COMPLETE) { + rc = -EIO; + ddcb_requ_free(req); + goto free_buffer; + } + + load->size -= tocopy; + flash += tocopy; + buf += tocopy; + blocks_to_flash--; + ddcb_requ_free(req); + } + + free_buffer: + __genwqe_free_consistent(cd, FLASH_BLOCK, xbuf, dma_addr); + return rc; +} + +static int do_flash_read(struct genwqe_file *cfile, + struct genwqe_bitstream *load) +{ + int rc, blocks_to_flash; + dma_addr_t dma_addr; + u64 flash = 0; + size_t tocopy = 0; + u8 __user *buf; + u8 *xbuf; + u8 cmdopts; + struct genwqe_dev *cd = cfile->cd; + struct pci_dev *pci_dev = cd->pci_dev; + struct genwqe_ddcb_cmd *cmd; + + if ((load->size & 0x3) != 0) + return -EINVAL; + + if (((unsigned long)(load->data_addr) & ~PAGE_MASK) != 0) + return -EINVAL; + + /* FIXME Bits have changed for new service layer! */ + switch ((char)load->partition) { + case '0': + cmdopts = 0x12; + break; /* upload/part_0 */ + case '1': + cmdopts = 0x1A; + break; /* upload/part_1 */ + case 'v': + cmdopts = 0x0A; + break; /* upload/vpd */ + default: + return -EINVAL; + } + + buf = (u8 __user *)load->data_addr; + xbuf = __genwqe_alloc_consistent(cd, FLASH_BLOCK, &dma_addr); + if (xbuf == NULL) + return -ENOMEM; + + blocks_to_flash = load->size / FLASH_BLOCK; + while (load->size) { + /* + * We must be 4 byte aligned. Buffer must be 0 appened + * to have defined values when calculating CRC. + */ + tocopy = min_t(size_t, load->size, FLASH_BLOCK); + + dev_dbg(&pci_dev->dev, + "[%s] DMA: %lx SZ: %ld %d\n", + __func__, (unsigned long)dma_addr, tocopy, + blocks_to_flash); + + /* prepare DDCB for SLU process */ + cmd = ddcb_requ_alloc(); + if (cmd == NULL) { + rc = -ENOMEM; + goto free_buffer; + } + cmd->cmd = SLCMD_MOVE_FLASH; + cmd->cmdopts = cmdopts; + + /* prepare invariant values */ + if (genwqe_get_slu_id(cd) <= 0x2) { + *(__be64 *)&cmd->__asiv[0] = cpu_to_be64(dma_addr); + *(__be64 *)&cmd->__asiv[8] = cpu_to_be64(tocopy); + *(__be64 *)&cmd->__asiv[16] = cpu_to_be64(flash); + *(__be32 *)&cmd->__asiv[24] = cpu_to_be32(0); + cmd->__asiv[24] = load->uid; + *(__be32 *)&cmd->__asiv[28] = cpu_to_be32(0) /* CRC */; + cmd->asiv_length = 32; /* bytes included in crc calc */ + } else { /* setup DDCB for ATS architecture */ + *(__be64 *)&cmd->asiv[0] = cpu_to_be64(dma_addr); + *(__be32 *)&cmd->asiv[8] = cpu_to_be32(tocopy); + *(__be32 *)&cmd->asiv[12] = cpu_to_be32(0); /* resvd */ + *(__be64 *)&cmd->asiv[16] = cpu_to_be64(flash); + *(__be32 *)&cmd->asiv[24] = cpu_to_be32(load->uid<<24); + *(__be32 *)&cmd->asiv[28] = cpu_to_be32(0); /* CRC */ + + /* rd/wr */ + cmd->ats = 0x5ULL << 44; + cmd->asiv_length = 40; /* bytes included in crc calc */ + } + cmd->asv_length = 8; + + /* we only get back the calculated CRC */ + *(u64 *)&cmd->asv[0] = 0ULL; /* 0x80 */ + + rc = __genwqe_execute_raw_ddcb(cd, cmd); + + load->retc = cmd->retc; + load->attn = cmd->attn; + load->progress = cmd->progress; + + if ((rc < 0) && (rc != -EBADMSG)) { + ddcb_requ_free(cmd); + goto free_buffer; + } + + rc = copy_to_user(buf, xbuf, tocopy); + if (rc) { + rc = -EFAULT; + ddcb_requ_free(cmd); + goto free_buffer; + } + + /* We know that we can get retc 0x104 with CRC err */ + if (((cmd->retc == DDCB_RETC_FAULT) && + (cmd->attn != 0x02)) || /* Normally ignore CRC error */ + ((cmd->retc == DDCB_RETC_COMPLETE) && + (cmd->attn != 0x00))) { /* Everything was fine */ + rc = -EIO; + ddcb_requ_free(cmd); + goto free_buffer; + } + + load->size -= tocopy; + flash += tocopy; + buf += tocopy; + blocks_to_flash--; + ddcb_requ_free(cmd); + } + rc = 0; + + free_buffer: + __genwqe_free_consistent(cd, FLASH_BLOCK, xbuf, dma_addr); + return rc; +} + +static int genwqe_pin_mem(struct genwqe_file *cfile, struct genwqe_mem *m) +{ + int rc; + struct genwqe_dev *cd = cfile->cd; + struct pci_dev *pci_dev = cfile->cd->pci_dev; + struct dma_mapping *dma_map; + unsigned long map_addr; + unsigned long map_size; + + if ((m->addr == 0x0) || (m->size == 0)) + return -EINVAL; + + map_addr = (m->addr & PAGE_MASK); + map_size = round_up(m->size + (m->addr & ~PAGE_MASK), PAGE_SIZE); + + dma_map = kzalloc(sizeof(struct dma_mapping), GFP_ATOMIC); + if (dma_map == NULL) + return -ENOMEM; + + genwqe_mapping_init(dma_map, GENWQE_MAPPING_SGL_PINNED); + rc = genwqe_user_vmap(cd, dma_map, (void *)map_addr, map_size, NULL); + if (rc != 0) { + dev_err(&pci_dev->dev, + "[%s] genwqe_user_vmap rc=%d\n", __func__, rc); + kfree(dma_map); + return rc; + } + + genwqe_add_pin(cfile, dma_map); + return 0; +} + +static int genwqe_unpin_mem(struct genwqe_file *cfile, struct genwqe_mem *m) +{ + struct genwqe_dev *cd = cfile->cd; + struct dma_mapping *dma_map; + unsigned long map_addr; + unsigned long map_size; + + if (m->addr == 0x0) + return -EINVAL; + + map_addr = (m->addr & PAGE_MASK); + map_size = round_up(m->size + (m->addr & ~PAGE_MASK), PAGE_SIZE); + + dma_map = genwqe_search_pin(cfile, map_addr, map_size, NULL); + if (dma_map == NULL) + return -ENOENT; + + genwqe_del_pin(cfile, dma_map); + genwqe_user_vunmap(cd, dma_map, NULL); + kfree(dma_map); + return 0; +} + +/** + * ddcb_cmd_cleanup() - Remove dynamically created fixup entries + * + * Only if there are any. Pinnings are not removed. + */ +static int ddcb_cmd_cleanup(struct genwqe_file *cfile, struct ddcb_requ *req) +{ + unsigned int i; + struct dma_mapping *dma_map; + struct genwqe_dev *cd = cfile->cd; + + for (i = 0; i < DDCB_FIXUPS; i++) { + dma_map = &req->dma_mappings[i]; + + if (dma_mapping_used(dma_map)) { + __genwqe_del_mapping(cfile, dma_map); + genwqe_user_vunmap(cd, dma_map, req); + } + if (req->sgls[i].sgl != NULL) + genwqe_free_sync_sgl(cd, &req->sgls[i]); + } + return 0; +} + +/** + * ddcb_cmd_fixups() - Establish DMA fixups/sglists for user memory references + * + * Before the DDCB gets executed we need to handle the fixups. We + * replace the user-space addresses with DMA addresses or do + * additional setup work e.g. generating a scatter-gather list which + * is used to describe the memory referred to in the fixup. + */ +static int ddcb_cmd_fixups(struct genwqe_file *cfile, struct ddcb_requ *req) +{ + int rc; + unsigned int asiv_offs, i; + struct genwqe_dev *cd = cfile->cd; + struct genwqe_ddcb_cmd *cmd = &req->cmd; + struct dma_mapping *m; + const char *type = "UNKNOWN"; + + for (i = 0, asiv_offs = 0x00; asiv_offs <= 0x58; + i++, asiv_offs += 0x08) { + + u64 u_addr; + dma_addr_t d_addr; + u32 u_size = 0; + u64 ats_flags; + + ats_flags = ATS_GET_FLAGS(cmd->ats, asiv_offs); + + switch (ats_flags) { + + case ATS_TYPE_DATA: + break; /* nothing to do here */ + + case ATS_TYPE_FLAT_RDWR: + case ATS_TYPE_FLAT_RD: { + u_addr = be64_to_cpu(*((__be64 *)&cmd-> + asiv[asiv_offs])); + u_size = be32_to_cpu(*((__be32 *)&cmd-> + asiv[asiv_offs + 0x08])); + + /* + * No data available. Ignore u_addr in this + * case and set addr to 0. Hardware must not + * fetch the buffer. + */ + if (u_size == 0x0) { + *((__be64 *)&cmd->asiv[asiv_offs]) = + cpu_to_be64(0x0); + break; + } + + m = __genwqe_search_mapping(cfile, u_addr, u_size, + &d_addr, NULL); + if (m == NULL) { + rc = -EFAULT; + goto err_out; + } + + *((__be64 *)&cmd->asiv[asiv_offs]) = + cpu_to_be64(d_addr); + break; + } + + case ATS_TYPE_SGL_RDWR: + case ATS_TYPE_SGL_RD: { + int page_offs; + + u_addr = be64_to_cpu(*((__be64 *) + &cmd->asiv[asiv_offs])); + u_size = be32_to_cpu(*((__be32 *) + &cmd->asiv[asiv_offs + 0x08])); + + /* + * No data available. Ignore u_addr in this + * case and set addr to 0. Hardware must not + * fetch the empty sgl. + */ + if (u_size == 0x0) { + *((__be64 *)&cmd->asiv[asiv_offs]) = + cpu_to_be64(0x0); + break; + } + + m = genwqe_search_pin(cfile, u_addr, u_size, NULL); + if (m != NULL) { + type = "PINNING"; + page_offs = (u_addr - + (u64)m->u_vaddr)/PAGE_SIZE; + } else { + type = "MAPPING"; + m = &req->dma_mappings[i]; + + genwqe_mapping_init(m, + GENWQE_MAPPING_SGL_TEMP); + rc = genwqe_user_vmap(cd, m, (void *)u_addr, + u_size, req); + if (rc != 0) + goto err_out; + + __genwqe_add_mapping(cfile, m); + page_offs = 0; + } + + /* create genwqe style scatter gather list */ + rc = genwqe_alloc_sync_sgl(cd, &req->sgls[i], + (void __user *)u_addr, + u_size); + if (rc != 0) + goto err_out; + + genwqe_setup_sgl(cd, &req->sgls[i], + &m->dma_list[page_offs]); + + *((__be64 *)&cmd->asiv[asiv_offs]) = + cpu_to_be64(req->sgls[i].sgl_dma_addr); + + break; + } + default: + rc = -EINVAL; + goto err_out; + } + } + return 0; + + err_out: + ddcb_cmd_cleanup(cfile, req); + return rc; +} + +/** + * genwqe_execute_ddcb() - Execute DDCB using userspace address fixups + * + * The code will build up the translation tables or lookup the + * contignous memory allocation table to find the right translations + * and DMA addresses. + */ +static int genwqe_execute_ddcb(struct genwqe_file *cfile, + struct genwqe_ddcb_cmd *cmd) +{ + int rc; + struct genwqe_dev *cd = cfile->cd; + struct ddcb_requ *req = container_of(cmd, struct ddcb_requ, cmd); + + rc = ddcb_cmd_fixups(cfile, req); + if (rc != 0) + return rc; + + rc = __genwqe_execute_raw_ddcb(cd, cmd); + ddcb_cmd_cleanup(cfile, req); + return rc; +} + +static int do_execute_ddcb(struct genwqe_file *cfile, + unsigned long arg, int raw) +{ + int rc; + struct genwqe_ddcb_cmd *cmd; + struct ddcb_requ *req; + struct genwqe_dev *cd = cfile->cd; + + cmd = ddcb_requ_alloc(); + if (cmd == NULL) + return -ENOMEM; + + req = container_of(cmd, struct ddcb_requ, cmd); + + if (copy_from_user(cmd, (void __user *)arg, sizeof(*cmd))) { + ddcb_requ_free(cmd); + return -EFAULT; + } + + if (!raw) + rc = genwqe_execute_ddcb(cfile, cmd); + else + rc = __genwqe_execute_raw_ddcb(cd, cmd); + + /* Copy back only the modifed fields. Do not copy ASIV + back since the copy got modified by the driver. */ + if (copy_to_user((void __user *)arg, cmd, + sizeof(*cmd) - DDCB_ASIV_LENGTH)) { + ddcb_requ_free(cmd); + return -EFAULT; + } + + ddcb_requ_free(cmd); + return rc; +} + +/** + * genwqe_ioctl() - IO control + * @filp: file handle + * @cmd: command identifier (passed from user) + * @arg: argument (passed from user) + * + * Return: 0 success + */ +static long genwqe_ioctl(struct file *filp, unsigned int cmd, + unsigned long arg) +{ + int rc = 0; + struct genwqe_file *cfile = (struct genwqe_file *)filp->private_data; + struct genwqe_dev *cd = cfile->cd; + struct genwqe_reg_io __user *io; + u64 val; + u32 reg_offs; + + if (_IOC_TYPE(cmd) != GENWQE_IOC_CODE) + return -EINVAL; + + switch (cmd) { + + case GENWQE_GET_CARD_STATE: + put_user(cd->card_state, (enum genwqe_card_state __user *)arg); + return 0; + + /* Register access */ + case GENWQE_READ_REG64: { + io = (struct genwqe_reg_io __user *)arg; + + if (get_user(reg_offs, &io->num)) + return -EFAULT; + + if ((reg_offs >= cd->mmio_len) || (reg_offs & 0x7)) + return -EINVAL; + + val = __genwqe_readq(cd, reg_offs); + put_user(val, &io->val64); + return 0; + } + + case GENWQE_WRITE_REG64: { + io = (struct genwqe_reg_io __user *)arg; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if ((filp->f_flags & O_ACCMODE) == O_RDONLY) + return -EPERM; + + if (get_user(reg_offs, &io->num)) + return -EFAULT; + + if ((reg_offs >= cd->mmio_len) || (reg_offs & 0x7)) + return -EINVAL; + + if (get_user(val, &io->val64)) + return -EFAULT; + + __genwqe_writeq(cd, reg_offs, val); + return 0; + } + + case GENWQE_READ_REG32: { + io = (struct genwqe_reg_io __user *)arg; + + if (get_user(reg_offs, &io->num)) + return -EFAULT; + + if ((reg_offs >= cd->mmio_len) || (reg_offs & 0x3)) + return -EINVAL; + + val = __genwqe_readl(cd, reg_offs); + put_user(val, &io->val64); + return 0; + } + + case GENWQE_WRITE_REG32: { + io = (struct genwqe_reg_io __user *)arg; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if ((filp->f_flags & O_ACCMODE) == O_RDONLY) + return -EPERM; + + if (get_user(reg_offs, &io->num)) + return -EFAULT; + + if ((reg_offs >= cd->mmio_len) || (reg_offs & 0x3)) + return -EINVAL; + + if (get_user(val, &io->val64)) + return -EFAULT; + + __genwqe_writel(cd, reg_offs, val); + return 0; + } + + /* Flash update/reading */ + case GENWQE_SLU_UPDATE: { + struct genwqe_bitstream load; + + if (!genwqe_is_privileged(cd)) + return -EPERM; + + if ((filp->f_flags & O_ACCMODE) == O_RDONLY) + return -EPERM; + + if (copy_from_user(&load, (void __user *)arg, + sizeof(load))) + return -EFAULT; + + rc = do_flash_update(cfile, &load); + + if (copy_to_user((void __user *)arg, &load, sizeof(load))) + return -EFAULT; + + return rc; + } + + case GENWQE_SLU_READ: { + struct genwqe_bitstream load; + + if (!genwqe_is_privileged(cd)) + return -EPERM; + + if (genwqe_flash_readback_fails(cd)) + return -ENOSPC; /* known to fail for old versions */ + + if (copy_from_user(&load, (void __user *)arg, sizeof(load))) + return -EFAULT; + + rc = do_flash_read(cfile, &load); + + if (copy_to_user((void __user *)arg, &load, sizeof(load))) + return -EFAULT; + + return rc; + } + + /* memory pinning and unpinning */ + case GENWQE_PIN_MEM: { + struct genwqe_mem m; + + if (copy_from_user(&m, (void __user *)arg, sizeof(m))) + return -EFAULT; + + return genwqe_pin_mem(cfile, &m); + } + + case GENWQE_UNPIN_MEM: { + struct genwqe_mem m; + + if (copy_from_user(&m, (void __user *)arg, sizeof(m))) + return -EFAULT; + + return genwqe_unpin_mem(cfile, &m); + } + + /* launch an DDCB and wait for completion */ + case GENWQE_EXECUTE_DDCB: + return do_execute_ddcb(cfile, arg, 0); + + case GENWQE_EXECUTE_RAW_DDCB: { + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + return do_execute_ddcb(cfile, arg, 1); + } + + default: + return -EINVAL; + } + + return rc; +} + +#if defined(CONFIG_COMPAT) +/** + * genwqe_compat_ioctl() - Compatibility ioctl + * + * Called whenever a 32-bit process running under a 64-bit kernel + * performs an ioctl on /dev/genwqe<n>_card. + * + * @filp: file pointer. + * @cmd: command. + * @arg: user argument. + * Return: zero on success or negative number on failure. + */ +static long genwqe_compat_ioctl(struct file *filp, unsigned int cmd, + unsigned long arg) +{ + return genwqe_ioctl(filp, cmd, arg); +} +#endif /* defined(CONFIG_COMPAT) */ + +static const struct file_operations genwqe_fops = { + .owner = THIS_MODULE, + .open = genwqe_open, + .fasync = genwqe_fasync, + .mmap = genwqe_mmap, + .unlocked_ioctl = genwqe_ioctl, +#if defined(CONFIG_COMPAT) + .compat_ioctl = genwqe_compat_ioctl, +#endif + .release = genwqe_release, +}; + +static int genwqe_device_initialized(struct genwqe_dev *cd) +{ + return cd->dev != NULL; +} + +/** + * genwqe_device_create() - Create and configure genwqe char device + * @cd: genwqe device descriptor + * + * This function must be called before we create any more genwqe + * character devices, because it is allocating the major and minor + * number which are supposed to be used by the client drivers. + */ +int genwqe_device_create(struct genwqe_dev *cd) +{ + int rc; + struct pci_dev *pci_dev = cd->pci_dev; + + /* + * Here starts the individual setup per client. It must + * initialize its own cdev data structure with its own fops. + * The appropriate devnum needs to be created. The ranges must + * not overlap. + */ + rc = alloc_chrdev_region(&cd->devnum_genwqe, 0, + GENWQE_MAX_MINOR, GENWQE_DEVNAME); + if (rc < 0) { + dev_err(&pci_dev->dev, "err: alloc_chrdev_region failed\n"); + goto err_dev; + } + + cdev_init(&cd->cdev_genwqe, &genwqe_fops); + cd->cdev_genwqe.owner = THIS_MODULE; + + rc = cdev_add(&cd->cdev_genwqe, cd->devnum_genwqe, 1); + if (rc < 0) { + dev_err(&pci_dev->dev, "err: cdev_add failed\n"); + goto err_add; + } + + /* + * Finally the device in /dev/... must be created. The rule is + * to use card%d_clientname for each created device. + */ + cd->dev = device_create_with_groups(cd->class_genwqe, + &cd->pci_dev->dev, + cd->devnum_genwqe, cd, + genwqe_attribute_groups, + GENWQE_DEVNAME "%u_card", + cd->card_idx); + if (IS_ERR(cd->dev)) { + rc = PTR_ERR(cd->dev); + goto err_cdev; + } + + rc = genwqe_init_debugfs(cd); + if (rc != 0) + goto err_debugfs; + + return 0; + + err_debugfs: + device_destroy(cd->class_genwqe, cd->devnum_genwqe); + err_cdev: + cdev_del(&cd->cdev_genwqe); + err_add: + unregister_chrdev_region(cd->devnum_genwqe, GENWQE_MAX_MINOR); + err_dev: + cd->dev = NULL; + return rc; +} + +static int genwqe_inform_and_stop_processes(struct genwqe_dev *cd) +{ + int rc; + unsigned int i; + struct pci_dev *pci_dev = cd->pci_dev; + + if (!genwqe_open_files(cd)) + return 0; + + dev_warn(&pci_dev->dev, "[%s] send SIGIO and wait ...\n", __func__); + + rc = genwqe_kill_fasync(cd, SIGIO); + if (rc > 0) { + /* give kill_timeout seconds to close file descriptors ... */ + for (i = 0; (i < genwqe_kill_timeout) && + genwqe_open_files(cd); i++) { + dev_info(&pci_dev->dev, " %d sec ...", i); + + cond_resched(); + msleep(1000); + } + + /* if no open files we can safely continue, else ... */ + if (!genwqe_open_files(cd)) + return 0; + + dev_warn(&pci_dev->dev, + "[%s] send SIGKILL and wait ...\n", __func__); + + rc = genwqe_force_sig(cd, SIGKILL); /* force terminate */ + if (rc) { + /* Give kill_timout more seconds to end processes */ + for (i = 0; (i < genwqe_kill_timeout) && + genwqe_open_files(cd); i++) { + dev_warn(&pci_dev->dev, " %d sec ...", i); + + cond_resched(); + msleep(1000); + } + } + } + return 0; +} + +/** + * genwqe_device_remove() - Remove genwqe's char device + * + * This function must be called after the client devices are removed + * because it will free the major/minor number range for the genwqe + * drivers. + * + * This function must be robust enough to be called twice. + */ +int genwqe_device_remove(struct genwqe_dev *cd) +{ + int rc; + struct pci_dev *pci_dev = cd->pci_dev; + + if (!genwqe_device_initialized(cd)) + return 1; + + genwqe_inform_and_stop_processes(cd); + + /* + * We currently do wait until all filedescriptors are + * closed. This leads to a problem when we abort the + * application which will decrease this reference from + * 1/unused to 0/illegal and not from 2/used 1/empty. + */ + rc = atomic_read(&cd->cdev_genwqe.kobj.kref.refcount); + if (rc != 1) { + dev_err(&pci_dev->dev, + "[%s] err: cdev_genwqe...refcount=%d\n", __func__, rc); + panic("Fatal err: cannot free resources with pending references!"); + } + + genqwe_exit_debugfs(cd); + device_destroy(cd->class_genwqe, cd->devnum_genwqe); + cdev_del(&cd->cdev_genwqe); + unregister_chrdev_region(cd->devnum_genwqe, GENWQE_MAX_MINOR); + cd->dev = NULL; + + return 0; +} diff --git a/drivers/misc/genwqe/card_sysfs.c b/drivers/misc/genwqe/card_sysfs.c new file mode 100644 index 00000000000..a72a99266c3 --- /dev/null +++ b/drivers/misc/genwqe/card_sysfs.c @@ -0,0 +1,288 @@ +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +/* + * Sysfs interfaces for the GenWQE card. There are attributes to query + * the version of the bitstream as well as some for the driver. For + * debugging, please also see the debugfs interfaces of this driver. + */ + +#include <linux/version.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/string.h> +#include <linux/fs.h> +#include <linux/sysfs.h> +#include <linux/ctype.h> +#include <linux/device.h> + +#include "card_base.h" +#include "card_ddcb.h" + +static const char * const genwqe_types[] = { + [GENWQE_TYPE_ALTERA_230] = "GenWQE4-230", + [GENWQE_TYPE_ALTERA_530] = "GenWQE4-530", + [GENWQE_TYPE_ALTERA_A4] = "GenWQE5-A4", + [GENWQE_TYPE_ALTERA_A7] = "GenWQE5-A7", +}; + +static ssize_t status_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct genwqe_dev *cd = dev_get_drvdata(dev); + const char *cs[GENWQE_CARD_STATE_MAX] = { "unused", "used", "error" }; + + return sprintf(buf, "%s\n", cs[cd->card_state]); +} +static DEVICE_ATTR_RO(status); + +static ssize_t appid_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + char app_name[5]; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + genwqe_read_app_id(cd, app_name, sizeof(app_name)); + return sprintf(buf, "%s\n", app_name); +} +static DEVICE_ATTR_RO(appid); + +static ssize_t version_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u64 slu_id, app_id; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + slu_id = __genwqe_readq(cd, IO_SLU_UNITCFG); + app_id = __genwqe_readq(cd, IO_APP_UNITCFG); + + return sprintf(buf, "%016llx.%016llx\n", slu_id, app_id); +} +static DEVICE_ATTR_RO(version); + +static ssize_t type_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u8 card_type; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + card_type = genwqe_card_type(cd); + return sprintf(buf, "%s\n", (card_type >= ARRAY_SIZE(genwqe_types)) ? + "invalid" : genwqe_types[card_type]); +} +static DEVICE_ATTR_RO(type); + +static ssize_t driver_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + return sprintf(buf, "%s\n", DRV_VERS_STRING); +} +static DEVICE_ATTR_RO(driver); + +static ssize_t tempsens_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u64 tempsens; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + tempsens = __genwqe_readq(cd, IO_SLU_TEMPERATURE_SENSOR); + return sprintf(buf, "%016llx\n", tempsens); +} +static DEVICE_ATTR_RO(tempsens); + +static ssize_t freerunning_timer_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + u64 t; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + t = __genwqe_readq(cd, IO_SLC_FREE_RUNNING_TIMER); + return sprintf(buf, "%016llx\n", t); +} +static DEVICE_ATTR_RO(freerunning_timer); + +static ssize_t queue_working_time_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + u64 t; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + t = __genwqe_readq(cd, IO_SLC_QUEUE_WTIME); + return sprintf(buf, "%016llx\n", t); +} +static DEVICE_ATTR_RO(queue_working_time); + +static ssize_t base_clock_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + u64 base_clock; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + base_clock = genwqe_base_clock_frequency(cd); + return sprintf(buf, "%lld\n", base_clock); +} +static DEVICE_ATTR_RO(base_clock); + +/** + * curr_bitstream_show() - Show the current bitstream id + * + * There is a bug in some old versions of the CPLD which selects the + * bitstream, which causes the IO_SLU_BITSTREAM register to report + * unreliable data in very rare cases. This makes this sysfs + * unreliable up to the point were a new CPLD version is being used. + * + * Unfortunately there is no automatic way yet to query the CPLD + * version, such that you need to manually ensure via programming + * tools that you have a recent version of the CPLD software. + * + * The proposed circumvention is to use a special recovery bitstream + * on the backup partition (0) to identify problems while loading the + * image. + */ +static ssize_t curr_bitstream_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int curr_bitstream; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + curr_bitstream = __genwqe_readq(cd, IO_SLU_BITSTREAM) & 0x1; + return sprintf(buf, "%d\n", curr_bitstream); +} +static DEVICE_ATTR_RO(curr_bitstream); + +/** + * next_bitstream_show() - Show the next activated bitstream + * + * IO_SLC_CFGREG_SOFTRESET: This register can only be accessed by the PF. + */ +static ssize_t next_bitstream_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + int next_bitstream; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + switch ((cd->softreset & 0xc) >> 2) { + case 0x2: + next_bitstream = 0; + break; + case 0x3: + next_bitstream = 1; + break; + default: + next_bitstream = -1; + break; /* error */ + } + return sprintf(buf, "%d\n", next_bitstream); +} + +static ssize_t next_bitstream_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int partition; + struct genwqe_dev *cd = dev_get_drvdata(dev); + + if (kstrtoint(buf, 0, &partition) < 0) + return -EINVAL; + + switch (partition) { + case 0x0: + cd->softreset = 0x78; + break; + case 0x1: + cd->softreset = 0x7c; + break; + default: + return -EINVAL; + } + + __genwqe_writeq(cd, IO_SLC_CFGREG_SOFTRESET, cd->softreset); + return count; +} +static DEVICE_ATTR_RW(next_bitstream); + +/* + * Create device_attribute structures / params: name, mode, show, store + * additional flag if valid in VF + */ +static struct attribute *genwqe_attributes[] = { + &dev_attr_tempsens.attr, + &dev_attr_next_bitstream.attr, + &dev_attr_curr_bitstream.attr, + &dev_attr_base_clock.attr, + &dev_attr_driver.attr, + &dev_attr_type.attr, + &dev_attr_version.attr, + &dev_attr_appid.attr, + &dev_attr_status.attr, + &dev_attr_freerunning_timer.attr, + &dev_attr_queue_working_time.attr, + NULL, +}; + +static struct attribute *genwqe_normal_attributes[] = { + &dev_attr_driver.attr, + &dev_attr_type.attr, + &dev_attr_version.attr, + &dev_attr_appid.attr, + &dev_attr_status.attr, + &dev_attr_freerunning_timer.attr, + &dev_attr_queue_working_time.attr, + NULL, +}; + +/** + * genwqe_is_visible() - Determine if sysfs attribute should be visible or not + * + * VFs have restricted mmio capabilities, so not all sysfs entries + * are allowed in VFs. + */ +static umode_t genwqe_is_visible(struct kobject *kobj, + struct attribute *attr, int n) +{ + unsigned int j; + struct device *dev = container_of(kobj, struct device, kobj); + struct genwqe_dev *cd = dev_get_drvdata(dev); + umode_t mode = attr->mode; + + if (genwqe_is_privileged(cd)) + return mode; + + for (j = 0; genwqe_normal_attributes[j] != NULL; j++) + if (genwqe_normal_attributes[j] == attr) + return mode; + + return 0; +} + +static struct attribute_group genwqe_attribute_group = { + .is_visible = genwqe_is_visible, + .attrs = genwqe_attributes, +}; + +const struct attribute_group *genwqe_attribute_groups[] = { + &genwqe_attribute_group, + NULL, +}; diff --git a/drivers/misc/genwqe/card_utils.c b/drivers/misc/genwqe/card_utils.c new file mode 100644 index 00000000000..62cc6bb3f62 --- /dev/null +++ b/drivers/misc/genwqe/card_utils.c @@ -0,0 +1,1034 @@ +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +/* + * Miscelanous functionality used in the other GenWQE driver parts. + */ + +#include <linux/kernel.h> +#include <linux/dma-mapping.h> +#include <linux/sched.h> +#include <linux/vmalloc.h> +#include <linux/page-flags.h> +#include <linux/scatterlist.h> +#include <linux/hugetlb.h> +#include <linux/iommu.h> +#include <linux/delay.h> +#include <linux/pci.h> +#include <linux/dma-mapping.h> +#include <linux/ctype.h> +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/delay.h> +#include <asm/pgtable.h> + +#include "genwqe_driver.h" +#include "card_base.h" +#include "card_ddcb.h" + +/** + * __genwqe_writeq() - Write 64-bit register + * @cd: genwqe device descriptor + * @byte_offs: byte offset within BAR + * @val: 64-bit value + * + * Return: 0 if success; < 0 if error + */ +int __genwqe_writeq(struct genwqe_dev *cd, u64 byte_offs, u64 val) +{ + if (cd->err_inject & GENWQE_INJECT_HARDWARE_FAILURE) + return -EIO; + + if (cd->mmio == NULL) + return -EIO; + + __raw_writeq((__force u64)cpu_to_be64(val), cd->mmio + byte_offs); + return 0; +} + +/** + * __genwqe_readq() - Read 64-bit register + * @cd: genwqe device descriptor + * @byte_offs: offset within BAR + * + * Return: value from register + */ +u64 __genwqe_readq(struct genwqe_dev *cd, u64 byte_offs) +{ + if (cd->err_inject & GENWQE_INJECT_HARDWARE_FAILURE) + return 0xffffffffffffffffull; + + if ((cd->err_inject & GENWQE_INJECT_GFIR_FATAL) && + (byte_offs == IO_SLC_CFGREG_GFIR)) + return 0x000000000000ffffull; + + if ((cd->err_inject & GENWQE_INJECT_GFIR_INFO) && + (byte_offs == IO_SLC_CFGREG_GFIR)) + return 0x00000000ffff0000ull; + + if (cd->mmio == NULL) + return 0xffffffffffffffffull; + + return be64_to_cpu((__force __be64)__raw_readq(cd->mmio + byte_offs)); +} + +/** + * __genwqe_writel() - Write 32-bit register + * @cd: genwqe device descriptor + * @byte_offs: byte offset within BAR + * @val: 32-bit value + * + * Return: 0 if success; < 0 if error + */ +int __genwqe_writel(struct genwqe_dev *cd, u64 byte_offs, u32 val) +{ + if (cd->err_inject & GENWQE_INJECT_HARDWARE_FAILURE) + return -EIO; + + if (cd->mmio == NULL) + return -EIO; + + __raw_writel((__force u32)cpu_to_be32(val), cd->mmio + byte_offs); + return 0; +} + +/** + * __genwqe_readl() - Read 32-bit register + * @cd: genwqe device descriptor + * @byte_offs: offset within BAR + * + * Return: Value from register + */ +u32 __genwqe_readl(struct genwqe_dev *cd, u64 byte_offs) +{ + if (cd->err_inject & GENWQE_INJECT_HARDWARE_FAILURE) + return 0xffffffff; + + if (cd->mmio == NULL) + return 0xffffffff; + + return be32_to_cpu((__force __be32)__raw_readl(cd->mmio + byte_offs)); +} + +/** + * genwqe_read_app_id() - Extract app_id + * + * app_unitcfg need to be filled with valid data first + */ +int genwqe_read_app_id(struct genwqe_dev *cd, char *app_name, int len) +{ + int i, j; + u32 app_id = (u32)cd->app_unitcfg; + + memset(app_name, 0, len); + for (i = 0, j = 0; j < min(len, 4); j++) { + char ch = (char)((app_id >> (24 - j*8)) & 0xff); + if (ch == ' ') + continue; + app_name[i++] = isprint(ch) ? ch : 'X'; + } + return i; +} + +/** + * genwqe_init_crc32() - Prepare a lookup table for fast crc32 calculations + * + * Existing kernel functions seem to use a different polynom, + * therefore we could not use them here. + * + * Genwqe's Polynomial = 0x20044009 + */ +#define CRC32_POLYNOMIAL 0x20044009 +static u32 crc32_tab[256]; /* crc32 lookup table */ + +void genwqe_init_crc32(void) +{ + int i, j; + u32 crc; + + for (i = 0; i < 256; i++) { + crc = i << 24; + for (j = 0; j < 8; j++) { + if (crc & 0x80000000) + crc = (crc << 1) ^ CRC32_POLYNOMIAL; + else + crc = (crc << 1); + } + crc32_tab[i] = crc; + } +} + +/** + * genwqe_crc32() - Generate 32-bit crc as required for DDCBs + * @buff: pointer to data buffer + * @len: length of data for calculation + * @init: initial crc (0xffffffff at start) + * + * polynomial = x^32 * + x^29 + x^18 + x^14 + x^3 + 1 (0x20044009) + + * Example: 4 bytes 0x01 0x02 0x03 0x04 with init=0xffffffff should + * result in a crc32 of 0xf33cb7d3. + * + * The existing kernel crc functions did not cover this polynom yet. + * + * Return: crc32 checksum. + */ +u32 genwqe_crc32(u8 *buff, size_t len, u32 init) +{ + int i; + u32 crc; + + crc = init; + while (len--) { + i = ((crc >> 24) ^ *buff++) & 0xFF; + crc = (crc << 8) ^ crc32_tab[i]; + } + return crc; +} + +void *__genwqe_alloc_consistent(struct genwqe_dev *cd, size_t size, + dma_addr_t *dma_handle) +{ + if (get_order(size) > MAX_ORDER) + return NULL; + + return pci_alloc_consistent(cd->pci_dev, size, dma_handle); +} + +void __genwqe_free_consistent(struct genwqe_dev *cd, size_t size, + void *vaddr, dma_addr_t dma_handle) +{ + if (vaddr == NULL) + return; + + pci_free_consistent(cd->pci_dev, size, vaddr, dma_handle); +} + +static void genwqe_unmap_pages(struct genwqe_dev *cd, dma_addr_t *dma_list, + int num_pages) +{ + int i; + struct pci_dev *pci_dev = cd->pci_dev; + + for (i = 0; (i < num_pages) && (dma_list[i] != 0x0); i++) { + pci_unmap_page(pci_dev, dma_list[i], + PAGE_SIZE, PCI_DMA_BIDIRECTIONAL); + dma_list[i] = 0x0; + } +} + +static int genwqe_map_pages(struct genwqe_dev *cd, + struct page **page_list, int num_pages, + dma_addr_t *dma_list) +{ + int i; + struct pci_dev *pci_dev = cd->pci_dev; + + /* establish DMA mapping for requested pages */ + for (i = 0; i < num_pages; i++) { + dma_addr_t daddr; + + dma_list[i] = 0x0; + daddr = pci_map_page(pci_dev, page_list[i], + 0, /* map_offs */ + PAGE_SIZE, + PCI_DMA_BIDIRECTIONAL); /* FIXME rd/rw */ + + if (pci_dma_mapping_error(pci_dev, daddr)) { + dev_err(&pci_dev->dev, + "[%s] err: no dma addr daddr=%016llx!\n", + __func__, (long long)daddr); + goto err; + } + + dma_list[i] = daddr; + } + return 0; + + err: + genwqe_unmap_pages(cd, dma_list, num_pages); + return -EIO; +} + +static int genwqe_sgl_size(int num_pages) +{ + int len, num_tlb = num_pages / 7; + + len = sizeof(struct sg_entry) * (num_pages+num_tlb + 1); + return roundup(len, PAGE_SIZE); +} + +/** + * genwqe_alloc_sync_sgl() - Allocate memory for sgl and overlapping pages + * + * Allocates memory for sgl and overlapping pages. Pages which might + * overlap other user-space memory blocks are being cached for DMAs, + * such that we do not run into syncronization issues. Data is copied + * from user-space into the cached pages. + */ +int genwqe_alloc_sync_sgl(struct genwqe_dev *cd, struct genwqe_sgl *sgl, + void __user *user_addr, size_t user_size) +{ + int rc; + struct pci_dev *pci_dev = cd->pci_dev; + + sgl->fpage_offs = offset_in_page((unsigned long)user_addr); + sgl->fpage_size = min_t(size_t, PAGE_SIZE-sgl->fpage_offs, user_size); + sgl->nr_pages = DIV_ROUND_UP(sgl->fpage_offs + user_size, PAGE_SIZE); + sgl->lpage_size = (user_size - sgl->fpage_size) % PAGE_SIZE; + + dev_dbg(&pci_dev->dev, "[%s] uaddr=%p usize=%8ld nr_pages=%ld " + "fpage_offs=%lx fpage_size=%ld lpage_size=%ld\n", + __func__, user_addr, user_size, sgl->nr_pages, + sgl->fpage_offs, sgl->fpage_size, sgl->lpage_size); + + sgl->user_addr = user_addr; + sgl->user_size = user_size; + sgl->sgl_size = genwqe_sgl_size(sgl->nr_pages); + + if (get_order(sgl->sgl_size) > MAX_ORDER) { + dev_err(&pci_dev->dev, + "[%s] err: too much memory requested!\n", __func__); + return -ENOMEM; + } + + sgl->sgl = __genwqe_alloc_consistent(cd, sgl->sgl_size, + &sgl->sgl_dma_addr); + if (sgl->sgl == NULL) { + dev_err(&pci_dev->dev, + "[%s] err: no memory available!\n", __func__); + return -ENOMEM; + } + + /* Only use buffering on incomplete pages */ + if ((sgl->fpage_size != 0) && (sgl->fpage_size != PAGE_SIZE)) { + sgl->fpage = __genwqe_alloc_consistent(cd, PAGE_SIZE, + &sgl->fpage_dma_addr); + if (sgl->fpage == NULL) + goto err_out; + + /* Sync with user memory */ + if (copy_from_user(sgl->fpage + sgl->fpage_offs, + user_addr, sgl->fpage_size)) { + rc = -EFAULT; + goto err_out; + } + } + if (sgl->lpage_size != 0) { + sgl->lpage = __genwqe_alloc_consistent(cd, PAGE_SIZE, + &sgl->lpage_dma_addr); + if (sgl->lpage == NULL) + goto err_out1; + + /* Sync with user memory */ + if (copy_from_user(sgl->lpage, user_addr + user_size - + sgl->lpage_size, sgl->lpage_size)) { + rc = -EFAULT; + goto err_out1; + } + } + return 0; + + err_out1: + __genwqe_free_consistent(cd, PAGE_SIZE, sgl->fpage, + sgl->fpage_dma_addr); + err_out: + __genwqe_free_consistent(cd, sgl->sgl_size, sgl->sgl, + sgl->sgl_dma_addr); + return -ENOMEM; +} + +int genwqe_setup_sgl(struct genwqe_dev *cd, struct genwqe_sgl *sgl, + dma_addr_t *dma_list) +{ + int i = 0, j = 0, p; + unsigned long dma_offs, map_offs; + dma_addr_t prev_daddr = 0; + struct sg_entry *s, *last_s = NULL; + size_t size = sgl->user_size; + + dma_offs = 128; /* next block if needed/dma_offset */ + map_offs = sgl->fpage_offs; /* offset in first page */ + + s = &sgl->sgl[0]; /* first set of 8 entries */ + p = 0; /* page */ + while (p < sgl->nr_pages) { + dma_addr_t daddr; + unsigned int size_to_map; + + /* always write the chaining entry, cleanup is done later */ + j = 0; + s[j].target_addr = cpu_to_be64(sgl->sgl_dma_addr + dma_offs); + s[j].len = cpu_to_be32(128); + s[j].flags = cpu_to_be32(SG_CHAINED); + j++; + + while (j < 8) { + /* DMA mapping for requested page, offs, size */ + size_to_map = min(size, PAGE_SIZE - map_offs); + + if ((p == 0) && (sgl->fpage != NULL)) { + daddr = sgl->fpage_dma_addr + map_offs; + + } else if ((p == sgl->nr_pages - 1) && + (sgl->lpage != NULL)) { + daddr = sgl->lpage_dma_addr; + } else { + daddr = dma_list[p] + map_offs; + } + + size -= size_to_map; + map_offs = 0; + + if (prev_daddr == daddr) { + u32 prev_len = be32_to_cpu(last_s->len); + + /* pr_info("daddr combining: " + "%016llx/%08x -> %016llx\n", + prev_daddr, prev_len, daddr); */ + + last_s->len = cpu_to_be32(prev_len + + size_to_map); + + p++; /* process next page */ + if (p == sgl->nr_pages) + goto fixup; /* nothing to do */ + + prev_daddr = daddr + size_to_map; + continue; + } + + /* start new entry */ + s[j].target_addr = cpu_to_be64(daddr); + s[j].len = cpu_to_be32(size_to_map); + s[j].flags = cpu_to_be32(SG_DATA); + prev_daddr = daddr + size_to_map; + last_s = &s[j]; + j++; + + p++; /* process next page */ + if (p == sgl->nr_pages) + goto fixup; /* nothing to do */ + } + dma_offs += 128; + s += 8; /* continue 8 elements further */ + } + fixup: + if (j == 1) { /* combining happend on last entry! */ + s -= 8; /* full shift needed on previous sgl block */ + j = 7; /* shift all elements */ + } + + for (i = 0; i < j; i++) /* move elements 1 up */ + s[i] = s[i + 1]; + + s[i].target_addr = cpu_to_be64(0); + s[i].len = cpu_to_be32(0); + s[i].flags = cpu_to_be32(SG_END_LIST); + return 0; +} + +/** + * genwqe_free_sync_sgl() - Free memory for sgl and overlapping pages + * + * After the DMA transfer has been completed we free the memory for + * the sgl and the cached pages. Data is being transfered from cached + * pages into user-space buffers. + */ +int genwqe_free_sync_sgl(struct genwqe_dev *cd, struct genwqe_sgl *sgl) +{ + int rc = 0; + struct pci_dev *pci_dev = cd->pci_dev; + + if (sgl->fpage) { + if (copy_to_user(sgl->user_addr, sgl->fpage + sgl->fpage_offs, + sgl->fpage_size)) { + dev_err(&pci_dev->dev, "[%s] err: copying fpage!\n", + __func__); + rc = -EFAULT; + } + __genwqe_free_consistent(cd, PAGE_SIZE, sgl->fpage, + sgl->fpage_dma_addr); + sgl->fpage = NULL; + sgl->fpage_dma_addr = 0; + } + if (sgl->lpage) { + if (copy_to_user(sgl->user_addr + sgl->user_size - + sgl->lpage_size, sgl->lpage, + sgl->lpage_size)) { + dev_err(&pci_dev->dev, "[%s] err: copying lpage!\n", + __func__); + rc = -EFAULT; + } + __genwqe_free_consistent(cd, PAGE_SIZE, sgl->lpage, + sgl->lpage_dma_addr); + sgl->lpage = NULL; + sgl->lpage_dma_addr = 0; + } + __genwqe_free_consistent(cd, sgl->sgl_size, sgl->sgl, + sgl->sgl_dma_addr); + + sgl->sgl = NULL; + sgl->sgl_dma_addr = 0x0; + sgl->sgl_size = 0; + return rc; +} + +/** + * free_user_pages() - Give pinned pages back + * + * Documentation of get_user_pages is in mm/memory.c: + * + * If the page is written to, set_page_dirty (or set_page_dirty_lock, + * as appropriate) must be called after the page is finished with, and + * before put_page is called. + * + * FIXME Could be of use to others and might belong in the generic + * code, if others agree. E.g. + * ll_free_user_pages in drivers/staging/lustre/lustre/llite/rw26.c + * ceph_put_page_vector in net/ceph/pagevec.c + * maybe more? + */ +static int free_user_pages(struct page **page_list, unsigned int nr_pages, + int dirty) +{ + unsigned int i; + + for (i = 0; i < nr_pages; i++) { + if (page_list[i] != NULL) { + if (dirty) + set_page_dirty_lock(page_list[i]); + put_page(page_list[i]); + } + } + return 0; +} + +/** + * genwqe_user_vmap() - Map user-space memory to virtual kernel memory + * @cd: pointer to genwqe device + * @m: mapping params + * @uaddr: user virtual address + * @size: size of memory to be mapped + * + * We need to think about how we could speed this up. Of course it is + * not a good idea to do this over and over again, like we are + * currently doing it. Nevertheless, I am curious where on the path + * the performance is spend. Most probably within the memory + * allocation functions, but maybe also in the DMA mapping code. + * + * Restrictions: The maximum size of the possible mapping currently depends + * on the amount of memory we can get using kzalloc() for the + * page_list and pci_alloc_consistent for the sg_list. + * The sg_list is currently itself not scattered, which could + * be fixed with some effort. The page_list must be split into + * PAGE_SIZE chunks too. All that will make the complicated + * code more complicated. + * + * Return: 0 if success + */ +int genwqe_user_vmap(struct genwqe_dev *cd, struct dma_mapping *m, void *uaddr, + unsigned long size, struct ddcb_requ *req) +{ + int rc = -EINVAL; + unsigned long data, offs; + struct pci_dev *pci_dev = cd->pci_dev; + + if ((uaddr == NULL) || (size == 0)) { + m->size = 0; /* mark unused and not added */ + return -EINVAL; + } + m->u_vaddr = uaddr; + m->size = size; + + /* determine space needed for page_list. */ + data = (unsigned long)uaddr; + offs = offset_in_page(data); + m->nr_pages = DIV_ROUND_UP(offs + size, PAGE_SIZE); + + m->page_list = kcalloc(m->nr_pages, + sizeof(struct page *) + sizeof(dma_addr_t), + GFP_KERNEL); + if (!m->page_list) { + dev_err(&pci_dev->dev, "err: alloc page_list failed\n"); + m->nr_pages = 0; + m->u_vaddr = NULL; + m->size = 0; /* mark unused and not added */ + return -ENOMEM; + } + m->dma_list = (dma_addr_t *)(m->page_list + m->nr_pages); + + /* pin user pages in memory */ + rc = get_user_pages_fast(data & PAGE_MASK, /* page aligned addr */ + m->nr_pages, + 1, /* write by caller */ + m->page_list); /* ptrs to pages */ + + /* assumption: get_user_pages can be killed by signals. */ + if (rc < m->nr_pages) { + free_user_pages(m->page_list, rc, 0); + rc = -EFAULT; + goto fail_get_user_pages; + } + + rc = genwqe_map_pages(cd, m->page_list, m->nr_pages, m->dma_list); + if (rc != 0) + goto fail_free_user_pages; + + return 0; + + fail_free_user_pages: + free_user_pages(m->page_list, m->nr_pages, 0); + + fail_get_user_pages: + kfree(m->page_list); + m->page_list = NULL; + m->dma_list = NULL; + m->nr_pages = 0; + m->u_vaddr = NULL; + m->size = 0; /* mark unused and not added */ + return rc; +} + +/** + * genwqe_user_vunmap() - Undo mapping of user-space mem to virtual kernel + * memory + * @cd: pointer to genwqe device + * @m: mapping params + */ +int genwqe_user_vunmap(struct genwqe_dev *cd, struct dma_mapping *m, + struct ddcb_requ *req) +{ + struct pci_dev *pci_dev = cd->pci_dev; + + if (!dma_mapping_used(m)) { + dev_err(&pci_dev->dev, "[%s] err: mapping %p not used!\n", + __func__, m); + return -EINVAL; + } + + if (m->dma_list) + genwqe_unmap_pages(cd, m->dma_list, m->nr_pages); + + if (m->page_list) { + free_user_pages(m->page_list, m->nr_pages, 1); + + kfree(m->page_list); + m->page_list = NULL; + m->dma_list = NULL; + m->nr_pages = 0; + } + + m->u_vaddr = NULL; + m->size = 0; /* mark as unused and not added */ + return 0; +} + +/** + * genwqe_card_type() - Get chip type SLU Configuration Register + * @cd: pointer to the genwqe device descriptor + * Return: 0: Altera Stratix-IV 230 + * 1: Altera Stratix-IV 530 + * 2: Altera Stratix-V A4 + * 3: Altera Stratix-V A7 + */ +u8 genwqe_card_type(struct genwqe_dev *cd) +{ + u64 card_type = cd->slu_unitcfg; + return (u8)((card_type & IO_SLU_UNITCFG_TYPE_MASK) >> 20); +} + +/** + * genwqe_card_reset() - Reset the card + * @cd: pointer to the genwqe device descriptor + */ +int genwqe_card_reset(struct genwqe_dev *cd) +{ + u64 softrst; + struct pci_dev *pci_dev = cd->pci_dev; + + if (!genwqe_is_privileged(cd)) + return -ENODEV; + + /* new SL */ + __genwqe_writeq(cd, IO_SLC_CFGREG_SOFTRESET, 0x1ull); + msleep(1000); + __genwqe_readq(cd, IO_HSU_FIR_CLR); + __genwqe_readq(cd, IO_APP_FIR_CLR); + __genwqe_readq(cd, IO_SLU_FIR_CLR); + + /* + * Read-modify-write to preserve the stealth bits + * + * For SL >= 039, Stealth WE bit allows removing + * the read-modify-wrote. + * r-m-w may require a mask 0x3C to avoid hitting hard + * reset again for error reset (should be 0, chicken). + */ + softrst = __genwqe_readq(cd, IO_SLC_CFGREG_SOFTRESET) & 0x3cull; + __genwqe_writeq(cd, IO_SLC_CFGREG_SOFTRESET, softrst | 0x2ull); + + /* give ERRORRESET some time to finish */ + msleep(50); + + if (genwqe_need_err_masking(cd)) { + dev_info(&pci_dev->dev, + "[%s] masking errors for old bitstreams\n", __func__); + __genwqe_writeq(cd, IO_SLC_MISC_DEBUG, 0x0aull); + } + return 0; +} + +int genwqe_read_softreset(struct genwqe_dev *cd) +{ + u64 bitstream; + + if (!genwqe_is_privileged(cd)) + return -ENODEV; + + bitstream = __genwqe_readq(cd, IO_SLU_BITSTREAM) & 0x1; + cd->softreset = (bitstream == 0) ? 0x8ull : 0xcull; + return 0; +} + +/** + * genwqe_set_interrupt_capability() - Configure MSI capability structure + * @cd: pointer to the device + * Return: 0 if no error + */ +int genwqe_set_interrupt_capability(struct genwqe_dev *cd, int count) +{ + int rc; + struct pci_dev *pci_dev = cd->pci_dev; + + rc = pci_enable_msi_exact(pci_dev, count); + if (rc == 0) + cd->flags |= GENWQE_FLAG_MSI_ENABLED; + return rc; +} + +/** + * genwqe_reset_interrupt_capability() - Undo genwqe_set_interrupt_capability() + * @cd: pointer to the device + */ +void genwqe_reset_interrupt_capability(struct genwqe_dev *cd) +{ + struct pci_dev *pci_dev = cd->pci_dev; + + if (cd->flags & GENWQE_FLAG_MSI_ENABLED) { + pci_disable_msi(pci_dev); + cd->flags &= ~GENWQE_FLAG_MSI_ENABLED; + } +} + +/** + * set_reg_idx() - Fill array with data. Ignore illegal offsets. + * @cd: card device + * @r: debug register array + * @i: index to desired entry + * @m: maximum possible entries + * @addr: addr which is read + * @index: index in debug array + * @val: read value + */ +static int set_reg_idx(struct genwqe_dev *cd, struct genwqe_reg *r, + unsigned int *i, unsigned int m, u32 addr, u32 idx, + u64 val) +{ + if (WARN_ON_ONCE(*i >= m)) + return -EFAULT; + + r[*i].addr = addr; + r[*i].idx = idx; + r[*i].val = val; + ++*i; + return 0; +} + +static int set_reg(struct genwqe_dev *cd, struct genwqe_reg *r, + unsigned int *i, unsigned int m, u32 addr, u64 val) +{ + return set_reg_idx(cd, r, i, m, addr, 0, val); +} + +int genwqe_read_ffdc_regs(struct genwqe_dev *cd, struct genwqe_reg *regs, + unsigned int max_regs, int all) +{ + unsigned int i, j, idx = 0; + u32 ufir_addr, ufec_addr, sfir_addr, sfec_addr; + u64 gfir, sluid, appid, ufir, ufec, sfir, sfec; + + /* Global FIR */ + gfir = __genwqe_readq(cd, IO_SLC_CFGREG_GFIR); + set_reg(cd, regs, &idx, max_regs, IO_SLC_CFGREG_GFIR, gfir); + + /* UnitCfg for SLU */ + sluid = __genwqe_readq(cd, IO_SLU_UNITCFG); /* 0x00000000 */ + set_reg(cd, regs, &idx, max_regs, IO_SLU_UNITCFG, sluid); + + /* UnitCfg for APP */ + appid = __genwqe_readq(cd, IO_APP_UNITCFG); /* 0x02000000 */ + set_reg(cd, regs, &idx, max_regs, IO_APP_UNITCFG, appid); + + /* Check all chip Units */ + for (i = 0; i < GENWQE_MAX_UNITS; i++) { + + /* Unit FIR */ + ufir_addr = (i << 24) | 0x008; + ufir = __genwqe_readq(cd, ufir_addr); + set_reg(cd, regs, &idx, max_regs, ufir_addr, ufir); + + /* Unit FEC */ + ufec_addr = (i << 24) | 0x018; + ufec = __genwqe_readq(cd, ufec_addr); + set_reg(cd, regs, &idx, max_regs, ufec_addr, ufec); + + for (j = 0; j < 64; j++) { + /* wherever there is a primary 1, read the 2ndary */ + if (!all && (!(ufir & (1ull << j)))) + continue; + + sfir_addr = (i << 24) | (0x100 + 8 * j); + sfir = __genwqe_readq(cd, sfir_addr); + set_reg(cd, regs, &idx, max_regs, sfir_addr, sfir); + + sfec_addr = (i << 24) | (0x300 + 8 * j); + sfec = __genwqe_readq(cd, sfec_addr); + set_reg(cd, regs, &idx, max_regs, sfec_addr, sfec); + } + } + + /* fill with invalid data until end */ + for (i = idx; i < max_regs; i++) { + regs[i].addr = 0xffffffff; + regs[i].val = 0xffffffffffffffffull; + } + return idx; +} + +/** + * genwqe_ffdc_buff_size() - Calculates the number of dump registers + */ +int genwqe_ffdc_buff_size(struct genwqe_dev *cd, int uid) +{ + int entries = 0, ring, traps, traces, trace_entries; + u32 eevptr_addr, l_addr, d_len, d_type; + u64 eevptr, val, addr; + + eevptr_addr = GENWQE_UID_OFFS(uid) | IO_EXTENDED_ERROR_POINTER; + eevptr = __genwqe_readq(cd, eevptr_addr); + + if ((eevptr != 0x0) && (eevptr != -1ull)) { + l_addr = GENWQE_UID_OFFS(uid) | eevptr; + + while (1) { + val = __genwqe_readq(cd, l_addr); + + if ((val == 0x0) || (val == -1ull)) + break; + + /* 38:24 */ + d_len = (val & 0x0000007fff000000ull) >> 24; + + /* 39 */ + d_type = (val & 0x0000008000000000ull) >> 36; + + if (d_type) { /* repeat */ + entries += d_len; + } else { /* size in bytes! */ + entries += d_len >> 3; + } + + l_addr += 8; + } + } + + for (ring = 0; ring < 8; ring++) { + addr = GENWQE_UID_OFFS(uid) | IO_EXTENDED_DIAG_MAP(ring); + val = __genwqe_readq(cd, addr); + + if ((val == 0x0ull) || (val == -1ull)) + continue; + + traps = (val >> 24) & 0xff; + traces = (val >> 16) & 0xff; + trace_entries = val & 0xffff; + + entries += traps + (traces * trace_entries); + } + return entries; +} + +/** + * genwqe_ffdc_buff_read() - Implements LogoutExtendedErrorRegisters procedure + */ +int genwqe_ffdc_buff_read(struct genwqe_dev *cd, int uid, + struct genwqe_reg *regs, unsigned int max_regs) +{ + int i, traps, traces, trace, trace_entries, trace_entry, ring; + unsigned int idx = 0; + u32 eevptr_addr, l_addr, d_addr, d_len, d_type; + u64 eevptr, e, val, addr; + + eevptr_addr = GENWQE_UID_OFFS(uid) | IO_EXTENDED_ERROR_POINTER; + eevptr = __genwqe_readq(cd, eevptr_addr); + + if ((eevptr != 0x0) && (eevptr != 0xffffffffffffffffull)) { + l_addr = GENWQE_UID_OFFS(uid) | eevptr; + while (1) { + e = __genwqe_readq(cd, l_addr); + if ((e == 0x0) || (e == 0xffffffffffffffffull)) + break; + + d_addr = (e & 0x0000000000ffffffull); /* 23:0 */ + d_len = (e & 0x0000007fff000000ull) >> 24; /* 38:24 */ + d_type = (e & 0x0000008000000000ull) >> 36; /* 39 */ + d_addr |= GENWQE_UID_OFFS(uid); + + if (d_type) { + for (i = 0; i < (int)d_len; i++) { + val = __genwqe_readq(cd, d_addr); + set_reg_idx(cd, regs, &idx, max_regs, + d_addr, i, val); + } + } else { + d_len >>= 3; /* Size in bytes! */ + for (i = 0; i < (int)d_len; i++, d_addr += 8) { + val = __genwqe_readq(cd, d_addr); + set_reg_idx(cd, regs, &idx, max_regs, + d_addr, 0, val); + } + } + l_addr += 8; + } + } + + /* + * To save time, there are only 6 traces poplulated on Uid=2, + * Ring=1. each with iters=512. + */ + for (ring = 0; ring < 8; ring++) { /* 0 is fls, 1 is fds, + 2...7 are ASI rings */ + addr = GENWQE_UID_OFFS(uid) | IO_EXTENDED_DIAG_MAP(ring); + val = __genwqe_readq(cd, addr); + + if ((val == 0x0ull) || (val == -1ull)) + continue; + + traps = (val >> 24) & 0xff; /* Number of Traps */ + traces = (val >> 16) & 0xff; /* Number of Traces */ + trace_entries = val & 0xffff; /* Entries per trace */ + + /* Note: This is a combined loop that dumps both the traps */ + /* (for the trace == 0 case) as well as the traces 1 to */ + /* 'traces'. */ + for (trace = 0; trace <= traces; trace++) { + u32 diag_sel = + GENWQE_EXTENDED_DIAG_SELECTOR(ring, trace); + + addr = (GENWQE_UID_OFFS(uid) | + IO_EXTENDED_DIAG_SELECTOR); + __genwqe_writeq(cd, addr, diag_sel); + + for (trace_entry = 0; + trace_entry < (trace ? trace_entries : traps); + trace_entry++) { + addr = (GENWQE_UID_OFFS(uid) | + IO_EXTENDED_DIAG_READ_MBX); + val = __genwqe_readq(cd, addr); + set_reg_idx(cd, regs, &idx, max_regs, addr, + (diag_sel<<16) | trace_entry, val); + } + } + } + return 0; +} + +/** + * genwqe_write_vreg() - Write register in virtual window + * + * Note, these registers are only accessible to the PF through the + * VF-window. It is not intended for the VF to access. + */ +int genwqe_write_vreg(struct genwqe_dev *cd, u32 reg, u64 val, int func) +{ + __genwqe_writeq(cd, IO_PF_SLC_VIRTUAL_WINDOW, func & 0xf); + __genwqe_writeq(cd, reg, val); + return 0; +} + +/** + * genwqe_read_vreg() - Read register in virtual window + * + * Note, these registers are only accessible to the PF through the + * VF-window. It is not intended for the VF to access. + */ +u64 genwqe_read_vreg(struct genwqe_dev *cd, u32 reg, int func) +{ + __genwqe_writeq(cd, IO_PF_SLC_VIRTUAL_WINDOW, func & 0xf); + return __genwqe_readq(cd, reg); +} + +/** + * genwqe_base_clock_frequency() - Deteremine base clock frequency of the card + * + * Note: From a design perspective it turned out to be a bad idea to + * use codes here to specifiy the frequency/speed values. An old + * driver cannot understand new codes and is therefore always a + * problem. Better is to measure out the value or put the + * speed/frequency directly into a register which is always a valid + * value for old as well as for new software. + * + * Return: Card clock in MHz + */ +int genwqe_base_clock_frequency(struct genwqe_dev *cd) +{ + u16 speed; /* MHz MHz MHz MHz */ + static const int speed_grade[] = { 250, 200, 166, 175 }; + + speed = (u16)((cd->slu_unitcfg >> 28) & 0x0full); + if (speed >= ARRAY_SIZE(speed_grade)) + return 0; /* illegal value */ + + return speed_grade[speed]; +} + +/** + * genwqe_stop_traps() - Stop traps + * + * Before reading out the analysis data, we need to stop the traps. + */ +void genwqe_stop_traps(struct genwqe_dev *cd) +{ + __genwqe_writeq(cd, IO_SLC_MISC_DEBUG_SET, 0xcull); +} + +/** + * genwqe_start_traps() - Start traps + * + * After having read the data, we can/must enable the traps again. + */ +void genwqe_start_traps(struct genwqe_dev *cd) +{ + __genwqe_writeq(cd, IO_SLC_MISC_DEBUG_CLR, 0xcull); + + if (genwqe_need_err_masking(cd)) + __genwqe_writeq(cd, IO_SLC_MISC_DEBUG, 0x0aull); +} diff --git a/drivers/misc/genwqe/genwqe_driver.h b/drivers/misc/genwqe/genwqe_driver.h new file mode 100644 index 00000000000..cd5263163a6 --- /dev/null +++ b/drivers/misc/genwqe/genwqe_driver.h @@ -0,0 +1,77 @@ +#ifndef __GENWQE_DRIVER_H__ +#define __GENWQE_DRIVER_H__ + +/** + * IBM Accelerator Family 'GenWQE' + * + * (C) Copyright IBM Corp. 2013 + * + * Author: Frank Haverkamp <haver@linux.vnet.ibm.com> + * Author: Joerg-Stephan Vogt <jsvogt@de.ibm.com> + * Author: Michael Jung <mijung@de.ibm.com> + * Author: Michael Ruettger <michael@ibmra.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/types.h> +#include <linux/stddef.h> +#include <linux/cdev.h> +#include <linux/list.h> +#include <linux/kthread.h> +#include <linux/scatterlist.h> +#include <linux/iommu.h> +#include <linux/spinlock.h> +#include <linux/mutex.h> +#include <linux/platform_device.h> +#include <linux/printk.h> + +#include <asm/byteorder.h> +#include <linux/genwqe/genwqe_card.h> + +#define DRV_VERS_STRING "2.0.15" + +/* + * Static minor number assignement, until we decide/implement + * something dynamic. + */ +#define GENWQE_MAX_MINOR 128 /* up to 128 possible genwqe devices */ + +/** + * genwqe_requ_alloc() - Allocate a new DDCB execution request + * + * This data structure contains the user visiable fields of the DDCB + * to be executed. + * + * Return: ptr to genwqe_ddcb_cmd data structure + */ +struct genwqe_ddcb_cmd *ddcb_requ_alloc(void); + +/** + * ddcb_requ_free() - Free DDCB execution request. + * @req: ptr to genwqe_ddcb_cmd data structure. + */ +void ddcb_requ_free(struct genwqe_ddcb_cmd *req); + +u32 genwqe_crc32(u8 *buff, size_t len, u32 init); + +static inline void genwqe_hexdump(struct pci_dev *pci_dev, + const void *buff, unsigned int size) +{ + char prefix[32]; + + scnprintf(prefix, sizeof(prefix), "%s %s: ", + GENWQE_DEVNAME, pci_name(pci_dev)); + + print_hex_dump_debug(prefix, DUMP_PREFIX_OFFSET, 16, 1, buff, + size, true); +} + +#endif /* __GENWQE_DRIVER_H__ */ diff --git a/drivers/misc/hmc6352.c b/drivers/misc/hmc6352.c index 423cd40f1c0..90520d76633 100644 --- a/drivers/misc/hmc6352.c +++ b/drivers/misc/hmc6352.c @@ -22,7 +22,6 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <linux/slab.h> #include <linux/i2c.h> #include <linux/err.h> @@ -46,8 +45,9 @@ static int compass_store(struct device *dev, const char *buf, size_t count, int ret; unsigned long val; - if (strict_strtoul(buf, 10, &val)) - return -EINVAL; + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; if (val >= strlen(map)) return -EINVAL; mutex_lock(&compass_mutex); diff --git a/drivers/misc/hpilo.c b/drivers/misc/hpilo.c index 12ccdf94e4f..b83e3ca12a4 100644 --- a/drivers/misc/hpilo.c +++ b/drivers/misc/hpilo.c @@ -30,7 +30,7 @@ static struct class *ilo_class; static unsigned int ilo_major; -static unsigned int max_ccb = MIN_CCB; +static unsigned int max_ccb = 16; static char ilo_hwdev[MAX_ILO_DEV]; static inline int get_entry_id(int entry) @@ -686,7 +686,7 @@ static void ilo_unmap_device(struct pci_dev *pdev, struct ilo_hwinfo *hw) pci_iounmap(pdev, hw->mmio_vaddr); } -static int __devinit ilo_map_device(struct pci_dev *pdev, struct ilo_hwinfo *hw) +static int ilo_map_device(struct pci_dev *pdev, struct ilo_hwinfo *hw) { int error = -ENOMEM; @@ -725,6 +725,9 @@ static void ilo_remove(struct pci_dev *pdev) int i, minor; struct ilo_hwinfo *ilo_hw = pci_get_drvdata(pdev); + if (!ilo_hw) + return; + clear_device(ilo_hw); minor = MINOR(ilo_hw->cdev.dev); @@ -748,12 +751,16 @@ static void ilo_remove(struct pci_dev *pdev) ilo_hwdev[(minor / max_ccb)] = 0; } -static int __devinit ilo_probe(struct pci_dev *pdev, +static int ilo_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { - int devnum, minor, start, error; + int devnum, minor, start, error = 0; struct ilo_hwinfo *ilo_hw; + /* Ignore subsystem_device = 0x1979 (set by BIOS) */ + if (pdev->subsystem_device == 0x1979) + return 0; + if (max_ccb > MAX_CCB) max_ccb = MAX_CCB; else if (max_ccb < MIN_CCB) @@ -852,7 +859,7 @@ static struct pci_driver ilo_driver = { .name = ILO_NAME, .id_table = ilo_devices, .probe = ilo_probe, - .remove = __devexit_p(ilo_remove), + .remove = ilo_remove, }; static int __init ilo_init(void) @@ -892,14 +899,14 @@ static void __exit ilo_exit(void) class_destroy(ilo_class); } -MODULE_VERSION("1.3"); +MODULE_VERSION("1.4.1"); MODULE_ALIAS(ILO_NAME); MODULE_DESCRIPTION(ILO_NAME); MODULE_AUTHOR("David Altobelli <david.altobelli@hp.com>"); MODULE_LICENSE("GPL v2"); module_param(max_ccb, uint, 0444); -MODULE_PARM_DESC(max_ccb, "Maximum number of HP iLO channels to attach (8)"); +MODULE_PARM_DESC(max_ccb, "Maximum number of HP iLO channels to attach (16)"); module_init(ilo_init); module_exit(ilo_exit); diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c index 6673e578b3e..e8b933111e0 100644 --- a/drivers/misc/ibmasm/ibmasmfs.c +++ b/drivers/misc/ibmasm/ibmasmfs.c @@ -110,6 +110,7 @@ static struct file_system_type ibmasmfs_type = { .mount = ibmasmfs_mount, .kill_sb = kill_litter_super, }; +MODULE_ALIAS_FS("ibmasmfs"); static int ibmasmfs_fill_super (struct super_block *sb, void *data, int silent) { @@ -148,8 +149,7 @@ static struct inode *ibmasmfs_make_inode(struct super_block *sb, int mode) return ret; } -static struct dentry *ibmasmfs_create_file (struct super_block *sb, - struct dentry *parent, +static struct dentry *ibmasmfs_create_file(struct dentry *parent, const char *name, const struct file_operations *fops, void *data, @@ -162,7 +162,7 @@ static struct dentry *ibmasmfs_create_file (struct super_block *sb, if (!dentry) return NULL; - inode = ibmasmfs_make_inode(sb, S_IFREG | mode); + inode = ibmasmfs_make_inode(parent->d_sb, S_IFREG | mode); if (!inode) { dput(dentry); return NULL; @@ -175,8 +175,7 @@ static struct dentry *ibmasmfs_create_file (struct super_block *sb, return dentry; } -static struct dentry *ibmasmfs_create_dir (struct super_block *sb, - struct dentry *parent, +static struct dentry *ibmasmfs_create_dir(struct dentry *parent, const char *name) { struct dentry *dentry; @@ -186,7 +185,7 @@ static struct dentry *ibmasmfs_create_dir (struct super_block *sb, if (!dentry) return NULL; - inode = ibmasmfs_make_inode(sb, S_IFDIR | 0500); + inode = ibmasmfs_make_inode(parent->d_sb, S_IFDIR | 0500); if (!inode) { dput(dentry); return NULL; @@ -611,20 +610,20 @@ static void ibmasmfs_create_files (struct super_block *sb) struct dentry *dir; struct dentry *remote_dir; sp = list_entry(entry, struct service_processor, node); - dir = ibmasmfs_create_dir(sb, sb->s_root, sp->dirname); + dir = ibmasmfs_create_dir(sb->s_root, sp->dirname); if (!dir) continue; - ibmasmfs_create_file(sb, dir, "command", &command_fops, sp, S_IRUSR|S_IWUSR); - ibmasmfs_create_file(sb, dir, "event", &event_fops, sp, S_IRUSR|S_IWUSR); - ibmasmfs_create_file(sb, dir, "reverse_heartbeat", &r_heartbeat_fops, sp, S_IRUSR|S_IWUSR); + ibmasmfs_create_file(dir, "command", &command_fops, sp, S_IRUSR|S_IWUSR); + ibmasmfs_create_file(dir, "event", &event_fops, sp, S_IRUSR|S_IWUSR); + ibmasmfs_create_file(dir, "reverse_heartbeat", &r_heartbeat_fops, sp, S_IRUSR|S_IWUSR); - remote_dir = ibmasmfs_create_dir(sb, dir, "remote_video"); + remote_dir = ibmasmfs_create_dir(dir, "remote_video"); if (!remote_dir) continue; - ibmasmfs_create_file(sb, remote_dir, "width", &remote_settings_fops, (void *)display_width(sp), S_IRUSR|S_IWUSR); - ibmasmfs_create_file(sb, remote_dir, "height", &remote_settings_fops, (void *)display_height(sp), S_IRUSR|S_IWUSR); - ibmasmfs_create_file(sb, remote_dir, "depth", &remote_settings_fops, (void *)display_depth(sp), S_IRUSR|S_IWUSR); + ibmasmfs_create_file(remote_dir, "width", &remote_settings_fops, (void *)display_width(sp), S_IRUSR|S_IWUSR); + ibmasmfs_create_file(remote_dir, "height", &remote_settings_fops, (void *)display_height(sp), S_IRUSR|S_IWUSR); + ibmasmfs_create_file(remote_dir, "depth", &remote_settings_fops, (void *)display_depth(sp), S_IRUSR|S_IWUSR); } } diff --git a/drivers/misc/ibmasm/module.c b/drivers/misc/ibmasm/module.c index 168d8008f46..6b3bf9ab051 100644 --- a/drivers/misc/ibmasm/module.c +++ b/drivers/misc/ibmasm/module.c @@ -62,7 +62,7 @@ module_param(ibmasm_debug, int , S_IRUGO | S_IWUSR); MODULE_PARM_DESC(ibmasm_debug, " Set debug mode on or off"); -static int __devinit ibmasm_init_one(struct pci_dev *pdev, const struct pci_device_id *id) +static int ibmasm_init_one(struct pci_dev *pdev, const struct pci_device_id *id) { int result; struct service_processor *sp; @@ -153,7 +153,6 @@ error_ioremap: error_heartbeat: ibmasm_event_buffer_exit(sp); error_eventbuffer: - pci_set_drvdata(pdev, NULL); kfree(sp); error_kmalloc: pci_release_regions(pdev); @@ -163,9 +162,9 @@ error_resources: return result; } -static void __devexit ibmasm_remove_one(struct pci_dev *pdev) +static void ibmasm_remove_one(struct pci_dev *pdev) { - struct service_processor *sp = (struct service_processor *)pci_get_drvdata(pdev); + struct service_processor *sp = pci_get_drvdata(pdev); dbg("Unregistering UART\n"); ibmasm_unregister_uart(sp); @@ -182,7 +181,6 @@ static void __devexit ibmasm_remove_one(struct pci_dev *pdev) ibmasm_free_remote_input_dev(sp); iounmap(sp->base_address); ibmasm_event_buffer_exit(sp); - pci_set_drvdata(pdev, NULL); kfree(sp); pci_release_regions(pdev); pci_disable_device(pdev); @@ -198,7 +196,7 @@ static struct pci_driver ibmasm_driver = { .name = DRIVER_NAME, .id_table = ibmasm_pci_table, .probe = ibmasm_init_one, - .remove = __devexit_p(ibmasm_remove_one), + .remove = ibmasm_remove_one, }; static void __exit ibmasm_exit (void) diff --git a/drivers/misc/ics932s401.c b/drivers/misc/ics932s401.c index 00295367c06..28f51e01fd2 100644 --- a/drivers/misc/ics932s401.c +++ b/drivers/misc/ics932s401.c @@ -2,7 +2,7 @@ * A driver for the Integrated Circuits ICS932S401 * Copyright (C) 2008 IBM * - * Author: Darrick J. Wong <djwong@us.ibm.com> + * Author: Darrick J. Wong <darrick.wong@oracle.com> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -482,7 +482,7 @@ static int ics932s401_remove(struct i2c_client *client) module_i2c_driver(ics932s401_driver); -MODULE_AUTHOR("Darrick J. Wong <djwong@us.ibm.com>"); +MODULE_AUTHOR("Darrick J. Wong <darrick.wong@oracle.com>"); MODULE_DESCRIPTION("ICS932S401 driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/misc/ioc4.c b/drivers/misc/ioc4.c index 6a7710603a9..06f6ad29cef 100644 --- a/drivers/misc/ioc4.c +++ b/drivers/misc/ioc4.c @@ -139,7 +139,7 @@ ioc4_unregister_submodule(struct ioc4_submodule *is) * even though the following code utilizes external interrupt registers * to perform the speed calculation. */ -static void __devinit +static void ioc4_clock_calibrate(struct ioc4_driver_data *idd) { union ioc4_int_out int_out; @@ -231,7 +231,7 @@ ioc4_clock_calibrate(struct ioc4_driver_data *idd) * on the same PCI bus at slot number 3 to differentiate IO9 from IO10. * If neither is present, it's a PCI-RT. */ -static unsigned int __devinit +static unsigned int ioc4_variant(struct ioc4_driver_data *idd) { struct pci_dev *pdev = NULL; @@ -279,7 +279,7 @@ ioc4_load_modules(struct work_struct *work) static DECLARE_WORK(ioc4_load_modules_work, ioc4_load_modules); /* Adds a new instance of an IOC4 card */ -static int __devinit +static int ioc4_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) { struct ioc4_driver_data *idd; @@ -415,7 +415,7 @@ out: } /* Removes a particular instance of an IOC4 card. */ -static void __devexit +static void ioc4_remove(struct pci_dev *pdev) { struct ioc4_submodule *is; @@ -466,7 +466,7 @@ static struct pci_driver ioc4_driver = { .name = "IOC4", .id_table = ioc4_id_table, .probe = ioc4_probe, - .remove = __devexit_p(ioc4_remove), + .remove = ioc4_remove, }; MODULE_DEVICE_TABLE(pci, ioc4_id_table); diff --git a/drivers/misc/isl29003.c b/drivers/misc/isl29003.c index eb5de2e210d..12c30b486b2 100644 --- a/drivers/misc/isl29003.c +++ b/drivers/misc/isl29003.c @@ -26,7 +26,6 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <linux/slab.h> #include <linux/i2c.h> #include <linux/mutex.h> @@ -208,7 +207,11 @@ static ssize_t isl29003_store_range(struct device *dev, unsigned long val; int ret; - if ((strict_strtoul(buf, 10, &val) < 0) || (val > 3)) + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + if (val > 3) return -EINVAL; ret = isl29003_set_range(client, val); @@ -239,7 +242,11 @@ static ssize_t isl29003_store_resolution(struct device *dev, unsigned long val; int ret; - if ((strict_strtoul(buf, 10, &val) < 0) || (val > 3)) + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + if (val > 3) return -EINVAL; ret = isl29003_set_resolution(client, val); @@ -267,7 +274,11 @@ static ssize_t isl29003_store_mode(struct device *dev, unsigned long val; int ret; - if ((strict_strtoul(buf, 10, &val) < 0) || (val > 2)) + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + if (val > 2) return -EINVAL; ret = isl29003_set_mode(client, val); @@ -298,7 +309,11 @@ static ssize_t isl29003_store_power_state(struct device *dev, unsigned long val; int ret; - if ((strict_strtoul(buf, 10, &val) < 0) || (val > 1)) + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + if (val > 1) return -EINVAL; ret = isl29003_set_power_state(client, val); @@ -365,7 +380,7 @@ static int isl29003_init_client(struct i2c_client *client) * I2C layer */ -static int __devinit isl29003_probe(struct i2c_client *client, +static int isl29003_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent); @@ -401,7 +416,7 @@ exit_kfree: return err; } -static int __devexit isl29003_remove(struct i2c_client *client) +static int isl29003_remove(struct i2c_client *client) { sysfs_remove_group(&client->dev.kobj, &isl29003_attr_group); isl29003_set_power_state(client, 0); @@ -409,18 +424,20 @@ static int __devexit isl29003_remove(struct i2c_client *client) return 0; } -#ifdef CONFIG_PM -static int isl29003_suspend(struct i2c_client *client, pm_message_t mesg) +#ifdef CONFIG_PM_SLEEP +static int isl29003_suspend(struct device *dev) { + struct i2c_client *client = to_i2c_client(dev); struct isl29003_data *data = i2c_get_clientdata(client); data->power_state_before_suspend = isl29003_get_power_state(client); return isl29003_set_power_state(client, 0); } -static int isl29003_resume(struct i2c_client *client) +static int isl29003_resume(struct device *dev) { int i; + struct i2c_client *client = to_i2c_client(dev); struct isl29003_data *data = i2c_get_clientdata(client); /* restore registers from cache */ @@ -432,10 +449,12 @@ static int isl29003_resume(struct i2c_client *client) data->power_state_before_suspend); } +static SIMPLE_DEV_PM_OPS(isl29003_pm_ops, isl29003_suspend, isl29003_resume); +#define ISL29003_PM_OPS (&isl29003_pm_ops) + #else -#define isl29003_suspend NULL -#define isl29003_resume NULL -#endif /* CONFIG_PM */ +#define ISL29003_PM_OPS NULL +#endif /* CONFIG_PM_SLEEP */ static const struct i2c_device_id isl29003_id[] = { { "isl29003", 0 }, @@ -447,11 +466,10 @@ static struct i2c_driver isl29003_driver = { .driver = { .name = ISL29003_DRV_NAME, .owner = THIS_MODULE, + .pm = ISL29003_PM_OPS, }, - .suspend = isl29003_suspend, - .resume = isl29003_resume, .probe = isl29003_probe, - .remove = __devexit_p(isl29003_remove), + .remove = isl29003_remove, .id_table = isl29003_id, }; diff --git a/drivers/misc/isl29020.c b/drivers/misc/isl29020.c index 0aa08c74646..4a9c50a43af 100644 --- a/drivers/misc/isl29020.c +++ b/drivers/misc/isl29020.c @@ -23,7 +23,6 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <linux/slab.h> #include <linux/i2c.h> #include <linux/err.h> @@ -90,8 +89,10 @@ static ssize_t als_sensing_range_store(struct device *dev, int ret_val; unsigned long val; - if (strict_strtoul(buf, 10, &val)) - return -EINVAL; + ret_val = kstrtoul(buf, 10, &val); + if (ret_val) + return ret_val; + if (val < 1 || val > 64000) return -EINVAL; diff --git a/drivers/misc/kgdbts.c b/drivers/misc/kgdbts.c index 3aa9a969b37..36f5d52775a 100644 --- a/drivers/misc/kgdbts.c +++ b/drivers/misc/kgdbts.c @@ -103,6 +103,7 @@ #include <linux/delay.h> #include <linux/kthread.h> #include <linux/module.h> +#include <asm/sections.h> #define v1printk(a...) do { \ if (verbose) \ @@ -222,6 +223,7 @@ static unsigned long lookup_addr(char *arg) addr = (unsigned long)do_fork; else if (!strcmp(arg, "hw_break_val")) addr = (unsigned long)&hw_break_val; + addr = (unsigned long) dereference_function_descriptor((void *)addr); return addr; } diff --git a/drivers/misc/lattice-ecp3-config.c b/drivers/misc/lattice-ecp3-config.c new file mode 100644 index 00000000000..0a1565e63c7 --- /dev/null +++ b/drivers/misc/lattice-ecp3-config.c @@ -0,0 +1,242 @@ +/* + * Copyright (C) 2012 Stefan Roese <sr@denx.de> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/device.h> +#include <linux/firmware.h> +#include <linux/module.h> +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/spi/spi.h> +#include <linux/platform_device.h> +#include <linux/delay.h> + +#define FIRMWARE_NAME "lattice-ecp3.bit" + +/* + * The JTAG ID's of the supported FPGA's. The ID is 32bit wide + * reversed as noted in the manual. + */ +#define ID_ECP3_17 0xc2088080 +#define ID_ECP3_35 0xc2048080 + +/* FPGA commands */ +#define FPGA_CMD_READ_ID 0x07 /* plus 24 bits */ +#define FPGA_CMD_READ_STATUS 0x09 /* plus 24 bits */ +#define FPGA_CMD_CLEAR 0x70 +#define FPGA_CMD_REFRESH 0x71 +#define FPGA_CMD_WRITE_EN 0x4a /* plus 2 bits */ +#define FPGA_CMD_WRITE_DIS 0x4f /* plus 8 bits */ +#define FPGA_CMD_WRITE_INC 0x41 /* plus 0 bits */ + +/* + * The status register is 32bit revered, DONE is bit 17 from the TN1222.pdf + * (LatticeECP3 Slave SPI Port User's Guide) + */ +#define FPGA_STATUS_DONE 0x00004000 +#define FPGA_STATUS_CLEARED 0x00010000 + +#define FPGA_CLEAR_TIMEOUT 5000 /* max. 5000ms for FPGA clear */ +#define FPGA_CLEAR_MSLEEP 10 +#define FPGA_CLEAR_LOOP_COUNT (FPGA_CLEAR_TIMEOUT / FPGA_CLEAR_MSLEEP) + +struct fpga_data { + struct completion fw_loaded; +}; + +struct ecp3_dev { + u32 jedec_id; + char *name; +}; + +static const struct ecp3_dev ecp3_dev[] = { + { + .jedec_id = ID_ECP3_17, + .name = "Lattice ECP3-17", + }, + { + .jedec_id = ID_ECP3_35, + .name = "Lattice ECP3-35", + }, +}; + +static void firmware_load(const struct firmware *fw, void *context) +{ + struct spi_device *spi = (struct spi_device *)context; + struct fpga_data *data = spi_get_drvdata(spi); + u8 *buffer; + int ret; + u8 txbuf[8]; + u8 rxbuf[8]; + int rx_len = 8; + int i; + u32 jedec_id; + u32 status; + + if (fw->size == 0) { + dev_err(&spi->dev, "Error: Firmware size is 0!\n"); + return; + } + + /* Fill dummy data (24 stuffing bits for commands) */ + txbuf[1] = 0x00; + txbuf[2] = 0x00; + txbuf[3] = 0x00; + + /* Trying to speak with the FPGA via SPI... */ + txbuf[0] = FPGA_CMD_READ_ID; + ret = spi_write_then_read(spi, txbuf, 8, rxbuf, rx_len); + dev_dbg(&spi->dev, "FPGA JTAG ID=%08x\n", *(u32 *)&rxbuf[4]); + jedec_id = *(u32 *)&rxbuf[4]; + + for (i = 0; i < ARRAY_SIZE(ecp3_dev); i++) { + if (jedec_id == ecp3_dev[i].jedec_id) + break; + } + if (i == ARRAY_SIZE(ecp3_dev)) { + dev_err(&spi->dev, + "Error: No supported FPGA detected (JEDEC_ID=%08x)!\n", + jedec_id); + return; + } + + dev_info(&spi->dev, "FPGA %s detected\n", ecp3_dev[i].name); + + txbuf[0] = FPGA_CMD_READ_STATUS; + ret = spi_write_then_read(spi, txbuf, 8, rxbuf, rx_len); + dev_dbg(&spi->dev, "FPGA Status=%08x\n", *(u32 *)&rxbuf[4]); + + buffer = kzalloc(fw->size + 8, GFP_KERNEL); + if (!buffer) { + dev_err(&spi->dev, "Error: Can't allocate memory!\n"); + return; + } + + /* + * Insert WRITE_INC command into stream (one SPI frame) + */ + buffer[0] = FPGA_CMD_WRITE_INC; + buffer[1] = 0xff; + buffer[2] = 0xff; + buffer[3] = 0xff; + memcpy(buffer + 4, fw->data, fw->size); + + txbuf[0] = FPGA_CMD_REFRESH; + ret = spi_write(spi, txbuf, 4); + + txbuf[0] = FPGA_CMD_WRITE_EN; + ret = spi_write(spi, txbuf, 4); + + txbuf[0] = FPGA_CMD_CLEAR; + ret = spi_write(spi, txbuf, 4); + + /* + * Wait for FPGA memory to become cleared + */ + for (i = 0; i < FPGA_CLEAR_LOOP_COUNT; i++) { + txbuf[0] = FPGA_CMD_READ_STATUS; + ret = spi_write_then_read(spi, txbuf, 8, rxbuf, rx_len); + status = *(u32 *)&rxbuf[4]; + if (status == FPGA_STATUS_CLEARED) + break; + + msleep(FPGA_CLEAR_MSLEEP); + } + + if (i == FPGA_CLEAR_LOOP_COUNT) { + dev_err(&spi->dev, + "Error: Timeout waiting for FPGA to clear (status=%08x)!\n", + status); + kfree(buffer); + return; + } + + dev_info(&spi->dev, "Configuring the FPGA...\n"); + ret = spi_write(spi, buffer, fw->size + 8); + + txbuf[0] = FPGA_CMD_WRITE_DIS; + ret = spi_write(spi, txbuf, 4); + + txbuf[0] = FPGA_CMD_READ_STATUS; + ret = spi_write_then_read(spi, txbuf, 8, rxbuf, rx_len); + dev_dbg(&spi->dev, "FPGA Status=%08x\n", *(u32 *)&rxbuf[4]); + status = *(u32 *)&rxbuf[4]; + + /* Check result */ + if (status & FPGA_STATUS_DONE) + dev_info(&spi->dev, "FPGA successfully configured!\n"); + else + dev_info(&spi->dev, "FPGA not configured (DONE not set)\n"); + + /* + * Don't forget to release the firmware again + */ + release_firmware(fw); + + kfree(buffer); + + complete(&data->fw_loaded); +} + +static int lattice_ecp3_probe(struct spi_device *spi) +{ + struct fpga_data *data; + int err; + + data = devm_kzalloc(&spi->dev, sizeof(*data), GFP_KERNEL); + if (!data) { + dev_err(&spi->dev, "Memory allocation for fpga_data failed\n"); + return -ENOMEM; + } + spi_set_drvdata(spi, data); + + init_completion(&data->fw_loaded); + err = request_firmware_nowait(THIS_MODULE, FW_ACTION_NOHOTPLUG, + FIRMWARE_NAME, &spi->dev, + GFP_KERNEL, spi, firmware_load); + if (err) { + dev_err(&spi->dev, "Firmware loading failed with %d!\n", err); + return err; + } + + dev_info(&spi->dev, "FPGA bitstream configuration driver registered\n"); + + return 0; +} + +static int lattice_ecp3_remove(struct spi_device *spi) +{ + struct fpga_data *data = spi_get_drvdata(spi); + + wait_for_completion(&data->fw_loaded); + + return 0; +} + +static const struct spi_device_id lattice_ecp3_id[] = { + { "ecp3-17", 0 }, + { "ecp3-35", 0 }, + { } +}; +MODULE_DEVICE_TABLE(spi, lattice_ecp3_id); + +static struct spi_driver lattice_ecp3_driver = { + .driver = { + .name = "lattice-ecp3", + .owner = THIS_MODULE, + }, + .probe = lattice_ecp3_probe, + .remove = lattice_ecp3_remove, + .id_table = lattice_ecp3_id, +}; + +module_spi_driver(lattice_ecp3_driver); + +MODULE_AUTHOR("Stefan Roese <sr@denx.de>"); +MODULE_DESCRIPTION("Lattice ECP3 FPGA configuration via SPI"); +MODULE_LICENSE("GPL"); diff --git a/drivers/misc/lis3lv02d/lis3lv02d.c b/drivers/misc/lis3lv02d/lis3lv02d.c index 4a87e5c0a32..3ef4627f9cb 100644 --- a/drivers/misc/lis3lv02d/lis3lv02d.c +++ b/drivers/misc/lis3lv02d/lis3lv02d.c @@ -23,7 +23,6 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/kernel.h> -#include <linux/init.h> #include <linux/dmi.h> #include <linux/module.h> #include <linux/types.h> @@ -593,7 +592,6 @@ static int lis3lv02d_misc_release(struct inode *inode, struct file *file) struct lis3lv02d *lis3 = container_of(file->private_data, struct lis3lv02d, miscdev); - fasync_helper(-1, file, 0, &lis3->async_queue); clear_bit(0, &lis3->misc_opened); /* release the device */ if (lis3->pm_dev) pm_runtime_put(lis3->pm_dev); @@ -832,9 +830,11 @@ static ssize_t lis3lv02d_rate_set(struct device *dev, { struct lis3lv02d *lis3 = dev_get_drvdata(dev); unsigned long rate; + int ret; - if (strict_strtoul(buf, 0, &rate)) - return -EINVAL; + ret = kstrtoul(buf, 0, &rate); + if (ret) + return ret; lis3lv02d_sysfs_poweron(lis3); if (lis3lv02d_set_odr(lis3, rate)) diff --git a/drivers/misc/lis3lv02d/lis3lv02d_i2c.c b/drivers/misc/lis3lv02d/lis3lv02d_i2c.c index 60ec8689d6e..d324f8a97b8 100644 --- a/drivers/misc/lis3lv02d/lis3lv02d_i2c.c +++ b/drivers/misc/lis3lv02d/lis3lv02d_i2c.c @@ -26,7 +26,6 @@ #include <linux/module.h> #include <linux/kernel.h> -#include <linux/init.h> #include <linux/err.h> #include <linux/i2c.h> #include <linux/pm_runtime.h> @@ -114,7 +113,7 @@ static struct of_device_id lis3lv02d_i2c_dt_ids[] = { MODULE_DEVICE_TABLE(of, lis3lv02d_i2c_dt_ids); #endif -static int __devinit lis3lv02d_i2c_probe(struct i2c_client *client, +static int lis3lv02d_i2c_probe(struct i2c_client *client, const struct i2c_device_id *id) { int ret = 0; @@ -191,7 +190,7 @@ fail: return ret; } -static int __devexit lis3lv02d_i2c_remove(struct i2c_client *client) +static int lis3lv02d_i2c_remove(struct i2c_client *client) { struct lis3lv02d *lis3 = i2c_get_clientdata(client); struct lis3lv02d_platform_data *pdata = client->dev.platform_data; @@ -280,7 +279,7 @@ static struct i2c_driver lis3lv02d_i2c_driver = { .of_match_table = of_match_ptr(lis3lv02d_i2c_dt_ids), }, .probe = lis3lv02d_i2c_probe, - .remove = __devexit_p(lis3lv02d_i2c_remove), + .remove = lis3lv02d_i2c_remove, .id_table = lis3lv02d_id, }; diff --git a/drivers/misc/lis3lv02d/lis3lv02d_spi.c b/drivers/misc/lis3lv02d/lis3lv02d_spi.c index ccb6475fa05..bd06d0cfac4 100644 --- a/drivers/misc/lis3lv02d/lis3lv02d_spi.c +++ b/drivers/misc/lis3lv02d/lis3lv02d_spi.c @@ -10,7 +10,6 @@ #include <linux/module.h> #include <linux/kernel.h> -#include <linux/init.h> #include <linux/err.h> #include <linux/input.h> #include <linux/interrupt.h> @@ -69,7 +68,7 @@ static struct of_device_id lis302dl_spi_dt_ids[] = { MODULE_DEVICE_TABLE(of, lis302dl_spi_dt_ids); #endif -static int __devinit lis302dl_spi_probe(struct spi_device *spi) +static int lis302dl_spi_probe(struct spi_device *spi) { int ret; @@ -100,7 +99,7 @@ static int __devinit lis302dl_spi_probe(struct spi_device *spi) return lis3lv02d_init_device(&lis3_dev); } -static int __devexit lis302dl_spi_remove(struct spi_device *spi) +static int lis302dl_spi_remove(struct spi_device *spi) { struct lis3lv02d *lis3 = spi_get_drvdata(spi); lis3lv02d_joystick_disable(lis3); @@ -144,7 +143,7 @@ static struct spi_driver lis302dl_spi_driver = { .of_match_table = of_match_ptr(lis302dl_spi_dt_ids), }, .probe = lis302dl_spi_probe, - .remove = __devexit_p(lis302dl_spi_remove), + .remove = lis302dl_spi_remove, }; module_spi_driver(lis302dl_spi_driver); diff --git a/drivers/misc/lkdtm.c b/drivers/misc/lkdtm.c index 08aad69c8da..d66a2f24f6b 100644 --- a/drivers/misc/lkdtm.c +++ b/drivers/misc/lkdtm.c @@ -30,6 +30,7 @@ * * See Documentation/fault-injection/provoke-crashes.txt for instructions */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/kernel.h> #include <linux/fs.h> @@ -43,13 +44,28 @@ #include <linux/slab.h> #include <scsi/scsi_cmnd.h> #include <linux/debugfs.h> +#include <linux/vmalloc.h> +#include <linux/mman.h> +#include <asm/cacheflush.h> #ifdef CONFIG_IDE #include <linux/ide.h> #endif +/* + * Make sure our attempts to over run the kernel stack doesn't trigger + * a compiler warning when CONFIG_FRAME_WARN is set. Then make sure we + * recurse past the end of THREAD_SIZE by default. + */ +#if defined(CONFIG_FRAME_WARN) && (CONFIG_FRAME_WARN > 0) +#define REC_STACK_SIZE (CONFIG_FRAME_WARN / 2) +#else +#define REC_STACK_SIZE (THREAD_SIZE / 8) +#endif +#define REC_NUM_DEFAULT ((THREAD_SIZE / REC_STACK_SIZE) * 2) + #define DEFAULT_COUNT 10 -#define REC_NUM_DEFAULT 10 +#define EXEC_SIZE 64 enum cname { CN_INVALID, @@ -68,6 +84,7 @@ enum ctype { CT_NONE, CT_PANIC, CT_BUG, + CT_WARNING, CT_EXCEPTION, CT_LOOP, CT_OVERFLOW, @@ -77,7 +94,16 @@ enum ctype { CT_WRITE_AFTER_FREE, CT_SOFTLOCKUP, CT_HARDLOCKUP, + CT_SPINLOCKUP, CT_HUNG_TASK, + CT_EXEC_DATA, + CT_EXEC_STACK, + CT_EXEC_KMALLOC, + CT_EXEC_VMALLOC, + CT_EXEC_USERSPACE, + CT_ACCESS_USERSPACE, + CT_WRITE_RO, + CT_WRITE_KERN, }; static char* cp_name[] = { @@ -95,6 +121,7 @@ static char* cp_name[] = { static char* cp_type[] = { "PANIC", "BUG", + "WARNING", "EXCEPTION", "LOOP", "OVERFLOW", @@ -104,7 +131,16 @@ static char* cp_type[] = { "WRITE_AFTER_FREE", "SOFTLOCKUP", "HARDLOCKUP", + "SPINLOCKUP", "HUNG_TASK", + "EXEC_DATA", + "EXEC_STACK", + "EXEC_KMALLOC", + "EXEC_VMALLOC", + "EXEC_USERSPACE", + "ACCESS_USERSPACE", + "WRITE_RO", + "WRITE_KERN", }; static struct jprobe lkdtm; @@ -121,10 +157,14 @@ static enum cname cpoint = CN_INVALID; static enum ctype cptype = CT_NONE; static int count = DEFAULT_COUNT; static DEFINE_SPINLOCK(count_lock); +static DEFINE_SPINLOCK(lock_me_up); + +static u8 data_area[EXEC_SIZE]; + +static const unsigned long rodata = 0xAA55AA55; module_param(recur_count, int, 0644); -MODULE_PARM_DESC(recur_count, " Recursion level for the stack overflow test, "\ - "default is 10"); +MODULE_PARM_DESC(recur_count, " Recursion level for the stack overflow test"); module_param(cpoint_name, charp, 0444); MODULE_PARM_DESC(cpoint_name, " Crash Point, where kernel is to be crashed"); module_param(cpoint_type, charp, 0444); @@ -188,7 +228,7 @@ static int jp_scsi_dispatch_cmd(struct scsi_cmnd *cmd) } #ifdef CONFIG_IDE -int jp_generic_ide_ioctl(ide_drive_t *drive, struct file *file, +static int jp_generic_ide_ioctl(ide_drive_t *drive, struct file *file, struct block_device *bdev, unsigned int cmd, unsigned long arg) { @@ -263,16 +303,64 @@ static int lkdtm_parse_commandline(void) return -EINVAL; } -static int recursive_loop(int a) +static int recursive_loop(int remaining) { - char buf[1024]; + char buf[REC_STACK_SIZE]; - memset(buf,0xFF,1024); - recur_count--; - if (!recur_count) + /* Make sure compiler does not optimize this away. */ + memset(buf, (remaining & 0xff) | 0x1, REC_STACK_SIZE); + if (!remaining) return 0; else - return recursive_loop(a); + return recursive_loop(remaining - 1); +} + +static void do_nothing(void) +{ + return; +} + +/* Must immediately follow do_nothing for size calculuations to work out. */ +static void do_overwritten(void) +{ + pr_info("do_overwritten wasn't overwritten!\n"); + return; +} + +static noinline void corrupt_stack(void) +{ + /* Use default char array length that triggers stack protection. */ + char data[8]; + + memset((void *)data, 0, 64); +} + +static void execute_location(void *dst) +{ + void (*func)(void) = dst; + + pr_info("attempting ok execution at %p\n", do_nothing); + do_nothing(); + + memcpy(dst, do_nothing, EXEC_SIZE); + flush_icache_range((unsigned long)dst, (unsigned long)dst + EXEC_SIZE); + pr_info("attempting bad execution at %p\n", func); + func(); +} + +static void execute_user_location(void *dst) +{ + /* Intentionally crossing kernel/user memory boundary. */ + void (*func)(void) = dst; + + pr_info("attempting ok execution at %p\n", do_nothing); + do_nothing(); + + if (copy_to_user((void __user *)dst, do_nothing, EXEC_SIZE)) + return; + flush_icache_range((unsigned long)dst, (unsigned long)dst + EXEC_SIZE); + pr_info("attempting bad execution at %p\n", func); + func(); } static void lkdtm_do_action(enum ctype which) @@ -284,6 +372,9 @@ static void lkdtm_do_action(enum ctype which) case CT_BUG: BUG(); break; + case CT_WARNING: + WARN_ON(1); + break; case CT_EXCEPTION: *((int *) 0) = 0; break; @@ -292,15 +383,11 @@ static void lkdtm_do_action(enum ctype which) ; break; case CT_OVERFLOW: - (void) recursive_loop(0); + (void) recursive_loop(recur_count); break; - case CT_CORRUPT_STACK: { - volatile u32 data[8]; - volatile u32 *p = data; - - p[12] = 0x12345678; + case CT_CORRUPT_STACK: + corrupt_stack(); break; - } case CT_UNALIGNED_LOAD_STORE_WRITE: { static u8 data[5] __attribute__((aligned(4))) = {1, 2, 3, 4, 5}; @@ -340,10 +427,101 @@ static void lkdtm_do_action(enum ctype which) for (;;) cpu_relax(); break; + case CT_SPINLOCKUP: + /* Must be called twice to trigger. */ + spin_lock(&lock_me_up); + /* Let sparse know we intended to exit holding the lock. */ + __release(&lock_me_up); + break; case CT_HUNG_TASK: set_current_state(TASK_UNINTERRUPTIBLE); schedule(); break; + case CT_EXEC_DATA: + execute_location(data_area); + break; + case CT_EXEC_STACK: { + u8 stack_area[EXEC_SIZE]; + execute_location(stack_area); + break; + } + case CT_EXEC_KMALLOC: { + u32 *kmalloc_area = kmalloc(EXEC_SIZE, GFP_KERNEL); + execute_location(kmalloc_area); + kfree(kmalloc_area); + break; + } + case CT_EXEC_VMALLOC: { + u32 *vmalloc_area = vmalloc(EXEC_SIZE); + execute_location(vmalloc_area); + vfree(vmalloc_area); + break; + } + case CT_EXEC_USERSPACE: { + unsigned long user_addr; + + user_addr = vm_mmap(NULL, 0, PAGE_SIZE, + PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_ANONYMOUS | MAP_PRIVATE, 0); + if (user_addr >= TASK_SIZE) { + pr_warn("Failed to allocate user memory\n"); + return; + } + execute_user_location((void *)user_addr); + vm_munmap(user_addr, PAGE_SIZE); + break; + } + case CT_ACCESS_USERSPACE: { + unsigned long user_addr, tmp; + unsigned long *ptr; + + user_addr = vm_mmap(NULL, 0, PAGE_SIZE, + PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_ANONYMOUS | MAP_PRIVATE, 0); + if (user_addr >= TASK_SIZE) { + pr_warn("Failed to allocate user memory\n"); + return; + } + + ptr = (unsigned long *)user_addr; + + pr_info("attempting bad read at %p\n", ptr); + tmp = *ptr; + tmp += 0xc0dec0de; + + pr_info("attempting bad write at %p\n", ptr); + *ptr = tmp; + + vm_munmap(user_addr, PAGE_SIZE); + + break; + } + case CT_WRITE_RO: { + unsigned long *ptr; + + ptr = (unsigned long *)&rodata; + + pr_info("attempting bad write at %p\n", ptr); + *ptr ^= 0xabcd1234; + + break; + } + case CT_WRITE_KERN: { + size_t size; + unsigned char *ptr; + + size = (unsigned long)do_overwritten - + (unsigned long)do_nothing; + ptr = (unsigned char *)do_overwritten; + + pr_info("attempting bad %zu byte write at %p\n", size, ptr); + memcpy(ptr, (unsigned char *)do_nothing, size); + flush_icache_range((unsigned long)ptr, + (unsigned long)(ptr + size)); + + do_overwritten(); + break; + } case CT_NONE: default: break; @@ -358,8 +536,8 @@ static void lkdtm_handler(void) spin_lock_irqsave(&count_lock, flags); count--; - printk(KERN_INFO "lkdtm: Crash point %s of type %s hit, trigger in %d rounds\n", - cp_name_to_str(cpoint), cp_type_to_str(cptype), count); + pr_info("Crash point %s of type %s hit, trigger in %d rounds\n", + cp_name_to_str(cpoint), cp_type_to_str(cptype), count); if (count == 0) { do_it = true; @@ -416,18 +594,18 @@ static int lkdtm_register_cpoint(enum cname which) lkdtm.kp.symbol_name = "generic_ide_ioctl"; lkdtm.entry = (kprobe_opcode_t*) jp_generic_ide_ioctl; #else - printk(KERN_INFO "lkdtm: Crash point not available\n"); + pr_info("Crash point not available\n"); return -EINVAL; #endif break; default: - printk(KERN_INFO "lkdtm: Invalid Crash Point\n"); + pr_info("Invalid Crash Point\n"); return -EINVAL; } cpoint = which; if ((ret = register_jprobe(&lkdtm)) < 0) { - printk(KERN_INFO "lkdtm: Couldn't register jprobe\n"); + pr_info("Couldn't register jprobe\n"); cpoint = CN_INVALID; } @@ -574,8 +752,7 @@ static ssize_t direct_entry(struct file *f, const char __user *user_buf, if (type == CT_NONE) return -EINVAL; - printk(KERN_INFO "lkdtm: Performing direct entry %s\n", - cp_type_to_str(type)); + pr_info("Performing direct entry %s\n", cp_type_to_str(type)); lkdtm_do_action(type); *off += count; @@ -637,7 +814,7 @@ static int __init lkdtm_module_init(void) /* Register debugfs interface */ lkdtm_debugfs_root = debugfs_create_dir("provoke-crash", NULL); if (!lkdtm_debugfs_root) { - printk(KERN_ERR "lkdtm: creating root dir failed\n"); + pr_err("creating root dir failed\n"); return -ENODEV; } @@ -652,28 +829,26 @@ static int __init lkdtm_module_init(void) de = debugfs_create_file(cur->name, 0644, lkdtm_debugfs_root, NULL, &cur->fops); if (de == NULL) { - printk(KERN_ERR "lkdtm: could not create %s\n", - cur->name); + pr_err("could not create %s\n", cur->name); goto out_err; } } if (lkdtm_parse_commandline() == -EINVAL) { - printk(KERN_INFO "lkdtm: Invalid command\n"); + pr_info("Invalid command\n"); goto out_err; } if (cpoint != CN_INVALID && cptype != CT_NONE) { ret = lkdtm_register_cpoint(cpoint); if (ret < 0) { - printk(KERN_INFO "lkdtm: Invalid crash point %d\n", - cpoint); + pr_info("Invalid crash point %d\n", cpoint); goto out_err; } - printk(KERN_INFO "lkdtm: Crash point %s of type %s registered\n", - cpoint_name, cpoint_type); + pr_info("Crash point %s of type %s registered\n", + cpoint_name, cpoint_type); } else { - printk(KERN_INFO "lkdtm: No crash points registered, enable through debugfs\n"); + pr_info("No crash points registered, enable through debugfs\n"); } return 0; @@ -688,7 +863,7 @@ static void __exit lkdtm_module_exit(void) debugfs_remove_recursive(lkdtm_debugfs_root); unregister_jprobe(&lkdtm); - printk(KERN_INFO "lkdtm: Crash point unregistered\n"); + pr_info("Crash point unregistered\n"); } module_init(lkdtm_module_init); diff --git a/drivers/misc/mei/Kconfig b/drivers/misc/mei/Kconfig index 5a79ccde2fd..d23384dde73 100644 --- a/drivers/misc/mei/Kconfig +++ b/drivers/misc/mei/Kconfig @@ -1,11 +1,21 @@ config INTEL_MEI - tristate "Intel Management Engine Interface (Intel MEI)" + tristate "Intel Management Engine Interface" depends on X86 && PCI && WATCHDOG_CORE help The Intel Management Engine (Intel ME) provides Manageability, Security and Media services for system containing Intel chipsets. if selected /dev/mei misc device will be created. + For more information see + <http://software.intel.com/en-us/manageability/> + +config INTEL_MEI_ME + tristate "ME Enabled Intel Chipsets" + select INTEL_MEI + depends on X86 && PCI && WATCHDOG_CORE + help + MEI support for ME Enabled Intel chipsets. + Supported Chipsets are: 7 Series Chipset Family 6 Series Chipset Family @@ -24,5 +34,12 @@ config INTEL_MEI 82Q33 Express 82X38/X48 Express - For more information see - <http://software.intel.com/en-us/manageability/> +config INTEL_MEI_TXE + tristate "Intel Trusted Execution Environment with ME Interface" + select INTEL_MEI + depends on X86 && PCI && WATCHDOG_CORE + help + MEI Support for Trusted Execution Environment device on Intel SoCs + + Supported SoCs: + Intel Bay Trail diff --git a/drivers/misc/mei/Makefile b/drivers/misc/mei/Makefile index 57168db6c7e..8ebc6cda137 100644 --- a/drivers/misc/mei/Makefile +++ b/drivers/misc/mei/Makefile @@ -1,11 +1,23 @@ # # Makefile - Intel Management Engine Interface (Intel MEI) Linux driver -# Copyright (c) 2010-2011, Intel Corporation. +# Copyright (c) 2010-2014, Intel Corporation. # obj-$(CONFIG_INTEL_MEI) += mei.o mei-objs := init.o +mei-objs += hbm.o mei-objs += interrupt.o -mei-objs += interface.o -mei-objs += iorw.o +mei-objs += client.o mei-objs += main.o +mei-objs += amthif.o mei-objs += wd.o +mei-objs += bus.o +mei-objs += nfc.o +mei-$(CONFIG_DEBUG_FS) += debugfs.o + +obj-$(CONFIG_INTEL_MEI_ME) += mei-me.o +mei-me-objs := pci-me.o +mei-me-objs += hw-me.o + +obj-$(CONFIG_INTEL_MEI_TXE) += mei-txe.o +mei-txe-objs := pci-txe.o +mei-txe-objs += hw-txe.o diff --git a/drivers/misc/mei/amthif.c b/drivers/misc/mei/amthif.c new file mode 100644 index 00000000000..0d6234db00f --- /dev/null +++ b/drivers/misc/mei/amthif.c @@ -0,0 +1,745 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/fcntl.h> +#include <linux/aio.h> +#include <linux/pci.h> +#include <linux/ioctl.h> +#include <linux/cdev.h> +#include <linux/list.h> +#include <linux/delay.h> +#include <linux/sched.h> +#include <linux/uuid.h> +#include <linux/jiffies.h> +#include <linux/uaccess.h> + +#include <linux/mei.h> + +#include "mei_dev.h" +#include "hbm.h" +#include "client.h" + +const uuid_le mei_amthif_guid = UUID_LE(0x12f80028, 0xb4b7, 0x4b2d, + 0xac, 0xa8, 0x46, 0xe0, + 0xff, 0x65, 0x81, 0x4c); + +/** + * mei_amthif_reset_params - initializes mei device iamthif + * + * @dev: the device structure + */ +void mei_amthif_reset_params(struct mei_device *dev) +{ + /* reset iamthif parameters. */ + dev->iamthif_current_cb = NULL; + dev->iamthif_msg_buf_size = 0; + dev->iamthif_msg_buf_index = 0; + dev->iamthif_canceled = false; + dev->iamthif_ioctl = false; + dev->iamthif_state = MEI_IAMTHIF_IDLE; + dev->iamthif_timer = 0; + dev->iamthif_stall_timer = 0; + dev->iamthif_open_count = 0; +} + +/** + * mei_amthif_host_init - mei initialization amthif client. + * + * @dev: the device structure + * + */ +int mei_amthif_host_init(struct mei_device *dev) +{ + struct mei_cl *cl = &dev->iamthif_cl; + unsigned char *msg_buf; + int ret, i; + + dev->iamthif_state = MEI_IAMTHIF_IDLE; + + mei_cl_init(cl, dev); + + i = mei_me_cl_by_uuid(dev, &mei_amthif_guid); + if (i < 0) { + dev_info(&dev->pdev->dev, + "amthif: failed to find the client %d\n", i); + return -ENOTTY; + } + + cl->me_client_id = dev->me_clients[i].client_id; + + /* Assign iamthif_mtu to the value received from ME */ + + dev->iamthif_mtu = dev->me_clients[i].props.max_msg_length; + dev_dbg(&dev->pdev->dev, "IAMTHIF_MTU = %d\n", + dev->me_clients[i].props.max_msg_length); + + kfree(dev->iamthif_msg_buf); + dev->iamthif_msg_buf = NULL; + + /* allocate storage for ME message buffer */ + msg_buf = kcalloc(dev->iamthif_mtu, + sizeof(unsigned char), GFP_KERNEL); + if (!msg_buf) { + dev_err(&dev->pdev->dev, "amthif: memory allocation for ME message buffer failed.\n"); + return -ENOMEM; + } + + dev->iamthif_msg_buf = msg_buf; + + ret = mei_cl_link(cl, MEI_IAMTHIF_HOST_CLIENT_ID); + + if (ret < 0) { + dev_err(&dev->pdev->dev, + "amthif: failed link client %d\n", ret); + return ret; + } + + ret = mei_cl_connect(cl, NULL); + + dev->iamthif_state = MEI_IAMTHIF_IDLE; + + return ret; +} + +/** + * mei_amthif_find_read_list_entry - finds a amthilist entry for current file + * + * @dev: the device structure + * @file: pointer to file object + * + * returns returned a list entry on success, NULL on failure. + */ +struct mei_cl_cb *mei_amthif_find_read_list_entry(struct mei_device *dev, + struct file *file) +{ + struct mei_cl_cb *cb; + + list_for_each_entry(cb, &dev->amthif_rd_complete_list.list, list) { + if (cb->cl && cb->cl == &dev->iamthif_cl && + cb->file_object == file) + return cb; + } + return NULL; +} + + +/** + * mei_amthif_read - read data from AMTHIF client + * + * @dev: the device structure + * @if_num: minor number + * @file: pointer to file object + * @*ubuf: pointer to user data in user space + * @length: data length to read + * @offset: data read offset + * + * Locking: called under "dev->device_lock" lock + * + * returns + * returned data length on success, + * zero if no data to read, + * negative on failure. + */ +int mei_amthif_read(struct mei_device *dev, struct file *file, + char __user *ubuf, size_t length, loff_t *offset) +{ + int rets; + int wait_ret; + struct mei_cl_cb *cb = NULL; + struct mei_cl *cl = file->private_data; + unsigned long timeout; + int i; + + /* Only possible if we are in timeout */ + if (!cl || cl != &dev->iamthif_cl) { + dev_dbg(&dev->pdev->dev, "bad file ext.\n"); + return -ETIME; + } + + i = mei_me_cl_by_id(dev, dev->iamthif_cl.me_client_id); + if (i < 0) { + dev_dbg(&dev->pdev->dev, "amthif client not found.\n"); + return -ENOTTY; + } + dev_dbg(&dev->pdev->dev, "checking amthif data\n"); + cb = mei_amthif_find_read_list_entry(dev, file); + + /* Check for if we can block or not*/ + if (cb == NULL && file->f_flags & O_NONBLOCK) + return -EAGAIN; + + + dev_dbg(&dev->pdev->dev, "waiting for amthif data\n"); + while (cb == NULL) { + /* unlock the Mutex */ + mutex_unlock(&dev->device_lock); + + wait_ret = wait_event_interruptible(dev->iamthif_cl.wait, + (cb = mei_amthif_find_read_list_entry(dev, file))); + + /* Locking again the Mutex */ + mutex_lock(&dev->device_lock); + + if (wait_ret) + return -ERESTARTSYS; + + dev_dbg(&dev->pdev->dev, "woke up from sleep\n"); + } + + + dev_dbg(&dev->pdev->dev, "Got amthif data\n"); + dev->iamthif_timer = 0; + + if (cb) { + timeout = cb->read_time + + mei_secs_to_jiffies(MEI_IAMTHIF_READ_TIMER); + dev_dbg(&dev->pdev->dev, "amthif timeout = %lud\n", + timeout); + + if (time_after(jiffies, timeout)) { + dev_dbg(&dev->pdev->dev, "amthif Time out\n"); + /* 15 sec for the message has expired */ + list_del(&cb->list); + rets = -ETIME; + goto free; + } + } + /* if the whole message will fit remove it from the list */ + if (cb->buf_idx >= *offset && length >= (cb->buf_idx - *offset)) + list_del(&cb->list); + else if (cb->buf_idx > 0 && cb->buf_idx <= *offset) { + /* end of the message has been reached */ + list_del(&cb->list); + rets = 0; + goto free; + } + /* else means that not full buffer will be read and do not + * remove message from deletion list + */ + + dev_dbg(&dev->pdev->dev, "amthif cb->response_buffer size - %d\n", + cb->response_buffer.size); + dev_dbg(&dev->pdev->dev, "amthif cb->buf_idx - %lu\n", cb->buf_idx); + + /* length is being truncated to PAGE_SIZE, however, + * the buf_idx may point beyond */ + length = min_t(size_t, length, (cb->buf_idx - *offset)); + + if (copy_to_user(ubuf, cb->response_buffer.data + *offset, length)) { + dev_dbg(&dev->pdev->dev, "failed to copy data to userland\n"); + rets = -EFAULT; + } else { + rets = length; + if ((*offset + length) < cb->buf_idx) { + *offset += length; + goto out; + } + } +free: + dev_dbg(&dev->pdev->dev, "free amthif cb memory.\n"); + *offset = 0; + mei_io_cb_free(cb); +out: + return rets; +} + +/** + * mei_amthif_send_cmd - send amthif command to the ME + * + * @dev: the device structure + * @cb: mei call back struct + * + * returns 0 on success, <0 on failure. + * + */ +static int mei_amthif_send_cmd(struct mei_device *dev, struct mei_cl_cb *cb) +{ + struct mei_msg_hdr mei_hdr; + int ret; + + if (!dev || !cb) + return -ENODEV; + + dev_dbg(&dev->pdev->dev, "write data to amthif client.\n"); + + dev->iamthif_state = MEI_IAMTHIF_WRITING; + dev->iamthif_current_cb = cb; + dev->iamthif_file_object = cb->file_object; + dev->iamthif_canceled = false; + dev->iamthif_ioctl = true; + dev->iamthif_msg_buf_size = cb->request_buffer.size; + memcpy(dev->iamthif_msg_buf, cb->request_buffer.data, + cb->request_buffer.size); + + ret = mei_cl_flow_ctrl_creds(&dev->iamthif_cl); + if (ret < 0) + return ret; + + if (ret && mei_hbuf_acquire(dev)) { + ret = 0; + if (cb->request_buffer.size > mei_hbuf_max_len(dev)) { + mei_hdr.length = mei_hbuf_max_len(dev); + mei_hdr.msg_complete = 0; + } else { + mei_hdr.length = cb->request_buffer.size; + mei_hdr.msg_complete = 1; + } + + mei_hdr.host_addr = dev->iamthif_cl.host_client_id; + mei_hdr.me_addr = dev->iamthif_cl.me_client_id; + mei_hdr.reserved = 0; + mei_hdr.internal = 0; + dev->iamthif_msg_buf_index += mei_hdr.length; + ret = mei_write_message(dev, &mei_hdr, dev->iamthif_msg_buf); + if (ret) + return ret; + + if (mei_hdr.msg_complete) { + if (mei_cl_flow_ctrl_reduce(&dev->iamthif_cl)) + return -EIO; + dev->iamthif_flow_control_pending = true; + dev->iamthif_state = MEI_IAMTHIF_FLOW_CONTROL; + dev_dbg(&dev->pdev->dev, "add amthif cb to write waiting list\n"); + dev->iamthif_current_cb = cb; + dev->iamthif_file_object = cb->file_object; + list_add_tail(&cb->list, &dev->write_waiting_list.list); + } else { + dev_dbg(&dev->pdev->dev, "message does not complete, so add amthif cb to write list.\n"); + list_add_tail(&cb->list, &dev->write_list.list); + } + } else { + list_add_tail(&cb->list, &dev->write_list.list); + } + return 0; +} + +/** + * mei_amthif_write - write amthif data to amthif client + * + * @dev: the device structure + * @cb: mei call back struct + * + * returns 0 on success, <0 on failure. + * + */ +int mei_amthif_write(struct mei_device *dev, struct mei_cl_cb *cb) +{ + int ret; + + if (!dev || !cb) + return -ENODEV; + + ret = mei_io_cb_alloc_resp_buf(cb, dev->iamthif_mtu); + if (ret) + return ret; + + cb->fop_type = MEI_FOP_WRITE; + + if (!list_empty(&dev->amthif_cmd_list.list) || + dev->iamthif_state != MEI_IAMTHIF_IDLE) { + dev_dbg(&dev->pdev->dev, + "amthif state = %d\n", dev->iamthif_state); + dev_dbg(&dev->pdev->dev, "AMTHIF: add cb to the wait list\n"); + list_add_tail(&cb->list, &dev->amthif_cmd_list.list); + return 0; + } + return mei_amthif_send_cmd(dev, cb); +} +/** + * mei_amthif_run_next_cmd + * + * @dev: the device structure + * + * returns 0 on success, <0 on failure. + */ +void mei_amthif_run_next_cmd(struct mei_device *dev) +{ + struct mei_cl_cb *pos = NULL; + struct mei_cl_cb *next = NULL; + int status; + + if (!dev) + return; + + dev->iamthif_msg_buf_size = 0; + dev->iamthif_msg_buf_index = 0; + dev->iamthif_canceled = false; + dev->iamthif_ioctl = true; + dev->iamthif_state = MEI_IAMTHIF_IDLE; + dev->iamthif_timer = 0; + dev->iamthif_file_object = NULL; + + dev_dbg(&dev->pdev->dev, "complete amthif cmd_list cb.\n"); + + list_for_each_entry_safe(pos, next, &dev->amthif_cmd_list.list, list) { + list_del(&pos->list); + + if (pos->cl && pos->cl == &dev->iamthif_cl) { + status = mei_amthif_send_cmd(dev, pos); + if (status) { + dev_dbg(&dev->pdev->dev, + "amthif write failed status = %d\n", + status); + return; + } + break; + } + } +} + + +unsigned int mei_amthif_poll(struct mei_device *dev, + struct file *file, poll_table *wait) +{ + unsigned int mask = 0; + + poll_wait(file, &dev->iamthif_cl.wait, wait); + + mutex_lock(&dev->device_lock); + if (!mei_cl_is_connected(&dev->iamthif_cl)) { + + mask = POLLERR; + + } else if (dev->iamthif_state == MEI_IAMTHIF_READ_COMPLETE && + dev->iamthif_file_object == file) { + + mask |= (POLLIN | POLLRDNORM); + dev_dbg(&dev->pdev->dev, "run next amthif cb\n"); + mei_amthif_run_next_cmd(dev); + } + mutex_unlock(&dev->device_lock); + + return mask; +} + + + +/** + * mei_amthif_irq_write - write iamthif command in irq thread context. + * + * @dev: the device structure. + * @cb_pos: callback block. + * @cl: private data of the file object. + * @cmpl_list: complete list. + * + * returns 0, OK; otherwise, error. + */ +int mei_amthif_irq_write(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list) +{ + struct mei_device *dev = cl->dev; + struct mei_msg_hdr mei_hdr; + size_t len = dev->iamthif_msg_buf_size - dev->iamthif_msg_buf_index; + u32 msg_slots = mei_data2slots(len); + int slots; + int rets; + + rets = mei_cl_flow_ctrl_creds(cl); + if (rets < 0) + return rets; + + if (rets == 0) { + cl_dbg(dev, cl, "No flow control credentials: not sending.\n"); + return 0; + } + + mei_hdr.host_addr = cl->host_client_id; + mei_hdr.me_addr = cl->me_client_id; + mei_hdr.reserved = 0; + mei_hdr.internal = 0; + + slots = mei_hbuf_empty_slots(dev); + + if (slots >= msg_slots) { + mei_hdr.length = len; + mei_hdr.msg_complete = 1; + /* Split the message only if we can write the whole host buffer */ + } else if (slots == dev->hbuf_depth) { + msg_slots = slots; + len = (slots * sizeof(u32)) - sizeof(struct mei_msg_hdr); + mei_hdr.length = len; + mei_hdr.msg_complete = 0; + } else { + /* wait for next time the host buffer is empty */ + return 0; + } + + dev_dbg(&dev->pdev->dev, MEI_HDR_FMT, MEI_HDR_PRM(&mei_hdr)); + + rets = mei_write_message(dev, &mei_hdr, + dev->iamthif_msg_buf + dev->iamthif_msg_buf_index); + if (rets) { + dev->iamthif_state = MEI_IAMTHIF_IDLE; + cl->status = rets; + list_del(&cb->list); + return rets; + } + + if (mei_cl_flow_ctrl_reduce(cl)) + return -EIO; + + dev->iamthif_msg_buf_index += mei_hdr.length; + cl->status = 0; + + if (mei_hdr.msg_complete) { + dev->iamthif_state = MEI_IAMTHIF_FLOW_CONTROL; + dev->iamthif_flow_control_pending = true; + + /* save iamthif cb sent to amthif client */ + cb->buf_idx = dev->iamthif_msg_buf_index; + dev->iamthif_current_cb = cb; + + list_move_tail(&cb->list, &dev->write_waiting_list.list); + } + + + return 0; +} + +/** + * mei_amthif_irq_read_message - read routine after ISR to + * handle the read amthif message + * + * @dev: the device structure + * @mei_hdr: header of amthif message + * @complete_list: An instance of our list structure + * + * returns 0 on success, <0 on failure. + */ +int mei_amthif_irq_read_msg(struct mei_device *dev, + struct mei_msg_hdr *mei_hdr, + struct mei_cl_cb *complete_list) +{ + struct mei_cl_cb *cb; + unsigned char *buffer; + + BUG_ON(mei_hdr->me_addr != dev->iamthif_cl.me_client_id); + BUG_ON(dev->iamthif_state != MEI_IAMTHIF_READING); + + buffer = dev->iamthif_msg_buf + dev->iamthif_msg_buf_index; + BUG_ON(dev->iamthif_mtu < dev->iamthif_msg_buf_index + mei_hdr->length); + + mei_read_slots(dev, buffer, mei_hdr->length); + + dev->iamthif_msg_buf_index += mei_hdr->length; + + if (!mei_hdr->msg_complete) + return 0; + + dev_dbg(&dev->pdev->dev, "amthif_message_buffer_index =%d\n", + mei_hdr->length); + + dev_dbg(&dev->pdev->dev, "completed amthif read.\n "); + if (!dev->iamthif_current_cb) + return -ENODEV; + + cb = dev->iamthif_current_cb; + dev->iamthif_current_cb = NULL; + + if (!cb->cl) + return -ENODEV; + + dev->iamthif_stall_timer = 0; + cb->buf_idx = dev->iamthif_msg_buf_index; + cb->read_time = jiffies; + if (dev->iamthif_ioctl && cb->cl == &dev->iamthif_cl) { + /* found the iamthif cb */ + dev_dbg(&dev->pdev->dev, "complete the amthif read cb.\n "); + dev_dbg(&dev->pdev->dev, "add the amthif read cb to complete.\n "); + list_add_tail(&cb->list, &complete_list->list); + } + return 0; +} + +/** + * mei_amthif_irq_read - prepares to read amthif data. + * + * @dev: the device structure. + * @slots: free slots. + * + * returns 0, OK; otherwise, error. + */ +int mei_amthif_irq_read(struct mei_device *dev, s32 *slots) +{ + u32 msg_slots = mei_data2slots(sizeof(struct hbm_flow_control)); + + if (*slots < msg_slots) + return -EMSGSIZE; + + *slots -= msg_slots; + + if (mei_hbm_cl_flow_control_req(dev, &dev->iamthif_cl)) { + dev_dbg(&dev->pdev->dev, "iamthif flow control failed\n"); + return -EIO; + } + + dev_dbg(&dev->pdev->dev, "iamthif flow control success\n"); + dev->iamthif_state = MEI_IAMTHIF_READING; + dev->iamthif_flow_control_pending = false; + dev->iamthif_msg_buf_index = 0; + dev->iamthif_msg_buf_size = 0; + dev->iamthif_stall_timer = MEI_IAMTHIF_STALL_TIMER; + dev->hbuf_is_ready = mei_hbuf_is_ready(dev); + return 0; +} + +/** + * mei_amthif_complete - complete amthif callback. + * + * @dev: the device structure. + * @cb_pos: callback block. + */ +void mei_amthif_complete(struct mei_device *dev, struct mei_cl_cb *cb) +{ + if (dev->iamthif_canceled != 1) { + dev->iamthif_state = MEI_IAMTHIF_READ_COMPLETE; + dev->iamthif_stall_timer = 0; + memcpy(cb->response_buffer.data, + dev->iamthif_msg_buf, + dev->iamthif_msg_buf_index); + list_add_tail(&cb->list, &dev->amthif_rd_complete_list.list); + dev_dbg(&dev->pdev->dev, "amthif read completed\n"); + dev->iamthif_timer = jiffies; + dev_dbg(&dev->pdev->dev, "dev->iamthif_timer = %ld\n", + dev->iamthif_timer); + } else { + mei_amthif_run_next_cmd(dev); + } + + dev_dbg(&dev->pdev->dev, "completing amthif call back.\n"); + wake_up_interruptible(&dev->iamthif_cl.wait); +} + +/** + * mei_clear_list - removes all callbacks associated with file + * from mei_cb_list + * + * @dev: device structure. + * @file: file structure + * @mei_cb_list: callbacks list + * + * mei_clear_list is called to clear resources associated with file + * when application calls close function or Ctrl-C was pressed + * + * returns true if callback removed from the list, false otherwise + */ +static bool mei_clear_list(struct mei_device *dev, + const struct file *file, struct list_head *mei_cb_list) +{ + struct mei_cl_cb *cb_pos = NULL; + struct mei_cl_cb *cb_next = NULL; + bool removed = false; + + /* list all list member */ + list_for_each_entry_safe(cb_pos, cb_next, mei_cb_list, list) { + /* check if list member associated with a file */ + if (file == cb_pos->file_object) { + /* remove member from the list */ + list_del(&cb_pos->list); + /* check if cb equal to current iamthif cb */ + if (dev->iamthif_current_cb == cb_pos) { + dev->iamthif_current_cb = NULL; + /* send flow control to iamthif client */ + mei_hbm_cl_flow_control_req(dev, + &dev->iamthif_cl); + } + /* free all allocated buffers */ + mei_io_cb_free(cb_pos); + cb_pos = NULL; + removed = true; + } + } + return removed; +} + +/** + * mei_clear_lists - removes all callbacks associated with file + * + * @dev: device structure + * @file: file structure + * + * mei_clear_lists is called to clear resources associated with file + * when application calls close function or Ctrl-C was pressed + * + * returns true if callback removed from the list, false otherwise + */ +static bool mei_clear_lists(struct mei_device *dev, struct file *file) +{ + bool removed = false; + + /* remove callbacks associated with a file */ + mei_clear_list(dev, file, &dev->amthif_cmd_list.list); + if (mei_clear_list(dev, file, &dev->amthif_rd_complete_list.list)) + removed = true; + + mei_clear_list(dev, file, &dev->ctrl_rd_list.list); + + if (mei_clear_list(dev, file, &dev->ctrl_wr_list.list)) + removed = true; + + if (mei_clear_list(dev, file, &dev->write_waiting_list.list)) + removed = true; + + if (mei_clear_list(dev, file, &dev->write_list.list)) + removed = true; + + /* check if iamthif_current_cb not NULL */ + if (dev->iamthif_current_cb && !removed) { + /* check file and iamthif current cb association */ + if (dev->iamthif_current_cb->file_object == file) { + /* remove cb */ + mei_io_cb_free(dev->iamthif_current_cb); + dev->iamthif_current_cb = NULL; + removed = true; + } + } + return removed; +} + +/** +* mei_amthif_release - the release function +* +* @dev: device structure +* @file: pointer to file structure +* +* returns 0 on success, <0 on error +*/ +int mei_amthif_release(struct mei_device *dev, struct file *file) +{ + if (dev->iamthif_open_count > 0) + dev->iamthif_open_count--; + + if (dev->iamthif_file_object == file && + dev->iamthif_state != MEI_IAMTHIF_IDLE) { + + dev_dbg(&dev->pdev->dev, "amthif canceled iamthif state %d\n", + dev->iamthif_state); + dev->iamthif_canceled = true; + if (dev->iamthif_state == MEI_IAMTHIF_READ_COMPLETE) { + dev_dbg(&dev->pdev->dev, "run next amthif iamthif cb\n"); + mei_amthif_run_next_cmd(dev); + } + } + + if (mei_clear_lists(dev, file)) + dev->iamthif_state = MEI_IAMTHIF_IDLE; + + return 0; +} diff --git a/drivers/misc/mei/bus.c b/drivers/misc/mei/bus.c new file mode 100644 index 00000000000..0e993ef28b9 --- /dev/null +++ b/drivers/misc/mei/bus.c @@ -0,0 +1,548 @@ +/* + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2012-2013, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/module.h> +#include <linux/device.h> +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/mutex.h> +#include <linux/interrupt.h> +#include <linux/pci.h> +#include <linux/mei_cl_bus.h> + +#include "mei_dev.h" +#include "client.h" + +#define to_mei_cl_driver(d) container_of(d, struct mei_cl_driver, driver) +#define to_mei_cl_device(d) container_of(d, struct mei_cl_device, dev) + +static int mei_cl_device_match(struct device *dev, struct device_driver *drv) +{ + struct mei_cl_device *device = to_mei_cl_device(dev); + struct mei_cl_driver *driver = to_mei_cl_driver(drv); + const struct mei_cl_device_id *id; + + if (!device) + return 0; + + if (!driver || !driver->id_table) + return 0; + + id = driver->id_table; + + while (id->name[0]) { + if (!strncmp(dev_name(dev), id->name, sizeof(id->name))) + return 1; + + id++; + } + + return 0; +} + +static int mei_cl_device_probe(struct device *dev) +{ + struct mei_cl_device *device = to_mei_cl_device(dev); + struct mei_cl_driver *driver; + struct mei_cl_device_id id; + + if (!device) + return 0; + + driver = to_mei_cl_driver(dev->driver); + if (!driver || !driver->probe) + return -ENODEV; + + dev_dbg(dev, "Device probe\n"); + + strncpy(id.name, dev_name(dev), sizeof(id.name)); + + return driver->probe(device, &id); +} + +static int mei_cl_device_remove(struct device *dev) +{ + struct mei_cl_device *device = to_mei_cl_device(dev); + struct mei_cl_driver *driver; + + if (!device || !dev->driver) + return 0; + + if (device->event_cb) { + device->event_cb = NULL; + cancel_work_sync(&device->event_work); + } + + driver = to_mei_cl_driver(dev->driver); + if (!driver->remove) { + dev->driver = NULL; + + return 0; + } + + return driver->remove(device); +} + +static ssize_t modalias_show(struct device *dev, struct device_attribute *a, + char *buf) +{ + int len; + + len = snprintf(buf, PAGE_SIZE, "mei:%s\n", dev_name(dev)); + + return (len >= PAGE_SIZE) ? (PAGE_SIZE - 1) : len; +} +static DEVICE_ATTR_RO(modalias); + +static struct attribute *mei_cl_dev_attrs[] = { + &dev_attr_modalias.attr, + NULL, +}; +ATTRIBUTE_GROUPS(mei_cl_dev); + +static int mei_cl_uevent(struct device *dev, struct kobj_uevent_env *env) +{ + if (add_uevent_var(env, "MODALIAS=mei:%s", dev_name(dev))) + return -ENOMEM; + + return 0; +} + +static struct bus_type mei_cl_bus_type = { + .name = "mei", + .dev_groups = mei_cl_dev_groups, + .match = mei_cl_device_match, + .probe = mei_cl_device_probe, + .remove = mei_cl_device_remove, + .uevent = mei_cl_uevent, +}; + +static void mei_cl_dev_release(struct device *dev) +{ + kfree(to_mei_cl_device(dev)); +} + +static struct device_type mei_cl_device_type = { + .release = mei_cl_dev_release, +}; + +static struct mei_cl *mei_bus_find_mei_cl_by_uuid(struct mei_device *dev, + uuid_le uuid) +{ + struct mei_cl *cl; + + list_for_each_entry(cl, &dev->device_list, device_link) { + if (!uuid_le_cmp(uuid, cl->device_uuid)) + return cl; + } + + return NULL; +} +struct mei_cl_device *mei_cl_add_device(struct mei_device *dev, + uuid_le uuid, char *name, + struct mei_cl_ops *ops) +{ + struct mei_cl_device *device; + struct mei_cl *cl; + int status; + + cl = mei_bus_find_mei_cl_by_uuid(dev, uuid); + if (cl == NULL) + return NULL; + + device = kzalloc(sizeof(struct mei_cl_device), GFP_KERNEL); + if (!device) + return NULL; + + device->cl = cl; + device->ops = ops; + + device->dev.parent = &dev->pdev->dev; + device->dev.bus = &mei_cl_bus_type; + device->dev.type = &mei_cl_device_type; + + dev_set_name(&device->dev, "%s", name); + + status = device_register(&device->dev); + if (status) { + dev_err(&dev->pdev->dev, "Failed to register MEI device\n"); + kfree(device); + return NULL; + } + + cl->device = device; + + dev_dbg(&device->dev, "client %s registered\n", name); + + return device; +} +EXPORT_SYMBOL_GPL(mei_cl_add_device); + +void mei_cl_remove_device(struct mei_cl_device *device) +{ + device_unregister(&device->dev); +} +EXPORT_SYMBOL_GPL(mei_cl_remove_device); + +int __mei_cl_driver_register(struct mei_cl_driver *driver, struct module *owner) +{ + int err; + + driver->driver.name = driver->name; + driver->driver.owner = owner; + driver->driver.bus = &mei_cl_bus_type; + + err = driver_register(&driver->driver); + if (err) + return err; + + pr_debug("mei: driver [%s] registered\n", driver->driver.name); + + return 0; +} +EXPORT_SYMBOL_GPL(__mei_cl_driver_register); + +void mei_cl_driver_unregister(struct mei_cl_driver *driver) +{ + driver_unregister(&driver->driver); + + pr_debug("mei: driver [%s] unregistered\n", driver->driver.name); +} +EXPORT_SYMBOL_GPL(mei_cl_driver_unregister); + +static int ___mei_cl_send(struct mei_cl *cl, u8 *buf, size_t length, + bool blocking) +{ + struct mei_device *dev; + struct mei_cl_cb *cb; + int id; + int rets; + + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + dev = cl->dev; + + if (cl->state != MEI_FILE_CONNECTED) + return -ENODEV; + + /* Check if we have an ME client device */ + id = mei_me_cl_by_id(dev, cl->me_client_id); + if (id < 0) + return id; + + if (length > dev->me_clients[id].props.max_msg_length) + return -EFBIG; + + cb = mei_io_cb_init(cl, NULL); + if (!cb) + return -ENOMEM; + + rets = mei_io_cb_alloc_req_buf(cb, length); + if (rets < 0) { + mei_io_cb_free(cb); + return rets; + } + + memcpy(cb->request_buffer.data, buf, length); + + mutex_lock(&dev->device_lock); + + rets = mei_cl_write(cl, cb, blocking); + + mutex_unlock(&dev->device_lock); + if (rets < 0) + mei_io_cb_free(cb); + + return rets; +} + +int __mei_cl_recv(struct mei_cl *cl, u8 *buf, size_t length) +{ + struct mei_device *dev; + struct mei_cl_cb *cb; + size_t r_length; + int err; + + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + dev = cl->dev; + + mutex_lock(&dev->device_lock); + + if (!cl->read_cb) { + err = mei_cl_read_start(cl, length); + if (err < 0) { + mutex_unlock(&dev->device_lock); + return err; + } + } + + if (cl->reading_state != MEI_READ_COMPLETE && + !waitqueue_active(&cl->rx_wait)) { + + mutex_unlock(&dev->device_lock); + + if (wait_event_interruptible(cl->rx_wait, + cl->reading_state == MEI_READ_COMPLETE || + mei_cl_is_transitioning(cl))) { + + if (signal_pending(current)) + return -EINTR; + return -ERESTARTSYS; + } + + mutex_lock(&dev->device_lock); + } + + cb = cl->read_cb; + + if (cl->reading_state != MEI_READ_COMPLETE) { + r_length = 0; + goto out; + } + + r_length = min_t(size_t, length, cb->buf_idx); + + memcpy(buf, cb->response_buffer.data, r_length); + + mei_io_cb_free(cb); + cl->reading_state = MEI_IDLE; + cl->read_cb = NULL; + +out: + mutex_unlock(&dev->device_lock); + + return r_length; +} + +inline int __mei_cl_async_send(struct mei_cl *cl, u8 *buf, size_t length) +{ + return ___mei_cl_send(cl, buf, length, 0); +} + +inline int __mei_cl_send(struct mei_cl *cl, u8 *buf, size_t length) +{ + return ___mei_cl_send(cl, buf, length, 1); +} + +int mei_cl_send(struct mei_cl_device *device, u8 *buf, size_t length) +{ + struct mei_cl *cl = device->cl; + + if (cl == NULL) + return -ENODEV; + + if (device->ops && device->ops->send) + return device->ops->send(device, buf, length); + + return __mei_cl_send(cl, buf, length); +} +EXPORT_SYMBOL_GPL(mei_cl_send); + +int mei_cl_recv(struct mei_cl_device *device, u8 *buf, size_t length) +{ + struct mei_cl *cl = device->cl; + + if (cl == NULL) + return -ENODEV; + + if (device->ops && device->ops->recv) + return device->ops->recv(device, buf, length); + + return __mei_cl_recv(cl, buf, length); +} +EXPORT_SYMBOL_GPL(mei_cl_recv); + +static void mei_bus_event_work(struct work_struct *work) +{ + struct mei_cl_device *device; + + device = container_of(work, struct mei_cl_device, event_work); + + if (device->event_cb) + device->event_cb(device, device->events, device->event_context); + + device->events = 0; + + /* Prepare for the next read */ + mei_cl_read_start(device->cl, 0); +} + +int mei_cl_register_event_cb(struct mei_cl_device *device, + mei_cl_event_cb_t event_cb, void *context) +{ + if (device->event_cb) + return -EALREADY; + + device->events = 0; + device->event_cb = event_cb; + device->event_context = context; + INIT_WORK(&device->event_work, mei_bus_event_work); + + mei_cl_read_start(device->cl, 0); + + return 0; +} +EXPORT_SYMBOL_GPL(mei_cl_register_event_cb); + +void *mei_cl_get_drvdata(const struct mei_cl_device *device) +{ + return dev_get_drvdata(&device->dev); +} +EXPORT_SYMBOL_GPL(mei_cl_get_drvdata); + +void mei_cl_set_drvdata(struct mei_cl_device *device, void *data) +{ + dev_set_drvdata(&device->dev, data); +} +EXPORT_SYMBOL_GPL(mei_cl_set_drvdata); + +int mei_cl_enable_device(struct mei_cl_device *device) +{ + int err; + struct mei_device *dev; + struct mei_cl *cl = device->cl; + + if (cl == NULL) + return -ENODEV; + + dev = cl->dev; + + mutex_lock(&dev->device_lock); + + err = mei_cl_connect(cl, NULL); + if (err < 0) { + mutex_unlock(&dev->device_lock); + dev_err(&dev->pdev->dev, "Could not connect to the ME client"); + + return err; + } + + mutex_unlock(&dev->device_lock); + + if (device->event_cb && !cl->read_cb) + mei_cl_read_start(device->cl, 0); + + if (!device->ops || !device->ops->enable) + return 0; + + return device->ops->enable(device); +} +EXPORT_SYMBOL_GPL(mei_cl_enable_device); + +int mei_cl_disable_device(struct mei_cl_device *device) +{ + int err; + struct mei_device *dev; + struct mei_cl *cl = device->cl; + + if (cl == NULL) + return -ENODEV; + + dev = cl->dev; + + mutex_lock(&dev->device_lock); + + if (cl->state != MEI_FILE_CONNECTED) { + mutex_unlock(&dev->device_lock); + dev_err(&dev->pdev->dev, "Already disconnected"); + + return 0; + } + + cl->state = MEI_FILE_DISCONNECTING; + + err = mei_cl_disconnect(cl); + if (err < 0) { + mutex_unlock(&dev->device_lock); + dev_err(&dev->pdev->dev, + "Could not disconnect from the ME client"); + + return err; + } + + /* Flush queues and remove any pending read */ + mei_cl_flush_queues(cl); + + if (cl->read_cb) { + struct mei_cl_cb *cb = NULL; + + cb = mei_cl_find_read_cb(cl); + /* Remove entry from read list */ + if (cb) + list_del(&cb->list); + + cb = cl->read_cb; + cl->read_cb = NULL; + + if (cb) { + mei_io_cb_free(cb); + cb = NULL; + } + } + + device->event_cb = NULL; + + mutex_unlock(&dev->device_lock); + + if (!device->ops || !device->ops->disable) + return 0; + + return device->ops->disable(device); +} +EXPORT_SYMBOL_GPL(mei_cl_disable_device); + +void mei_cl_bus_rx_event(struct mei_cl *cl) +{ + struct mei_cl_device *device = cl->device; + + if (!device || !device->event_cb) + return; + + set_bit(MEI_CL_EVENT_RX, &device->events); + + schedule_work(&device->event_work); +} + +void mei_cl_bus_remove_devices(struct mei_device *dev) +{ + struct mei_cl *cl, *next; + + mutex_lock(&dev->device_lock); + list_for_each_entry_safe(cl, next, &dev->device_list, device_link) { + if (cl->device) + mei_cl_remove_device(cl->device); + + list_del(&cl->device_link); + mei_cl_unlink(cl); + kfree(cl); + } + mutex_unlock(&dev->device_lock); +} + +int __init mei_cl_bus_init(void) +{ + return bus_register(&mei_cl_bus_type); +} + +void __exit mei_cl_bus_exit(void) +{ + bus_unregister(&mei_cl_bus_type); +} diff --git a/drivers/misc/mei/client.c b/drivers/misc/mei/client.c new file mode 100644 index 00000000000..59d20c599b1 --- /dev/null +++ b/drivers/misc/mei/client.c @@ -0,0 +1,1067 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/pci.h> +#include <linux/sched.h> +#include <linux/wait.h> +#include <linux/delay.h> +#include <linux/pm_runtime.h> + +#include <linux/mei.h> + +#include "mei_dev.h" +#include "hbm.h" +#include "client.h" + +/** + * mei_me_cl_by_uuid - locate index of me client + * + * @dev: mei device + * + * Locking: called under "dev->device_lock" lock + * + * returns me client index or -ENOENT if not found + */ +int mei_me_cl_by_uuid(const struct mei_device *dev, const uuid_le *uuid) +{ + int i; + + for (i = 0; i < dev->me_clients_num; ++i) + if (uuid_le_cmp(*uuid, + dev->me_clients[i].props.protocol_name) == 0) + return i; + + return -ENOENT; +} + + +/** + * mei_me_cl_by_id return index to me_clients for client_id + * + * @dev: the device structure + * @client_id: me client id + * + * Locking: called under "dev->device_lock" lock + * + * returns index on success, -ENOENT on failure. + */ + +int mei_me_cl_by_id(struct mei_device *dev, u8 client_id) +{ + int i; + + for (i = 0; i < dev->me_clients_num; i++) + if (dev->me_clients[i].client_id == client_id) + return i; + + return -ENOENT; +} + + +/** + * mei_cl_cmp_id - tells if the clients are the same + * + * @cl1: host client 1 + * @cl2: host client 2 + * + * returns true - if the clients has same host and me ids + * false - otherwise + */ +static inline bool mei_cl_cmp_id(const struct mei_cl *cl1, + const struct mei_cl *cl2) +{ + return cl1 && cl2 && + (cl1->host_client_id == cl2->host_client_id) && + (cl1->me_client_id == cl2->me_client_id); +} + +/** + * mei_io_list_flush - removes cbs belonging to cl. + * + * @list: an instance of our list structure + * @cl: host client, can be NULL for flushing the whole list + * @free: whether to free the cbs + */ +static void __mei_io_list_flush(struct mei_cl_cb *list, + struct mei_cl *cl, bool free) +{ + struct mei_cl_cb *cb; + struct mei_cl_cb *next; + + /* enable removing everything if no cl is specified */ + list_for_each_entry_safe(cb, next, &list->list, list) { + if (!cl || (cb->cl && mei_cl_cmp_id(cl, cb->cl))) { + list_del(&cb->list); + if (free) + mei_io_cb_free(cb); + } + } +} + +/** + * mei_io_list_flush - removes list entry belonging to cl. + * + * @list: An instance of our list structure + * @cl: host client + */ +static inline void mei_io_list_flush(struct mei_cl_cb *list, struct mei_cl *cl) +{ + __mei_io_list_flush(list, cl, false); +} + + +/** + * mei_io_list_free - removes cb belonging to cl and free them + * + * @list: An instance of our list structure + * @cl: host client + */ +static inline void mei_io_list_free(struct mei_cl_cb *list, struct mei_cl *cl) +{ + __mei_io_list_flush(list, cl, true); +} + +/** + * mei_io_cb_free - free mei_cb_private related memory + * + * @cb: mei callback struct + */ +void mei_io_cb_free(struct mei_cl_cb *cb) +{ + if (cb == NULL) + return; + + kfree(cb->request_buffer.data); + kfree(cb->response_buffer.data); + kfree(cb); +} + +/** + * mei_io_cb_init - allocate and initialize io callback + * + * @cl - mei client + * @fp: pointer to file structure + * + * returns mei_cl_cb pointer or NULL; + */ +struct mei_cl_cb *mei_io_cb_init(struct mei_cl *cl, struct file *fp) +{ + struct mei_cl_cb *cb; + + cb = kzalloc(sizeof(struct mei_cl_cb), GFP_KERNEL); + if (!cb) + return NULL; + + mei_io_list_init(cb); + + cb->file_object = fp; + cb->cl = cl; + cb->buf_idx = 0; + return cb; +} + +/** + * mei_io_cb_alloc_req_buf - allocate request buffer + * + * @cb: io callback structure + * @length: size of the buffer + * + * returns 0 on success + * -EINVAL if cb is NULL + * -ENOMEM if allocation failed + */ +int mei_io_cb_alloc_req_buf(struct mei_cl_cb *cb, size_t length) +{ + if (!cb) + return -EINVAL; + + if (length == 0) + return 0; + + cb->request_buffer.data = kmalloc(length, GFP_KERNEL); + if (!cb->request_buffer.data) + return -ENOMEM; + cb->request_buffer.size = length; + return 0; +} +/** + * mei_io_cb_alloc_resp_buf - allocate response buffer + * + * @cb: io callback structure + * @length: size of the buffer + * + * returns 0 on success + * -EINVAL if cb is NULL + * -ENOMEM if allocation failed + */ +int mei_io_cb_alloc_resp_buf(struct mei_cl_cb *cb, size_t length) +{ + if (!cb) + return -EINVAL; + + if (length == 0) + return 0; + + cb->response_buffer.data = kmalloc(length, GFP_KERNEL); + if (!cb->response_buffer.data) + return -ENOMEM; + cb->response_buffer.size = length; + return 0; +} + + + +/** + * mei_cl_flush_queues - flushes queue lists belonging to cl. + * + * @cl: host client + */ +int mei_cl_flush_queues(struct mei_cl *cl) +{ + struct mei_device *dev; + + if (WARN_ON(!cl || !cl->dev)) + return -EINVAL; + + dev = cl->dev; + + cl_dbg(dev, cl, "remove list entry belonging to cl\n"); + mei_io_list_flush(&cl->dev->read_list, cl); + mei_io_list_free(&cl->dev->write_list, cl); + mei_io_list_free(&cl->dev->write_waiting_list, cl); + mei_io_list_flush(&cl->dev->ctrl_wr_list, cl); + mei_io_list_flush(&cl->dev->ctrl_rd_list, cl); + mei_io_list_flush(&cl->dev->amthif_cmd_list, cl); + mei_io_list_flush(&cl->dev->amthif_rd_complete_list, cl); + return 0; +} + + +/** + * mei_cl_init - initializes cl. + * + * @cl: host client to be initialized + * @dev: mei device + */ +void mei_cl_init(struct mei_cl *cl, struct mei_device *dev) +{ + memset(cl, 0, sizeof(struct mei_cl)); + init_waitqueue_head(&cl->wait); + init_waitqueue_head(&cl->rx_wait); + init_waitqueue_head(&cl->tx_wait); + INIT_LIST_HEAD(&cl->link); + INIT_LIST_HEAD(&cl->device_link); + cl->reading_state = MEI_IDLE; + cl->writing_state = MEI_IDLE; + cl->dev = dev; +} + +/** + * mei_cl_allocate - allocates cl structure and sets it up. + * + * @dev: mei device + * returns The allocated file or NULL on failure + */ +struct mei_cl *mei_cl_allocate(struct mei_device *dev) +{ + struct mei_cl *cl; + + cl = kmalloc(sizeof(struct mei_cl), GFP_KERNEL); + if (!cl) + return NULL; + + mei_cl_init(cl, dev); + + return cl; +} + +/** + * mei_cl_find_read_cb - find this cl's callback in the read list + * + * @cl: host client + * + * returns cb on success, NULL on error + */ +struct mei_cl_cb *mei_cl_find_read_cb(struct mei_cl *cl) +{ + struct mei_device *dev = cl->dev; + struct mei_cl_cb *cb; + + list_for_each_entry(cb, &dev->read_list.list, list) + if (mei_cl_cmp_id(cl, cb->cl)) + return cb; + return NULL; +} + +/** mei_cl_link: allocate host id in the host map + * + * @cl - host client + * @id - fixed host id or -1 for generic one + * + * returns 0 on success + * -EINVAL on incorrect values + * -ENONET if client not found + */ +int mei_cl_link(struct mei_cl *cl, int id) +{ + struct mei_device *dev; + long open_handle_count; + + if (WARN_ON(!cl || !cl->dev)) + return -EINVAL; + + dev = cl->dev; + + /* If Id is not assigned get one*/ + if (id == MEI_HOST_CLIENT_ID_ANY) + id = find_first_zero_bit(dev->host_clients_map, + MEI_CLIENTS_MAX); + + if (id >= MEI_CLIENTS_MAX) { + dev_err(&dev->pdev->dev, "id exceeded %d", MEI_CLIENTS_MAX); + return -EMFILE; + } + + open_handle_count = dev->open_handle_count + dev->iamthif_open_count; + if (open_handle_count >= MEI_MAX_OPEN_HANDLE_COUNT) { + dev_err(&dev->pdev->dev, "open_handle_count exceeded %d", + MEI_MAX_OPEN_HANDLE_COUNT); + return -EMFILE; + } + + dev->open_handle_count++; + + cl->host_client_id = id; + list_add_tail(&cl->link, &dev->file_list); + + set_bit(id, dev->host_clients_map); + + cl->state = MEI_FILE_INITIALIZING; + + cl_dbg(dev, cl, "link cl\n"); + return 0; +} + +/** + * mei_cl_unlink - remove me_cl from the list + * + * @cl: host client + */ +int mei_cl_unlink(struct mei_cl *cl) +{ + struct mei_device *dev; + + /* don't shout on error exit path */ + if (!cl) + return 0; + + /* wd and amthif might not be initialized */ + if (!cl->dev) + return 0; + + dev = cl->dev; + + cl_dbg(dev, cl, "unlink client"); + + if (dev->open_handle_count > 0) + dev->open_handle_count--; + + /* never clear the 0 bit */ + if (cl->host_client_id) + clear_bit(cl->host_client_id, dev->host_clients_map); + + list_del_init(&cl->link); + + cl->state = MEI_FILE_INITIALIZING; + + return 0; +} + + +void mei_host_client_init(struct work_struct *work) +{ + struct mei_device *dev = container_of(work, + struct mei_device, init_work); + struct mei_client_properties *client_props; + int i; + + mutex_lock(&dev->device_lock); + + for (i = 0; i < dev->me_clients_num; i++) { + client_props = &dev->me_clients[i].props; + + if (!uuid_le_cmp(client_props->protocol_name, mei_amthif_guid)) + mei_amthif_host_init(dev); + else if (!uuid_le_cmp(client_props->protocol_name, mei_wd_guid)) + mei_wd_host_init(dev); + else if (!uuid_le_cmp(client_props->protocol_name, mei_nfc_guid)) + mei_nfc_host_init(dev); + + } + + dev->dev_state = MEI_DEV_ENABLED; + dev->reset_count = 0; + + mutex_unlock(&dev->device_lock); + + pm_runtime_mark_last_busy(&dev->pdev->dev); + dev_dbg(&dev->pdev->dev, "rpm: autosuspend\n"); + pm_runtime_autosuspend(&dev->pdev->dev); +} + +/** + * mei_hbuf_acquire: try to acquire host buffer + * + * @dev: the device structure + * returns true if host buffer was acquired + */ +bool mei_hbuf_acquire(struct mei_device *dev) +{ + if (mei_pg_state(dev) == MEI_PG_ON || + dev->pg_event == MEI_PG_EVENT_WAIT) { + dev_dbg(&dev->pdev->dev, "device is in pg\n"); + return false; + } + + if (!dev->hbuf_is_ready) { + dev_dbg(&dev->pdev->dev, "hbuf is not ready\n"); + return false; + } + + dev->hbuf_is_ready = false; + + return true; +} + +/** + * mei_cl_disconnect - disconnect host client from the me one + * + * @cl: host client + * + * Locking: called under "dev->device_lock" lock + * + * returns 0 on success, <0 on failure. + */ +int mei_cl_disconnect(struct mei_cl *cl) +{ + struct mei_device *dev; + struct mei_cl_cb *cb; + int rets, err; + + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + dev = cl->dev; + + cl_dbg(dev, cl, "disconnecting"); + + if (cl->state != MEI_FILE_DISCONNECTING) + return 0; + + rets = pm_runtime_get(&dev->pdev->dev); + if (rets < 0 && rets != -EINPROGRESS) { + pm_runtime_put_noidle(&dev->pdev->dev); + cl_err(dev, cl, "rpm: get failed %d\n", rets); + return rets; + } + + cb = mei_io_cb_init(cl, NULL); + if (!cb) { + rets = -ENOMEM; + goto free; + } + + cb->fop_type = MEI_FOP_CLOSE; + if (mei_hbuf_acquire(dev)) { + if (mei_hbm_cl_disconnect_req(dev, cl)) { + rets = -ENODEV; + cl_err(dev, cl, "failed to disconnect.\n"); + goto free; + } + mdelay(10); /* Wait for hardware disconnection ready */ + list_add_tail(&cb->list, &dev->ctrl_rd_list.list); + } else { + cl_dbg(dev, cl, "add disconnect cb to control write list\n"); + list_add_tail(&cb->list, &dev->ctrl_wr_list.list); + + } + mutex_unlock(&dev->device_lock); + + err = wait_event_timeout(dev->wait_recvd_msg, + MEI_FILE_DISCONNECTED == cl->state, + mei_secs_to_jiffies(MEI_CL_CONNECT_TIMEOUT)); + + mutex_lock(&dev->device_lock); + if (MEI_FILE_DISCONNECTED == cl->state) { + rets = 0; + cl_dbg(dev, cl, "successfully disconnected from FW client.\n"); + } else { + rets = -ENODEV; + if (MEI_FILE_DISCONNECTED != cl->state) + cl_err(dev, cl, "wrong status client disconnect.\n"); + + if (err) + cl_dbg(dev, cl, "wait failed disconnect err=%d\n", err); + + cl_err(dev, cl, "failed to disconnect from FW client.\n"); + } + + mei_io_list_flush(&dev->ctrl_rd_list, cl); + mei_io_list_flush(&dev->ctrl_wr_list, cl); +free: + cl_dbg(dev, cl, "rpm: autosuspend\n"); + pm_runtime_mark_last_busy(&dev->pdev->dev); + pm_runtime_put_autosuspend(&dev->pdev->dev); + + mei_io_cb_free(cb); + return rets; +} + + +/** + * mei_cl_is_other_connecting - checks if other + * client with the same me client id is connecting + * + * @cl: private data of the file object + * + * returns true if other client is connected, false - otherwise. + */ +bool mei_cl_is_other_connecting(struct mei_cl *cl) +{ + struct mei_device *dev; + struct mei_cl *ocl; /* the other client */ + + if (WARN_ON(!cl || !cl->dev)) + return false; + + dev = cl->dev; + + list_for_each_entry(ocl, &dev->file_list, link) { + if (ocl->state == MEI_FILE_CONNECTING && + ocl != cl && + cl->me_client_id == ocl->me_client_id) + return true; + + } + + return false; +} + +/** + * mei_cl_connect - connect host client to the me one + * + * @cl: host client + * + * Locking: called under "dev->device_lock" lock + * + * returns 0 on success, <0 on failure. + */ +int mei_cl_connect(struct mei_cl *cl, struct file *file) +{ + struct mei_device *dev; + struct mei_cl_cb *cb; + int rets; + + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + dev = cl->dev; + + rets = pm_runtime_get(&dev->pdev->dev); + if (rets < 0 && rets != -EINPROGRESS) { + pm_runtime_put_noidle(&dev->pdev->dev); + cl_err(dev, cl, "rpm: get failed %d\n", rets); + return rets; + } + + cb = mei_io_cb_init(cl, file); + if (!cb) { + rets = -ENOMEM; + goto out; + } + + cb->fop_type = MEI_FOP_CONNECT; + + /* run hbuf acquire last so we don't have to undo */ + if (!mei_cl_is_other_connecting(cl) && mei_hbuf_acquire(dev)) { + cl->state = MEI_FILE_CONNECTING; + if (mei_hbm_cl_connect_req(dev, cl)) { + rets = -ENODEV; + goto out; + } + cl->timer_count = MEI_CONNECT_TIMEOUT; + list_add_tail(&cb->list, &dev->ctrl_rd_list.list); + } else { + list_add_tail(&cb->list, &dev->ctrl_wr_list.list); + } + + mutex_unlock(&dev->device_lock); + wait_event_timeout(dev->wait_recvd_msg, + (cl->state == MEI_FILE_CONNECTED || + cl->state == MEI_FILE_DISCONNECTED), + mei_secs_to_jiffies(MEI_CL_CONNECT_TIMEOUT)); + mutex_lock(&dev->device_lock); + + if (cl->state != MEI_FILE_CONNECTED) { + /* something went really wrong */ + if (!cl->status) + cl->status = -EFAULT; + + mei_io_list_flush(&dev->ctrl_rd_list, cl); + mei_io_list_flush(&dev->ctrl_wr_list, cl); + } + + rets = cl->status; + +out: + cl_dbg(dev, cl, "rpm: autosuspend\n"); + pm_runtime_mark_last_busy(&dev->pdev->dev); + pm_runtime_put_autosuspend(&dev->pdev->dev); + + mei_io_cb_free(cb); + return rets; +} + +/** + * mei_cl_flow_ctrl_creds - checks flow_control credits for cl. + * + * @cl: private data of the file object + * + * returns 1 if mei_flow_ctrl_creds >0, 0 - otherwise. + * -ENOENT if mei_cl is not present + * -EINVAL if single_recv_buf == 0 + */ +int mei_cl_flow_ctrl_creds(struct mei_cl *cl) +{ + struct mei_device *dev; + struct mei_me_client *me_cl; + int id; + + if (WARN_ON(!cl || !cl->dev)) + return -EINVAL; + + dev = cl->dev; + + if (!dev->me_clients_num) + return 0; + + if (cl->mei_flow_ctrl_creds > 0) + return 1; + + id = mei_me_cl_by_id(dev, cl->me_client_id); + if (id < 0) { + cl_err(dev, cl, "no such me client %d\n", cl->me_client_id); + return id; + } + + me_cl = &dev->me_clients[id]; + if (me_cl->mei_flow_ctrl_creds) { + if (WARN_ON(me_cl->props.single_recv_buf == 0)) + return -EINVAL; + return 1; + } + return 0; +} + +/** + * mei_cl_flow_ctrl_reduce - reduces flow_control. + * + * @cl: private data of the file object + * + * @returns + * 0 on success + * -ENOENT when me client is not found + * -EINVAL when ctrl credits are <= 0 + */ +int mei_cl_flow_ctrl_reduce(struct mei_cl *cl) +{ + struct mei_device *dev; + struct mei_me_client *me_cl; + int id; + + if (WARN_ON(!cl || !cl->dev)) + return -EINVAL; + + dev = cl->dev; + + id = mei_me_cl_by_id(dev, cl->me_client_id); + if (id < 0) { + cl_err(dev, cl, "no such me client %d\n", cl->me_client_id); + return id; + } + + me_cl = &dev->me_clients[id]; + if (me_cl->props.single_recv_buf != 0) { + if (WARN_ON(me_cl->mei_flow_ctrl_creds <= 0)) + return -EINVAL; + me_cl->mei_flow_ctrl_creds--; + } else { + if (WARN_ON(cl->mei_flow_ctrl_creds <= 0)) + return -EINVAL; + cl->mei_flow_ctrl_creds--; + } + return 0; +} + +/** + * mei_cl_read_start - the start read client message function. + * + * @cl: host client + * + * returns 0 on success, <0 on failure. + */ +int mei_cl_read_start(struct mei_cl *cl, size_t length) +{ + struct mei_device *dev; + struct mei_cl_cb *cb; + int rets; + int i; + + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + dev = cl->dev; + + if (!mei_cl_is_connected(cl)) + return -ENODEV; + + if (cl->read_cb) { + cl_dbg(dev, cl, "read is pending.\n"); + return -EBUSY; + } + i = mei_me_cl_by_id(dev, cl->me_client_id); + if (i < 0) { + cl_err(dev, cl, "no such me client %d\n", cl->me_client_id); + return -ENOTTY; + } + + rets = pm_runtime_get(&dev->pdev->dev); + if (rets < 0 && rets != -EINPROGRESS) { + pm_runtime_put_noidle(&dev->pdev->dev); + cl_err(dev, cl, "rpm: get failed %d\n", rets); + return rets; + } + + cb = mei_io_cb_init(cl, NULL); + if (!cb) { + rets = -ENOMEM; + goto out; + } + + /* always allocate at least client max message */ + length = max_t(size_t, length, dev->me_clients[i].props.max_msg_length); + rets = mei_io_cb_alloc_resp_buf(cb, length); + if (rets) + goto out; + + cb->fop_type = MEI_FOP_READ; + if (mei_hbuf_acquire(dev)) { + rets = mei_hbm_cl_flow_control_req(dev, cl); + if (rets < 0) + goto out; + + list_add_tail(&cb->list, &dev->read_list.list); + } else { + list_add_tail(&cb->list, &dev->ctrl_wr_list.list); + } + + cl->read_cb = cb; + +out: + cl_dbg(dev, cl, "rpm: autosuspend\n"); + pm_runtime_mark_last_busy(&dev->pdev->dev); + pm_runtime_put_autosuspend(&dev->pdev->dev); + + if (rets) + mei_io_cb_free(cb); + + return rets; +} + +/** + * mei_cl_irq_write - write a message to device + * from the interrupt thread context + * + * @cl: client + * @cb: callback block. + * @cmpl_list: complete list. + * + * returns 0, OK; otherwise error. + */ +int mei_cl_irq_write(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list) +{ + struct mei_device *dev; + struct mei_msg_data *buf; + struct mei_msg_hdr mei_hdr; + size_t len; + u32 msg_slots; + int slots; + int rets; + + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + dev = cl->dev; + + buf = &cb->request_buffer; + + rets = mei_cl_flow_ctrl_creds(cl); + if (rets < 0) + return rets; + + if (rets == 0) { + cl_dbg(dev, cl, "No flow control credentials: not sending.\n"); + return 0; + } + + slots = mei_hbuf_empty_slots(dev); + len = buf->size - cb->buf_idx; + msg_slots = mei_data2slots(len); + + mei_hdr.host_addr = cl->host_client_id; + mei_hdr.me_addr = cl->me_client_id; + mei_hdr.reserved = 0; + mei_hdr.internal = cb->internal; + + if (slots >= msg_slots) { + mei_hdr.length = len; + mei_hdr.msg_complete = 1; + /* Split the message only if we can write the whole host buffer */ + } else if (slots == dev->hbuf_depth) { + msg_slots = slots; + len = (slots * sizeof(u32)) - sizeof(struct mei_msg_hdr); + mei_hdr.length = len; + mei_hdr.msg_complete = 0; + } else { + /* wait for next time the host buffer is empty */ + return 0; + } + + cl_dbg(dev, cl, "buf: size = %d idx = %lu\n", + cb->request_buffer.size, cb->buf_idx); + + rets = mei_write_message(dev, &mei_hdr, buf->data + cb->buf_idx); + if (rets) { + cl->status = rets; + list_move_tail(&cb->list, &cmpl_list->list); + return rets; + } + + cl->status = 0; + cl->writing_state = MEI_WRITING; + cb->buf_idx += mei_hdr.length; + + if (mei_hdr.msg_complete) { + if (mei_cl_flow_ctrl_reduce(cl)) + return -EIO; + list_move_tail(&cb->list, &dev->write_waiting_list.list); + } + + return 0; +} + +/** + * mei_cl_write - submit a write cb to mei device + assumes device_lock is locked + * + * @cl: host client + * @cl: write callback with filled data + * + * returns number of bytes sent on success, <0 on failure. + */ +int mei_cl_write(struct mei_cl *cl, struct mei_cl_cb *cb, bool blocking) +{ + struct mei_device *dev; + struct mei_msg_data *buf; + struct mei_msg_hdr mei_hdr; + int rets; + + + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + if (WARN_ON(!cb)) + return -EINVAL; + + dev = cl->dev; + + + buf = &cb->request_buffer; + + cl_dbg(dev, cl, "mei_cl_write %d\n", buf->size); + + rets = pm_runtime_get(&dev->pdev->dev); + if (rets < 0 && rets != -EINPROGRESS) { + pm_runtime_put_noidle(&dev->pdev->dev); + cl_err(dev, cl, "rpm: get failed %d\n", rets); + return rets; + } + + cb->fop_type = MEI_FOP_WRITE; + cb->buf_idx = 0; + cl->writing_state = MEI_IDLE; + + mei_hdr.host_addr = cl->host_client_id; + mei_hdr.me_addr = cl->me_client_id; + mei_hdr.reserved = 0; + mei_hdr.msg_complete = 0; + mei_hdr.internal = cb->internal; + + rets = mei_cl_flow_ctrl_creds(cl); + if (rets < 0) + goto err; + + if (rets == 0) { + cl_dbg(dev, cl, "No flow control credentials: not sending.\n"); + rets = buf->size; + goto out; + } + if (!mei_hbuf_acquire(dev)) { + cl_dbg(dev, cl, "Cannot acquire the host buffer: not sending.\n"); + rets = buf->size; + goto out; + } + + /* Check for a maximum length */ + if (buf->size > mei_hbuf_max_len(dev)) { + mei_hdr.length = mei_hbuf_max_len(dev); + mei_hdr.msg_complete = 0; + } else { + mei_hdr.length = buf->size; + mei_hdr.msg_complete = 1; + } + + rets = mei_write_message(dev, &mei_hdr, buf->data); + if (rets) + goto err; + + cl->writing_state = MEI_WRITING; + cb->buf_idx = mei_hdr.length; + +out: + if (mei_hdr.msg_complete) { + rets = mei_cl_flow_ctrl_reduce(cl); + if (rets < 0) + goto err; + + list_add_tail(&cb->list, &dev->write_waiting_list.list); + } else { + list_add_tail(&cb->list, &dev->write_list.list); + } + + + if (blocking && cl->writing_state != MEI_WRITE_COMPLETE) { + + mutex_unlock(&dev->device_lock); + rets = wait_event_interruptible(cl->tx_wait, + cl->writing_state == MEI_WRITE_COMPLETE); + mutex_lock(&dev->device_lock); + /* wait_event_interruptible returns -ERESTARTSYS */ + if (rets) { + if (signal_pending(current)) + rets = -EINTR; + goto err; + } + } + + rets = buf->size; +err: + cl_dbg(dev, cl, "rpm: autosuspend\n"); + pm_runtime_mark_last_busy(&dev->pdev->dev); + pm_runtime_put_autosuspend(&dev->pdev->dev); + + return rets; +} + + +/** + * mei_cl_complete - processes completed operation for a client + * + * @cl: private data of the file object. + * @cb: callback block. + */ +void mei_cl_complete(struct mei_cl *cl, struct mei_cl_cb *cb) +{ + if (cb->fop_type == MEI_FOP_WRITE) { + mei_io_cb_free(cb); + cb = NULL; + cl->writing_state = MEI_WRITE_COMPLETE; + if (waitqueue_active(&cl->tx_wait)) + wake_up_interruptible(&cl->tx_wait); + + } else if (cb->fop_type == MEI_FOP_READ && + MEI_READING == cl->reading_state) { + cl->reading_state = MEI_READ_COMPLETE; + if (waitqueue_active(&cl->rx_wait)) + wake_up_interruptible(&cl->rx_wait); + else + mei_cl_bus_rx_event(cl); + + } +} + + +/** + * mei_cl_all_disconnect - disconnect forcefully all connected clients + * + * @dev - mei device + */ + +void mei_cl_all_disconnect(struct mei_device *dev) +{ + struct mei_cl *cl; + + list_for_each_entry(cl, &dev->file_list, link) { + cl->state = MEI_FILE_DISCONNECTED; + cl->mei_flow_ctrl_creds = 0; + cl->timer_count = 0; + } +} + + +/** + * mei_cl_all_wakeup - wake up all readers and writers they can be interrupted + * + * @dev - mei device + */ +void mei_cl_all_wakeup(struct mei_device *dev) +{ + struct mei_cl *cl; + list_for_each_entry(cl, &dev->file_list, link) { + if (waitqueue_active(&cl->rx_wait)) { + cl_dbg(dev, cl, "Waking up reading client!\n"); + wake_up_interruptible(&cl->rx_wait); + } + if (waitqueue_active(&cl->tx_wait)) { + cl_dbg(dev, cl, "Waking up writing client!\n"); + wake_up_interruptible(&cl->tx_wait); + } + } +} + +/** + * mei_cl_all_write_clear - clear all pending writes + + * @dev - mei device + */ +void mei_cl_all_write_clear(struct mei_device *dev) +{ + mei_io_list_free(&dev->write_list, NULL); + mei_io_list_free(&dev->write_waiting_list, NULL); +} + + diff --git a/drivers/misc/mei/client.h b/drivers/misc/mei/client.h new file mode 100644 index 00000000000..96d5de0389f --- /dev/null +++ b/drivers/misc/mei/client.h @@ -0,0 +1,109 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#ifndef _MEI_CLIENT_H_ +#define _MEI_CLIENT_H_ + +#include <linux/types.h> +#include <linux/watchdog.h> +#include <linux/poll.h> +#include <linux/mei.h> + +#include "mei_dev.h" + +int mei_me_cl_by_uuid(const struct mei_device *dev, const uuid_le *cuuid); +int mei_me_cl_by_id(struct mei_device *dev, u8 client_id); + +/* + * MEI IO Functions + */ +struct mei_cl_cb *mei_io_cb_init(struct mei_cl *cl, struct file *fp); +void mei_io_cb_free(struct mei_cl_cb *priv_cb); +int mei_io_cb_alloc_req_buf(struct mei_cl_cb *cb, size_t length); +int mei_io_cb_alloc_resp_buf(struct mei_cl_cb *cb, size_t length); + + +/** + * mei_io_list_init - Sets up a queue list. + * + * @list: An instance cl callback structure + */ +static inline void mei_io_list_init(struct mei_cl_cb *list) +{ + INIT_LIST_HEAD(&list->list); +} +/* + * MEI Host Client Functions + */ + +struct mei_cl *mei_cl_allocate(struct mei_device *dev); +void mei_cl_init(struct mei_cl *cl, struct mei_device *dev); + + +int mei_cl_link(struct mei_cl *cl, int id); +int mei_cl_unlink(struct mei_cl *cl); + +int mei_cl_flush_queues(struct mei_cl *cl); +struct mei_cl_cb *mei_cl_find_read_cb(struct mei_cl *cl); + + +int mei_cl_flow_ctrl_creds(struct mei_cl *cl); + +int mei_cl_flow_ctrl_reduce(struct mei_cl *cl); +/* + * MEI input output function prototype + */ +static inline bool mei_cl_is_connected(struct mei_cl *cl) +{ + return cl->dev && + cl->dev->dev_state == MEI_DEV_ENABLED && + cl->state == MEI_FILE_CONNECTED; +} +static inline bool mei_cl_is_transitioning(struct mei_cl *cl) +{ + return MEI_FILE_INITIALIZING == cl->state || + MEI_FILE_DISCONNECTED == cl->state || + MEI_FILE_DISCONNECTING == cl->state; +} + +bool mei_cl_is_other_connecting(struct mei_cl *cl); +int mei_cl_disconnect(struct mei_cl *cl); +int mei_cl_connect(struct mei_cl *cl, struct file *file); +int mei_cl_read_start(struct mei_cl *cl, size_t length); +int mei_cl_write(struct mei_cl *cl, struct mei_cl_cb *cb, bool blocking); +int mei_cl_irq_write(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list); + +void mei_cl_complete(struct mei_cl *cl, struct mei_cl_cb *cb); + +void mei_host_client_init(struct work_struct *work); + + + +void mei_cl_all_disconnect(struct mei_device *dev); +void mei_cl_all_wakeup(struct mei_device *dev); +void mei_cl_all_write_clear(struct mei_device *dev); + +#define MEI_CL_FMT "cl:host=%02d me=%02d " +#define MEI_CL_PRM(cl) (cl)->host_client_id, (cl)->me_client_id + +#define cl_dbg(dev, cl, format, arg...) \ + dev_dbg(&(dev)->pdev->dev, MEI_CL_FMT format, MEI_CL_PRM(cl), ##arg) + +#define cl_err(dev, cl, format, arg...) \ + dev_err(&(dev)->pdev->dev, MEI_CL_FMT format, MEI_CL_PRM(cl), ##arg) + +#endif /* _MEI_CLIENT_H_ */ diff --git a/drivers/misc/mei/debugfs.c b/drivers/misc/mei/debugfs.c new file mode 100644 index 00000000000..ced5b777c70 --- /dev/null +++ b/drivers/misc/mei/debugfs.c @@ -0,0 +1,197 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2012-2013, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ +#include <linux/slab.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/debugfs.h> +#include <linux/pci.h> + +#include <linux/mei.h> + +#include "mei_dev.h" +#include "hw.h" + +static ssize_t mei_dbgfs_read_meclients(struct file *fp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct mei_device *dev = fp->private_data; + struct mei_me_client *cl; + const size_t bufsz = 1024; + char *buf = kzalloc(bufsz, GFP_KERNEL); + int i; + int pos = 0; + int ret; + + if (!buf) + return -ENOMEM; + + pos += scnprintf(buf + pos, bufsz - pos, + " |id|addr| UUID |con|msg len|\n"); + + mutex_lock(&dev->device_lock); + + /* if the driver is not enabled the list won't be consistent */ + if (dev->dev_state != MEI_DEV_ENABLED) + goto out; + + for (i = 0; i < dev->me_clients_num; i++) { + cl = &dev->me_clients[i]; + + /* skip me clients that cannot be connected */ + if (cl->props.max_number_of_connections == 0) + continue; + + pos += scnprintf(buf + pos, bufsz - pos, + "%2d|%2d|%4d|%pUl|%3d|%7d|\n", + i, cl->client_id, + cl->props.fixed_address, + &cl->props.protocol_name, + cl->props.max_number_of_connections, + cl->props.max_msg_length); + } +out: + mutex_unlock(&dev->device_lock); + ret = simple_read_from_buffer(ubuf, cnt, ppos, buf, pos); + kfree(buf); + return ret; +} + +static const struct file_operations mei_dbgfs_fops_meclients = { + .open = simple_open, + .read = mei_dbgfs_read_meclients, + .llseek = generic_file_llseek, +}; + +static ssize_t mei_dbgfs_read_active(struct file *fp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct mei_device *dev = fp->private_data; + struct mei_cl *cl; + const size_t bufsz = 1024; + char *buf; + int i = 0; + int pos = 0; + int ret; + + if (!dev) + return -ENODEV; + + buf = kzalloc(bufsz, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + pos += scnprintf(buf + pos, bufsz - pos, + " |me|host|state|rd|wr|\n"); + + mutex_lock(&dev->device_lock); + + /* if the driver is not enabled the list won't b consitent */ + if (dev->dev_state != MEI_DEV_ENABLED) + goto out; + + list_for_each_entry(cl, &dev->file_list, link) { + + pos += scnprintf(buf + pos, bufsz - pos, + "%2d|%2d|%4d|%5d|%2d|%2d|\n", + i, cl->me_client_id, cl->host_client_id, cl->state, + cl->reading_state, cl->writing_state); + i++; + } +out: + mutex_unlock(&dev->device_lock); + ret = simple_read_from_buffer(ubuf, cnt, ppos, buf, pos); + kfree(buf); + return ret; +} + +static const struct file_operations mei_dbgfs_fops_active = { + .open = simple_open, + .read = mei_dbgfs_read_active, + .llseek = generic_file_llseek, +}; + +static ssize_t mei_dbgfs_read_devstate(struct file *fp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct mei_device *dev = fp->private_data; + const size_t bufsz = 1024; + char *buf = kzalloc(bufsz, GFP_KERNEL); + int pos = 0; + int ret; + + if (!buf) + return -ENOMEM; + + pos += scnprintf(buf + pos, bufsz - pos, "%s\n", + mei_dev_state_str(dev->dev_state)); + ret = simple_read_from_buffer(ubuf, cnt, ppos, buf, pos); + kfree(buf); + return ret; +} +static const struct file_operations mei_dbgfs_fops_devstate = { + .open = simple_open, + .read = mei_dbgfs_read_devstate, + .llseek = generic_file_llseek, +}; + +/** + * mei_dbgfs_deregister - Remove the debugfs files and directories + * @mei - pointer to mei device private data + */ +void mei_dbgfs_deregister(struct mei_device *dev) +{ + if (!dev->dbgfs_dir) + return; + debugfs_remove_recursive(dev->dbgfs_dir); + dev->dbgfs_dir = NULL; +} + +/** + * Add the debugfs files + * + */ +int mei_dbgfs_register(struct mei_device *dev, const char *name) +{ + struct dentry *dir, *f; + dir = debugfs_create_dir(name, NULL); + if (!dir) + return -ENOMEM; + + f = debugfs_create_file("meclients", S_IRUSR, dir, + dev, &mei_dbgfs_fops_meclients); + if (!f) { + dev_err(&dev->pdev->dev, "meclients: registration failed\n"); + goto err; + } + f = debugfs_create_file("active", S_IRUSR, dir, + dev, &mei_dbgfs_fops_active); + if (!f) { + dev_err(&dev->pdev->dev, "meclients: registration failed\n"); + goto err; + } + f = debugfs_create_file("devstate", S_IRUSR, dir, + dev, &mei_dbgfs_fops_devstate); + if (!f) { + dev_err(&dev->pdev->dev, "devstate: registration failed\n"); + goto err; + } + dev->dbgfs_dir = dir; + return 0; +err: + mei_dbgfs_deregister(dev); + return -ENODEV; +} + diff --git a/drivers/misc/mei/hbm.c b/drivers/misc/mei/hbm.c new file mode 100644 index 00000000000..804106209d7 --- /dev/null +++ b/drivers/misc/mei/hbm.c @@ -0,0 +1,879 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/export.h> +#include <linux/pci.h> +#include <linux/sched.h> +#include <linux/wait.h> +#include <linux/mei.h> +#include <linux/pm_runtime.h> + +#include "mei_dev.h" +#include "hbm.h" +#include "client.h" + +static const char *mei_cl_conn_status_str(enum mei_cl_connect_status status) +{ +#define MEI_CL_CS(status) case MEI_CL_CONN_##status: return #status + switch (status) { + MEI_CL_CS(SUCCESS); + MEI_CL_CS(NOT_FOUND); + MEI_CL_CS(ALREADY_STARTED); + MEI_CL_CS(OUT_OF_RESOURCES); + MEI_CL_CS(MESSAGE_SMALL); + default: return "unknown"; + } +#undef MEI_CL_CCS +} + +/** + * mei_cl_conn_status_to_errno - convert client connect response + * status to error code + * + * @status: client connect response status + * + * returns corresponding error code + */ +static int mei_cl_conn_status_to_errno(enum mei_cl_connect_status status) +{ + switch (status) { + case MEI_CL_CONN_SUCCESS: return 0; + case MEI_CL_CONN_NOT_FOUND: return -ENOTTY; + case MEI_CL_CONN_ALREADY_STARTED: return -EBUSY; + case MEI_CL_CONN_OUT_OF_RESOURCES: return -EBUSY; + case MEI_CL_CONN_MESSAGE_SMALL: return -EINVAL; + default: return -EINVAL; + } +} + +/** + * mei_hbm_idle - set hbm to idle state + * + * @dev: the device structure + */ +void mei_hbm_idle(struct mei_device *dev) +{ + dev->init_clients_timer = 0; + dev->hbm_state = MEI_HBM_IDLE; +} + +/** + * mei_hbm_reset - reset hbm counters and book keeping data structurs + * + * @dev: the device structure + */ +void mei_hbm_reset(struct mei_device *dev) +{ + dev->me_clients_num = 0; + dev->me_client_presentation_num = 0; + dev->me_client_index = 0; + + kfree(dev->me_clients); + dev->me_clients = NULL; + + mei_hbm_idle(dev); +} + +/** + * mei_hbm_me_cl_allocate - allocates storage for me clients + * + * @dev: the device structure + * + * returns 0 on success -ENOMEM on allocation failure + */ +static int mei_hbm_me_cl_allocate(struct mei_device *dev) +{ + struct mei_me_client *clients; + int b; + + mei_hbm_reset(dev); + + /* count how many ME clients we have */ + for_each_set_bit(b, dev->me_clients_map, MEI_CLIENTS_MAX) + dev->me_clients_num++; + + if (dev->me_clients_num == 0) + return 0; + + dev_dbg(&dev->pdev->dev, "memory allocation for ME clients size=%ld.\n", + dev->me_clients_num * sizeof(struct mei_me_client)); + /* allocate storage for ME clients representation */ + clients = kcalloc(dev->me_clients_num, + sizeof(struct mei_me_client), GFP_KERNEL); + if (!clients) { + dev_err(&dev->pdev->dev, "memory allocation for ME clients failed.\n"); + return -ENOMEM; + } + dev->me_clients = clients; + return 0; +} + +/** + * mei_hbm_cl_hdr - construct client hbm header + * + * @cl: - client + * @hbm_cmd: host bus message command + * @buf: buffer for cl header + * @len: buffer length + */ +static inline +void mei_hbm_cl_hdr(struct mei_cl *cl, u8 hbm_cmd, void *buf, size_t len) +{ + struct mei_hbm_cl_cmd *cmd = buf; + + memset(cmd, 0, len); + + cmd->hbm_cmd = hbm_cmd; + cmd->host_addr = cl->host_client_id; + cmd->me_addr = cl->me_client_id; +} + +/** + * mei_hbm_cl_addr_equal - tells if they have the same address + * + * @cl: - client + * @buf: buffer with cl header + * + * returns true if addresses are the same + */ +static inline +bool mei_hbm_cl_addr_equal(struct mei_cl *cl, void *buf) +{ + struct mei_hbm_cl_cmd *cmd = buf; + return cl->host_client_id == cmd->host_addr && + cl->me_client_id == cmd->me_addr; +} + + +int mei_hbm_start_wait(struct mei_device *dev) +{ + int ret; + if (dev->hbm_state > MEI_HBM_START) + return 0; + + mutex_unlock(&dev->device_lock); + ret = wait_event_interruptible_timeout(dev->wait_recvd_msg, + dev->hbm_state == MEI_HBM_IDLE || + dev->hbm_state >= MEI_HBM_STARTED, + mei_secs_to_jiffies(MEI_HBM_TIMEOUT)); + mutex_lock(&dev->device_lock); + + if (ret <= 0 && (dev->hbm_state <= MEI_HBM_START)) { + dev->hbm_state = MEI_HBM_IDLE; + dev_err(&dev->pdev->dev, "waiting for mei start failed\n"); + return -ETIME; + } + return 0; +} + +/** + * mei_hbm_start_req - sends start request message. + * + * @dev: the device structure + * + * returns 0 on success and < 0 on failure + */ +int mei_hbm_start_req(struct mei_device *dev) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + struct hbm_host_version_request *start_req; + const size_t len = sizeof(struct hbm_host_version_request); + int ret; + + mei_hbm_hdr(mei_hdr, len); + + /* host start message */ + start_req = (struct hbm_host_version_request *)dev->wr_msg.data; + memset(start_req, 0, len); + start_req->hbm_cmd = HOST_START_REQ_CMD; + start_req->host_version.major_version = HBM_MAJOR_VERSION; + start_req->host_version.minor_version = HBM_MINOR_VERSION; + + dev->hbm_state = MEI_HBM_IDLE; + ret = mei_write_message(dev, mei_hdr, dev->wr_msg.data); + if (ret) { + dev_err(&dev->pdev->dev, "version message write failed: ret = %d\n", + ret); + return ret; + } + + dev->hbm_state = MEI_HBM_START; + dev->init_clients_timer = MEI_CLIENTS_INIT_TIMEOUT; + return 0; +} + +/* + * mei_hbm_enum_clients_req - sends enumeration client request message. + * + * @dev: the device structure + * + * returns 0 on success and < 0 on failure + */ +static int mei_hbm_enum_clients_req(struct mei_device *dev) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + struct hbm_host_enum_request *enum_req; + const size_t len = sizeof(struct hbm_host_enum_request); + int ret; + + /* enumerate clients */ + mei_hbm_hdr(mei_hdr, len); + + enum_req = (struct hbm_host_enum_request *)dev->wr_msg.data; + memset(enum_req, 0, len); + enum_req->hbm_cmd = HOST_ENUM_REQ_CMD; + + ret = mei_write_message(dev, mei_hdr, dev->wr_msg.data); + if (ret) { + dev_err(&dev->pdev->dev, "enumeration request write failed: ret = %d.\n", + ret); + return ret; + } + dev->hbm_state = MEI_HBM_ENUM_CLIENTS; + dev->init_clients_timer = MEI_CLIENTS_INIT_TIMEOUT; + return 0; +} + +/** + * mei_hbm_prop_req - request property for a single client + * + * @dev: the device structure + * + * returns 0 on success and < 0 on failure + */ + +static int mei_hbm_prop_req(struct mei_device *dev) +{ + + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + struct hbm_props_request *prop_req; + const size_t len = sizeof(struct hbm_props_request); + unsigned long next_client_index; + unsigned long client_num; + int ret; + + client_num = dev->me_client_presentation_num; + + next_client_index = find_next_bit(dev->me_clients_map, MEI_CLIENTS_MAX, + dev->me_client_index); + + /* We got all client properties */ + if (next_client_index == MEI_CLIENTS_MAX) { + dev->hbm_state = MEI_HBM_STARTED; + schedule_work(&dev->init_work); + + return 0; + } + + dev->me_clients[client_num].client_id = next_client_index; + dev->me_clients[client_num].mei_flow_ctrl_creds = 0; + + mei_hbm_hdr(mei_hdr, len); + prop_req = (struct hbm_props_request *)dev->wr_msg.data; + + memset(prop_req, 0, sizeof(struct hbm_props_request)); + + + prop_req->hbm_cmd = HOST_CLIENT_PROPERTIES_REQ_CMD; + prop_req->address = next_client_index; + + ret = mei_write_message(dev, mei_hdr, dev->wr_msg.data); + if (ret) { + dev_err(&dev->pdev->dev, "properties request write failed: ret = %d\n", + ret); + return ret; + } + + dev->init_clients_timer = MEI_CLIENTS_INIT_TIMEOUT; + dev->me_client_index = next_client_index; + + return 0; +} + +/* + * mei_hbm_pg - sends pg command + * + * @dev: the device structure + * @pg_cmd: the pg command code + * + * This function returns -EIO on write failure + */ +int mei_hbm_pg(struct mei_device *dev, u8 pg_cmd) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + struct hbm_power_gate *req; + const size_t len = sizeof(struct hbm_power_gate); + int ret; + + mei_hbm_hdr(mei_hdr, len); + + req = (struct hbm_power_gate *)dev->wr_msg.data; + memset(req, 0, len); + req->hbm_cmd = pg_cmd; + + ret = mei_write_message(dev, mei_hdr, dev->wr_msg.data); + if (ret) + dev_err(&dev->pdev->dev, "power gate command write failed.\n"); + return ret; +} +EXPORT_SYMBOL_GPL(mei_hbm_pg); + +/** + * mei_hbm_stop_req - send stop request message + * + * @dev - mei device + * @cl: client info + * + * This function returns -EIO on write failure + */ +static int mei_hbm_stop_req(struct mei_device *dev) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + struct hbm_host_stop_request *req = + (struct hbm_host_stop_request *)dev->wr_msg.data; + const size_t len = sizeof(struct hbm_host_stop_request); + + mei_hbm_hdr(mei_hdr, len); + + memset(req, 0, len); + req->hbm_cmd = HOST_STOP_REQ_CMD; + req->reason = DRIVER_STOP_REQUEST; + + return mei_write_message(dev, mei_hdr, dev->wr_msg.data); +} + +/** + * mei_hbm_cl_flow_control_req - sends flow control request. + * + * @dev: the device structure + * @cl: client info + * + * This function returns -EIO on write failure + */ +int mei_hbm_cl_flow_control_req(struct mei_device *dev, struct mei_cl *cl) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + const size_t len = sizeof(struct hbm_flow_control); + + mei_hbm_hdr(mei_hdr, len); + mei_hbm_cl_hdr(cl, MEI_FLOW_CONTROL_CMD, dev->wr_msg.data, len); + + cl_dbg(dev, cl, "sending flow control\n"); + + return mei_write_message(dev, mei_hdr, dev->wr_msg.data); +} + +/** + * mei_hbm_add_single_flow_creds - adds single buffer credentials. + * + * @dev: the device structure + * @flow: flow control. + * + * return 0 on success, < 0 otherwise + */ +static int mei_hbm_add_single_flow_creds(struct mei_device *dev, + struct hbm_flow_control *flow) +{ + struct mei_me_client *me_cl; + int id; + + id = mei_me_cl_by_id(dev, flow->me_addr); + if (id < 0) { + dev_err(&dev->pdev->dev, "no such me client %d\n", + flow->me_addr); + return id; + } + + me_cl = &dev->me_clients[id]; + if (me_cl->props.single_recv_buf) { + me_cl->mei_flow_ctrl_creds++; + dev_dbg(&dev->pdev->dev, "recv flow ctrl msg ME %d (single).\n", + flow->me_addr); + dev_dbg(&dev->pdev->dev, "flow control credentials =%d.\n", + me_cl->mei_flow_ctrl_creds); + } else { + BUG(); /* error in flow control */ + } + + return 0; +} + +/** + * mei_hbm_cl_flow_control_res - flow control response from me + * + * @dev: the device structure + * @flow_control: flow control response bus message + */ +static void mei_hbm_cl_flow_control_res(struct mei_device *dev, + struct hbm_flow_control *flow_control) +{ + struct mei_cl *cl; + + if (!flow_control->host_addr) { + /* single receive buffer */ + mei_hbm_add_single_flow_creds(dev, flow_control); + return; + } + + /* normal connection */ + list_for_each_entry(cl, &dev->file_list, link) { + if (mei_hbm_cl_addr_equal(cl, flow_control)) { + cl->mei_flow_ctrl_creds++; + dev_dbg(&dev->pdev->dev, "flow ctrl msg for host %d ME %d.\n", + flow_control->host_addr, flow_control->me_addr); + dev_dbg(&dev->pdev->dev, "flow control credentials = %d.\n", + cl->mei_flow_ctrl_creds); + break; + } + } +} + + +/** + * mei_hbm_cl_disconnect_req - sends disconnect message to fw. + * + * @dev: the device structure + * @cl: a client to disconnect from + * + * This function returns -EIO on write failure + */ +int mei_hbm_cl_disconnect_req(struct mei_device *dev, struct mei_cl *cl) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + const size_t len = sizeof(struct hbm_client_connect_request); + + mei_hbm_hdr(mei_hdr, len); + mei_hbm_cl_hdr(cl, CLIENT_DISCONNECT_REQ_CMD, dev->wr_msg.data, len); + + return mei_write_message(dev, mei_hdr, dev->wr_msg.data); +} + +/** + * mei_hbm_cl_disconnect_rsp - sends disconnect respose to the FW + * + * @dev: the device structure + * @cl: a client to disconnect from + * + * This function returns -EIO on write failure + */ +int mei_hbm_cl_disconnect_rsp(struct mei_device *dev, struct mei_cl *cl) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + const size_t len = sizeof(struct hbm_client_connect_response); + + mei_hbm_hdr(mei_hdr, len); + mei_hbm_cl_hdr(cl, CLIENT_DISCONNECT_RES_CMD, dev->wr_msg.data, len); + + return mei_write_message(dev, mei_hdr, dev->wr_msg.data); +} + +/** + * mei_hbm_cl_disconnect_res - disconnect response from ME + * + * @dev: the device structure + * @rs: disconnect response bus message + */ +static void mei_hbm_cl_disconnect_res(struct mei_device *dev, + struct hbm_client_connect_response *rs) +{ + struct mei_cl *cl; + struct mei_cl_cb *cb, *next; + + dev_dbg(&dev->pdev->dev, "hbm: disconnect response cl:host=%02d me=%02d status=%d\n", + rs->me_addr, rs->host_addr, rs->status); + + list_for_each_entry_safe(cb, next, &dev->ctrl_rd_list.list, list) { + cl = cb->cl; + + /* this should not happen */ + if (WARN_ON(!cl)) { + list_del(&cb->list); + return; + } + + if (mei_hbm_cl_addr_equal(cl, rs)) { + list_del(&cb->list); + if (rs->status == MEI_CL_DISCONN_SUCCESS) + cl->state = MEI_FILE_DISCONNECTED; + + cl->status = 0; + cl->timer_count = 0; + break; + } + } +} + +/** + * mei_hbm_cl_connect_req - send connection request to specific me client + * + * @dev: the device structure + * @cl: a client to connect to + * + * returns -EIO on write failure + */ +int mei_hbm_cl_connect_req(struct mei_device *dev, struct mei_cl *cl) +{ + struct mei_msg_hdr *mei_hdr = &dev->wr_msg.hdr; + const size_t len = sizeof(struct hbm_client_connect_request); + + mei_hbm_hdr(mei_hdr, len); + mei_hbm_cl_hdr(cl, CLIENT_CONNECT_REQ_CMD, dev->wr_msg.data, len); + + return mei_write_message(dev, mei_hdr, dev->wr_msg.data); +} + +/** + * mei_hbm_cl_connect_res - connect response from the ME + * + * @dev: the device structure + * @rs: connect response bus message + */ +static void mei_hbm_cl_connect_res(struct mei_device *dev, + struct hbm_client_connect_response *rs) +{ + + struct mei_cl *cl; + struct mei_cl_cb *cb, *next; + + dev_dbg(&dev->pdev->dev, "hbm: connect response cl:host=%02d me=%02d status=%s\n", + rs->me_addr, rs->host_addr, + mei_cl_conn_status_str(rs->status)); + + cl = NULL; + + list_for_each_entry_safe(cb, next, &dev->ctrl_rd_list.list, list) { + + cl = cb->cl; + /* this should not happen */ + if (WARN_ON(!cl)) { + list_del_init(&cb->list); + continue; + } + + if (cb->fop_type != MEI_FOP_CONNECT) + continue; + + if (mei_hbm_cl_addr_equal(cl, rs)) { + list_del(&cb->list); + break; + } + } + + if (!cl) + return; + + cl->timer_count = 0; + if (rs->status == MEI_CL_CONN_SUCCESS) + cl->state = MEI_FILE_CONNECTED; + else + cl->state = MEI_FILE_DISCONNECTED; + cl->status = mei_cl_conn_status_to_errno(rs->status); +} + + +/** + * mei_hbm_fw_disconnect_req - disconnect request initiated by ME firmware + * host sends disconnect response + * + * @dev: the device structure. + * @disconnect_req: disconnect request bus message from the me + * + * returns -ENOMEM on allocation failure + */ +static int mei_hbm_fw_disconnect_req(struct mei_device *dev, + struct hbm_client_connect_request *disconnect_req) +{ + struct mei_cl *cl; + struct mei_cl_cb *cb; + + list_for_each_entry(cl, &dev->file_list, link) { + if (mei_hbm_cl_addr_equal(cl, disconnect_req)) { + dev_dbg(&dev->pdev->dev, "disconnect request host client %d ME client %d.\n", + disconnect_req->host_addr, + disconnect_req->me_addr); + cl->state = MEI_FILE_DISCONNECTED; + cl->timer_count = 0; + + cb = mei_io_cb_init(cl, NULL); + if (!cb) + return -ENOMEM; + cb->fop_type = MEI_FOP_DISCONNECT_RSP; + cl_dbg(dev, cl, "add disconnect response as first\n"); + list_add(&cb->list, &dev->ctrl_wr_list.list); + + break; + } + } + return 0; +} + + +/** + * mei_hbm_version_is_supported - checks whether the driver can + * support the hbm version of the device + * + * @dev: the device structure + * returns true if driver can support hbm version of the device + */ +bool mei_hbm_version_is_supported(struct mei_device *dev) +{ + return (dev->version.major_version < HBM_MAJOR_VERSION) || + (dev->version.major_version == HBM_MAJOR_VERSION && + dev->version.minor_version <= HBM_MINOR_VERSION); +} + +/** + * mei_hbm_dispatch - bottom half read routine after ISR to + * handle the read bus message cmd processing. + * + * @dev: the device structure + * @mei_hdr: header of bus message + * + * returns 0 on success and < 0 on failure + */ +int mei_hbm_dispatch(struct mei_device *dev, struct mei_msg_hdr *hdr) +{ + struct mei_bus_message *mei_msg; + struct mei_me_client *me_client; + struct hbm_host_version_response *version_res; + struct hbm_client_connect_response *connect_res; + struct hbm_client_connect_response *disconnect_res; + struct hbm_client_connect_request *disconnect_req; + struct hbm_flow_control *flow_control; + struct hbm_props_response *props_res; + struct hbm_host_enum_response *enum_res; + + /* read the message to our buffer */ + BUG_ON(hdr->length >= sizeof(dev->rd_msg_buf)); + mei_read_slots(dev, dev->rd_msg_buf, hdr->length); + mei_msg = (struct mei_bus_message *)dev->rd_msg_buf; + + /* ignore spurious message and prevent reset nesting + * hbm is put to idle during system reset + */ + if (dev->hbm_state == MEI_HBM_IDLE) { + dev_dbg(&dev->pdev->dev, "hbm: state is idle ignore spurious messages\n"); + return 0; + } + + switch (mei_msg->hbm_cmd) { + case HOST_START_RES_CMD: + dev_dbg(&dev->pdev->dev, "hbm: start: response message received.\n"); + + dev->init_clients_timer = 0; + + version_res = (struct hbm_host_version_response *)mei_msg; + + dev_dbg(&dev->pdev->dev, "HBM VERSION: DRIVER=%02d:%02d DEVICE=%02d:%02d\n", + HBM_MAJOR_VERSION, HBM_MINOR_VERSION, + version_res->me_max_version.major_version, + version_res->me_max_version.minor_version); + + if (version_res->host_version_supported) { + dev->version.major_version = HBM_MAJOR_VERSION; + dev->version.minor_version = HBM_MINOR_VERSION; + } else { + dev->version.major_version = + version_res->me_max_version.major_version; + dev->version.minor_version = + version_res->me_max_version.minor_version; + } + + if (!mei_hbm_version_is_supported(dev)) { + dev_warn(&dev->pdev->dev, "hbm: start: version mismatch - stopping the driver.\n"); + + dev->hbm_state = MEI_HBM_STOPPED; + if (mei_hbm_stop_req(dev)) { + dev_err(&dev->pdev->dev, "hbm: start: failed to send stop request\n"); + return -EIO; + } + break; + } + + if (dev->dev_state != MEI_DEV_INIT_CLIENTS || + dev->hbm_state != MEI_HBM_START) { + dev_err(&dev->pdev->dev, "hbm: start: state mismatch, [%d, %d]\n", + dev->dev_state, dev->hbm_state); + return -EPROTO; + } + + dev->hbm_state = MEI_HBM_STARTED; + + if (mei_hbm_enum_clients_req(dev)) { + dev_err(&dev->pdev->dev, "hbm: start: failed to send enumeration request\n"); + return -EIO; + } + + wake_up_interruptible(&dev->wait_recvd_msg); + break; + + case CLIENT_CONNECT_RES_CMD: + dev_dbg(&dev->pdev->dev, "hbm: client connect response: message received.\n"); + + connect_res = (struct hbm_client_connect_response *) mei_msg; + mei_hbm_cl_connect_res(dev, connect_res); + wake_up(&dev->wait_recvd_msg); + break; + + case CLIENT_DISCONNECT_RES_CMD: + dev_dbg(&dev->pdev->dev, "hbm: client disconnect response: message received.\n"); + + disconnect_res = (struct hbm_client_connect_response *) mei_msg; + mei_hbm_cl_disconnect_res(dev, disconnect_res); + wake_up(&dev->wait_recvd_msg); + break; + + case MEI_FLOW_CONTROL_CMD: + dev_dbg(&dev->pdev->dev, "hbm: client flow control response: message received.\n"); + + flow_control = (struct hbm_flow_control *) mei_msg; + mei_hbm_cl_flow_control_res(dev, flow_control); + break; + + case MEI_PG_ISOLATION_ENTRY_RES_CMD: + dev_dbg(&dev->pdev->dev, "power gate isolation entry response received\n"); + dev->pg_event = MEI_PG_EVENT_RECEIVED; + if (waitqueue_active(&dev->wait_pg)) + wake_up(&dev->wait_pg); + break; + + case MEI_PG_ISOLATION_EXIT_REQ_CMD: + dev_dbg(&dev->pdev->dev, "power gate isolation exit request received\n"); + dev->pg_event = MEI_PG_EVENT_RECEIVED; + if (waitqueue_active(&dev->wait_pg)) + wake_up(&dev->wait_pg); + else + /* + * If the driver is not waiting on this then + * this is HW initiated exit from PG. + * Start runtime pm resume sequence to exit from PG. + */ + pm_request_resume(&dev->pdev->dev); + break; + + case HOST_CLIENT_PROPERTIES_RES_CMD: + dev_dbg(&dev->pdev->dev, "hbm: properties response: message received.\n"); + + dev->init_clients_timer = 0; + + if (dev->me_clients == NULL) { + dev_err(&dev->pdev->dev, "hbm: properties response: mei_clients not allocated\n"); + return -EPROTO; + } + + props_res = (struct hbm_props_response *)mei_msg; + me_client = &dev->me_clients[dev->me_client_presentation_num]; + + if (props_res->status) { + dev_err(&dev->pdev->dev, "hbm: properties response: wrong status = %d\n", + props_res->status); + return -EPROTO; + } + + if (me_client->client_id != props_res->address) { + dev_err(&dev->pdev->dev, "hbm: properties response: address mismatch %d ?= %d\n", + me_client->client_id, props_res->address); + return -EPROTO; + } + + if (dev->dev_state != MEI_DEV_INIT_CLIENTS || + dev->hbm_state != MEI_HBM_CLIENT_PROPERTIES) { + dev_err(&dev->pdev->dev, "hbm: properties response: state mismatch, [%d, %d]\n", + dev->dev_state, dev->hbm_state); + return -EPROTO; + } + + me_client->props = props_res->client_properties; + dev->me_client_index++; + dev->me_client_presentation_num++; + + /* request property for the next client */ + if (mei_hbm_prop_req(dev)) + return -EIO; + + break; + + case HOST_ENUM_RES_CMD: + dev_dbg(&dev->pdev->dev, "hbm: enumeration response: message received\n"); + + dev->init_clients_timer = 0; + + enum_res = (struct hbm_host_enum_response *) mei_msg; + BUILD_BUG_ON(sizeof(dev->me_clients_map) + < sizeof(enum_res->valid_addresses)); + memcpy(dev->me_clients_map, enum_res->valid_addresses, + sizeof(enum_res->valid_addresses)); + + if (dev->dev_state != MEI_DEV_INIT_CLIENTS || + dev->hbm_state != MEI_HBM_ENUM_CLIENTS) { + dev_err(&dev->pdev->dev, "hbm: enumeration response: state mismatch, [%d, %d]\n", + dev->dev_state, dev->hbm_state); + return -EPROTO; + } + + if (mei_hbm_me_cl_allocate(dev)) { + dev_err(&dev->pdev->dev, "hbm: enumeration response: cannot allocate clients array\n"); + return -ENOMEM; + } + + dev->hbm_state = MEI_HBM_CLIENT_PROPERTIES; + + /* first property request */ + if (mei_hbm_prop_req(dev)) + return -EIO; + + break; + + case HOST_STOP_RES_CMD: + dev_dbg(&dev->pdev->dev, "hbm: stop response: message received\n"); + + dev->init_clients_timer = 0; + + if (dev->hbm_state != MEI_HBM_STOPPED) { + dev_err(&dev->pdev->dev, "hbm: stop response: state mismatch, [%d, %d]\n", + dev->dev_state, dev->hbm_state); + return -EPROTO; + } + + dev->dev_state = MEI_DEV_POWER_DOWN; + dev_info(&dev->pdev->dev, "hbm: stop response: resetting.\n"); + /* force the reset */ + return -EPROTO; + break; + + case CLIENT_DISCONNECT_REQ_CMD: + dev_dbg(&dev->pdev->dev, "hbm: disconnect request: message received\n"); + + disconnect_req = (struct hbm_client_connect_request *)mei_msg; + mei_hbm_fw_disconnect_req(dev, disconnect_req); + break; + + case ME_STOP_REQ_CMD: + dev_dbg(&dev->pdev->dev, "hbm: stop request: message received\n"); + dev->hbm_state = MEI_HBM_STOPPED; + if (mei_hbm_stop_req(dev)) { + dev_err(&dev->pdev->dev, "hbm: start: failed to send stop request\n"); + return -EIO; + } + break; + default: + BUG(); + break; + + } + return 0; +} + diff --git a/drivers/misc/mei/hbm.h b/drivers/misc/mei/hbm.h new file mode 100644 index 00000000000..683eb2835ce --- /dev/null +++ b/drivers/misc/mei/hbm.h @@ -0,0 +1,64 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#ifndef _MEI_HBM_H_ +#define _MEI_HBM_H_ + +struct mei_device; +struct mei_msg_hdr; +struct mei_cl; + +/** + * enum mei_hbm_state - host bus message protocol state + * + * @MEI_HBM_IDLE : protocol not started + * @MEI_HBM_START : start request message was sent + * @MEI_HBM_ENUM_CLIENTS : enumeration request was sent + * @MEI_HBM_CLIENT_PROPERTIES : acquiring clients properties + */ +enum mei_hbm_state { + MEI_HBM_IDLE = 0, + MEI_HBM_START, + MEI_HBM_STARTED, + MEI_HBM_ENUM_CLIENTS, + MEI_HBM_CLIENT_PROPERTIES, + MEI_HBM_STOPPED, +}; + +int mei_hbm_dispatch(struct mei_device *dev, struct mei_msg_hdr *hdr); + +static inline void mei_hbm_hdr(struct mei_msg_hdr *hdr, size_t length) +{ + hdr->host_addr = 0; + hdr->me_addr = 0; + hdr->length = length; + hdr->msg_complete = 1; + hdr->reserved = 0; +} + +void mei_hbm_idle(struct mei_device *dev); +void mei_hbm_reset(struct mei_device *dev); +int mei_hbm_start_req(struct mei_device *dev); +int mei_hbm_start_wait(struct mei_device *dev); +int mei_hbm_cl_flow_control_req(struct mei_device *dev, struct mei_cl *cl); +int mei_hbm_cl_disconnect_req(struct mei_device *dev, struct mei_cl *cl); +int mei_hbm_cl_disconnect_rsp(struct mei_device *dev, struct mei_cl *cl); +int mei_hbm_cl_connect_req(struct mei_device *dev, struct mei_cl *cl); +bool mei_hbm_version_is_supported(struct mei_device *dev); +int mei_hbm_pg(struct mei_device *dev, u8 pg_cmd); + +#endif /* _MEI_HBM_H_ */ + diff --git a/drivers/misc/mei/hw-me-regs.h b/drivers/misc/mei/hw-me-regs.h new file mode 100644 index 00000000000..a7856c0ac57 --- /dev/null +++ b/drivers/misc/mei/hw-me-regs.h @@ -0,0 +1,185 @@ +/****************************************************************************** + * Intel Management Engine Interface (Intel MEI) Linux driver + * Intel MEI Interface Header + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * Copyright(c) 2003 - 2012 Intel Corporation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110, + * USA + * + * The full GNU General Public License is included in this distribution + * in the file called LICENSE.GPL. + * + * Contact Information: + * Intel Corporation. + * linux-mei@linux.intel.com + * http://www.intel.com + * + * BSD LICENSE + * + * Copyright(c) 2003 - 2012 Intel Corporation. All rights reserved. + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * * Neither the name Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + *****************************************************************************/ +#ifndef _MEI_HW_MEI_REGS_H_ +#define _MEI_HW_MEI_REGS_H_ + +/* + * MEI device IDs + */ +#define MEI_DEV_ID_82946GZ 0x2974 /* 82946GZ/GL */ +#define MEI_DEV_ID_82G35 0x2984 /* 82G35 Express */ +#define MEI_DEV_ID_82Q965 0x2994 /* 82Q963/Q965 */ +#define MEI_DEV_ID_82G965 0x29A4 /* 82P965/G965 */ + +#define MEI_DEV_ID_82GM965 0x2A04 /* Mobile PM965/GM965 */ +#define MEI_DEV_ID_82GME965 0x2A14 /* Mobile GME965/GLE960 */ + +#define MEI_DEV_ID_ICH9_82Q35 0x29B4 /* 82Q35 Express */ +#define MEI_DEV_ID_ICH9_82G33 0x29C4 /* 82G33/G31/P35/P31 Express */ +#define MEI_DEV_ID_ICH9_82Q33 0x29D4 /* 82Q33 Express */ +#define MEI_DEV_ID_ICH9_82X38 0x29E4 /* 82X38/X48 Express */ +#define MEI_DEV_ID_ICH9_3200 0x29F4 /* 3200/3210 Server */ + +#define MEI_DEV_ID_ICH9_6 0x28B4 /* Bearlake */ +#define MEI_DEV_ID_ICH9_7 0x28C4 /* Bearlake */ +#define MEI_DEV_ID_ICH9_8 0x28D4 /* Bearlake */ +#define MEI_DEV_ID_ICH9_9 0x28E4 /* Bearlake */ +#define MEI_DEV_ID_ICH9_10 0x28F4 /* Bearlake */ + +#define MEI_DEV_ID_ICH9M_1 0x2A44 /* Cantiga */ +#define MEI_DEV_ID_ICH9M_2 0x2A54 /* Cantiga */ +#define MEI_DEV_ID_ICH9M_3 0x2A64 /* Cantiga */ +#define MEI_DEV_ID_ICH9M_4 0x2A74 /* Cantiga */ + +#define MEI_DEV_ID_ICH10_1 0x2E04 /* Eaglelake */ +#define MEI_DEV_ID_ICH10_2 0x2E14 /* Eaglelake */ +#define MEI_DEV_ID_ICH10_3 0x2E24 /* Eaglelake */ +#define MEI_DEV_ID_ICH10_4 0x2E34 /* Eaglelake */ + +#define MEI_DEV_ID_IBXPK_1 0x3B64 /* Calpella */ +#define MEI_DEV_ID_IBXPK_2 0x3B65 /* Calpella */ + +#define MEI_DEV_ID_CPT_1 0x1C3A /* Couger Point */ +#define MEI_DEV_ID_PBG_1 0x1D3A /* C600/X79 Patsburg */ + +#define MEI_DEV_ID_PPT_1 0x1E3A /* Panther Point */ +#define MEI_DEV_ID_PPT_2 0x1CBA /* Panther Point */ +#define MEI_DEV_ID_PPT_3 0x1DBA /* Panther Point */ + +#define MEI_DEV_ID_LPT_H 0x8C3A /* Lynx Point H */ +#define MEI_DEV_ID_LPT_W 0x8D3A /* Lynx Point - Wellsburg */ +#define MEI_DEV_ID_LPT_LP 0x9C3A /* Lynx Point LP */ +#define MEI_DEV_ID_LPT_HR 0x8CBA /* Lynx Point H Refresh */ + +#define MEI_DEV_ID_WPT_LP 0x9CBA /* Wildcat Point LP */ + +/* Host Firmware Status Registers in PCI Config Space */ +#define PCI_CFG_HFS_1 0x40 +#define PCI_CFG_HFS_2 0x48 + +/* + * MEI HW Section + */ + +/* MEI registers */ +/* H_CB_WW - Host Circular Buffer (CB) Write Window register */ +#define H_CB_WW 0 +/* H_CSR - Host Control Status register */ +#define H_CSR 4 +/* ME_CB_RW - ME Circular Buffer Read Window register (read only) */ +#define ME_CB_RW 8 +/* ME_CSR_HA - ME Control Status Host Access register (read only) */ +#define ME_CSR_HA 0xC +/* H_HGC_CSR - PGI register */ +#define H_HPG_CSR 0x10 + + +/* register bits of H_CSR (Host Control Status register) */ +/* Host Circular Buffer Depth - maximum number of 32-bit entries in CB */ +#define H_CBD 0xFF000000 +/* Host Circular Buffer Write Pointer */ +#define H_CBWP 0x00FF0000 +/* Host Circular Buffer Read Pointer */ +#define H_CBRP 0x0000FF00 +/* Host Reset */ +#define H_RST 0x00000010 +/* Host Ready */ +#define H_RDY 0x00000008 +/* Host Interrupt Generate */ +#define H_IG 0x00000004 +/* Host Interrupt Status */ +#define H_IS 0x00000002 +/* Host Interrupt Enable */ +#define H_IE 0x00000001 + + +/* register bits of ME_CSR_HA (ME Control Status Host Access register) */ +/* ME CB (Circular Buffer) Depth HRA (Host Read Access) - host read only +access to ME_CBD */ +#define ME_CBD_HRA 0xFF000000 +/* ME CB Write Pointer HRA - host read only access to ME_CBWP */ +#define ME_CBWP_HRA 0x00FF0000 +/* ME CB Read Pointer HRA - host read only access to ME_CBRP */ +#define ME_CBRP_HRA 0x0000FF00 +/* ME Power Gate Isolation Capability HRA - host ready only access */ +#define ME_PGIC_HRA 0x00000040 +/* ME Reset HRA - host read only access to ME_RST */ +#define ME_RST_HRA 0x00000010 +/* ME Ready HRA - host read only access to ME_RDY */ +#define ME_RDY_HRA 0x00000008 +/* ME Interrupt Generate HRA - host read only access to ME_IG */ +#define ME_IG_HRA 0x00000004 +/* ME Interrupt Status HRA - host read only access to ME_IS */ +#define ME_IS_HRA 0x00000002 +/* ME Interrupt Enable HRA - host read only access to ME_IE */ +#define ME_IE_HRA 0x00000001 + + +/* register bits - H_HPG_CSR */ +#define H_HPG_CSR_PGIHEXR 0x00000001 +#define H_HPG_CSR_PGI 0x00000002 + +#endif /* _MEI_HW_MEI_REGS_H_ */ diff --git a/drivers/misc/mei/hw-me.c b/drivers/misc/mei/hw-me.c new file mode 100644 index 00000000000..6a2d272cea4 --- /dev/null +++ b/drivers/misc/mei/hw-me.c @@ -0,0 +1,885 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/pci.h> + +#include <linux/kthread.h> +#include <linux/interrupt.h> + +#include "mei_dev.h" +#include "hbm.h" + +#include "hw-me.h" +#include "hw-me-regs.h" + +/** + * mei_me_reg_read - Reads 32bit data from the mei device + * + * @dev: the device structure + * @offset: offset from which to read the data + * + * returns register value (u32) + */ +static inline u32 mei_me_reg_read(const struct mei_me_hw *hw, + unsigned long offset) +{ + return ioread32(hw->mem_addr + offset); +} + + +/** + * mei_me_reg_write - Writes 32bit data to the mei device + * + * @dev: the device structure + * @offset: offset from which to write the data + * @value: register value to write (u32) + */ +static inline void mei_me_reg_write(const struct mei_me_hw *hw, + unsigned long offset, u32 value) +{ + iowrite32(value, hw->mem_addr + offset); +} + +/** + * mei_me_mecbrw_read - Reads 32bit data from ME circular buffer + * read window register + * + * @dev: the device structure + * + * returns ME_CB_RW register value (u32) + */ +static u32 mei_me_mecbrw_read(const struct mei_device *dev) +{ + return mei_me_reg_read(to_me_hw(dev), ME_CB_RW); +} +/** + * mei_me_mecsr_read - Reads 32bit data from the ME CSR + * + * @dev: the device structure + * + * returns ME_CSR_HA register value (u32) + */ +static inline u32 mei_me_mecsr_read(const struct mei_me_hw *hw) +{ + return mei_me_reg_read(hw, ME_CSR_HA); +} + +/** + * mei_hcsr_read - Reads 32bit data from the host CSR + * + * @dev: the device structure + * + * returns H_CSR register value (u32) + */ +static inline u32 mei_hcsr_read(const struct mei_me_hw *hw) +{ + return mei_me_reg_read(hw, H_CSR); +} + +/** + * mei_hcsr_set - writes H_CSR register to the mei device, + * and ignores the H_IS bit for it is write-one-to-zero. + * + * @dev: the device structure + */ +static inline void mei_hcsr_set(struct mei_me_hw *hw, u32 hcsr) +{ + hcsr &= ~H_IS; + mei_me_reg_write(hw, H_CSR, hcsr); +} + + +/** + * mei_me_hw_config - configure hw dependent settings + * + * @dev: mei device + */ +static void mei_me_hw_config(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 hcsr = mei_hcsr_read(to_me_hw(dev)); + /* Doesn't change in runtime */ + dev->hbuf_depth = (hcsr & H_CBD) >> 24; + + hw->pg_state = MEI_PG_OFF; +} + +/** + * mei_me_pg_state - translate internal pg state + * to the mei power gating state + * + * @hw - me hardware + * returns: MEI_PG_OFF if aliveness is on and MEI_PG_ON otherwise + */ +static inline enum mei_pg_state mei_me_pg_state(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + return hw->pg_state; +} + +/** + * mei_clear_interrupts - clear and stop interrupts + * + * @dev: the device structure + */ +static void mei_me_intr_clear(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 hcsr = mei_hcsr_read(hw); + if ((hcsr & H_IS) == H_IS) + mei_me_reg_write(hw, H_CSR, hcsr); +} +/** + * mei_me_intr_enable - enables mei device interrupts + * + * @dev: the device structure + */ +static void mei_me_intr_enable(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 hcsr = mei_hcsr_read(hw); + hcsr |= H_IE; + mei_hcsr_set(hw, hcsr); +} + +/** + * mei_disable_interrupts - disables mei device interrupts + * + * @dev: the device structure + */ +static void mei_me_intr_disable(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 hcsr = mei_hcsr_read(hw); + hcsr &= ~H_IE; + mei_hcsr_set(hw, hcsr); +} + +/** + * mei_me_hw_reset_release - release device from the reset + * + * @dev: the device structure + */ +static void mei_me_hw_reset_release(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 hcsr = mei_hcsr_read(hw); + + hcsr |= H_IG; + hcsr &= ~H_RST; + mei_hcsr_set(hw, hcsr); + + /* complete this write before we set host ready on another CPU */ + mmiowb(); +} +/** + * mei_me_hw_reset - resets fw via mei csr register. + * + * @dev: the device structure + * @intr_enable: if interrupt should be enabled after reset. + */ +static int mei_me_hw_reset(struct mei_device *dev, bool intr_enable) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 hcsr = mei_hcsr_read(hw); + + hcsr |= H_RST | H_IG | H_IS; + + if (intr_enable) + hcsr |= H_IE; + else + hcsr &= ~H_IE; + + dev->recvd_hw_ready = false; + mei_me_reg_write(hw, H_CSR, hcsr); + + /* + * Host reads the H_CSR once to ensure that the + * posted write to H_CSR completes. + */ + hcsr = mei_hcsr_read(hw); + + if ((hcsr & H_RST) == 0) + dev_warn(&dev->pdev->dev, "H_RST is not set = 0x%08X", hcsr); + + if ((hcsr & H_RDY) == H_RDY) + dev_warn(&dev->pdev->dev, "H_RDY is not cleared 0x%08X", hcsr); + + if (intr_enable == false) + mei_me_hw_reset_release(dev); + + return 0; +} + +/** + * mei_me_host_set_ready - enable device + * + * @dev - mei device + * returns bool + */ + +static void mei_me_host_set_ready(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + hw->host_hw_state = mei_hcsr_read(hw); + hw->host_hw_state |= H_IE | H_IG | H_RDY; + mei_hcsr_set(hw, hw->host_hw_state); +} +/** + * mei_me_host_is_ready - check whether the host has turned ready + * + * @dev - mei device + * returns bool + */ +static bool mei_me_host_is_ready(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + hw->host_hw_state = mei_hcsr_read(hw); + return (hw->host_hw_state & H_RDY) == H_RDY; +} + +/** + * mei_me_hw_is_ready - check whether the me(hw) has turned ready + * + * @dev - mei device + * returns bool + */ +static bool mei_me_hw_is_ready(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + hw->me_hw_state = mei_me_mecsr_read(hw); + return (hw->me_hw_state & ME_RDY_HRA) == ME_RDY_HRA; +} + +static int mei_me_hw_ready_wait(struct mei_device *dev) +{ + int err; + + mutex_unlock(&dev->device_lock); + err = wait_event_interruptible_timeout(dev->wait_hw_ready, + dev->recvd_hw_ready, + mei_secs_to_jiffies(MEI_HW_READY_TIMEOUT)); + mutex_lock(&dev->device_lock); + if (!err && !dev->recvd_hw_ready) { + if (!err) + err = -ETIME; + dev_err(&dev->pdev->dev, + "wait hw ready failed. status = %d\n", err); + return err; + } + + dev->recvd_hw_ready = false; + return 0; +} + +static int mei_me_hw_start(struct mei_device *dev) +{ + int ret = mei_me_hw_ready_wait(dev); + if (ret) + return ret; + dev_dbg(&dev->pdev->dev, "hw is ready\n"); + + mei_me_host_set_ready(dev); + return ret; +} + + +/** + * mei_hbuf_filled_slots - gets number of device filled buffer slots + * + * @dev: the device structure + * + * returns number of filled slots + */ +static unsigned char mei_hbuf_filled_slots(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + char read_ptr, write_ptr; + + hw->host_hw_state = mei_hcsr_read(hw); + + read_ptr = (char) ((hw->host_hw_state & H_CBRP) >> 8); + write_ptr = (char) ((hw->host_hw_state & H_CBWP) >> 16); + + return (unsigned char) (write_ptr - read_ptr); +} + +/** + * mei_me_hbuf_is_empty - checks if host buffer is empty. + * + * @dev: the device structure + * + * returns true if empty, false - otherwise. + */ +static bool mei_me_hbuf_is_empty(struct mei_device *dev) +{ + return mei_hbuf_filled_slots(dev) == 0; +} + +/** + * mei_me_hbuf_empty_slots - counts write empty slots. + * + * @dev: the device structure + * + * returns -EOVERFLOW if overflow, otherwise empty slots count + */ +static int mei_me_hbuf_empty_slots(struct mei_device *dev) +{ + unsigned char filled_slots, empty_slots; + + filled_slots = mei_hbuf_filled_slots(dev); + empty_slots = dev->hbuf_depth - filled_slots; + + /* check for overflow */ + if (filled_slots > dev->hbuf_depth) + return -EOVERFLOW; + + return empty_slots; +} + +static size_t mei_me_hbuf_max_len(const struct mei_device *dev) +{ + return dev->hbuf_depth * sizeof(u32) - sizeof(struct mei_msg_hdr); +} + + +/** + * mei_me_write_message - writes a message to mei device. + * + * @dev: the device structure + * @header: mei HECI header of message + * @buf: message payload will be written + * + * This function returns -EIO if write has failed + */ +static int mei_me_write_message(struct mei_device *dev, + struct mei_msg_hdr *header, + unsigned char *buf) +{ + struct mei_me_hw *hw = to_me_hw(dev); + unsigned long rem; + unsigned long length = header->length; + u32 *reg_buf = (u32 *)buf; + u32 hcsr; + u32 dw_cnt; + int i; + int empty_slots; + + dev_dbg(&dev->pdev->dev, MEI_HDR_FMT, MEI_HDR_PRM(header)); + + empty_slots = mei_hbuf_empty_slots(dev); + dev_dbg(&dev->pdev->dev, "empty slots = %hu.\n", empty_slots); + + dw_cnt = mei_data2slots(length); + if (empty_slots < 0 || dw_cnt > empty_slots) + return -EMSGSIZE; + + mei_me_reg_write(hw, H_CB_WW, *((u32 *) header)); + + for (i = 0; i < length / 4; i++) + mei_me_reg_write(hw, H_CB_WW, reg_buf[i]); + + rem = length & 0x3; + if (rem > 0) { + u32 reg = 0; + memcpy(®, &buf[length - rem], rem); + mei_me_reg_write(hw, H_CB_WW, reg); + } + + hcsr = mei_hcsr_read(hw) | H_IG; + mei_hcsr_set(hw, hcsr); + if (!mei_me_hw_is_ready(dev)) + return -EIO; + + return 0; +} + +/** + * mei_me_count_full_read_slots - counts read full slots. + * + * @dev: the device structure + * + * returns -EOVERFLOW if overflow, otherwise filled slots count + */ +static int mei_me_count_full_read_slots(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + char read_ptr, write_ptr; + unsigned char buffer_depth, filled_slots; + + hw->me_hw_state = mei_me_mecsr_read(hw); + buffer_depth = (unsigned char)((hw->me_hw_state & ME_CBD_HRA) >> 24); + read_ptr = (char) ((hw->me_hw_state & ME_CBRP_HRA) >> 8); + write_ptr = (char) ((hw->me_hw_state & ME_CBWP_HRA) >> 16); + filled_slots = (unsigned char) (write_ptr - read_ptr); + + /* check for overflow */ + if (filled_slots > buffer_depth) + return -EOVERFLOW; + + dev_dbg(&dev->pdev->dev, "filled_slots =%08x\n", filled_slots); + return (int)filled_slots; +} + +/** + * mei_me_read_slots - reads a message from mei device. + * + * @dev: the device structure + * @buffer: message buffer will be written + * @buffer_length: message size will be read + */ +static int mei_me_read_slots(struct mei_device *dev, unsigned char *buffer, + unsigned long buffer_length) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 *reg_buf = (u32 *)buffer; + u32 hcsr; + + for (; buffer_length >= sizeof(u32); buffer_length -= sizeof(u32)) + *reg_buf++ = mei_me_mecbrw_read(dev); + + if (buffer_length > 0) { + u32 reg = mei_me_mecbrw_read(dev); + memcpy(reg_buf, ®, buffer_length); + } + + hcsr = mei_hcsr_read(hw) | H_IG; + mei_hcsr_set(hw, hcsr); + return 0; +} + +/** + * mei_me_pg_enter - write pg enter register to mei device. + * + * @dev: the device structure + */ +static void mei_me_pg_enter(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 reg = mei_me_reg_read(hw, H_HPG_CSR); + reg |= H_HPG_CSR_PGI; + mei_me_reg_write(hw, H_HPG_CSR, reg); +} + +/** + * mei_me_pg_enter - write pg enter register to mei device. + * + * @dev: the device structure + */ +static void mei_me_pg_exit(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 reg = mei_me_reg_read(hw, H_HPG_CSR); + + WARN(!(reg & H_HPG_CSR_PGI), "PGI is not set\n"); + + reg |= H_HPG_CSR_PGIHEXR; + mei_me_reg_write(hw, H_HPG_CSR, reg); +} + +/** + * mei_me_pg_set_sync - perform pg entry procedure + * + * @dev: the device structure + * + * returns 0 on success an error code otherwise + */ +int mei_me_pg_set_sync(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + unsigned long timeout = mei_secs_to_jiffies(MEI_PGI_TIMEOUT); + int ret; + + dev->pg_event = MEI_PG_EVENT_WAIT; + + ret = mei_hbm_pg(dev, MEI_PG_ISOLATION_ENTRY_REQ_CMD); + if (ret) + return ret; + + mutex_unlock(&dev->device_lock); + wait_event_timeout(dev->wait_pg, + dev->pg_event == MEI_PG_EVENT_RECEIVED, timeout); + mutex_lock(&dev->device_lock); + + if (dev->pg_event == MEI_PG_EVENT_RECEIVED) { + mei_me_pg_enter(dev); + ret = 0; + } else { + ret = -ETIME; + } + + dev->pg_event = MEI_PG_EVENT_IDLE; + hw->pg_state = MEI_PG_ON; + + return ret; +} + +/** + * mei_me_pg_unset_sync - perform pg exit procedure + * + * @dev: the device structure + * + * returns 0 on success an error code otherwise + */ +int mei_me_pg_unset_sync(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + unsigned long timeout = mei_secs_to_jiffies(MEI_PGI_TIMEOUT); + int ret; + + if (dev->pg_event == MEI_PG_EVENT_RECEIVED) + goto reply; + + dev->pg_event = MEI_PG_EVENT_WAIT; + + mei_me_pg_exit(dev); + + mutex_unlock(&dev->device_lock); + wait_event_timeout(dev->wait_pg, + dev->pg_event == MEI_PG_EVENT_RECEIVED, timeout); + mutex_lock(&dev->device_lock); + +reply: + if (dev->pg_event == MEI_PG_EVENT_RECEIVED) + ret = mei_hbm_pg(dev, MEI_PG_ISOLATION_EXIT_RES_CMD); + else + ret = -ETIME; + + dev->pg_event = MEI_PG_EVENT_IDLE; + hw->pg_state = MEI_PG_OFF; + + return ret; +} + +/** + * mei_me_pg_is_enabled - detect if PG is supported by HW + * + * @dev: the device structure + * + * returns: true is pg supported, false otherwise + */ +static bool mei_me_pg_is_enabled(struct mei_device *dev) +{ + struct mei_me_hw *hw = to_me_hw(dev); + u32 reg = mei_me_reg_read(hw, ME_CSR_HA); + + if ((reg & ME_PGIC_HRA) == 0) + goto notsupported; + + if (dev->version.major_version < HBM_MAJOR_VERSION_PGI) + goto notsupported; + + if (dev->version.major_version == HBM_MAJOR_VERSION_PGI && + dev->version.minor_version < HBM_MINOR_VERSION_PGI) + goto notsupported; + + return true; + +notsupported: + dev_dbg(&dev->pdev->dev, "pg: not supported: HGP = %d hbm version %d.%d ?= %d.%d\n", + !!(reg & ME_PGIC_HRA), + dev->version.major_version, + dev->version.minor_version, + HBM_MAJOR_VERSION_PGI, + HBM_MINOR_VERSION_PGI); + + return false; +} + +/** + * mei_me_irq_quick_handler - The ISR of the MEI device + * + * @irq: The irq number + * @dev_id: pointer to the device structure + * + * returns irqreturn_t + */ + +irqreturn_t mei_me_irq_quick_handler(int irq, void *dev_id) +{ + struct mei_device *dev = (struct mei_device *) dev_id; + struct mei_me_hw *hw = to_me_hw(dev); + u32 csr_reg = mei_hcsr_read(hw); + + if ((csr_reg & H_IS) != H_IS) + return IRQ_NONE; + + /* clear H_IS bit in H_CSR */ + mei_me_reg_write(hw, H_CSR, csr_reg); + + return IRQ_WAKE_THREAD; +} + +/** + * mei_me_irq_thread_handler - function called after ISR to handle the interrupt + * processing. + * + * @irq: The irq number + * @dev_id: pointer to the device structure + * + * returns irqreturn_t + * + */ +irqreturn_t mei_me_irq_thread_handler(int irq, void *dev_id) +{ + struct mei_device *dev = (struct mei_device *) dev_id; + struct mei_cl_cb complete_list; + s32 slots; + int rets = 0; + + dev_dbg(&dev->pdev->dev, "function called after ISR to handle the interrupt processing.\n"); + /* initialize our complete list */ + mutex_lock(&dev->device_lock); + mei_io_list_init(&complete_list); + + /* Ack the interrupt here + * In case of MSI we don't go through the quick handler */ + if (pci_dev_msi_enabled(dev->pdev)) + mei_clear_interrupts(dev); + + /* check if ME wants a reset */ + if (!mei_hw_is_ready(dev) && dev->dev_state != MEI_DEV_RESETTING) { + dev_warn(&dev->pdev->dev, "FW not ready: resetting.\n"); + schedule_work(&dev->reset_work); + goto end; + } + + /* check if we need to start the dev */ + if (!mei_host_is_ready(dev)) { + if (mei_hw_is_ready(dev)) { + mei_me_hw_reset_release(dev); + dev_dbg(&dev->pdev->dev, "we need to start the dev.\n"); + + dev->recvd_hw_ready = true; + wake_up_interruptible(&dev->wait_hw_ready); + } else { + dev_dbg(&dev->pdev->dev, "Spurious Interrupt\n"); + } + goto end; + } + /* check slots available for reading */ + slots = mei_count_full_read_slots(dev); + while (slots > 0) { + dev_dbg(&dev->pdev->dev, "slots to read = %08x\n", slots); + rets = mei_irq_read_handler(dev, &complete_list, &slots); + /* There is a race between ME write and interrupt delivery: + * Not all data is always available immediately after the + * interrupt, so try to read again on the next interrupt. + */ + if (rets == -ENODATA) + break; + + if (rets && dev->dev_state != MEI_DEV_RESETTING) { + dev_err(&dev->pdev->dev, "mei_irq_read_handler ret = %d.\n", + rets); + schedule_work(&dev->reset_work); + goto end; + } + } + + dev->hbuf_is_ready = mei_hbuf_is_ready(dev); + + /* + * During PG handshake only allowed write is the replay to the + * PG exit message, so block calling write function + * if the pg state is not idle + */ + if (dev->pg_event == MEI_PG_EVENT_IDLE) { + rets = mei_irq_write_handler(dev, &complete_list); + dev->hbuf_is_ready = mei_hbuf_is_ready(dev); + } + + mei_irq_compl_handler(dev, &complete_list); + +end: + dev_dbg(&dev->pdev->dev, "interrupt thread end ret = %d\n", rets); + mutex_unlock(&dev->device_lock); + return IRQ_HANDLED; +} + +/** + * mei_me_fw_status - retrieve fw status from the pci config space + * + * @dev: the device structure + * @fw_status: fw status registers storage + * + * returns 0 on success an error code otherwise + */ +static int mei_me_fw_status(struct mei_device *dev, + struct mei_fw_status *fw_status) +{ + const u32 pci_cfg_reg[] = {PCI_CFG_HFS_1, PCI_CFG_HFS_2}; + int i; + + if (!fw_status) + return -EINVAL; + + switch (dev->pdev->device) { + case MEI_DEV_ID_IBXPK_1: + case MEI_DEV_ID_IBXPK_2: + case MEI_DEV_ID_CPT_1: + case MEI_DEV_ID_PBG_1: + case MEI_DEV_ID_PPT_1: + case MEI_DEV_ID_PPT_2: + case MEI_DEV_ID_PPT_3: + case MEI_DEV_ID_LPT_H: + case MEI_DEV_ID_LPT_W: + case MEI_DEV_ID_LPT_LP: + case MEI_DEV_ID_LPT_HR: + case MEI_DEV_ID_WPT_LP: + fw_status->count = 2; + break; + case MEI_DEV_ID_ICH10_1: + case MEI_DEV_ID_ICH10_2: + case MEI_DEV_ID_ICH10_3: + case MEI_DEV_ID_ICH10_4: + fw_status->count = 1; + break; + default: + fw_status->count = 0; + break; + } + + for (i = 0; i < fw_status->count && i < MEI_FW_STATUS_MAX; i++) { + int ret; + ret = pci_read_config_dword(dev->pdev, + pci_cfg_reg[i], &fw_status->status[i]); + if (ret) + return ret; + } + return 0; +} + +static const struct mei_hw_ops mei_me_hw_ops = { + + .pg_state = mei_me_pg_state, + + .fw_status = mei_me_fw_status, + .host_is_ready = mei_me_host_is_ready, + + .hw_is_ready = mei_me_hw_is_ready, + .hw_reset = mei_me_hw_reset, + .hw_config = mei_me_hw_config, + .hw_start = mei_me_hw_start, + + .pg_is_enabled = mei_me_pg_is_enabled, + + .intr_clear = mei_me_intr_clear, + .intr_enable = mei_me_intr_enable, + .intr_disable = mei_me_intr_disable, + + .hbuf_free_slots = mei_me_hbuf_empty_slots, + .hbuf_is_ready = mei_me_hbuf_is_empty, + .hbuf_max_len = mei_me_hbuf_max_len, + + .write = mei_me_write_message, + + .rdbuf_full_slots = mei_me_count_full_read_slots, + .read_hdr = mei_me_mecbrw_read, + .read = mei_me_read_slots +}; + +static bool mei_me_fw_type_nm(struct pci_dev *pdev) +{ + u32 reg; + pci_read_config_dword(pdev, PCI_CFG_HFS_2, ®); + /* make sure that bit 9 (NM) is up and bit 10 (DM) is down */ + return (reg & 0x600) == 0x200; +} + +#define MEI_CFG_FW_NM \ + .quirk_probe = mei_me_fw_type_nm + +static bool mei_me_fw_type_sps(struct pci_dev *pdev) +{ + u32 reg; + /* Read ME FW Status check for SPS Firmware */ + pci_read_config_dword(pdev, PCI_CFG_HFS_1, ®); + /* if bits [19:16] = 15, running SPS Firmware */ + return (reg & 0xf0000) == 0xf0000; +} + +#define MEI_CFG_FW_SPS \ + .quirk_probe = mei_me_fw_type_sps + + +#define MEI_CFG_LEGACY_HFS \ + .fw_status.count = 0 + +#define MEI_CFG_ICH_HFS \ + .fw_status.count = 1, \ + .fw_status.status[0] = PCI_CFG_HFS_1 + +#define MEI_CFG_PCH_HFS \ + .fw_status.count = 2, \ + .fw_status.status[0] = PCI_CFG_HFS_1, \ + .fw_status.status[1] = PCI_CFG_HFS_2 + + +/* ICH Legacy devices */ +const struct mei_cfg mei_me_legacy_cfg = { + MEI_CFG_LEGACY_HFS, +}; + +/* ICH devices */ +const struct mei_cfg mei_me_ich_cfg = { + MEI_CFG_ICH_HFS, +}; + +/* PCH devices */ +const struct mei_cfg mei_me_pch_cfg = { + MEI_CFG_PCH_HFS, +}; + + +/* PCH Cougar Point and Patsburg with quirk for Node Manager exclusion */ +const struct mei_cfg mei_me_pch_cpt_pbg_cfg = { + MEI_CFG_PCH_HFS, + MEI_CFG_FW_NM, +}; + +/* PCH Lynx Point with quirk for SPS Firmware exclusion */ +const struct mei_cfg mei_me_lpt_cfg = { + MEI_CFG_PCH_HFS, + MEI_CFG_FW_SPS, +}; + +/** + * mei_me_dev_init - allocates and initializes the mei device structure + * + * @pdev: The pci device structure + * @cfg: per device generation config + * + * returns The mei_device_device pointer on success, NULL on failure. + */ +struct mei_device *mei_me_dev_init(struct pci_dev *pdev, + const struct mei_cfg *cfg) +{ + struct mei_device *dev; + + dev = kzalloc(sizeof(struct mei_device) + + sizeof(struct mei_me_hw), GFP_KERNEL); + if (!dev) + return NULL; + + mei_device_init(dev, cfg); + + dev->ops = &mei_me_hw_ops; + + dev->pdev = pdev; + return dev; +} + diff --git a/drivers/misc/mei/hw-me.h b/drivers/misc/mei/hw-me.h new file mode 100644 index 00000000000..12b0f4bbe1f --- /dev/null +++ b/drivers/misc/mei/hw-me.h @@ -0,0 +1,56 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + + + +#ifndef _MEI_INTERFACE_H_ +#define _MEI_INTERFACE_H_ + +#include <linux/mei.h> +#include <linux/irqreturn.h> +#include "mei_dev.h" +#include "client.h" + +#define MEI_ME_RPM_TIMEOUT 500 /* ms */ + +struct mei_me_hw { + void __iomem *mem_addr; + /* + * hw states of host and fw(ME) + */ + u32 host_hw_state; + u32 me_hw_state; + enum mei_pg_state pg_state; +}; + +#define to_me_hw(dev) (struct mei_me_hw *)((dev)->hw) + +extern const struct mei_cfg mei_me_legacy_cfg; +extern const struct mei_cfg mei_me_ich_cfg; +extern const struct mei_cfg mei_me_pch_cfg; +extern const struct mei_cfg mei_me_pch_cpt_pbg_cfg; +extern const struct mei_cfg mei_me_lpt_cfg; + +struct mei_device *mei_me_dev_init(struct pci_dev *pdev, + const struct mei_cfg *cfg); + +int mei_me_pg_set_sync(struct mei_device *dev); +int mei_me_pg_unset_sync(struct mei_device *dev); + +irqreturn_t mei_me_irq_quick_handler(int irq, void *dev_id); +irqreturn_t mei_me_irq_thread_handler(int irq, void *dev_id); + +#endif /* _MEI_INTERFACE_H_ */ diff --git a/drivers/misc/mei/hw-txe-regs.h b/drivers/misc/mei/hw-txe-regs.h new file mode 100644 index 00000000000..f19229c4e65 --- /dev/null +++ b/drivers/misc/mei/hw-txe-regs.h @@ -0,0 +1,294 @@ +/****************************************************************************** + * Intel Management Engine Interface (Intel MEI) Linux driver + * Intel MEI Interface Header + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * Copyright(c) 2013 - 2014 Intel Corporation. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution + * in the file called COPYING + * + * Contact Information: + * Intel Corporation. + * linux-mei@linux.intel.com + * http://www.intel.com + * + * BSD LICENSE + * + * Copyright(c) 2013 - 2014 Intel Corporation. All rights reserved. + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * * Neither the name Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + *****************************************************************************/ +#ifndef _MEI_HW_TXE_REGS_H_ +#define _MEI_HW_TXE_REGS_H_ + +#include "hw.h" + +#define SEC_ALIVENESS_TIMER_TIMEOUT (5 * MSEC_PER_SEC) +#define SEC_ALIVENESS_WAIT_TIMEOUT (1 * MSEC_PER_SEC) +#define SEC_RESET_WAIT_TIMEOUT (1 * MSEC_PER_SEC) +#define SEC_READY_WAIT_TIMEOUT (5 * MSEC_PER_SEC) +#define START_MESSAGE_RESPONSE_WAIT_TIMEOUT (5 * MSEC_PER_SEC) +#define RESET_CANCEL_WAIT_TIMEOUT (1 * MSEC_PER_SEC) + +enum { + SEC_BAR, + BRIDGE_BAR, + + NUM_OF_MEM_BARS +}; + +/* SeC FW Status Register + * + * FW uses this register in order to report its status to host. + * This register resides in PCI-E config space. + */ +#define PCI_CFG_TXE_FW_STS0 0x40 +# define PCI_CFG_TXE_FW_STS0_WRK_ST_MSK 0x0000000F +# define PCI_CFG_TXE_FW_STS0_OP_ST_MSK 0x000001C0 +# define PCI_CFG_TXE_FW_STS0_FW_INIT_CMPLT 0x00000200 +# define PCI_CFG_TXE_FW_STS0_ERR_CODE_MSK 0x0000F000 +# define PCI_CFG_TXE_FW_STS0_OP_MODE_MSK 0x000F0000 +# define PCI_CFG_TXE_FW_STS0_RST_CNT_MSK 0x00F00000 +#define PCI_CFG_TXE_FW_STS1 0x48 + +#define IPC_BASE_ADDR 0x80400 /* SeC IPC Base Address */ + +/* IPC Input Doorbell Register */ +#define SEC_IPC_INPUT_DOORBELL_REG (0x0000 + IPC_BASE_ADDR) + +/* IPC Input Status Register + * This register indicates whether or not processing of + * the most recent command has been completed by the SEC + * New commands and payloads should not be written by the Host + * until this indicates that the previous command has been processed. + */ +#define SEC_IPC_INPUT_STATUS_REG (0x0008 + IPC_BASE_ADDR) +# define SEC_IPC_INPUT_STATUS_RDY BIT(0) + +/* IPC Host Interrupt Status Register */ +#define SEC_IPC_HOST_INT_STATUS_REG (0x0010 + IPC_BASE_ADDR) +#define SEC_IPC_HOST_INT_STATUS_OUT_DB BIT(0) +#define SEC_IPC_HOST_INT_STATUS_IN_RDY BIT(1) +#define SEC_IPC_HOST_INT_STATUS_HDCP_M0_RCVD BIT(5) +#define SEC_IPC_HOST_INT_STATUS_ILL_MEM_ACCESS BIT(17) +#define SEC_IPC_HOST_INT_STATUS_AES_HKEY_ERR BIT(18) +#define SEC_IPC_HOST_INT_STATUS_DES_HKEY_ERR BIT(19) +#define SEC_IPC_HOST_INT_STATUS_TMRMTB_OVERFLOW BIT(21) + +/* Convenient mask for pending interrupts */ +#define SEC_IPC_HOST_INT_STATUS_PENDING \ + (SEC_IPC_HOST_INT_STATUS_OUT_DB| \ + SEC_IPC_HOST_INT_STATUS_IN_RDY) + +/* IPC Host Interrupt Mask Register */ +#define SEC_IPC_HOST_INT_MASK_REG (0x0014 + IPC_BASE_ADDR) + +# define SEC_IPC_HOST_INT_MASK_OUT_DB BIT(0) /* Output Doorbell Int Mask */ +# define SEC_IPC_HOST_INT_MASK_IN_RDY BIT(1) /* Input Ready Int Mask */ + +/* IPC Input Payload RAM */ +#define SEC_IPC_INPUT_PAYLOAD_REG (0x0100 + IPC_BASE_ADDR) +/* IPC Shared Payload RAM */ +#define IPC_SHARED_PAYLOAD_REG (0x0200 + IPC_BASE_ADDR) + +/* SeC Address Translation Table Entry 2 - Ctrl + * + * This register resides also in SeC's PCI-E Memory space. + */ +#define SATT2_CTRL_REG 0x1040 +# define SATT2_CTRL_VALID_MSK BIT(0) +# define SATT2_CTRL_BR_BASE_ADDR_REG_SHIFT 8 +# define SATT2_CTRL_BRIDGE_HOST_EN_MSK BIT(12) + +/* SATT Table Entry 2 SAP Base Address Register */ +#define SATT2_SAP_BA_REG 0x1044 +/* SATT Table Entry 2 SAP Size Register. */ +#define SATT2_SAP_SIZE_REG 0x1048 + /* SATT Table Entry 2 SAP Bridge Address - LSB Register */ +#define SATT2_BRG_BA_LSB_REG 0x104C + +/* Host High-level Interrupt Status Register */ +#define HHISR_REG 0x2020 +/* Host High-level Interrupt Enable Register + * + * Resides in PCI memory space. This is the top hierarchy for + * interrupts from SeC to host, aggregating both interrupts that + * arrive through HICR registers as well as interrupts + * that arrive via IPC. + */ +#define HHIER_REG 0x2024 +#define IPC_HHIER_SEC BIT(0) +#define IPC_HHIER_BRIDGE BIT(1) +#define IPC_HHIER_MSK (IPC_HHIER_SEC | IPC_HHIER_BRIDGE) + +/* Host High-level Interrupt Mask Register. + * + * Resides in PCI memory space. + * This is the top hierarchy for masking interrupts from SeC to host. + */ +#define HHIMR_REG 0x2028 +#define IPC_HHIMR_SEC BIT(0) +#define IPC_HHIMR_BRIDGE BIT(1) + +/* Host High-level IRQ Status Register */ +#define HHIRQSR_REG 0x202C + +/* Host Interrupt Cause Register 0 - SeC IPC Readiness + * + * This register is both an ICR to Host from PCI Memory Space + * and it is also exposed in the SeC memory space. + * This register is used by SeC's IPC driver in order + * to synchronize with host about IPC interface state. + */ +#define HICR_SEC_IPC_READINESS_REG 0x2040 +#define HICR_SEC_IPC_READINESS_HOST_RDY BIT(0) +#define HICR_SEC_IPC_READINESS_SEC_RDY BIT(1) +#define HICR_SEC_IPC_READINESS_SYS_RDY \ + (HICR_SEC_IPC_READINESS_HOST_RDY | \ + HICR_SEC_IPC_READINESS_SEC_RDY) +#define HICR_SEC_IPC_READINESS_RDY_CLR BIT(2) + +/* Host Interrupt Cause Register 1 - Aliveness Response */ +/* This register is both an ICR to Host from PCI Memory Space + * and it is also exposed in the SeC memory space. + * The register may be used by SeC to ACK a host request for aliveness. + */ +#define HICR_HOST_ALIVENESS_RESP_REG 0x2044 +#define HICR_HOST_ALIVENESS_RESP_ACK BIT(0) + +/* Host Interrupt Cause Register 2 - SeC IPC Output Doorbell */ +#define HICR_SEC_IPC_OUTPUT_DOORBELL_REG 0x2048 + +/* Host Interrupt Status Register. + * + * Resides in PCI memory space. + * This is the main register involved in generating interrupts + * from SeC to host via HICRs. + * The interrupt generation rules are as follows: + * An interrupt will be generated whenever for any i, + * there is a transition from a state where at least one of + * the following conditions did not hold, to a state where + * ALL the following conditions hold: + * A) HISR.INT[i]_STS == 1. + * B) HIER.INT[i]_EN == 1. + */ +#define HISR_REG 0x2060 +#define HISR_INT_0_STS BIT(0) +#define HISR_INT_1_STS BIT(1) +#define HISR_INT_2_STS BIT(2) +#define HISR_INT_3_STS BIT(3) +#define HISR_INT_4_STS BIT(4) +#define HISR_INT_5_STS BIT(5) +#define HISR_INT_6_STS BIT(6) +#define HISR_INT_7_STS BIT(7) +#define HISR_INT_STS_MSK \ + (HISR_INT_0_STS | HISR_INT_1_STS | HISR_INT_2_STS) + +/* Host Interrupt Enable Register. Resides in PCI memory space. */ +#define HIER_REG 0x2064 +#define HIER_INT_0_EN BIT(0) +#define HIER_INT_1_EN BIT(1) +#define HIER_INT_2_EN BIT(2) +#define HIER_INT_3_EN BIT(3) +#define HIER_INT_4_EN BIT(4) +#define HIER_INT_5_EN BIT(5) +#define HIER_INT_6_EN BIT(6) +#define HIER_INT_7_EN BIT(7) + +#define HIER_INT_EN_MSK \ + (HIER_INT_0_EN | HIER_INT_1_EN | HIER_INT_2_EN) + + +/* SEC Memory Space IPC output payload. + * + * This register is part of the output payload which SEC provides to host. + */ +#define BRIDGE_IPC_OUTPUT_PAYLOAD_REG 0x20C0 + +/* SeC Interrupt Cause Register - Host Aliveness Request + * This register is both an ICR to SeC and it is also exposed + * in the host-visible PCI memory space. + * The register is used by host to request SeC aliveness. + */ +#define SICR_HOST_ALIVENESS_REQ_REG 0x214C +#define SICR_HOST_ALIVENESS_REQ_REQUESTED BIT(0) + + +/* SeC Interrupt Cause Register - Host IPC Readiness + * + * This register is both an ICR to SeC and it is also exposed + * in the host-visible PCI memory space. + * This register is used by the host's SeC driver uses in order + * to synchronize with SeC about IPC interface state. + */ +#define SICR_HOST_IPC_READINESS_REQ_REG 0x2150 + + +#define SICR_HOST_IPC_READINESS_HOST_RDY BIT(0) +#define SICR_HOST_IPC_READINESS_SEC_RDY BIT(1) +#define SICR_HOST_IPC_READINESS_SYS_RDY \ + (SICR_HOST_IPC_READINESS_HOST_RDY | \ + SICR_HOST_IPC_READINESS_SEC_RDY) +#define SICR_HOST_IPC_READINESS_RDY_CLR BIT(2) + +/* SeC Interrupt Cause Register - SeC IPC Output Status + * + * This register indicates whether or not processing of the most recent + * command has been completed by the Host. + * New commands and payloads should not be written by SeC until this + * register indicates that the previous command has been processed. + */ +#define SICR_SEC_IPC_OUTPUT_STATUS_REG 0x2154 +# define SEC_IPC_OUTPUT_STATUS_RDY BIT(0) + + + +/* MEI IPC Message payload size 64 bytes */ +#define PAYLOAD_SIZE 64 + +/* MAX size for SATT range 32MB */ +#define SATT_RANGE_MAX (32 << 20) + + +#endif /* _MEI_HW_TXE_REGS_H_ */ + diff --git a/drivers/misc/mei/hw-txe.c b/drivers/misc/mei/hw-txe.c new file mode 100644 index 00000000000..93273783dec --- /dev/null +++ b/drivers/misc/mei/hw-txe.c @@ -0,0 +1,1190 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2013-2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/pci.h> +#include <linux/jiffies.h> +#include <linux/delay.h> +#include <linux/kthread.h> +#include <linux/irqreturn.h> + +#include <linux/mei.h> + +#include "mei_dev.h" +#include "hw-txe.h" +#include "client.h" +#include "hbm.h" + +/** + * mei_txe_reg_read - Reads 32bit data from the device + * + * @base_addr: registers base address + * @offset: register offset + * + */ +static inline u32 mei_txe_reg_read(void __iomem *base_addr, + unsigned long offset) +{ + return ioread32(base_addr + offset); +} + +/** + * mei_txe_reg_write - Writes 32bit data to the device + * + * @base_addr: registers base address + * @offset: register offset + * @value: the value to write + */ +static inline void mei_txe_reg_write(void __iomem *base_addr, + unsigned long offset, u32 value) +{ + iowrite32(value, base_addr + offset); +} + +/** + * mei_txe_sec_reg_read_silent - Reads 32bit data from the SeC BAR + * + * @dev: the device structure + * @offset: register offset + * + * Doesn't check for aliveness while Reads 32bit data from the SeC BAR + */ +static inline u32 mei_txe_sec_reg_read_silent(struct mei_txe_hw *hw, + unsigned long offset) +{ + return mei_txe_reg_read(hw->mem_addr[SEC_BAR], offset); +} + +/** + * mei_txe_sec_reg_read - Reads 32bit data from the SeC BAR + * + * @dev: the device structure + * @offset: register offset + * + * Reads 32bit data from the SeC BAR and shout loud if aliveness is not set + */ +static inline u32 mei_txe_sec_reg_read(struct mei_txe_hw *hw, + unsigned long offset) +{ + WARN(!hw->aliveness, "sec read: aliveness not asserted\n"); + return mei_txe_sec_reg_read_silent(hw, offset); +} +/** + * mei_txe_sec_reg_write_silent - Writes 32bit data to the SeC BAR + * doesn't check for aliveness + * + * @dev: the device structure + * @offset: register offset + * @value: value to write + * + * Doesn't check for aliveness while writes 32bit data from to the SeC BAR + */ +static inline void mei_txe_sec_reg_write_silent(struct mei_txe_hw *hw, + unsigned long offset, u32 value) +{ + mei_txe_reg_write(hw->mem_addr[SEC_BAR], offset, value); +} + +/** + * mei_txe_sec_reg_write - Writes 32bit data to the SeC BAR + * + * @dev: the device structure + * @offset: register offset + * @value: value to write + * + * Writes 32bit data from the SeC BAR and shout loud if aliveness is not set + */ +static inline void mei_txe_sec_reg_write(struct mei_txe_hw *hw, + unsigned long offset, u32 value) +{ + WARN(!hw->aliveness, "sec write: aliveness not asserted\n"); + mei_txe_sec_reg_write_silent(hw, offset, value); +} +/** + * mei_txe_br_reg_read - Reads 32bit data from the Bridge BAR + * + * @hw: the device structure + * @offset: offset from which to read the data + * + */ +static inline u32 mei_txe_br_reg_read(struct mei_txe_hw *hw, + unsigned long offset) +{ + return mei_txe_reg_read(hw->mem_addr[BRIDGE_BAR], offset); +} + +/** + * mei_txe_br_reg_write - Writes 32bit data to the Bridge BAR + * + * @hw: the device structure + * @offset: offset from which to write the data + * @value: the byte to write + */ +static inline void mei_txe_br_reg_write(struct mei_txe_hw *hw, + unsigned long offset, u32 value) +{ + mei_txe_reg_write(hw->mem_addr[BRIDGE_BAR], offset, value); +} + +/** + * mei_txe_aliveness_set - request for aliveness change + * + * @dev: the device structure + * @req: requested aliveness value + * + * Request for aliveness change and returns true if the change is + * really needed and false if aliveness is already + * in the requested state + * Requires device lock to be held + */ +static bool mei_txe_aliveness_set(struct mei_device *dev, u32 req) +{ + + struct mei_txe_hw *hw = to_txe_hw(dev); + bool do_req = hw->aliveness != req; + + dev_dbg(&dev->pdev->dev, "Aliveness current=%d request=%d\n", + hw->aliveness, req); + if (do_req) { + dev->pg_event = MEI_PG_EVENT_WAIT; + mei_txe_br_reg_write(hw, SICR_HOST_ALIVENESS_REQ_REG, req); + } + return do_req; +} + + +/** + * mei_txe_aliveness_req_get - get aliveness requested register value + * + * @dev: the device structure + * + * Extract HICR_HOST_ALIVENESS_RESP_ACK bit from + * from HICR_HOST_ALIVENESS_REQ register value + */ +static u32 mei_txe_aliveness_req_get(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + u32 reg; + reg = mei_txe_br_reg_read(hw, SICR_HOST_ALIVENESS_REQ_REG); + return reg & SICR_HOST_ALIVENESS_REQ_REQUESTED; +} + +/** + * mei_txe_aliveness_get - get aliveness response register value + * @dev: the device structure + * + * Extract HICR_HOST_ALIVENESS_RESP_ACK bit + * from HICR_HOST_ALIVENESS_RESP register value + */ +static u32 mei_txe_aliveness_get(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + u32 reg; + reg = mei_txe_br_reg_read(hw, HICR_HOST_ALIVENESS_RESP_REG); + return reg & HICR_HOST_ALIVENESS_RESP_ACK; +} + +/** + * mei_txe_aliveness_poll - waits for aliveness to settle + * + * @dev: the device structure + * @expected: expected aliveness value + * + * Polls for HICR_HOST_ALIVENESS_RESP.ALIVENESS_RESP to be set + * returns > 0 if the expected value was received, -ETIME otherwise + */ +static int mei_txe_aliveness_poll(struct mei_device *dev, u32 expected) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + int t = 0; + + do { + hw->aliveness = mei_txe_aliveness_get(dev); + if (hw->aliveness == expected) { + dev->pg_event = MEI_PG_EVENT_IDLE; + dev_dbg(&dev->pdev->dev, + "aliveness settled after %d msecs\n", t); + return t; + } + mutex_unlock(&dev->device_lock); + msleep(MSEC_PER_SEC / 5); + mutex_lock(&dev->device_lock); + t += MSEC_PER_SEC / 5; + } while (t < SEC_ALIVENESS_WAIT_TIMEOUT); + + dev->pg_event = MEI_PG_EVENT_IDLE; + dev_err(&dev->pdev->dev, "aliveness timed out\n"); + return -ETIME; +} + +/** + * mei_txe_aliveness_wait - waits for aliveness to settle + * + * @dev: the device structure + * @expected: expected aliveness value + * + * Waits for HICR_HOST_ALIVENESS_RESP.ALIVENESS_RESP to be set + * returns returns 0 on success and < 0 otherwise + */ +static int mei_txe_aliveness_wait(struct mei_device *dev, u32 expected) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + const unsigned long timeout = + msecs_to_jiffies(SEC_ALIVENESS_WAIT_TIMEOUT); + long err; + int ret; + + hw->aliveness = mei_txe_aliveness_get(dev); + if (hw->aliveness == expected) + return 0; + + mutex_unlock(&dev->device_lock); + err = wait_event_timeout(hw->wait_aliveness_resp, + dev->pg_event == MEI_PG_EVENT_RECEIVED, timeout); + mutex_lock(&dev->device_lock); + + hw->aliveness = mei_txe_aliveness_get(dev); + ret = hw->aliveness == expected ? 0 : -ETIME; + + if (ret) + dev_warn(&dev->pdev->dev, "aliveness timed out = %ld aliveness = %d event = %d\n", + err, hw->aliveness, dev->pg_event); + else + dev_dbg(&dev->pdev->dev, "aliveness settled after = %d msec aliveness = %d event = %d\n", + jiffies_to_msecs(timeout - err), + hw->aliveness, dev->pg_event); + + dev->pg_event = MEI_PG_EVENT_IDLE; + return ret; +} + +/** + * mei_txe_aliveness_set_sync - sets an wait for aliveness to complete + * + * @dev: the device structure + * + * returns returns 0 on success and < 0 otherwise + */ +int mei_txe_aliveness_set_sync(struct mei_device *dev, u32 req) +{ + if (mei_txe_aliveness_set(dev, req)) + return mei_txe_aliveness_wait(dev, req); + return 0; +} + +/** + * mei_txe_pg_is_enabled - detect if PG is supported by HW + * + * @dev: the device structure + * + * returns: true is pg supported, false otherwise + */ +static bool mei_txe_pg_is_enabled(struct mei_device *dev) +{ + return true; +} + +/** + * mei_txe_pg_state - translate aliveness register value + * to the mei power gating state + * + * @dev: the device structure + * + * returns: MEI_PG_OFF if aliveness is on and MEI_PG_ON otherwise + */ +static inline enum mei_pg_state mei_txe_pg_state(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + return hw->aliveness ? MEI_PG_OFF : MEI_PG_ON; +} + +/** + * mei_txe_input_ready_interrupt_enable - sets the Input Ready Interrupt + * + * @dev: the device structure + */ +static void mei_txe_input_ready_interrupt_enable(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + u32 hintmsk; + /* Enable the SEC_IPC_HOST_INT_MASK_IN_RDY interrupt */ + hintmsk = mei_txe_sec_reg_read(hw, SEC_IPC_HOST_INT_MASK_REG); + hintmsk |= SEC_IPC_HOST_INT_MASK_IN_RDY; + mei_txe_sec_reg_write(hw, SEC_IPC_HOST_INT_MASK_REG, hintmsk); +} + +/** + * mei_txe_input_doorbell_set + * - Sets bit 0 in SEC_IPC_INPUT_DOORBELL.IPC_INPUT_DOORBELL. + * @dev: the device structure + */ +static void mei_txe_input_doorbell_set(struct mei_txe_hw *hw) +{ + /* Clear the interrupt cause */ + clear_bit(TXE_INTR_IN_READY_BIT, &hw->intr_cause); + mei_txe_sec_reg_write(hw, SEC_IPC_INPUT_DOORBELL_REG, 1); +} + +/** + * mei_txe_output_ready_set - Sets the SICR_SEC_IPC_OUTPUT_STATUS bit to 1 + * + * @dev: the device structure + */ +static void mei_txe_output_ready_set(struct mei_txe_hw *hw) +{ + mei_txe_br_reg_write(hw, + SICR_SEC_IPC_OUTPUT_STATUS_REG, + SEC_IPC_OUTPUT_STATUS_RDY); +} + +/** + * mei_txe_is_input_ready - check if TXE is ready for receiving data + * + * @dev: the device structure + */ +static bool mei_txe_is_input_ready(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + u32 status; + status = mei_txe_sec_reg_read(hw, SEC_IPC_INPUT_STATUS_REG); + return !!(SEC_IPC_INPUT_STATUS_RDY & status); +} + +/** + * mei_txe_intr_clear - clear all interrupts + * + * @dev: the device structure + */ +static inline void mei_txe_intr_clear(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + mei_txe_sec_reg_write_silent(hw, SEC_IPC_HOST_INT_STATUS_REG, + SEC_IPC_HOST_INT_STATUS_PENDING); + mei_txe_br_reg_write(hw, HISR_REG, HISR_INT_STS_MSK); + mei_txe_br_reg_write(hw, HHISR_REG, IPC_HHIER_MSK); +} + +/** + * mei_txe_intr_disable - disable all interrupts + * + * @dev: the device structure + */ +static void mei_txe_intr_disable(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + mei_txe_br_reg_write(hw, HHIER_REG, 0); + mei_txe_br_reg_write(hw, HIER_REG, 0); +} +/** + * mei_txe_intr_disable - enable all interrupts + * + * @dev: the device structure + */ +static void mei_txe_intr_enable(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + mei_txe_br_reg_write(hw, HHIER_REG, IPC_HHIER_MSK); + mei_txe_br_reg_write(hw, HIER_REG, HIER_INT_EN_MSK); +} + +/** + * mei_txe_pending_interrupts - check if there are pending interrupts + * only Aliveness, Input ready, and output doorbell are of relevance + * + * @dev: the device structure + * + * Checks if there are pending interrupts + * only Aliveness, Readiness, Input ready, and Output doorbell are relevant + */ +static bool mei_txe_pending_interrupts(struct mei_device *dev) +{ + + struct mei_txe_hw *hw = to_txe_hw(dev); + bool ret = (hw->intr_cause & (TXE_INTR_READINESS | + TXE_INTR_ALIVENESS | + TXE_INTR_IN_READY | + TXE_INTR_OUT_DB)); + + if (ret) { + dev_dbg(&dev->pdev->dev, + "Pending Interrupts InReady=%01d Readiness=%01d, Aliveness=%01d, OutDoor=%01d\n", + !!(hw->intr_cause & TXE_INTR_IN_READY), + !!(hw->intr_cause & TXE_INTR_READINESS), + !!(hw->intr_cause & TXE_INTR_ALIVENESS), + !!(hw->intr_cause & TXE_INTR_OUT_DB)); + } + return ret; +} + +/** + * mei_txe_input_payload_write - write a dword to the host buffer + * at offset idx + * + * @dev: the device structure + * @idx: index in the host buffer + * @value: value + */ +static void mei_txe_input_payload_write(struct mei_device *dev, + unsigned long idx, u32 value) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + mei_txe_sec_reg_write(hw, SEC_IPC_INPUT_PAYLOAD_REG + + (idx * sizeof(u32)), value); +} + +/** + * mei_txe_out_data_read - read dword from the device buffer + * at offset idx + * + * @dev: the device structure + * @idx: index in the device buffer + * + * returns register value at index + */ +static u32 mei_txe_out_data_read(const struct mei_device *dev, + unsigned long idx) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + return mei_txe_br_reg_read(hw, + BRIDGE_IPC_OUTPUT_PAYLOAD_REG + (idx * sizeof(u32))); +} + +/* Readiness */ + +/** + * mei_txe_readiness_set_host_rdy + * + * @dev: the device structure + */ +static void mei_txe_readiness_set_host_rdy(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + mei_txe_br_reg_write(hw, + SICR_HOST_IPC_READINESS_REQ_REG, + SICR_HOST_IPC_READINESS_HOST_RDY); +} + +/** + * mei_txe_readiness_clear + * + * @dev: the device structure + */ +static void mei_txe_readiness_clear(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + mei_txe_br_reg_write(hw, SICR_HOST_IPC_READINESS_REQ_REG, + SICR_HOST_IPC_READINESS_RDY_CLR); +} +/** + * mei_txe_readiness_get - Reads and returns + * the HICR_SEC_IPC_READINESS register value + * + * @dev: the device structure + */ +static u32 mei_txe_readiness_get(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + return mei_txe_br_reg_read(hw, HICR_SEC_IPC_READINESS_REG); +} + + +/** + * mei_txe_readiness_is_sec_rdy - check readiness + * for HICR_SEC_IPC_READINESS_SEC_RDY + * + * @readiness - cached readiness state + */ +static inline bool mei_txe_readiness_is_sec_rdy(u32 readiness) +{ + return !!(readiness & HICR_SEC_IPC_READINESS_SEC_RDY); +} + +/** + * mei_txe_hw_is_ready - check if the hw is ready + * + * @dev: the device structure + */ +static bool mei_txe_hw_is_ready(struct mei_device *dev) +{ + u32 readiness = mei_txe_readiness_get(dev); + return mei_txe_readiness_is_sec_rdy(readiness); +} + +/** + * mei_txe_host_is_ready - check if the host is ready + * + * @dev: the device structure + */ +static inline bool mei_txe_host_is_ready(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + u32 reg = mei_txe_br_reg_read(hw, HICR_SEC_IPC_READINESS_REG); + return !!(reg & HICR_SEC_IPC_READINESS_HOST_RDY); +} + +/** + * mei_txe_readiness_wait - wait till readiness settles + * + * @dev: the device structure + * + * returns 0 on success and -ETIME on timeout + */ +static int mei_txe_readiness_wait(struct mei_device *dev) +{ + if (mei_txe_hw_is_ready(dev)) + return 0; + + mutex_unlock(&dev->device_lock); + wait_event_timeout(dev->wait_hw_ready, dev->recvd_hw_ready, + msecs_to_jiffies(SEC_RESET_WAIT_TIMEOUT)); + mutex_lock(&dev->device_lock); + if (!dev->recvd_hw_ready) { + dev_err(&dev->pdev->dev, "wait for readiness failed\n"); + return -ETIME; + } + + dev->recvd_hw_ready = false; + return 0; +} + +/** + * mei_txe_hw_config - configure hardware at the start of the devices + * + * @dev: the device structure + * + * Configure hardware at the start of the device should be done only + * once at the device probe time + */ +static void mei_txe_hw_config(struct mei_device *dev) +{ + + struct mei_txe_hw *hw = to_txe_hw(dev); + /* Doesn't change in runtime */ + dev->hbuf_depth = PAYLOAD_SIZE / 4; + + hw->aliveness = mei_txe_aliveness_get(dev); + hw->readiness = mei_txe_readiness_get(dev); + + dev_dbg(&dev->pdev->dev, "aliveness_resp = 0x%08x, readiness = 0x%08x.\n", + hw->aliveness, hw->readiness); +} + + +/** + * mei_txe_write - writes a message to device. + * + * @dev: the device structure + * @header: header of message + * @buf: message buffer will be written + * returns 1 if success, 0 - otherwise. + */ + +static int mei_txe_write(struct mei_device *dev, + struct mei_msg_hdr *header, unsigned char *buf) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + unsigned long rem; + unsigned long length; + int slots = dev->hbuf_depth; + u32 *reg_buf = (u32 *)buf; + u32 dw_cnt; + int i; + + if (WARN_ON(!header || !buf)) + return -EINVAL; + + length = header->length; + + dev_dbg(&dev->pdev->dev, MEI_HDR_FMT, MEI_HDR_PRM(header)); + + dw_cnt = mei_data2slots(length); + if (dw_cnt > slots) + return -EMSGSIZE; + + if (WARN(!hw->aliveness, "txe write: aliveness not asserted\n")) + return -EAGAIN; + + /* Enable Input Ready Interrupt. */ + mei_txe_input_ready_interrupt_enable(dev); + + if (!mei_txe_is_input_ready(dev)) { + struct mei_fw_status fw_status; + mei_fw_status(dev, &fw_status); + dev_err(&dev->pdev->dev, "Input is not ready " FW_STS_FMT "\n", + FW_STS_PRM(fw_status)); + return -EAGAIN; + } + + mei_txe_input_payload_write(dev, 0, *((u32 *)header)); + + for (i = 0; i < length / 4; i++) + mei_txe_input_payload_write(dev, i + 1, reg_buf[i]); + + rem = length & 0x3; + if (rem > 0) { + u32 reg = 0; + memcpy(®, &buf[length - rem], rem); + mei_txe_input_payload_write(dev, i + 1, reg); + } + + /* after each write the whole buffer is consumed */ + hw->slots = 0; + + /* Set Input-Doorbell */ + mei_txe_input_doorbell_set(hw); + + return 0; +} + +/** + * mei_txe_hbuf_max_len - mimics the me hbuf circular buffer + * + * @dev: the device structure + * + * returns the PAYLOAD_SIZE - 4 + */ +static size_t mei_txe_hbuf_max_len(const struct mei_device *dev) +{ + return PAYLOAD_SIZE - sizeof(struct mei_msg_hdr); +} + +/** + * mei_txe_hbuf_empty_slots - mimics the me hbuf circular buffer + * + * @dev: the device structure + * + * returns always hbuf_depth + */ +static int mei_txe_hbuf_empty_slots(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + return hw->slots; +} + +/** + * mei_txe_count_full_read_slots - mimics the me device circular buffer + * + * @dev: the device structure + * + * returns always buffer size in dwords count + */ +static int mei_txe_count_full_read_slots(struct mei_device *dev) +{ + /* read buffers has static size */ + return PAYLOAD_SIZE / 4; +} + +/** + * mei_txe_read_hdr - read message header which is always in 4 first bytes + * + * @dev: the device structure + * + * returns mei message header + */ + +static u32 mei_txe_read_hdr(const struct mei_device *dev) +{ + return mei_txe_out_data_read(dev, 0); +} +/** + * mei_txe_read - reads a message from the txe device. + * + * @dev: the device structure + * @buf: message buffer will be written + * @len: message size will be read + * + * returns -EINVAL on error wrong argument and 0 on success + */ +static int mei_txe_read(struct mei_device *dev, + unsigned char *buf, unsigned long len) +{ + + struct mei_txe_hw *hw = to_txe_hw(dev); + u32 i; + u32 *reg_buf = (u32 *)buf; + u32 rem = len & 0x3; + + if (WARN_ON(!buf || !len)) + return -EINVAL; + + dev_dbg(&dev->pdev->dev, + "buffer-length = %lu buf[0]0x%08X\n", + len, mei_txe_out_data_read(dev, 0)); + + for (i = 0; i < len / 4; i++) { + /* skip header: index starts from 1 */ + u32 reg = mei_txe_out_data_read(dev, i + 1); + dev_dbg(&dev->pdev->dev, "buf[%d] = 0x%08X\n", i, reg); + *reg_buf++ = reg; + } + + if (rem) { + u32 reg = mei_txe_out_data_read(dev, i + 1); + memcpy(reg_buf, ®, rem); + } + + mei_txe_output_ready_set(hw); + return 0; +} + +/** + * mei_txe_hw_reset - resets host and fw. + * + * @dev: the device structure + * @intr_enable: if interrupt should be enabled after reset. + * + * returns 0 on success and < 0 in case of error + */ +static int mei_txe_hw_reset(struct mei_device *dev, bool intr_enable) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + + u32 aliveness_req; + /* + * read input doorbell to ensure consistency between Bridge and SeC + * return value might be garbage return + */ + (void)mei_txe_sec_reg_read_silent(hw, SEC_IPC_INPUT_DOORBELL_REG); + + aliveness_req = mei_txe_aliveness_req_get(dev); + hw->aliveness = mei_txe_aliveness_get(dev); + + /* Disable interrupts in this stage we will poll */ + mei_txe_intr_disable(dev); + + /* + * If Aliveness Request and Aliveness Response are not equal then + * wait for them to be equal + * Since we might have interrupts disabled - poll for it + */ + if (aliveness_req != hw->aliveness) + if (mei_txe_aliveness_poll(dev, aliveness_req) < 0) { + dev_err(&dev->pdev->dev, + "wait for aliveness settle failed ... bailing out\n"); + return -EIO; + } + + /* + * If Aliveness Request and Aliveness Response are set then clear them + */ + if (aliveness_req) { + mei_txe_aliveness_set(dev, 0); + if (mei_txe_aliveness_poll(dev, 0) < 0) { + dev_err(&dev->pdev->dev, + "wait for aliveness failed ... bailing out\n"); + return -EIO; + } + } + + /* + * Set rediness RDY_CLR bit + */ + mei_txe_readiness_clear(dev); + + return 0; +} + +/** + * mei_txe_hw_start - start the hardware after reset + * + * @dev: the device structure + * + * returns 0 on success and < 0 in case of error + */ +static int mei_txe_hw_start(struct mei_device *dev) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + int ret; + + u32 hisr; + + /* bring back interrupts */ + mei_txe_intr_enable(dev); + + ret = mei_txe_readiness_wait(dev); + if (ret < 0) { + dev_err(&dev->pdev->dev, "wating for readiness failed\n"); + return ret; + } + + /* + * If HISR.INT2_STS interrupt status bit is set then clear it. + */ + hisr = mei_txe_br_reg_read(hw, HISR_REG); + if (hisr & HISR_INT_2_STS) + mei_txe_br_reg_write(hw, HISR_REG, HISR_INT_2_STS); + + /* Clear the interrupt cause of OutputDoorbell */ + clear_bit(TXE_INTR_OUT_DB_BIT, &hw->intr_cause); + + ret = mei_txe_aliveness_set_sync(dev, 1); + if (ret < 0) { + dev_err(&dev->pdev->dev, "wait for aliveness failed ... bailing out\n"); + return ret; + } + + /* enable input ready interrupts: + * SEC_IPC_HOST_INT_MASK.IPC_INPUT_READY_INT_MASK + */ + mei_txe_input_ready_interrupt_enable(dev); + + + /* Set the SICR_SEC_IPC_OUTPUT_STATUS.IPC_OUTPUT_READY bit */ + mei_txe_output_ready_set(hw); + + /* Set bit SICR_HOST_IPC_READINESS.HOST_RDY + */ + mei_txe_readiness_set_host_rdy(dev); + + return 0; +} + +/** + * mei_txe_check_and_ack_intrs - translate multi BAR interrupt into + * single bit mask and acknowledge the interrupts + * + * @dev: the device structure + * @do_ack: acknowledge interrupts + */ +static bool mei_txe_check_and_ack_intrs(struct mei_device *dev, bool do_ack) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + u32 hisr; + u32 hhisr; + u32 ipc_isr; + u32 aliveness; + bool generated; + + /* read interrupt registers */ + hhisr = mei_txe_br_reg_read(hw, HHISR_REG); + generated = (hhisr & IPC_HHIER_MSK); + if (!generated) + goto out; + + hisr = mei_txe_br_reg_read(hw, HISR_REG); + + aliveness = mei_txe_aliveness_get(dev); + if (hhisr & IPC_HHIER_SEC && aliveness) + ipc_isr = mei_txe_sec_reg_read_silent(hw, + SEC_IPC_HOST_INT_STATUS_REG); + else + ipc_isr = 0; + + generated = generated || + (hisr & HISR_INT_STS_MSK) || + (ipc_isr & SEC_IPC_HOST_INT_STATUS_PENDING); + + if (generated && do_ack) { + /* Save the interrupt causes */ + hw->intr_cause |= hisr & HISR_INT_STS_MSK; + if (ipc_isr & SEC_IPC_HOST_INT_STATUS_IN_RDY) + hw->intr_cause |= TXE_INTR_IN_READY; + + + mei_txe_intr_disable(dev); + /* Clear the interrupts in hierarchy: + * IPC and Bridge, than the High Level */ + mei_txe_sec_reg_write_silent(hw, + SEC_IPC_HOST_INT_STATUS_REG, ipc_isr); + mei_txe_br_reg_write(hw, HISR_REG, hisr); + mei_txe_br_reg_write(hw, HHISR_REG, hhisr); + } + +out: + return generated; +} + +/** + * mei_txe_irq_quick_handler - The ISR of the MEI device + * + * @irq: The irq number + * @dev_id: pointer to the device structure + * + * returns irqreturn_t + */ +irqreturn_t mei_txe_irq_quick_handler(int irq, void *dev_id) +{ + struct mei_device *dev = dev_id; + + if (mei_txe_check_and_ack_intrs(dev, true)) + return IRQ_WAKE_THREAD; + return IRQ_NONE; +} + + +/** + * mei_txe_irq_thread_handler - txe interrupt thread + * + * @irq: The irq number + * @dev_id: pointer to the device structure + * + * returns irqreturn_t + * + */ +irqreturn_t mei_txe_irq_thread_handler(int irq, void *dev_id) +{ + struct mei_device *dev = (struct mei_device *) dev_id; + struct mei_txe_hw *hw = to_txe_hw(dev); + struct mei_cl_cb complete_list; + s32 slots; + int rets = 0; + + dev_dbg(&dev->pdev->dev, "irq thread: Interrupt Registers HHISR|HISR|SEC=%02X|%04X|%02X\n", + mei_txe_br_reg_read(hw, HHISR_REG), + mei_txe_br_reg_read(hw, HISR_REG), + mei_txe_sec_reg_read_silent(hw, SEC_IPC_HOST_INT_STATUS_REG)); + + + /* initialize our complete list */ + mutex_lock(&dev->device_lock); + mei_io_list_init(&complete_list); + + if (pci_dev_msi_enabled(dev->pdev)) + mei_txe_check_and_ack_intrs(dev, true); + + /* show irq events */ + mei_txe_pending_interrupts(dev); + + hw->aliveness = mei_txe_aliveness_get(dev); + hw->readiness = mei_txe_readiness_get(dev); + + /* Readiness: + * Detection of TXE driver going through reset + * or TXE driver resetting the HECI interface. + */ + if (test_and_clear_bit(TXE_INTR_READINESS_BIT, &hw->intr_cause)) { + dev_dbg(&dev->pdev->dev, "Readiness Interrupt was received...\n"); + + /* Check if SeC is going through reset */ + if (mei_txe_readiness_is_sec_rdy(hw->readiness)) { + dev_dbg(&dev->pdev->dev, "we need to start the dev.\n"); + dev->recvd_hw_ready = true; + } else { + dev->recvd_hw_ready = false; + if (dev->dev_state != MEI_DEV_RESETTING) { + + dev_warn(&dev->pdev->dev, "FW not ready: resetting.\n"); + schedule_work(&dev->reset_work); + goto end; + + } + } + wake_up(&dev->wait_hw_ready); + } + + /************************************************************/ + /* Check interrupt cause: + * Aliveness: Detection of SeC acknowledge of host request that + * it remain alive or host cancellation of that request. + */ + + if (test_and_clear_bit(TXE_INTR_ALIVENESS_BIT, &hw->intr_cause)) { + /* Clear the interrupt cause */ + dev_dbg(&dev->pdev->dev, + "Aliveness Interrupt: Status: %d\n", hw->aliveness); + dev->pg_event = MEI_PG_EVENT_RECEIVED; + if (waitqueue_active(&hw->wait_aliveness_resp)) + wake_up(&hw->wait_aliveness_resp); + } + + + /* Output Doorbell: + * Detection of SeC having sent output to host + */ + slots = mei_count_full_read_slots(dev); + if (test_and_clear_bit(TXE_INTR_OUT_DB_BIT, &hw->intr_cause)) { + /* Read from TXE */ + rets = mei_irq_read_handler(dev, &complete_list, &slots); + if (rets && dev->dev_state != MEI_DEV_RESETTING) { + dev_err(&dev->pdev->dev, + "mei_irq_read_handler ret = %d.\n", rets); + + schedule_work(&dev->reset_work); + goto end; + } + } + /* Input Ready: Detection if host can write to SeC */ + if (test_and_clear_bit(TXE_INTR_IN_READY_BIT, &hw->intr_cause)) { + dev->hbuf_is_ready = true; + hw->slots = dev->hbuf_depth; + } + + if (hw->aliveness && dev->hbuf_is_ready) { + /* get the real register value */ + dev->hbuf_is_ready = mei_hbuf_is_ready(dev); + rets = mei_irq_write_handler(dev, &complete_list); + if (rets && rets != -EMSGSIZE) + dev_err(&dev->pdev->dev, "mei_irq_write_handler ret = %d.\n", + rets); + dev->hbuf_is_ready = mei_hbuf_is_ready(dev); + } + + mei_irq_compl_handler(dev, &complete_list); + +end: + dev_dbg(&dev->pdev->dev, "interrupt thread end ret = %d\n", rets); + + mutex_unlock(&dev->device_lock); + + mei_enable_interrupts(dev); + return IRQ_HANDLED; +} + + +/** + * mei_txe_fw_status - retrieve fw status from the pci config space + * + * @dev: the device structure + * @fw_status: fw status registers storage + * + * returns: 0 on success an error code otherwise + */ +static int mei_txe_fw_status(struct mei_device *dev, + struct mei_fw_status *fw_status) +{ + const u32 pci_cfg_reg[] = {PCI_CFG_TXE_FW_STS0, PCI_CFG_TXE_FW_STS1}; + int i; + + if (!fw_status) + return -EINVAL; + + fw_status->count = 2; + + for (i = 0; i < fw_status->count && i < MEI_FW_STATUS_MAX; i++) { + int ret; + ret = pci_read_config_dword(dev->pdev, + pci_cfg_reg[i], &fw_status->status[i]); + if (ret) + return ret; + } + + return 0; +} + +static const struct mei_hw_ops mei_txe_hw_ops = { + + .fw_status = mei_txe_fw_status, + .host_is_ready = mei_txe_host_is_ready, + + .pg_state = mei_txe_pg_state, + + .hw_is_ready = mei_txe_hw_is_ready, + .hw_reset = mei_txe_hw_reset, + .hw_config = mei_txe_hw_config, + .hw_start = mei_txe_hw_start, + + .pg_is_enabled = mei_txe_pg_is_enabled, + + .intr_clear = mei_txe_intr_clear, + .intr_enable = mei_txe_intr_enable, + .intr_disable = mei_txe_intr_disable, + + .hbuf_free_slots = mei_txe_hbuf_empty_slots, + .hbuf_is_ready = mei_txe_is_input_ready, + .hbuf_max_len = mei_txe_hbuf_max_len, + + .write = mei_txe_write, + + .rdbuf_full_slots = mei_txe_count_full_read_slots, + .read_hdr = mei_txe_read_hdr, + + .read = mei_txe_read, + +}; + +#define MEI_CFG_TXE_FW_STS \ + .fw_status.count = 2, \ + .fw_status.status[0] = PCI_CFG_TXE_FW_STS0, \ + .fw_status.status[1] = PCI_CFG_TXE_FW_STS1 + +const struct mei_cfg mei_txe_cfg = { + MEI_CFG_TXE_FW_STS, +}; + + +/** + * mei_txe_dev_init - allocates and initializes txe hardware specific structure + * + * @pdev - pci device + * @cfg - per device generation config + * + * returns struct mei_device * on success or NULL; + * + */ +struct mei_device *mei_txe_dev_init(struct pci_dev *pdev, + const struct mei_cfg *cfg) +{ + struct mei_device *dev; + struct mei_txe_hw *hw; + + dev = kzalloc(sizeof(struct mei_device) + + sizeof(struct mei_txe_hw), GFP_KERNEL); + if (!dev) + return NULL; + + mei_device_init(dev, cfg); + + hw = to_txe_hw(dev); + + init_waitqueue_head(&hw->wait_aliveness_resp); + + dev->ops = &mei_txe_hw_ops; + + dev->pdev = pdev; + return dev; +} + +/** + * mei_txe_setup_satt2 - SATT2 configuration for DMA support. + * + * @dev: the device structure + * @addr: physical address start of the range + * @range: physical range size + */ +int mei_txe_setup_satt2(struct mei_device *dev, phys_addr_t addr, u32 range) +{ + struct mei_txe_hw *hw = to_txe_hw(dev); + + u32 lo32 = lower_32_bits(addr); + u32 hi32 = upper_32_bits(addr); + u32 ctrl; + + /* SATT is limited to 36 Bits */ + if (hi32 & ~0xF) + return -EINVAL; + + /* SATT has to be 16Byte aligned */ + if (lo32 & 0xF) + return -EINVAL; + + /* SATT range has to be 4Bytes aligned */ + if (range & 0x4) + return -EINVAL; + + /* SATT is limited to 32 MB range*/ + if (range > SATT_RANGE_MAX) + return -EINVAL; + + ctrl = SATT2_CTRL_VALID_MSK; + ctrl |= hi32 << SATT2_CTRL_BR_BASE_ADDR_REG_SHIFT; + + mei_txe_br_reg_write(hw, SATT2_SAP_SIZE_REG, range); + mei_txe_br_reg_write(hw, SATT2_BRG_BA_LSB_REG, lo32); + mei_txe_br_reg_write(hw, SATT2_CTRL_REG, ctrl); + dev_dbg(&dev->pdev->dev, "SATT2: SAP_SIZE_OFFSET=0x%08X, BRG_BA_LSB_OFFSET=0x%08X, CTRL_OFFSET=0x%08X\n", + range, lo32, ctrl); + + return 0; +} diff --git a/drivers/misc/mei/hw-txe.h b/drivers/misc/mei/hw-txe.h new file mode 100644 index 00000000000..e244af79167 --- /dev/null +++ b/drivers/misc/mei/hw-txe.h @@ -0,0 +1,77 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2013-2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#ifndef _MEI_HW_TXE_H_ +#define _MEI_HW_TXE_H_ + +#include <linux/irqreturn.h> + +#include "hw.h" +#include "hw-txe-regs.h" + +#define MEI_TXI_RPM_TIMEOUT 500 /* ms */ + +/* Flatten Hierarchy interrupt cause */ +#define TXE_INTR_READINESS_BIT 0 /* HISR_INT_0_STS */ +#define TXE_INTR_READINESS HISR_INT_0_STS +#define TXE_INTR_ALIVENESS_BIT 1 /* HISR_INT_1_STS */ +#define TXE_INTR_ALIVENESS HISR_INT_1_STS +#define TXE_INTR_OUT_DB_BIT 2 /* HISR_INT_2_STS */ +#define TXE_INTR_OUT_DB HISR_INT_2_STS +#define TXE_INTR_IN_READY_BIT 8 /* beyond HISR */ +#define TXE_INTR_IN_READY BIT(8) + +/** + * struct mei_txe_hw - txe hardware specifics + * + * @mem_addr: SeC and BRIDGE bars + * @aliveness: aliveness (power gating) state of the hardware + * @readiness: readiness state of the hardware + * @wait_aliveness_resp: aliveness wait queue + * @intr_cause: translated interrupt cause + */ +struct mei_txe_hw { + void __iomem *mem_addr[NUM_OF_MEM_BARS]; + u32 aliveness; + u32 readiness; + u32 slots; + + wait_queue_head_t wait_aliveness_resp; + + unsigned long intr_cause; +}; + +#define to_txe_hw(dev) (struct mei_txe_hw *)((dev)->hw) + +static inline struct mei_device *hw_txe_to_mei(struct mei_txe_hw *hw) +{ + return container_of((void *)hw, struct mei_device, hw); +} + +extern const struct mei_cfg mei_txe_cfg; + +struct mei_device *mei_txe_dev_init(struct pci_dev *pdev, + const struct mei_cfg *cfg); + +irqreturn_t mei_txe_irq_quick_handler(int irq, void *dev_id); +irqreturn_t mei_txe_irq_thread_handler(int irq, void *dev_id); + +int mei_txe_aliveness_set_sync(struct mei_device *dev, u32 req); + +int mei_txe_setup_satt2(struct mei_device *dev, phys_addr_t addr, u32 range); + + +#endif /* _MEI_HW_TXE_H_ */ diff --git a/drivers/misc/mei/hw.h b/drivers/misc/mei/hw.h index 9700532f02f..dd448e58cc8 100644 --- a/drivers/misc/mei/hw.h +++ b/drivers/misc/mei/hw.h @@ -20,127 +20,31 @@ #include <linux/uuid.h> /* - * Timeouts + * Timeouts in Seconds */ -#define MEI_INTEROP_TIMEOUT (HZ * 7) -#define MEI_CONNECT_TIMEOUT 3 /* at least 2 seconds */ +#define MEI_HW_READY_TIMEOUT 2 /* Timeout on ready message */ +#define MEI_CONNECT_TIMEOUT 3 /* HPS: at least 2 seconds */ -#define CONNECT_TIMEOUT 15 /* HPS definition */ -#define INIT_CLIENTS_TIMEOUT 15 /* HPS definition */ +#define MEI_CL_CONNECT_TIMEOUT 15 /* HPS: Client Connect Timeout */ +#define MEI_CLIENTS_INIT_TIMEOUT 15 /* HPS: Clients Enumeration Timeout */ -#define IAMTHIF_STALL_TIMER 12 /* seconds */ -#define IAMTHIF_READ_TIMER 10000 /* ms */ +#define MEI_IAMTHIF_STALL_TIMER 12 /* HPS */ +#define MEI_IAMTHIF_READ_TIMER 10 /* HPS */ -/* - * Internal Clients Number - */ -#define MEI_WD_HOST_CLIENT_ID 1 -#define MEI_IAMTHIF_HOST_CLIENT_ID 2 +#define MEI_PGI_TIMEOUT 1 /* PG Isolation time response 1 sec */ +#define MEI_HBM_TIMEOUT 1 /* 1 second */ /* - * MEI device IDs - */ -#define MEI_DEV_ID_82946GZ 0x2974 /* 82946GZ/GL */ -#define MEI_DEV_ID_82G35 0x2984 /* 82G35 Express */ -#define MEI_DEV_ID_82Q965 0x2994 /* 82Q963/Q965 */ -#define MEI_DEV_ID_82G965 0x29A4 /* 82P965/G965 */ - -#define MEI_DEV_ID_82GM965 0x2A04 /* Mobile PM965/GM965 */ -#define MEI_DEV_ID_82GME965 0x2A14 /* Mobile GME965/GLE960 */ - -#define MEI_DEV_ID_ICH9_82Q35 0x29B4 /* 82Q35 Express */ -#define MEI_DEV_ID_ICH9_82G33 0x29C4 /* 82G33/G31/P35/P31 Express */ -#define MEI_DEV_ID_ICH9_82Q33 0x29D4 /* 82Q33 Express */ -#define MEI_DEV_ID_ICH9_82X38 0x29E4 /* 82X38/X48 Express */ -#define MEI_DEV_ID_ICH9_3200 0x29F4 /* 3200/3210 Server */ - -#define MEI_DEV_ID_ICH9_6 0x28B4 /* Bearlake */ -#define MEI_DEV_ID_ICH9_7 0x28C4 /* Bearlake */ -#define MEI_DEV_ID_ICH9_8 0x28D4 /* Bearlake */ -#define MEI_DEV_ID_ICH9_9 0x28E4 /* Bearlake */ -#define MEI_DEV_ID_ICH9_10 0x28F4 /* Bearlake */ - -#define MEI_DEV_ID_ICH9M_1 0x2A44 /* Cantiga */ -#define MEI_DEV_ID_ICH9M_2 0x2A54 /* Cantiga */ -#define MEI_DEV_ID_ICH9M_3 0x2A64 /* Cantiga */ -#define MEI_DEV_ID_ICH9M_4 0x2A74 /* Cantiga */ - -#define MEI_DEV_ID_ICH10_1 0x2E04 /* Eaglelake */ -#define MEI_DEV_ID_ICH10_2 0x2E14 /* Eaglelake */ -#define MEI_DEV_ID_ICH10_3 0x2E24 /* Eaglelake */ -#define MEI_DEV_ID_ICH10_4 0x2E34 /* Eaglelake */ - -#define MEI_DEV_ID_IBXPK_1 0x3B64 /* Calpella */ -#define MEI_DEV_ID_IBXPK_2 0x3B65 /* Calpella */ - -#define MEI_DEV_ID_CPT_1 0x1C3A /* Couger Point */ -#define MEI_DEV_ID_PBG_1 0x1D3A /* C600/X79 Patsburg */ - -#define MEI_DEV_ID_PPT_1 0x1E3A /* Panther Point */ -#define MEI_DEV_ID_PPT_2 0x1CBA /* Panther Point */ -#define MEI_DEV_ID_PPT_3 0x1DBA /* Panther Point */ - -#define MEI_DEV_ID_LPT 0x8C3A /* Lynx Point */ -#define MEI_DEV_ID_LPT_LP 0x9C3A /* Lynx Point LP */ -/* - * MEI HW Section + * MEI Version */ - -/* MEI registers */ -/* H_CB_WW - Host Circular Buffer (CB) Write Window register */ -#define H_CB_WW 0 -/* H_CSR - Host Control Status register */ -#define H_CSR 4 -/* ME_CB_RW - ME Circular Buffer Read Window register (read only) */ -#define ME_CB_RW 8 -/* ME_CSR_HA - ME Control Status Host Access register (read only) */ -#define ME_CSR_HA 0xC - - -/* register bits of H_CSR (Host Control Status register) */ -/* Host Circular Buffer Depth - maximum number of 32-bit entries in CB */ -#define H_CBD 0xFF000000 -/* Host Circular Buffer Write Pointer */ -#define H_CBWP 0x00FF0000 -/* Host Circular Buffer Read Pointer */ -#define H_CBRP 0x0000FF00 -/* Host Reset */ -#define H_RST 0x00000010 -/* Host Ready */ -#define H_RDY 0x00000008 -/* Host Interrupt Generate */ -#define H_IG 0x00000004 -/* Host Interrupt Status */ -#define H_IS 0x00000002 -/* Host Interrupt Enable */ -#define H_IE 0x00000001 - - -/* register bits of ME_CSR_HA (ME Control Status Host Access register) */ -/* ME CB (Circular Buffer) Depth HRA (Host Read Access) - host read only -access to ME_CBD */ -#define ME_CBD_HRA 0xFF000000 -/* ME CB Write Pointer HRA - host read only access to ME_CBWP */ -#define ME_CBWP_HRA 0x00FF0000 -/* ME CB Read Pointer HRA - host read only access to ME_CBRP */ -#define ME_CBRP_HRA 0x0000FF00 -/* ME Reset HRA - host read only access to ME_RST */ -#define ME_RST_HRA 0x00000010 -/* ME Ready HRA - host read only access to ME_RDY */ -#define ME_RDY_HRA 0x00000008 -/* ME Interrupt Generate HRA - host read only access to ME_IG */ -#define ME_IG_HRA 0x00000004 -/* ME Interrupt Status HRA - host read only access to ME_IS */ -#define ME_IS_HRA 0x00000002 -/* ME Interrupt Enable HRA - host read only access to ME_IE */ -#define ME_IE_HRA 0x00000001 +#define HBM_MINOR_VERSION 1 +#define HBM_MAJOR_VERSION 1 /* - * MEI Version + * MEI version with PGI support */ -#define HBM_MINOR_VERSION 0 -#define HBM_MAJOR_VERSION 1 -#define HBM_TIMEOUT 1 /* 1 second */ +#define HBM_MINOR_VERSION_PGI 1 +#define HBM_MAJOR_VERSION_PGI 1 /* Host bus message command opcode */ #define MEI_HBM_CMD_OP_MSK 0x7f @@ -172,6 +76,11 @@ access to ME_CBD */ #define MEI_FLOW_CONTROL_CMD 0x08 +#define MEI_PG_ISOLATION_ENTRY_REQ_CMD 0x0a +#define MEI_PG_ISOLATION_ENTRY_RES_CMD 0x8a +#define MEI_PG_ISOLATION_EXIT_REQ_CMD 0x0b +#define MEI_PG_ISOLATION_EXIT_RES_CMD 0x8b + /* * MEI Stop Reason * used by hbm_host_stop_request.reason @@ -192,19 +101,19 @@ enum mei_stop_reason_types { * Client Connect Status * used by hbm_client_connect_response.status */ -enum client_connect_status_types { - CCS_SUCCESS = 0x00, - CCS_NOT_FOUND = 0x01, - CCS_ALREADY_STARTED = 0x02, - CCS_OUT_OF_RESOURCES = 0x03, - CCS_MESSAGE_SMALL = 0x04 +enum mei_cl_connect_status { + MEI_CL_CONN_SUCCESS = 0x00, + MEI_CL_CONN_NOT_FOUND = 0x01, + MEI_CL_CONN_ALREADY_STARTED = 0x02, + MEI_CL_CONN_OUT_OF_RESOURCES = 0x03, + MEI_CL_CONN_MESSAGE_SMALL = 0x04 }; /* * Client Disconnect Status */ -enum client_disconnect_status_types { - CDS_SUCCESS = 0x00 +enum mei_cl_disconnect_status { + MEI_CL_DISCONN_SUCCESS = 0x00 }; /* @@ -214,7 +123,8 @@ struct mei_msg_hdr { u32 me_addr:8; u32 host_addr:8; u32 length:9; - u32 reserved:6; + u32 reserved:5; + u32 internal:1; u32 msg_complete:1; } __packed; @@ -224,6 +134,22 @@ struct mei_bus_message { u8 data[0]; } __packed; +/** + * struct hbm_cl_cmd - client specific host bus command + * CONNECT, DISCONNECT, and FlOW CONTROL + * + * @hbm_cmd - bus message command header + * @me_addr - address of the client in ME + * @host_addr - address of the client in the driver + * @data + */ +struct mei_hbm_cl_cmd { + u8 hbm_cmd; + u8 me_addr; + u8 host_addr; + u8 data; +}; + struct hbm_version { u8 minor_version; u8 major_version; @@ -293,6 +219,25 @@ struct hbm_props_response { struct mei_client_properties client_properties; } __packed; +/** + * struct hbm_power_gate - power gate request/response + * + * @hbm_cmd - bus message command header + * @reserved[3] + */ +struct hbm_power_gate { + u8 hbm_cmd; + u8 reserved[3]; +} __packed; + +/** + * struct hbm_client_connect_request - connect/disconnect request + * + * @hbm_cmd - bus message command header + * @me_addr - address of the client in ME + * @host_addr - address of the client in the driver + * @reserved + */ struct hbm_client_connect_request { u8 hbm_cmd; u8 me_addr; @@ -300,6 +245,14 @@ struct hbm_client_connect_request { u8 reserved; } __packed; +/** + * struct hbm_client_connect_response - connect/disconnect response + * + * @hbm_cmd - bus message command header + * @me_addr - address of the client in ME + * @host_addr - address of the client in the driver + * @status - status of the request + */ struct hbm_client_connect_response { u8 hbm_cmd; u8 me_addr; @@ -307,12 +260,6 @@ struct hbm_client_connect_response { u8 status; } __packed; -struct hbm_client_disconnect_request { - u8 hbm_cmd; - u8 me_addr; - u8 host_addr; - u8 reserved[1]; -} __packed; #define MEI_FC_MESSAGE_RESERVED_LENGTH 5 @@ -323,11 +270,5 @@ struct hbm_flow_control { u8 reserved[MEI_FC_MESSAGE_RESERVED_LENGTH]; } __packed; -struct mei_me_client { - struct mei_client_properties props; - u8 client_id; - u8 mei_flow_ctrl_creds; -} __packed; - #endif diff --git a/drivers/misc/mei/init.c b/drivers/misc/mei/init.c index 98f1430e3e1..00692922248 100644 --- a/drivers/misc/mei/init.c +++ b/drivers/misc/mei/init.c @@ -14,16 +14,18 @@ * */ +#include <linux/export.h> #include <linux/pci.h> #include <linux/sched.h> #include <linux/wait.h> #include <linux/delay.h> -#include "mei_dev.h" -#include "hw.h" -#include "interface.h" #include <linux/mei.h> +#include "mei_dev.h" +#include "hbm.h" +#include "client.h" + const char *mei_dev_state_str(int state) { #define MEI_DEV_STATE(state) case MEI_DEV_##state: return #state @@ -31,726 +33,363 @@ const char *mei_dev_state_str(int state) MEI_DEV_STATE(INITIALIZING); MEI_DEV_STATE(INIT_CLIENTS); MEI_DEV_STATE(ENABLED); - MEI_DEV_STATE(RESETING); + MEI_DEV_STATE(RESETTING); MEI_DEV_STATE(DISABLED); - MEI_DEV_STATE(RECOVERING_FROM_RESET); MEI_DEV_STATE(POWER_DOWN); MEI_DEV_STATE(POWER_UP); default: - return "unkown"; + return "unknown"; } #undef MEI_DEV_STATE } -const uuid_le mei_amthi_guid = UUID_LE(0x12f80028, 0xb4b7, 0x4b2d, 0xac, - 0xa8, 0x46, 0xe0, 0xff, 0x65, - 0x81, 0x4c); - /** - * mei_io_list_init - Sets up a queue list. + * mei_cancel_work. Cancel mei background jobs * - * @list: An instance io list structure * @dev: the device structure - */ -void mei_io_list_init(struct mei_io_list *list) -{ - /* initialize our queue list */ - INIT_LIST_HEAD(&list->mei_cb.cb_list); -} - -/** - * mei_io_list_flush - removes list entry belonging to cl. - * - * @list: An instance of our list structure - * @cl: private data of the file object - */ -void mei_io_list_flush(struct mei_io_list *list, struct mei_cl *cl) -{ - struct mei_cl_cb *pos; - struct mei_cl_cb *next; - - list_for_each_entry_safe(pos, next, &list->mei_cb.cb_list, cb_list) { - if (pos->file_private) { - struct mei_cl *cl_tmp; - cl_tmp = (struct mei_cl *)pos->file_private; - if (mei_cl_cmp_id(cl, cl_tmp)) - list_del(&pos->cb_list); - } - } -} -/** - * mei_cl_flush_queues - flushes queue lists belonging to cl. * - * @dev: the device structure - * @cl: private data of the file object + * returns 0 on success or < 0 if the reset hasn't succeeded */ -int mei_cl_flush_queues(struct mei_cl *cl) +void mei_cancel_work(struct mei_device *dev) { - if (!cl || !cl->dev) - return -EINVAL; + cancel_work_sync(&dev->init_work); + cancel_work_sync(&dev->reset_work); - dev_dbg(&cl->dev->pdev->dev, "remove list entry belonging to cl\n"); - mei_io_list_flush(&cl->dev->read_list, cl); - mei_io_list_flush(&cl->dev->write_list, cl); - mei_io_list_flush(&cl->dev->write_waiting_list, cl); - mei_io_list_flush(&cl->dev->ctrl_wr_list, cl); - mei_io_list_flush(&cl->dev->ctrl_rd_list, cl); - mei_io_list_flush(&cl->dev->amthi_cmd_list, cl); - mei_io_list_flush(&cl->dev->amthi_read_complete_list, cl); - return 0; + cancel_delayed_work(&dev->timer_work); } - - +EXPORT_SYMBOL_GPL(mei_cancel_work); /** - * mei_reset_iamthif_params - initializes mei device iamthif + * mei_reset - resets host and fw. * * @dev: the device structure */ -static void mei_reset_iamthif_params(struct mei_device *dev) +int mei_reset(struct mei_device *dev) { - /* reset iamthif parameters. */ - dev->iamthif_current_cb = NULL; - dev->iamthif_msg_buf_size = 0; - dev->iamthif_msg_buf_index = 0; - dev->iamthif_canceled = false; - dev->iamthif_ioctl = false; - dev->iamthif_state = MEI_IAMTHIF_IDLE; - dev->iamthif_timer = 0; -} - -/** - * init_mei_device - allocates and initializes the mei device structure - * - * @pdev: The pci device structure - * - * returns The mei_device_device pointer on success, NULL on failure. - */ -struct mei_device *mei_device_init(struct pci_dev *pdev) -{ - struct mei_device *dev; + enum mei_dev_state state = dev->dev_state; + bool interrupts_enabled; + int ret; - dev = kzalloc(sizeof(struct mei_device), GFP_KERNEL); - if (!dev) - return NULL; + if (state != MEI_DEV_INITIALIZING && + state != MEI_DEV_DISABLED && + state != MEI_DEV_POWER_DOWN && + state != MEI_DEV_POWER_UP) { + struct mei_fw_status fw_status; + mei_fw_status(dev, &fw_status); + dev_warn(&dev->pdev->dev, + "unexpected reset: dev_state = %s " FW_STS_FMT "\n", + mei_dev_state_str(state), FW_STS_PRM(fw_status)); + } - /* setup our list array */ - INIT_LIST_HEAD(&dev->file_list); - INIT_LIST_HEAD(&dev->wd_cl.link); - INIT_LIST_HEAD(&dev->iamthif_cl.link); - mutex_init(&dev->device_lock); - init_waitqueue_head(&dev->wait_recvd_msg); - init_waitqueue_head(&dev->wait_stop_wd); - dev->dev_state = MEI_DEV_INITIALIZING; - dev->iamthif_state = MEI_IAMTHIF_IDLE; - dev->wd_interface_reg = false; + /* we're already in reset, cancel the init timer + * if the reset was called due the hbm protocol error + * we need to call it before hw start + * so the hbm watchdog won't kick in + */ + mei_hbm_idle(dev); + /* enter reset flow */ + interrupts_enabled = state != MEI_DEV_POWER_DOWN; + dev->dev_state = MEI_DEV_RESETTING; - mei_io_list_init(&dev->read_list); - mei_io_list_init(&dev->write_list); - mei_io_list_init(&dev->write_waiting_list); - mei_io_list_init(&dev->ctrl_wr_list); - mei_io_list_init(&dev->ctrl_rd_list); - mei_io_list_init(&dev->amthi_cmd_list); - mei_io_list_init(&dev->amthi_read_complete_list); - dev->pdev = pdev; - return dev; -} + dev->reset_count++; + if (dev->reset_count > MEI_MAX_CONSEC_RESET) { + dev_err(&dev->pdev->dev, "reset: reached maximal consecutive resets: disabling the device\n"); + dev->dev_state = MEI_DEV_DISABLED; + return -ENODEV; + } -/** - * mei_hw_init - initializes host and fw to start work. - * - * @dev: the device structure - * - * returns 0 on success, <0 on failure. - */ -int mei_hw_init(struct mei_device *dev) -{ - int err = 0; - int ret; + ret = mei_hw_reset(dev, interrupts_enabled); + /* fall through and remove the sw state even if hw reset has failed */ - mutex_lock(&dev->device_lock); + /* no need to clean up software state in case of power up */ + if (state != MEI_DEV_INITIALIZING && + state != MEI_DEV_POWER_UP) { - dev->host_hw_state = mei_hcsr_read(dev); - dev->me_hw_state = mei_mecsr_read(dev); - dev_dbg(&dev->pdev->dev, "host_hw_state = 0x%08x, mestate = 0x%08x.\n", - dev->host_hw_state, dev->me_hw_state); + /* remove all waiting requests */ + mei_cl_all_write_clear(dev); - /* acknowledge interrupt and stop interupts */ - if ((dev->host_hw_state & H_IS) == H_IS) - mei_reg_write(dev, H_CSR, dev->host_hw_state); + mei_cl_all_disconnect(dev); - /* Doesn't change in runtime */ - dev->hbuf_depth = (dev->host_hw_state & H_CBD) >> 24; + /* wake up all readers and writers so they can be interrupted */ + mei_cl_all_wakeup(dev); - dev->recvd_msg = false; - dev_dbg(&dev->pdev->dev, "reset in start the mei device.\n"); + /* remove entry if already in list */ + dev_dbg(&dev->pdev->dev, "remove iamthif and wd from the file list.\n"); + mei_cl_unlink(&dev->wd_cl); + mei_cl_unlink(&dev->iamthif_cl); + mei_amthif_reset_params(dev); + } - mei_reset(dev, 1); + mei_hbm_reset(dev); - dev_dbg(&dev->pdev->dev, "host_hw_state = 0x%08x, me_hw_state = 0x%08x.\n", - dev->host_hw_state, dev->me_hw_state); + dev->rd_msg_hdr = 0; + dev->wd_pending = false; - /* wait for ME to turn on ME_RDY */ - if (!dev->recvd_msg) { - mutex_unlock(&dev->device_lock); - err = wait_event_interruptible_timeout(dev->wait_recvd_msg, - dev->recvd_msg, MEI_INTEROP_TIMEOUT); - mutex_lock(&dev->device_lock); + if (ret) { + dev_err(&dev->pdev->dev, "hw_reset failed ret = %d\n", ret); + return ret; } - if (err <= 0 && !dev->recvd_msg) { + if (state == MEI_DEV_POWER_DOWN) { + dev_dbg(&dev->pdev->dev, "powering down: end of reset\n"); dev->dev_state = MEI_DEV_DISABLED; - dev_dbg(&dev->pdev->dev, - "wait_event_interruptible_timeout failed" - "on wait for ME to turn on ME_RDY.\n"); - ret = -ENODEV; - goto out; + return 0; } - if (!(((dev->host_hw_state & H_RDY) == H_RDY) && - ((dev->me_hw_state & ME_RDY_HRA) == ME_RDY_HRA))) { - dev->dev_state = MEI_DEV_DISABLED; - dev_dbg(&dev->pdev->dev, - "host_hw_state = 0x%08x, me_hw_state = 0x%08x.\n", - dev->host_hw_state, dev->me_hw_state); - - if (!(dev->host_hw_state & H_RDY)) - dev_dbg(&dev->pdev->dev, "host turn off H_RDY.\n"); - - if (!(dev->me_hw_state & ME_RDY_HRA)) - dev_dbg(&dev->pdev->dev, "ME turn off ME_RDY.\n"); - - dev_err(&dev->pdev->dev, "link layer initialization failed.\n"); - ret = -ENODEV; - goto out; + ret = mei_hw_start(dev); + if (ret) { + dev_err(&dev->pdev->dev, "hw_start failed ret = %d\n", ret); + return ret; } - if (dev->version.major_version != HBM_MAJOR_VERSION || - dev->version.minor_version != HBM_MINOR_VERSION) { - dev_dbg(&dev->pdev->dev, "MEI start failed.\n"); - ret = -ENODEV; - goto out; - } + dev_dbg(&dev->pdev->dev, "link is established start sending messages.\n"); - dev->recvd_msg = false; - dev_dbg(&dev->pdev->dev, "host_hw_state = 0x%08x, me_hw_state = 0x%08x.\n", - dev->host_hw_state, dev->me_hw_state); - dev_dbg(&dev->pdev->dev, "ME turn on ME_RDY and host turn on H_RDY.\n"); - dev_dbg(&dev->pdev->dev, "link layer has been established.\n"); - dev_dbg(&dev->pdev->dev, "MEI start success.\n"); - ret = 0; + dev->dev_state = MEI_DEV_INIT_CLIENTS; + ret = mei_hbm_start_req(dev); + if (ret) { + dev_err(&dev->pdev->dev, "hbm_start failed ret = %d\n", ret); + dev->dev_state = MEI_DEV_RESETTING; + return ret; + } -out: - mutex_unlock(&dev->device_lock); - return ret; + return 0; } +EXPORT_SYMBOL_GPL(mei_reset); /** - * mei_hw_reset - resets fw via mei csr register. + * mei_start - initializes host and fw to start work. * * @dev: the device structure - * @interrupts_enabled: if interrupt should be enabled after reset. - */ -static void mei_hw_reset(struct mei_device *dev, int interrupts_enabled) -{ - dev->host_hw_state |= (H_RST | H_IG); - - if (interrupts_enabled) - mei_enable_interrupts(dev); - else - mei_disable_interrupts(dev); -} - -/** - * mei_reset - resets host and fw. * - * @dev: the device structure - * @interrupts_enabled: if interrupt should be enabled after reset. + * returns 0 on success, <0 on failure. */ -void mei_reset(struct mei_device *dev, int interrupts_enabled) +int mei_start(struct mei_device *dev) { - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; - struct mei_cl_cb *cb_pos = NULL; - struct mei_cl_cb *cb_next = NULL; - bool unexpected; - - if (dev->dev_state == MEI_DEV_RECOVERING_FROM_RESET) { - dev->need_reset = true; - return; - } - - unexpected = (dev->dev_state != MEI_DEV_INITIALIZING && - dev->dev_state != MEI_DEV_DISABLED && - dev->dev_state != MEI_DEV_POWER_DOWN && - dev->dev_state != MEI_DEV_POWER_UP); - - dev->host_hw_state = mei_hcsr_read(dev); - - dev_dbg(&dev->pdev->dev, "before reset host_hw_state = 0x%08x.\n", - dev->host_hw_state); - - mei_hw_reset(dev, interrupts_enabled); - - dev->host_hw_state &= ~H_RST; - dev->host_hw_state |= H_IG; + int ret; + mutex_lock(&dev->device_lock); - mei_hcsr_set(dev); + /* acknowledge interrupt and stop interrupts */ + mei_clear_interrupts(dev); - dev_dbg(&dev->pdev->dev, "currently saved host_hw_state = 0x%08x.\n", - dev->host_hw_state); + mei_hw_config(dev); - dev->need_reset = false; + dev_dbg(&dev->pdev->dev, "reset in start the mei device.\n"); - if (dev->dev_state != MEI_DEV_INITIALIZING) { - if (dev->dev_state != MEI_DEV_DISABLED && - dev->dev_state != MEI_DEV_POWER_DOWN) - dev->dev_state = MEI_DEV_RESETING; + dev->reset_count = 0; + do { + dev->dev_state = MEI_DEV_INITIALIZING; + ret = mei_reset(dev); - list_for_each_entry_safe(cl_pos, - cl_next, &dev->file_list, link) { - cl_pos->state = MEI_FILE_DISCONNECTED; - cl_pos->mei_flow_ctrl_creds = 0; - cl_pos->read_cb = NULL; - cl_pos->timer_count = 0; + if (ret == -ENODEV || dev->dev_state == MEI_DEV_DISABLED) { + dev_err(&dev->pdev->dev, "reset failed ret = %d", ret); + goto err; } - /* remove entry if already in list */ - dev_dbg(&dev->pdev->dev, "list del iamthif and wd file list.\n"); - mei_remove_client_from_file_list(dev, - dev->wd_cl.host_client_id); - - mei_remove_client_from_file_list(dev, - dev->iamthif_cl.host_client_id); + } while (ret); - mei_reset_iamthif_params(dev); - dev->extra_write_index = 0; + /* we cannot start the device w/o hbm start message completed */ + if (dev->dev_state == MEI_DEV_DISABLED) { + dev_err(&dev->pdev->dev, "reset failed"); + goto err; } - dev->me_clients_num = 0; - dev->rd_msg_hdr = 0; - dev->wd_pending = false; - - /* update the state of the registers after reset */ - dev->host_hw_state = mei_hcsr_read(dev); - dev->me_hw_state = mei_mecsr_read(dev); - - dev_dbg(&dev->pdev->dev, "after reset host_hw_state = 0x%08x, me_hw_state = 0x%08x.\n", - dev->host_hw_state, dev->me_hw_state); - - if (unexpected) - dev_warn(&dev->pdev->dev, "unexpected reset: dev_state = %s\n", - mei_dev_state_str(dev->dev_state)); + if (mei_hbm_start_wait(dev)) { + dev_err(&dev->pdev->dev, "HBM haven't started"); + goto err; + } - /* Wake up all readings so they can be interrupted */ - list_for_each_entry_safe(cl_pos, cl_next, &dev->file_list, link) { - if (waitqueue_active(&cl_pos->rx_wait)) { - dev_dbg(&dev->pdev->dev, "Waking up client!\n"); - wake_up_interruptible(&cl_pos->rx_wait); - } + if (!mei_host_is_ready(dev)) { + dev_err(&dev->pdev->dev, "host is not ready.\n"); + goto err; } - /* remove all waiting requests */ - list_for_each_entry_safe(cb_pos, cb_next, - &dev->write_list.mei_cb.cb_list, cb_list) { - list_del(&cb_pos->cb_list); - mei_free_cb_private(cb_pos); + + if (!mei_hw_is_ready(dev)) { + dev_err(&dev->pdev->dev, "ME is not ready.\n"); + goto err; } -} + if (!mei_hbm_version_is_supported(dev)) { + dev_dbg(&dev->pdev->dev, "MEI start failed.\n"); + goto err; + } + dev_dbg(&dev->pdev->dev, "link layer has been established.\n"); -/** - * host_start_message - mei host sends start message. - * - * @dev: the device structure - * - * returns none. - */ -void mei_host_start_message(struct mei_device *dev) -{ - struct mei_msg_hdr *mei_hdr; - struct hbm_host_version_request *host_start_req; - - /* host start message */ - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = sizeof(struct hbm_host_version_request); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - host_start_req = - (struct hbm_host_version_request *) &dev->wr_msg_buf[1]; - memset(host_start_req, 0, sizeof(struct hbm_host_version_request)); - host_start_req->hbm_cmd = HOST_START_REQ_CMD; - host_start_req->host_version.major_version = HBM_MAJOR_VERSION; - host_start_req->host_version.minor_version = HBM_MINOR_VERSION; - dev->recvd_msg = false; - if (mei_write_message(dev, mei_hdr, (unsigned char *)host_start_req, - mei_hdr->length)) { - dev_dbg(&dev->pdev->dev, "write send version message to FW fail.\n"); - dev->dev_state = MEI_DEV_RESETING; - mei_reset(dev, 1); - } - dev->init_clients_state = MEI_START_MESSAGE; - dev->init_clients_timer = INIT_CLIENTS_TIMEOUT; - return ; + mutex_unlock(&dev->device_lock); + return 0; +err: + dev_err(&dev->pdev->dev, "link layer initialization failed.\n"); + dev->dev_state = MEI_DEV_DISABLED; + mutex_unlock(&dev->device_lock); + return -ENODEV; } +EXPORT_SYMBOL_GPL(mei_start); /** - * host_enum_clients_message - host sends enumeration client request message. + * mei_restart - restart device after suspend * * @dev: the device structure * - * returns none. + * returns 0 on success or -ENODEV if the restart hasn't succeeded */ -void mei_host_enum_clients_message(struct mei_device *dev) +int mei_restart(struct mei_device *dev) { - struct mei_msg_hdr *mei_hdr; - struct hbm_host_enum_request *host_enum_req; - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - /* enumerate clients */ - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = sizeof(struct hbm_host_enum_request); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - host_enum_req = (struct hbm_host_enum_request *) &dev->wr_msg_buf[1]; - memset(host_enum_req, 0, sizeof(struct hbm_host_enum_request)); - host_enum_req->hbm_cmd = HOST_ENUM_REQ_CMD; - if (mei_write_message(dev, mei_hdr, (unsigned char *)host_enum_req, - mei_hdr->length)) { - dev->dev_state = MEI_DEV_RESETING; - dev_dbg(&dev->pdev->dev, "write send enumeration request message to FW fail.\n"); - mei_reset(dev, 1); - } - dev->init_clients_state = MEI_ENUM_CLIENTS_MESSAGE; - dev->init_clients_timer = INIT_CLIENTS_TIMEOUT; - return; -} + int err; + mutex_lock(&dev->device_lock); -/** - * allocate_me_clients_storage - allocates storage for me clients - * - * @dev: the device structure - * - * returns none. - */ -void mei_allocate_me_clients_storage(struct mei_device *dev) -{ - struct mei_me_client *clients; - int b; + mei_clear_interrupts(dev); - /* count how many ME clients we have */ - for_each_set_bit(b, dev->me_clients_map, MEI_CLIENTS_MAX) - dev->me_clients_num++; + dev->dev_state = MEI_DEV_POWER_UP; + dev->reset_count = 0; - if (dev->me_clients_num <= 0) - return ; + err = mei_reset(dev); + mutex_unlock(&dev->device_lock); - if (dev->me_clients != NULL) { - kfree(dev->me_clients); - dev->me_clients = NULL; - } - dev_dbg(&dev->pdev->dev, "memory allocation for ME clients size=%zd.\n", - dev->me_clients_num * sizeof(struct mei_me_client)); - /* allocate storage for ME clients representation */ - clients = kcalloc(dev->me_clients_num, - sizeof(struct mei_me_client), GFP_KERNEL); - if (!clients) { - dev_dbg(&dev->pdev->dev, "memory allocation for ME clients failed.\n"); - dev->dev_state = MEI_DEV_RESETING; - mei_reset(dev, 1); - return ; + if (err == -ENODEV || dev->dev_state == MEI_DEV_DISABLED) { + dev_err(&dev->pdev->dev, "device disabled = %d\n", err); + return -ENODEV; } - dev->me_clients = clients; - return ; -} -/** - * host_client_properties - reads properties for client - * - * @dev: the device structure - * - * returns: - * < 0 - Error. - * = 0 - no more clients. - * = 1 - still have clients to send properties request. - */ -int mei_host_client_properties(struct mei_device *dev) -{ - struct mei_msg_hdr *mei_header; - struct hbm_props_request *host_cli_req; - int b; - u8 client_num = dev->me_client_presentation_num; - - b = dev->me_client_index; - b = find_next_bit(dev->me_clients_map, MEI_CLIENTS_MAX, b); - if (b < MEI_CLIENTS_MAX) { - dev->me_clients[client_num].client_id = b; - dev->me_clients[client_num].mei_flow_ctrl_creds = 0; - mei_header = (struct mei_msg_hdr *)&dev->wr_msg_buf[0]; - mei_header->host_addr = 0; - mei_header->me_addr = 0; - mei_header->length = sizeof(struct hbm_props_request); - mei_header->msg_complete = 1; - mei_header->reserved = 0; - - host_cli_req = (struct hbm_props_request *)&dev->wr_msg_buf[1]; - - memset(host_cli_req, 0, sizeof(struct hbm_props_request)); - - host_cli_req->hbm_cmd = HOST_CLIENT_PROPERTIES_REQ_CMD; - host_cli_req->address = b; - - if (mei_write_message(dev, mei_header, - (unsigned char *)host_cli_req, - mei_header->length)) { - dev->dev_state = MEI_DEV_RESETING; - dev_dbg(&dev->pdev->dev, "write send enumeration request message to FW fail.\n"); - mei_reset(dev, 1); - return -EIO; - } - dev->init_clients_timer = INIT_CLIENTS_TIMEOUT; - dev->me_client_index = b; - return 1; - } + /* try to start again */ + if (err) + schedule_work(&dev->reset_work); - return 0; -} -/** - * mei_init_file_private - initializes private file structure. - * - * @priv: private file structure to be initialized - * @file: the file structure - */ -void mei_cl_init(struct mei_cl *priv, struct mei_device *dev) -{ - memset(priv, 0, sizeof(struct mei_cl)); - init_waitqueue_head(&priv->wait); - init_waitqueue_head(&priv->rx_wait); - init_waitqueue_head(&priv->tx_wait); - INIT_LIST_HEAD(&priv->link); - priv->reading_state = MEI_IDLE; - priv->writing_state = MEI_IDLE; - priv->dev = dev; + return 0; } +EXPORT_SYMBOL_GPL(mei_restart); -int mei_me_cl_by_uuid(const struct mei_device *dev, const uuid_le *cuuid) +static void mei_reset_work(struct work_struct *work) { - int i, res = -ENOENT; - - for (i = 0; i < dev->me_clients_num; ++i) - if (uuid_le_cmp(*cuuid, - dev->me_clients[i].props.protocol_name) == 0) { - res = i; - break; - } - - return res; -} - + struct mei_device *dev = + container_of(work, struct mei_device, reset_work); + int ret; -/** - * mei_me_cl_update_filext - searches for ME client guid - * sets client_id in mei_file_private if found - * @dev: the device structure - * @cl: private file structure to set client_id in - * @cuuid: searched uuid of ME client - * @client_id: id of host client to be set in file private structure - * - * returns ME client index - */ -int mei_me_cl_update_filext(struct mei_device *dev, struct mei_cl *cl, - const uuid_le *cuuid, u8 host_cl_id) -{ - int i; + mutex_lock(&dev->device_lock); - if (!dev || !cl || !cuuid) - return -EINVAL; + ret = mei_reset(dev); - /* check for valid client id */ - i = mei_me_cl_by_uuid(dev, cuuid); - if (i >= 0) { - cl->me_client_id = dev->me_clients[i].client_id; - cl->state = MEI_FILE_CONNECTING; - cl->host_client_id = host_cl_id; + mutex_unlock(&dev->device_lock); - list_add_tail(&cl->link, &dev->file_list); - return (u8)i; + if (dev->dev_state == MEI_DEV_DISABLED) { + dev_err(&dev->pdev->dev, "device disabled = %d\n", ret); + return; } - return -ENOENT; + /* retry reset in case of failure */ + if (ret) + schedule_work(&dev->reset_work); } -/** - * host_init_iamthif - mei initialization iamthif client. - * - * @dev: the device structure - * - */ -void mei_host_init_iamthif(struct mei_device *dev) +void mei_stop(struct mei_device *dev) { - int i; - unsigned char *msg_buf; + dev_dbg(&dev->pdev->dev, "stopping the device.\n"); - mei_cl_init(&dev->iamthif_cl, dev); - dev->iamthif_cl.state = MEI_FILE_DISCONNECTED; + mei_cancel_work(dev); - /* find ME amthi client */ - i = mei_me_cl_update_filext(dev, &dev->iamthif_cl, - &mei_amthi_guid, MEI_IAMTHIF_HOST_CLIENT_ID); - if (i < 0) { - dev_dbg(&dev->pdev->dev, "failed to find iamthif client.\n"); - return; - } + mei_nfc_host_exit(dev); - /* Assign iamthif_mtu to the value received from ME */ + mei_cl_bus_remove_devices(dev); - dev->iamthif_mtu = dev->me_clients[i].props.max_msg_length; - dev_dbg(&dev->pdev->dev, "IAMTHIF_MTU = %d\n", - dev->me_clients[i].props.max_msg_length); + mutex_lock(&dev->device_lock); - kfree(dev->iamthif_msg_buf); - dev->iamthif_msg_buf = NULL; + mei_wd_stop(dev); - /* allocate storage for ME message buffer */ - msg_buf = kcalloc(dev->iamthif_mtu, - sizeof(unsigned char), GFP_KERNEL); - if (!msg_buf) { - dev_dbg(&dev->pdev->dev, "memory allocation for ME message buffer failed.\n"); - return; - } + dev->dev_state = MEI_DEV_POWER_DOWN; + mei_reset(dev); - dev->iamthif_msg_buf = msg_buf; + mutex_unlock(&dev->device_lock); - if (mei_connect(dev, &dev->iamthif_cl)) { - dev_dbg(&dev->pdev->dev, "Failed to connect to AMTHI client\n"); - dev->iamthif_cl.state = MEI_FILE_DISCONNECTED; - dev->iamthif_cl.host_client_id = 0; - } else { - dev->iamthif_cl.timer_count = CONNECT_TIMEOUT; - } + mei_watchdog_unregister(dev); } +EXPORT_SYMBOL_GPL(mei_stop); /** - * mei_alloc_file_private - allocates a private file structure and sets it up. - * @file: the file structure + * mei_write_is_idle - check if the write queues are idle + * + * @dev: the device structure * - * returns The allocated file or NULL on failure + * returns true of there is no pending write */ -struct mei_cl *mei_cl_allocate(struct mei_device *dev) +bool mei_write_is_idle(struct mei_device *dev) { - struct mei_cl *cl; - - cl = kmalloc(sizeof(struct mei_cl), GFP_KERNEL); - if (!cl) - return NULL; + bool idle = (dev->dev_state == MEI_DEV_ENABLED && + list_empty(&dev->ctrl_wr_list.list) && + list_empty(&dev->write_list.list)); - mei_cl_init(cl, dev); + dev_dbg(&dev->pdev->dev, "write pg: is idle[%d] state=%s ctrl=%d write=%d\n", + idle, + mei_dev_state_str(dev->dev_state), + list_empty(&dev->ctrl_wr_list.list), + list_empty(&dev->write_list.list)); - return cl; + return idle; } +EXPORT_SYMBOL_GPL(mei_write_is_idle); +int mei_fw_status(struct mei_device *dev, struct mei_fw_status *fw_status) +{ + int i; + const struct mei_fw_status *fw_src = &dev->cfg->fw_status; + if (!fw_status) + return -EINVAL; -/** - * mei_disconnect_host_client - sends disconnect message to fw from host client. - * - * @dev: the device structure - * @cl: private data of the file object - * - * Locking: called under "dev->device_lock" lock - * - * returns 0 on success, <0 on failure. - */ -int mei_disconnect_host_client(struct mei_device *dev, struct mei_cl *cl) + fw_status->count = fw_src->count; + for (i = 0; i < fw_src->count && i < MEI_FW_STATUS_MAX; i++) { + int ret; + ret = pci_read_config_dword(dev->pdev, + fw_src->status[i], &fw_status->status[i]); + if (ret) + return ret; + } + + return 0; +} +EXPORT_SYMBOL_GPL(mei_fw_status); + +void mei_device_init(struct mei_device *dev, const struct mei_cfg *cfg) { - int rets, err; - long timeout = 15; /* 15 seconds */ - struct mei_cl_cb *cb; + /* setup our list array */ + INIT_LIST_HEAD(&dev->file_list); + INIT_LIST_HEAD(&dev->device_list); + mutex_init(&dev->device_lock); + init_waitqueue_head(&dev->wait_hw_ready); + init_waitqueue_head(&dev->wait_pg); + init_waitqueue_head(&dev->wait_recvd_msg); + init_waitqueue_head(&dev->wait_stop_wd); + dev->dev_state = MEI_DEV_INITIALIZING; + dev->reset_count = 0; - if (!dev || !cl) - return -ENODEV; + mei_io_list_init(&dev->read_list); + mei_io_list_init(&dev->write_list); + mei_io_list_init(&dev->write_waiting_list); + mei_io_list_init(&dev->ctrl_wr_list); + mei_io_list_init(&dev->ctrl_rd_list); - if (cl->state != MEI_FILE_DISCONNECTING) - return 0; + INIT_DELAYED_WORK(&dev->timer_work, mei_timer); + INIT_WORK(&dev->init_work, mei_host_client_init); + INIT_WORK(&dev->reset_work, mei_reset_work); - cb = kzalloc(sizeof(struct mei_cl_cb), GFP_KERNEL); - if (!cb) - return -ENOMEM; - - INIT_LIST_HEAD(&cb->cb_list); - cb->file_private = cl; - cb->major_file_operations = MEI_CLOSE; - if (dev->mei_host_buffer_is_empty) { - dev->mei_host_buffer_is_empty = false; - if (mei_disconnect(dev, cl)) { - rets = -ENODEV; - dev_dbg(&dev->pdev->dev, "failed to call mei_disconnect.\n"); - goto free; - } - mdelay(10); /* Wait for hardware disconnection ready */ - list_add_tail(&cb->cb_list, &dev->ctrl_rd_list.mei_cb.cb_list); - } else { - dev_dbg(&dev->pdev->dev, "add disconnect cb to control write list\n"); - list_add_tail(&cb->cb_list, - &dev->ctrl_wr_list.mei_cb.cb_list); - } - mutex_unlock(&dev->device_lock); + INIT_LIST_HEAD(&dev->wd_cl.link); + INIT_LIST_HEAD(&dev->iamthif_cl.link); + mei_io_list_init(&dev->amthif_cmd_list); + mei_io_list_init(&dev->amthif_rd_complete_list); - err = wait_event_timeout(dev->wait_recvd_msg, - (MEI_FILE_DISCONNECTED == cl->state), - timeout * HZ); + bitmap_zero(dev->host_clients_map, MEI_CLIENTS_MAX); + dev->open_handle_count = 0; - mutex_lock(&dev->device_lock); - if (MEI_FILE_DISCONNECTED == cl->state) { - rets = 0; - dev_dbg(&dev->pdev->dev, "successfully disconnected from FW client.\n"); - } else { - rets = -ENODEV; - if (MEI_FILE_DISCONNECTED != cl->state) - dev_dbg(&dev->pdev->dev, "wrong status client disconnect.\n"); - - if (err) - dev_dbg(&dev->pdev->dev, - "wait failed disconnect err=%08x\n", - err); - - dev_dbg(&dev->pdev->dev, "failed to disconnect from FW client.\n"); - } + /* + * Reserving the first client ID + * 0: Reserved for MEI Bus Message communications + */ + bitmap_set(dev->host_clients_map, 0, 1); - mei_io_list_flush(&dev->ctrl_rd_list, cl); - mei_io_list_flush(&dev->ctrl_wr_list, cl); -free: - mei_free_cb_private(cb); - return rets; + dev->pg_event = MEI_PG_EVENT_IDLE; + dev->cfg = cfg; } +EXPORT_SYMBOL_GPL(mei_device_init); -/** - * mei_remove_client_from_file_list - - * removes file private data from device file list - * - * @dev: the device structure - * @host_client_id: host client id to be removed - */ -void mei_remove_client_from_file_list(struct mei_device *dev, - u8 host_client_id) -{ - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; - list_for_each_entry_safe(cl_pos, cl_next, &dev->file_list, link) { - if (host_client_id == cl_pos->host_client_id) { - dev_dbg(&dev->pdev->dev, "remove host client = %d, ME client = %d\n", - cl_pos->host_client_id, - cl_pos->me_client_id); - list_del_init(&cl_pos->link); - break; - } - } -} diff --git a/drivers/misc/mei/interface.c b/drivers/misc/mei/interface.c deleted file mode 100644 index 509c3957ff4..00000000000 --- a/drivers/misc/mei/interface.c +++ /dev/null @@ -1,407 +0,0 @@ -/* - * - * Intel Management Engine Interface (Intel MEI) Linux driver - * Copyright (c) 2003-2012, Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - */ - -#include <linux/pci.h> -#include "mei_dev.h" -#include <linux/mei.h> -#include "interface.h" - - - -/** - * mei_set_csr_register - writes H_CSR register to the mei device, - * and ignores the H_IS bit for it is write-one-to-zero. - * - * @dev: the device structure - */ -void mei_hcsr_set(struct mei_device *dev) -{ - if ((dev->host_hw_state & H_IS) == H_IS) - dev->host_hw_state &= ~H_IS; - mei_reg_write(dev, H_CSR, dev->host_hw_state); - dev->host_hw_state = mei_hcsr_read(dev); -} - -/** - * mei_csr_enable_interrupts - enables mei device interrupts - * - * @dev: the device structure - */ -void mei_enable_interrupts(struct mei_device *dev) -{ - dev->host_hw_state |= H_IE; - mei_hcsr_set(dev); -} - -/** - * mei_csr_disable_interrupts - disables mei device interrupts - * - * @dev: the device structure - */ -void mei_disable_interrupts(struct mei_device *dev) -{ - dev->host_hw_state &= ~H_IE; - mei_hcsr_set(dev); -} - -/** - * mei_hbuf_filled_slots - gets number of device filled buffer slots - * - * @device: the device structure - * - * returns number of filled slots - */ -static unsigned char mei_hbuf_filled_slots(struct mei_device *dev) -{ - char read_ptr, write_ptr; - - dev->host_hw_state = mei_hcsr_read(dev); - - read_ptr = (char) ((dev->host_hw_state & H_CBRP) >> 8); - write_ptr = (char) ((dev->host_hw_state & H_CBWP) >> 16); - - return (unsigned char) (write_ptr - read_ptr); -} - -/** - * mei_hbuf_is_empty - checks if host buffer is empty. - * - * @dev: the device structure - * - * returns true if empty, false - otherwise. - */ -bool mei_hbuf_is_empty(struct mei_device *dev) -{ - return mei_hbuf_filled_slots(dev) == 0; -} - -/** - * mei_hbuf_empty_slots - counts write empty slots. - * - * @dev: the device structure - * - * returns -1(ESLOTS_OVERFLOW) if overflow, otherwise empty slots count - */ -int mei_hbuf_empty_slots(struct mei_device *dev) -{ - unsigned char filled_slots, empty_slots; - - filled_slots = mei_hbuf_filled_slots(dev); - empty_slots = dev->hbuf_depth - filled_slots; - - /* check for overflow */ - if (filled_slots > dev->hbuf_depth) - return -EOVERFLOW; - - return empty_slots; -} - -/** - * mei_write_message - writes a message to mei device. - * - * @dev: the device structure - * @header: header of message - * @write_buffer: message buffer will be written - * @write_length: message size will be written - * - * This function returns -EIO if write has failed - */ -int mei_write_message(struct mei_device *dev, struct mei_msg_hdr *header, - unsigned char *buf, unsigned long length) -{ - unsigned long rem, dw_cnt; - u32 *reg_buf = (u32 *)buf; - int i; - int empty_slots; - - - dev_dbg(&dev->pdev->dev, - "mei_write_message header=%08x.\n", - *((u32 *) header)); - - empty_slots = mei_hbuf_empty_slots(dev); - dev_dbg(&dev->pdev->dev, "empty slots = %hu.\n", empty_slots); - - dw_cnt = mei_data2slots(length); - if (empty_slots < 0 || dw_cnt > empty_slots) - return -EIO; - - mei_reg_write(dev, H_CB_WW, *((u32 *) header)); - - for (i = 0; i < length / 4; i++) - mei_reg_write(dev, H_CB_WW, reg_buf[i]); - - rem = length & 0x3; - if (rem > 0) { - u32 reg = 0; - memcpy(®, &buf[length - rem], rem); - mei_reg_write(dev, H_CB_WW, reg); - } - - dev->host_hw_state = mei_hcsr_read(dev); - dev->host_hw_state |= H_IG; - mei_hcsr_set(dev); - dev->me_hw_state = mei_mecsr_read(dev); - if ((dev->me_hw_state & ME_RDY_HRA) != ME_RDY_HRA) - return -EIO; - - return 0; -} - -/** - * mei_count_full_read_slots - counts read full slots. - * - * @dev: the device structure - * - * returns -1(ESLOTS_OVERFLOW) if overflow, otherwise filled slots count - */ -int mei_count_full_read_slots(struct mei_device *dev) -{ - char read_ptr, write_ptr; - unsigned char buffer_depth, filled_slots; - - dev->me_hw_state = mei_mecsr_read(dev); - buffer_depth = (unsigned char)((dev->me_hw_state & ME_CBD_HRA) >> 24); - read_ptr = (char) ((dev->me_hw_state & ME_CBRP_HRA) >> 8); - write_ptr = (char) ((dev->me_hw_state & ME_CBWP_HRA) >> 16); - filled_slots = (unsigned char) (write_ptr - read_ptr); - - /* check for overflow */ - if (filled_slots > buffer_depth) - return -EOVERFLOW; - - dev_dbg(&dev->pdev->dev, "filled_slots =%08x\n", filled_slots); - return (int)filled_slots; -} - -/** - * mei_read_slots - reads a message from mei device. - * - * @dev: the device structure - * @buffer: message buffer will be written - * @buffer_length: message size will be read - */ -void mei_read_slots(struct mei_device *dev, unsigned char *buffer, - unsigned long buffer_length) -{ - u32 *reg_buf = (u32 *)buffer; - - for (; buffer_length >= sizeof(u32); buffer_length -= sizeof(u32)) - *reg_buf++ = mei_mecbrw_read(dev); - - if (buffer_length > 0) { - u32 reg = mei_mecbrw_read(dev); - memcpy(reg_buf, ®, buffer_length); - } - - dev->host_hw_state |= H_IG; - mei_hcsr_set(dev); -} - -/** - * mei_flow_ctrl_creds - checks flow_control credentials. - * - * @dev: the device structure - * @cl: private data of the file object - * - * returns 1 if mei_flow_ctrl_creds >0, 0 - otherwise. - * -ENOENT if mei_cl is not present - * -EINVAL if single_recv_buf == 0 - */ -int mei_flow_ctrl_creds(struct mei_device *dev, struct mei_cl *cl) -{ - int i; - - if (!dev->me_clients_num) - return 0; - - if (cl->mei_flow_ctrl_creds > 0) - return 1; - - for (i = 0; i < dev->me_clients_num; i++) { - struct mei_me_client *me_cl = &dev->me_clients[i]; - if (me_cl->client_id == cl->me_client_id) { - if (me_cl->mei_flow_ctrl_creds) { - if (WARN_ON(me_cl->props.single_recv_buf == 0)) - return -EINVAL; - return 1; - } else { - return 0; - } - } - } - return -ENOENT; -} - -/** - * mei_flow_ctrl_reduce - reduces flow_control. - * - * @dev: the device structure - * @cl: private data of the file object - * @returns - * 0 on success - * -ENOENT when me client is not found - * -EINVAL when ctrl credits are <= 0 - */ -int mei_flow_ctrl_reduce(struct mei_device *dev, struct mei_cl *cl) -{ - int i; - - if (!dev->me_clients_num) - return -ENOENT; - - for (i = 0; i < dev->me_clients_num; i++) { - struct mei_me_client *me_cl = &dev->me_clients[i]; - if (me_cl->client_id == cl->me_client_id) { - if (me_cl->props.single_recv_buf != 0) { - if (WARN_ON(me_cl->mei_flow_ctrl_creds <= 0)) - return -EINVAL; - dev->me_clients[i].mei_flow_ctrl_creds--; - } else { - if (WARN_ON(cl->mei_flow_ctrl_creds <= 0)) - return -EINVAL; - cl->mei_flow_ctrl_creds--; - } - return 0; - } - } - return -ENOENT; -} - -/** - * mei_send_flow_control - sends flow control to fw. - * - * @dev: the device structure - * @cl: private data of the file object - * - * This function returns -EIO on write failure - */ -int mei_send_flow_control(struct mei_device *dev, struct mei_cl *cl) -{ - struct mei_msg_hdr *mei_hdr; - struct hbm_flow_control *mei_flow_control; - - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = sizeof(struct hbm_flow_control); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - mei_flow_control = (struct hbm_flow_control *) &dev->wr_msg_buf[1]; - memset(mei_flow_control, 0, sizeof(*mei_flow_control)); - mei_flow_control->host_addr = cl->host_client_id; - mei_flow_control->me_addr = cl->me_client_id; - mei_flow_control->hbm_cmd = MEI_FLOW_CONTROL_CMD; - memset(mei_flow_control->reserved, 0, - sizeof(mei_flow_control->reserved)); - dev_dbg(&dev->pdev->dev, "sending flow control host client = %d, ME client = %d\n", - cl->host_client_id, cl->me_client_id); - - return mei_write_message(dev, mei_hdr, - (unsigned char *) mei_flow_control, - sizeof(struct hbm_flow_control)); -} - -/** - * mei_other_client_is_connecting - checks if other - * client with the same client id is connected. - * - * @dev: the device structure - * @cl: private data of the file object - * - * returns 1 if other client is connected, 0 - otherwise. - */ -int mei_other_client_is_connecting(struct mei_device *dev, - struct mei_cl *cl) -{ - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; - - list_for_each_entry_safe(cl_pos, cl_next, &dev->file_list, link) { - if ((cl_pos->state == MEI_FILE_CONNECTING) && - (cl_pos != cl) && - cl->me_client_id == cl_pos->me_client_id) - return 1; - - } - return 0; -} - -/** - * mei_disconnect - sends disconnect message to fw. - * - * @dev: the device structure - * @cl: private data of the file object - * - * This function returns -EIO on write failure - */ -int mei_disconnect(struct mei_device *dev, struct mei_cl *cl) -{ - struct mei_msg_hdr *mei_hdr; - struct hbm_client_disconnect_request *mei_cli_disconnect; - - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = sizeof(struct hbm_client_disconnect_request); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - mei_cli_disconnect = - (struct hbm_client_disconnect_request *) &dev->wr_msg_buf[1]; - memset(mei_cli_disconnect, 0, sizeof(*mei_cli_disconnect)); - mei_cli_disconnect->host_addr = cl->host_client_id; - mei_cli_disconnect->me_addr = cl->me_client_id; - mei_cli_disconnect->hbm_cmd = CLIENT_DISCONNECT_REQ_CMD; - mei_cli_disconnect->reserved[0] = 0; - - return mei_write_message(dev, mei_hdr, - (unsigned char *) mei_cli_disconnect, - sizeof(struct hbm_client_disconnect_request)); -} - -/** - * mei_connect - sends connect message to fw. - * - * @dev: the device structure - * @cl: private data of the file object - * - * This function returns -EIO on write failure - */ -int mei_connect(struct mei_device *dev, struct mei_cl *cl) -{ - struct mei_msg_hdr *mei_hdr; - struct hbm_client_connect_request *mei_cli_connect; - - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = sizeof(struct hbm_client_connect_request); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - mei_cli_connect = - (struct hbm_client_connect_request *) &dev->wr_msg_buf[1]; - mei_cli_connect->host_addr = cl->host_client_id; - mei_cli_connect->me_addr = cl->me_client_id; - mei_cli_connect->hbm_cmd = CLIENT_CONNECT_REQ_CMD; - mei_cli_connect->reserved = 0; - - return mei_write_message(dev, mei_hdr, - (unsigned char *) mei_cli_connect, - sizeof(struct hbm_client_connect_request)); -} diff --git a/drivers/misc/mei/interface.h b/drivers/misc/mei/interface.h deleted file mode 100644 index ec6c785a396..00000000000 --- a/drivers/misc/mei/interface.h +++ /dev/null @@ -1,81 +0,0 @@ -/* - * - * Intel Management Engine Interface (Intel MEI) Linux driver - * Copyright (c) 2003-2012, Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - */ - - - -#ifndef _MEI_INTERFACE_H_ -#define _MEI_INTERFACE_H_ - -#include <linux/mei.h> -#include "mei_dev.h" - - - -void mei_read_slots(struct mei_device *dev, - unsigned char *buffer, - unsigned long buffer_length); - -int mei_write_message(struct mei_device *dev, - struct mei_msg_hdr *header, - unsigned char *write_buffer, - unsigned long write_length); - -bool mei_hbuf_is_empty(struct mei_device *dev); - -int mei_hbuf_empty_slots(struct mei_device *dev); - -static inline size_t mei_hbuf_max_data(const struct mei_device *dev) -{ - return dev->hbuf_depth * sizeof(u32) - sizeof(struct mei_msg_hdr); -} - -/* get slots (dwords) from a message length + header (bytes) */ -static inline unsigned char mei_data2slots(size_t length) -{ - return DIV_ROUND_UP(sizeof(struct mei_msg_hdr) + length, 4); -} - -int mei_count_full_read_slots(struct mei_device *dev); - - -int mei_flow_ctrl_creds(struct mei_device *dev, struct mei_cl *cl); - - - -int mei_wd_send(struct mei_device *dev); -int mei_wd_stop(struct mei_device *dev); -int mei_wd_host_init(struct mei_device *dev); -/* - * mei_watchdog_register - Registering watchdog interface - * once we got connection to the WD Client - * @dev - mei device - */ -void mei_watchdog_register(struct mei_device *dev); -/* - * mei_watchdog_unregister - Unregistering watchdog interface - * @dev - mei device - */ -void mei_watchdog_unregister(struct mei_device *dev); - -int mei_flow_ctrl_reduce(struct mei_device *dev, struct mei_cl *cl); - -int mei_send_flow_control(struct mei_device *dev, struct mei_cl *cl); - -int mei_disconnect(struct mei_device *dev, struct mei_cl *cl); -int mei_other_client_is_connecting(struct mei_device *dev, struct mei_cl *cl); -int mei_connect(struct mei_device *dev, struct mei_cl *cl); - -#endif /* _MEI_INTERFACE_H_ */ diff --git a/drivers/misc/mei/interrupt.c b/drivers/misc/mei/interrupt.c index 3533edde04a..4e3cba6da3f 100644 --- a/drivers/misc/mei/interrupt.c +++ b/drivers/misc/mei/interrupt.c @@ -15,1142 +15,399 @@ */ +#include <linux/export.h> #include <linux/pci.h> #include <linux/kthread.h> #include <linux/interrupt.h> #include <linux/fs.h> #include <linux/jiffies.h> -#include "mei_dev.h" #include <linux/mei.h> -#include "hw.h" -#include "interface.h" - -/** - * mei_interrupt_quick_handler - The ISR of the MEI device - * - * @irq: The irq number - * @dev_id: pointer to the device structure - * - * returns irqreturn_t - */ -irqreturn_t mei_interrupt_quick_handler(int irq, void *dev_id) -{ - struct mei_device *dev = (struct mei_device *) dev_id; - u32 csr_reg = mei_hcsr_read(dev); - - if ((csr_reg & H_IS) != H_IS) - return IRQ_NONE; - - /* clear H_IS bit in H_CSR */ - mei_reg_write(dev, H_CSR, csr_reg); +#include "mei_dev.h" +#include "hbm.h" +#include "client.h" - return IRQ_WAKE_THREAD; -} /** - * _mei_cmpl - processes completed operation. + * mei_irq_compl_handler - dispatch complete handlers + * for the completed callbacks * - * @cl: private data of the file object. - * @cb_pos: callback block. + * @dev - mei device + * @compl_list - list of completed cbs */ -static void _mei_cmpl(struct mei_cl *cl, struct mei_cl_cb *cb_pos) +void mei_irq_compl_handler(struct mei_device *dev, struct mei_cl_cb *compl_list) { - if (cb_pos->major_file_operations == MEI_WRITE) { - mei_free_cb_private(cb_pos); - cb_pos = NULL; - cl->writing_state = MEI_WRITE_COMPLETE; - if (waitqueue_active(&cl->tx_wait)) - wake_up_interruptible(&cl->tx_wait); - - } else if (cb_pos->major_file_operations == MEI_READ && - MEI_READING == cl->reading_state) { - cl->reading_state = MEI_READ_COMPLETE; - if (waitqueue_active(&cl->rx_wait)) - wake_up_interruptible(&cl->rx_wait); + struct mei_cl_cb *cb, *next; + struct mei_cl *cl; - } -} + list_for_each_entry_safe(cb, next, &compl_list->list, list) { + cl = cb->cl; + list_del(&cb->list); + if (!cl) + continue; -/** - * _mei_cmpl_iamthif - processes completed iamthif operation. - * - * @dev: the device structure. - * @cb_pos: callback block. - */ -static void _mei_cmpl_iamthif(struct mei_device *dev, struct mei_cl_cb *cb_pos) -{ - if (dev->iamthif_canceled != 1) { - dev->iamthif_state = MEI_IAMTHIF_READ_COMPLETE; - dev->iamthif_stall_timer = 0; - memcpy(cb_pos->response_buffer.data, - dev->iamthif_msg_buf, - dev->iamthif_msg_buf_index); - list_add_tail(&cb_pos->cb_list, - &dev->amthi_read_complete_list.mei_cb.cb_list); - dev_dbg(&dev->pdev->dev, "amthi read completed.\n"); - dev->iamthif_timer = jiffies; - dev_dbg(&dev->pdev->dev, "dev->iamthif_timer = %ld\n", - dev->iamthif_timer); - } else { - mei_run_next_iamthif_cmd(dev); + dev_dbg(&dev->pdev->dev, "completing call back.\n"); + if (cl == &dev->iamthif_cl) + mei_amthif_complete(dev, cb); + else + mei_cl_complete(cl, cb); } - - dev_dbg(&dev->pdev->dev, "completing amthi call back.\n"); - wake_up_interruptible(&dev->iamthif_cl.wait); } - +EXPORT_SYMBOL_GPL(mei_irq_compl_handler); /** - * mei_irq_thread_read_amthi_message - bottom half read routine after ISR to - * handle the read amthi message data processing. + * mei_cl_hbm_equal - check if hbm is addressed to the client * - * @complete_list: An instance of our list structure - * @dev: the device structure - * @mei_hdr: header of amthi message + * @cl: host client + * @mei_hdr: header of mei client message * - * returns 0 on success, <0 on failure. + * returns true if matches, false otherwise */ -static int mei_irq_thread_read_amthi_message(struct mei_io_list *complete_list, - struct mei_device *dev, - struct mei_msg_hdr *mei_hdr) +static inline int mei_cl_hbm_equal(struct mei_cl *cl, + struct mei_msg_hdr *mei_hdr) { - struct mei_cl *cl; - struct mei_cl_cb *cb; - unsigned char *buffer; - - BUG_ON(mei_hdr->me_addr != dev->iamthif_cl.me_client_id); - BUG_ON(dev->iamthif_state != MEI_IAMTHIF_READING); - - buffer = dev->iamthif_msg_buf + dev->iamthif_msg_buf_index; - BUG_ON(dev->iamthif_mtu < dev->iamthif_msg_buf_index + mei_hdr->length); - - mei_read_slots(dev, buffer, mei_hdr->length); - - dev->iamthif_msg_buf_index += mei_hdr->length; - - if (!mei_hdr->msg_complete) - return 0; - - dev_dbg(&dev->pdev->dev, - "amthi_message_buffer_index =%d\n", - mei_hdr->length); - - dev_dbg(&dev->pdev->dev, "completed amthi read.\n "); - if (!dev->iamthif_current_cb) - return -ENODEV; - - cb = dev->iamthif_current_cb; - dev->iamthif_current_cb = NULL; - - cl = (struct mei_cl *)cb->file_private; - if (!cl) - return -ENODEV; - - dev->iamthif_stall_timer = 0; - cb->information = dev->iamthif_msg_buf_index; - cb->read_time = jiffies; - if (dev->iamthif_ioctl && cl == &dev->iamthif_cl) { - /* found the iamthif cb */ - dev_dbg(&dev->pdev->dev, "complete the amthi read cb.\n "); - dev_dbg(&dev->pdev->dev, "add the amthi read cb to complete.\n "); - list_add_tail(&cb->cb_list, - &complete_list->mei_cb.cb_list); - } - return 0; + return cl->host_client_id == mei_hdr->host_addr && + cl->me_client_id == mei_hdr->me_addr; } - /** - * _mei_irq_thread_state_ok - checks if mei header matches file private data + * mei_cl_is_reading - checks if the client + is the one to read this message * - * @cl: private data of the file object - * @mei_hdr: header of mei client message + * @cl: mei client + * @mei_hdr: header of mei message * - * returns !=0 if matches, 0 if no match. + * returns true on match and false otherwise */ -static int _mei_irq_thread_state_ok(struct mei_cl *cl, - struct mei_msg_hdr *mei_hdr) +static bool mei_cl_is_reading(struct mei_cl *cl, struct mei_msg_hdr *mei_hdr) { - return (cl->host_client_id == mei_hdr->host_addr && - cl->me_client_id == mei_hdr->me_addr && + return mei_cl_hbm_equal(cl, mei_hdr) && cl->state == MEI_FILE_CONNECTED && - MEI_READ_COMPLETE != cl->reading_state); + cl->reading_state != MEI_READ_COMPLETE; } /** - * mei_irq_thread_read_client_message - bottom half read routine after ISR to - * handle the read mei client message data processing. + * mei_irq_read_client_message - process client message * - * @complete_list: An instance of our list structure * @dev: the device structure * @mei_hdr: header of mei client message + * @complete_list: An instance of our list structure * * returns 0 on success, <0 on failure. */ -static int mei_irq_thread_read_client_message(struct mei_io_list *complete_list, - struct mei_device *dev, - struct mei_msg_hdr *mei_hdr) +static int mei_cl_irq_read_msg(struct mei_device *dev, + struct mei_msg_hdr *mei_hdr, + struct mei_cl_cb *complete_list) { struct mei_cl *cl; - struct mei_cl_cb *cb_pos = NULL, *cb_next = NULL; + struct mei_cl_cb *cb, *next; unsigned char *buffer = NULL; - dev_dbg(&dev->pdev->dev, "start client msg\n"); - if (list_empty(&dev->read_list.mei_cb.cb_list)) - goto quit; - - list_for_each_entry_safe(cb_pos, cb_next, - &dev->read_list.mei_cb.cb_list, cb_list) { - cl = (struct mei_cl *)cb_pos->file_private; - if (cl && _mei_irq_thread_state_ok(cl, mei_hdr)) { - cl->reading_state = MEI_READING; - buffer = cb_pos->response_buffer.data + cb_pos->information; - - if (cb_pos->response_buffer.size < - mei_hdr->length + cb_pos->information) { - dev_dbg(&dev->pdev->dev, "message overflow.\n"); - list_del(&cb_pos->cb_list); + list_for_each_entry_safe(cb, next, &dev->read_list.list, list) { + cl = cb->cl; + if (!cl || !mei_cl_is_reading(cl, mei_hdr)) + continue; + + cl->reading_state = MEI_READING; + + if (cb->response_buffer.size == 0 || + cb->response_buffer.data == NULL) { + cl_err(dev, cl, "response buffer is not allocated.\n"); + list_del(&cb->list); + return -ENOMEM; + } + + if (cb->response_buffer.size < mei_hdr->length + cb->buf_idx) { + cl_dbg(dev, cl, "message overflow. size %d len %d idx %ld\n", + cb->response_buffer.size, + mei_hdr->length, cb->buf_idx); + buffer = krealloc(cb->response_buffer.data, + mei_hdr->length + cb->buf_idx, + GFP_KERNEL); + + if (!buffer) { + cl_err(dev, cl, "allocation failed.\n"); + list_del(&cb->list); return -ENOMEM; } - if (buffer) - mei_read_slots(dev, buffer, mei_hdr->length); - - cb_pos->information += mei_hdr->length; - if (mei_hdr->msg_complete) { - cl->status = 0; - list_del(&cb_pos->cb_list); - dev_dbg(&dev->pdev->dev, - "completed read H cl = %d, ME cl = %d, length = %lu\n", - cl->host_client_id, - cl->me_client_id, - cb_pos->information); - list_add_tail(&cb_pos->cb_list, - &complete_list->mei_cb.cb_list); - } - - break; + cb->response_buffer.data = buffer; + cb->response_buffer.size = + mei_hdr->length + cb->buf_idx; } + buffer = cb->response_buffer.data + cb->buf_idx; + mei_read_slots(dev, buffer, mei_hdr->length); + + cb->buf_idx += mei_hdr->length; + if (mei_hdr->msg_complete) { + cl->status = 0; + list_del(&cb->list); + cl_dbg(dev, cl, "completed read length = %lu\n", + cb->buf_idx); + list_add_tail(&cb->list, &complete_list->list); + } + break; } -quit: dev_dbg(&dev->pdev->dev, "message read\n"); if (!buffer) { mei_read_slots(dev, dev->rd_msg_buf, mei_hdr->length); - dev_dbg(&dev->pdev->dev, "discarding message, header =%08x.\n", - *(u32 *) dev->rd_msg_buf); - } - - return 0; -} - -/** - * _mei_irq_thread_iamthif_read - prepares to read iamthif data. - * - * @dev: the device structure. - * @slots: free slots. - * - * returns 0, OK; otherwise, error. - */ -static int _mei_irq_thread_iamthif_read(struct mei_device *dev, s32 *slots) -{ - - if (((*slots) * sizeof(u32)) < (sizeof(struct mei_msg_hdr) - + sizeof(struct hbm_flow_control))) { - return -EMSGSIZE; - } - *slots -= mei_data2slots(sizeof(struct hbm_flow_control)); - if (mei_send_flow_control(dev, &dev->iamthif_cl)) { - dev_dbg(&dev->pdev->dev, "iamthif flow control failed\n"); - return -EIO; + dev_dbg(&dev->pdev->dev, "discarding message " MEI_HDR_FMT "\n", + MEI_HDR_PRM(mei_hdr)); } - dev_dbg(&dev->pdev->dev, "iamthif flow control success\n"); - dev->iamthif_state = MEI_IAMTHIF_READING; - dev->iamthif_flow_control_pending = false; - dev->iamthif_msg_buf_index = 0; - dev->iamthif_msg_buf_size = 0; - dev->iamthif_stall_timer = IAMTHIF_STALL_TIMER; - dev->mei_host_buffer_is_empty = mei_hbuf_is_empty(dev); return 0; } /** - * _mei_irq_thread_close - processes close related operation. + * mei_cl_irq_disconnect_rsp - send disconnection response message * - * @dev: the device structure. - * @slots: free slots. - * @cb_pos: callback block. - * @cl: private data of the file object. + * @cl: client + * @cb: callback block. * @cmpl_list: complete list. * * returns 0, OK; otherwise, error. */ -static int _mei_irq_thread_close(struct mei_device *dev, s32 *slots, - struct mei_cl_cb *cb_pos, - struct mei_cl *cl, - struct mei_io_list *cmpl_list) +static int mei_cl_irq_disconnect_rsp(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list) { - if ((*slots * sizeof(u32)) < (sizeof(struct mei_msg_hdr) + - sizeof(struct hbm_client_disconnect_request))) - return -EBADMSG; + struct mei_device *dev = cl->dev; + u32 msg_slots; + int slots; + int ret; - *slots -= mei_data2slots(sizeof(struct hbm_client_disconnect_request)); + slots = mei_hbuf_empty_slots(dev); + msg_slots = mei_data2slots(sizeof(struct hbm_client_connect_response)); - if (mei_disconnect(dev, cl)) { - cl->status = 0; - cb_pos->information = 0; - list_move_tail(&cb_pos->cb_list, - &cmpl_list->mei_cb.cb_list); + if (slots < msg_slots) return -EMSGSIZE; - } else { - cl->state = MEI_FILE_DISCONNECTING; - cl->status = 0; - cb_pos->information = 0; - list_move_tail(&cb_pos->cb_list, - &dev->ctrl_rd_list.mei_cb.cb_list); - cl->timer_count = MEI_CONNECT_TIMEOUT; - } - - return 0; -} - -/** - * is_treat_specially_client - checks if the message belongs - * to the file private data. - * - * @cl: private data of the file object - * @rs: connect response bus message - * - */ -static bool is_treat_specially_client(struct mei_cl *cl, - struct hbm_client_connect_response *rs) -{ - - if (cl->host_client_id == rs->host_addr && - cl->me_client_id == rs->me_addr) { - if (!rs->status) { - cl->state = MEI_FILE_CONNECTED; - cl->status = 0; - } else { - cl->state = MEI_FILE_DISCONNECTED; - cl->status = -ENODEV; - } - cl->timer_count = 0; - - return true; - } - return false; -} - -/** - * mei_client_connect_response - connects to response irq routine - * - * @dev: the device structure - * @rs: connect response bus message - */ -static void mei_client_connect_response(struct mei_device *dev, - struct hbm_client_connect_response *rs) -{ - - struct mei_cl *cl; - struct mei_cl_cb *cb_pos = NULL, *cb_next = NULL; - - dev_dbg(&dev->pdev->dev, - "connect_response:\n" - "ME Client = %d\n" - "Host Client = %d\n" - "Status = %d\n", - rs->me_addr, - rs->host_addr, - rs->status); - - /* if WD or iamthif client treat specially */ - - if (is_treat_specially_client(&(dev->wd_cl), rs)) { - dev_dbg(&dev->pdev->dev, "successfully connected to WD client.\n"); - mei_watchdog_register(dev); - - /* next step in the state maching */ - mei_host_init_iamthif(dev); - return; - } + ret = mei_hbm_cl_disconnect_rsp(dev, cl); - if (is_treat_specially_client(&(dev->iamthif_cl), rs)) { - dev->iamthif_state = MEI_IAMTHIF_IDLE; - return; - } - list_for_each_entry_safe(cb_pos, cb_next, - &dev->ctrl_rd_list.mei_cb.cb_list, cb_list) { + cl->state = MEI_FILE_DISCONNECTED; + cl->status = 0; + list_del(&cb->list); + mei_io_cb_free(cb); - cl = (struct mei_cl *)cb_pos->file_private; - if (!cl) { - list_del(&cb_pos->cb_list); - return; - } - if (MEI_IOCTL == cb_pos->major_file_operations) { - if (is_treat_specially_client(cl, rs)) { - list_del(&cb_pos->cb_list); - cl->status = 0; - cl->timer_count = 0; - break; - } - } - } + return ret; } -/** - * mei_client_disconnect_response - disconnects from response irq routine - * - * @dev: the device structure - * @rs: disconnect response bus message - */ -static void mei_client_disconnect_response(struct mei_device *dev, - struct hbm_client_connect_response *rs) -{ - struct mei_cl *cl; - struct mei_cl_cb *cb_pos = NULL, *cb_next = NULL; - - dev_dbg(&dev->pdev->dev, - "disconnect_response:\n" - "ME Client = %d\n" - "Host Client = %d\n" - "Status = %d\n", - rs->me_addr, - rs->host_addr, - rs->status); - list_for_each_entry_safe(cb_pos, cb_next, - &dev->ctrl_rd_list.mei_cb.cb_list, cb_list) { - cl = (struct mei_cl *)cb_pos->file_private; - - if (!cl) { - list_del(&cb_pos->cb_list); - return; - } - - dev_dbg(&dev->pdev->dev, "list_for_each_entry_safe in ctrl_rd_list.\n"); - if (cl->host_client_id == rs->host_addr && - cl->me_client_id == rs->me_addr) { - - list_del(&cb_pos->cb_list); - if (!rs->status) - cl->state = MEI_FILE_DISCONNECTED; - - cl->status = 0; - cl->timer_count = 0; - break; - } - } -} /** - * same_flow_addr - tells if they have the same address. + * mei_cl_irq_close - processes close related operation from + * interrupt thread context - send disconnect request * - * @file: private data of the file object. - * @flow: flow control. - * - * returns !=0, same; 0,not. - */ -static int same_flow_addr(struct mei_cl *cl, struct hbm_flow_control *flow) -{ - return (cl->host_client_id == flow->host_addr && - cl->me_client_id == flow->me_addr); -} - -/** - * add_single_flow_creds - adds single buffer credentials. - * - * @file: private data ot the file object. - * @flow: flow control. - */ -static void add_single_flow_creds(struct mei_device *dev, - struct hbm_flow_control *flow) -{ - struct mei_me_client *client; - int i; - - for (i = 0; i < dev->me_clients_num; i++) { - client = &dev->me_clients[i]; - if (client && flow->me_addr == client->client_id) { - if (client->props.single_recv_buf) { - client->mei_flow_ctrl_creds++; - dev_dbg(&dev->pdev->dev, "recv flow ctrl msg ME %d (single).\n", - flow->me_addr); - dev_dbg(&dev->pdev->dev, "flow control credentials =%d.\n", - client->mei_flow_ctrl_creds); - } else { - BUG(); /* error in flow control */ - } - } - } -} - -/** - * mei_client_flow_control_response - flow control response irq routine + * @cl: client + * @cb: callback block. + * @cmpl_list: complete list. * - * @dev: the device structure - * @flow_control: flow control response bus message + * returns 0, OK; otherwise, error. */ -static void mei_client_flow_control_response(struct mei_device *dev, - struct hbm_flow_control *flow_control) +static int mei_cl_irq_close(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list) { - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; + struct mei_device *dev = cl->dev; + u32 msg_slots; + int slots; - if (!flow_control->host_addr) { - /* single receive buffer */ - add_single_flow_creds(dev, flow_control); - } else { - /* normal connection */ - list_for_each_entry_safe(cl_pos, cl_next, - &dev->file_list, link) { - dev_dbg(&dev->pdev->dev, "list_for_each_entry_safe in file_list\n"); - - dev_dbg(&dev->pdev->dev, "cl of host client %d ME client %d.\n", - cl_pos->host_client_id, - cl_pos->me_client_id); - dev_dbg(&dev->pdev->dev, "flow ctrl msg for host %d ME %d.\n", - flow_control->host_addr, - flow_control->me_addr); - if (same_flow_addr(cl_pos, flow_control)) { - dev_dbg(&dev->pdev->dev, "recv ctrl msg for host %d ME %d.\n", - flow_control->host_addr, - flow_control->me_addr); - cl_pos->mei_flow_ctrl_creds++; - dev_dbg(&dev->pdev->dev, "flow control credentials = %d.\n", - cl_pos->mei_flow_ctrl_creds); - break; - } - } - } -} + msg_slots = mei_data2slots(sizeof(struct hbm_client_connect_request)); + slots = mei_hbuf_empty_slots(dev); -/** - * same_disconn_addr - tells if they have the same address - * - * @file: private data of the file object. - * @disconn: disconnection request. - * - * returns !=0, same; 0,not. - */ -static int same_disconn_addr(struct mei_cl *cl, - struct hbm_client_disconnect_request *disconn) -{ - return (cl->host_client_id == disconn->host_addr && - cl->me_client_id == disconn->me_addr); -} + if (slots < msg_slots) + return -EMSGSIZE; -/** - * mei_client_disconnect_request - disconnects from request irq routine - * - * @dev: the device structure. - * @disconnect_req: disconnect request bus message. - */ -static void mei_client_disconnect_request(struct mei_device *dev, - struct hbm_client_disconnect_request *disconnect_req) -{ - struct mei_msg_hdr *mei_hdr; - struct hbm_client_connect_response *disconnect_res; - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; - - list_for_each_entry_safe(cl_pos, cl_next, &dev->file_list, link) { - if (same_disconn_addr(cl_pos, disconnect_req)) { - dev_dbg(&dev->pdev->dev, "disconnect request host client %d ME client %d.\n", - disconnect_req->host_addr, - disconnect_req->me_addr); - cl_pos->state = MEI_FILE_DISCONNECTED; - cl_pos->timer_count = 0; - if (cl_pos == &dev->wd_cl) - dev->wd_pending = false; - else if (cl_pos == &dev->iamthif_cl) - dev->iamthif_timer = 0; - - /* prepare disconnect response */ - mei_hdr = - (struct mei_msg_hdr *) &dev->ext_msg_buf[0]; - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = - sizeof(struct hbm_client_connect_response); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - disconnect_res = - (struct hbm_client_connect_response *) - &dev->ext_msg_buf[1]; - disconnect_res->host_addr = cl_pos->host_client_id; - disconnect_res->me_addr = cl_pos->me_client_id; - disconnect_res->hbm_cmd = CLIENT_DISCONNECT_RES_CMD; - disconnect_res->status = 0; - dev->extra_write_index = 2; - break; - } + if (mei_hbm_cl_disconnect_req(dev, cl)) { + cl->status = 0; + cb->buf_idx = 0; + list_move_tail(&cb->list, &cmpl_list->list); + return -EIO; } -} - - -/** - * mei_irq_thread_read_bus_message - bottom half read routine after ISR to - * handle the read bus message cmd processing. - * - * @dev: the device structure - * @mei_hdr: header of bus message - */ -static void mei_irq_thread_read_bus_message(struct mei_device *dev, - struct mei_msg_hdr *mei_hdr) -{ - struct mei_bus_message *mei_msg; - struct hbm_host_version_response *version_res; - struct hbm_client_connect_response *connect_res; - struct hbm_client_connect_response *disconnect_res; - struct hbm_flow_control *flow_control; - struct hbm_props_response *props_res; - struct hbm_host_enum_response *enum_res; - struct hbm_client_disconnect_request *disconnect_req; - struct hbm_host_stop_request *host_stop_req; - int res; - - - /* read the message to our buffer */ - BUG_ON(mei_hdr->length >= sizeof(dev->rd_msg_buf)); - mei_read_slots(dev, dev->rd_msg_buf, mei_hdr->length); - mei_msg = (struct mei_bus_message *)dev->rd_msg_buf; - - switch (mei_msg->hbm_cmd) { - case HOST_START_RES_CMD: - version_res = (struct hbm_host_version_response *) mei_msg; - if (version_res->host_version_supported) { - dev->version.major_version = HBM_MAJOR_VERSION; - dev->version.minor_version = HBM_MINOR_VERSION; - if (dev->dev_state == MEI_DEV_INIT_CLIENTS && - dev->init_clients_state == MEI_START_MESSAGE) { - dev->init_clients_timer = 0; - mei_host_enum_clients_message(dev); - } else { - dev->recvd_msg = false; - dev_dbg(&dev->pdev->dev, "IMEI reset due to received host start response bus message.\n"); - mei_reset(dev, 1); - return; - } - } else { - dev->version = version_res->me_max_version; - /* send stop message */ - mei_hdr = (struct mei_msg_hdr *)&dev->wr_msg_buf[0]; - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = sizeof(struct hbm_host_stop_request); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - host_stop_req = (struct hbm_host_stop_request *) - &dev->wr_msg_buf[1]; - - memset(host_stop_req, - 0, - sizeof(struct hbm_host_stop_request)); - host_stop_req->hbm_cmd = HOST_STOP_REQ_CMD; - host_stop_req->reason = DRIVER_STOP_REQUEST; - mei_write_message(dev, mei_hdr, - (unsigned char *) (host_stop_req), - mei_hdr->length); - dev_dbg(&dev->pdev->dev, "version mismatch.\n"); - return; - } - - dev->recvd_msg = true; - dev_dbg(&dev->pdev->dev, "host start response message received.\n"); - break; - - case CLIENT_CONNECT_RES_CMD: - connect_res = - (struct hbm_client_connect_response *) mei_msg; - mei_client_connect_response(dev, connect_res); - dev_dbg(&dev->pdev->dev, "client connect response message received.\n"); - wake_up(&dev->wait_recvd_msg); - break; - - case CLIENT_DISCONNECT_RES_CMD: - disconnect_res = - (struct hbm_client_connect_response *) mei_msg; - mei_client_disconnect_response(dev, disconnect_res); - dev_dbg(&dev->pdev->dev, "client disconnect response message received.\n"); - wake_up(&dev->wait_recvd_msg); - break; - - case MEI_FLOW_CONTROL_CMD: - flow_control = (struct hbm_flow_control *) mei_msg; - mei_client_flow_control_response(dev, flow_control); - dev_dbg(&dev->pdev->dev, "client flow control response message received.\n"); - break; - - case HOST_CLIENT_PROPERTIES_RES_CMD: - props_res = (struct hbm_props_response *)mei_msg; - if (props_res->status || !dev->me_clients) { - dev_dbg(&dev->pdev->dev, "reset due to received host client properties response bus message wrong status.\n"); - mei_reset(dev, 1); - return; - } - if (dev->me_clients[dev->me_client_presentation_num] - .client_id == props_res->address) { - - dev->me_clients[dev->me_client_presentation_num].props - = props_res->client_properties; - - if (dev->dev_state == MEI_DEV_INIT_CLIENTS && - dev->init_clients_state == - MEI_CLIENT_PROPERTIES_MESSAGE) { - dev->me_client_index++; - dev->me_client_presentation_num++; - - /** Send Client Properties request **/ - res = mei_host_client_properties(dev); - if (res < 0) { - dev_dbg(&dev->pdev->dev, "mei_host_client_properties() failed"); - return; - } else if (!res) { - /* - * No more clients to send to. - * Clear Map for indicating now ME clients - * with associated host client - */ - bitmap_zero(dev->host_clients_map, MEI_CLIENTS_MAX); - dev->open_handle_count = 0; - - /* - * Reserving the first three client IDs - * Client Id 0 - Reserved for MEI Bus Message communications - * Client Id 1 - Reserved for Watchdog - * Client ID 2 - Reserved for AMTHI - */ - bitmap_set(dev->host_clients_map, 0, 3); - dev->dev_state = MEI_DEV_ENABLED; - - /* if wd initialization fails, initialization the AMTHI client, - * otherwise the AMTHI client will be initialized after the WD client connect response - * will be received - */ - if (mei_wd_host_init(dev)) - mei_host_init_iamthif(dev); - } - - } else { - dev_dbg(&dev->pdev->dev, "reset due to received host client properties response bus message"); - mei_reset(dev, 1); - return; - } - } else { - dev_dbg(&dev->pdev->dev, "reset due to received host client properties response bus message for wrong client ID\n"); - mei_reset(dev, 1); - return; - } - break; - case HOST_ENUM_RES_CMD: - enum_res = (struct hbm_host_enum_response *) mei_msg; - memcpy(dev->me_clients_map, enum_res->valid_addresses, 32); - if (dev->dev_state == MEI_DEV_INIT_CLIENTS && - dev->init_clients_state == MEI_ENUM_CLIENTS_MESSAGE) { - dev->init_clients_timer = 0; - dev->me_client_presentation_num = 0; - dev->me_client_index = 0; - mei_allocate_me_clients_storage(dev); - dev->init_clients_state = - MEI_CLIENT_PROPERTIES_MESSAGE; - mei_host_client_properties(dev); - } else { - dev_dbg(&dev->pdev->dev, "reset due to received host enumeration clients response bus message.\n"); - mei_reset(dev, 1); - return; - } - break; + cl->state = MEI_FILE_DISCONNECTING; + cl->status = 0; + cb->buf_idx = 0; + list_move_tail(&cb->list, &dev->ctrl_rd_list.list); + cl->timer_count = MEI_CONNECT_TIMEOUT; - case HOST_STOP_RES_CMD: - dev->dev_state = MEI_DEV_DISABLED; - dev_dbg(&dev->pdev->dev, "resetting because of FW stop response.\n"); - mei_reset(dev, 1); - break; - - case CLIENT_DISCONNECT_REQ_CMD: - /* search for client */ - disconnect_req = - (struct hbm_client_disconnect_request *) mei_msg; - mei_client_disconnect_request(dev, disconnect_req); - break; - - case ME_STOP_REQ_CMD: - /* prepare stop request */ - mei_hdr = (struct mei_msg_hdr *) &dev->ext_msg_buf[0]; - mei_hdr->host_addr = 0; - mei_hdr->me_addr = 0; - mei_hdr->length = sizeof(struct hbm_host_stop_request); - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - host_stop_req = - (struct hbm_host_stop_request *) &dev->ext_msg_buf[1]; - memset(host_stop_req, 0, sizeof(struct hbm_host_stop_request)); - host_stop_req->hbm_cmd = HOST_STOP_REQ_CMD; - host_stop_req->reason = DRIVER_STOP_REQUEST; - host_stop_req->reserved[0] = 0; - host_stop_req->reserved[1] = 0; - dev->extra_write_index = 2; - break; - - default: - BUG(); - break; - - } + return 0; } /** - * _mei_hb_read - processes read related operation. + * mei_cl_irq_close - processes client read related operation from the + * interrupt thread context - request for flow control credits * - * @dev: the device structure. - * @slots: free slots. - * @cb_pos: callback block. - * @cl: private data of the file object. + * @cl: client + * @cb: callback block. * @cmpl_list: complete list. * * returns 0, OK; otherwise, error. */ -static int _mei_irq_thread_read(struct mei_device *dev, s32 *slots, - struct mei_cl_cb *cb_pos, - struct mei_cl *cl, - struct mei_io_list *cmpl_list) +static int mei_cl_irq_read(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list) { - if ((*slots * sizeof(u32)) < (sizeof(struct mei_msg_hdr) + - sizeof(struct hbm_flow_control))) { - /* return the cancel routine */ - list_del(&cb_pos->cb_list); - return -EBADMSG; - } + struct mei_device *dev = cl->dev; + u32 msg_slots; + int slots; + int ret; + + msg_slots = mei_data2slots(sizeof(struct hbm_flow_control)); + slots = mei_hbuf_empty_slots(dev); - *slots -= mei_data2slots(sizeof(struct hbm_flow_control)); + if (slots < msg_slots) + return -EMSGSIZE; - if (mei_send_flow_control(dev, cl)) { - cl->status = -ENODEV; - cb_pos->information = 0; - list_move_tail(&cb_pos->cb_list, &cmpl_list->mei_cb.cb_list); - return -ENODEV; + ret = mei_hbm_cl_flow_control_req(dev, cl); + if (ret) { + cl->status = ret; + cb->buf_idx = 0; + list_move_tail(&cb->list, &cmpl_list->list); + return ret; } - list_move_tail(&cb_pos->cb_list, &dev->read_list.mei_cb.cb_list); + + list_move_tail(&cb->list, &dev->read_list.list); return 0; } /** - * _mei_irq_thread_ioctl - processes ioctl related operation. + * mei_cl_irq_connect - send connect request in irq_thread context * - * @dev: the device structure. - * @slots: free slots. - * @cb_pos: callback block. - * @cl: private data of the file object. + * @cl: client + * @cb: callback block. * @cmpl_list: complete list. * * returns 0, OK; otherwise, error. */ -static int _mei_irq_thread_ioctl(struct mei_device *dev, s32 *slots, - struct mei_cl_cb *cb_pos, - struct mei_cl *cl, - struct mei_io_list *cmpl_list) +static int mei_cl_irq_connect(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list) { - if ((*slots * sizeof(u32)) < (sizeof(struct mei_msg_hdr) + - sizeof(struct hbm_client_connect_request))) { - /* return the cancel routine */ - list_del(&cb_pos->cb_list); - return -EBADMSG; - } + struct mei_device *dev = cl->dev; + u32 msg_slots; + int slots; + int ret; - cl->state = MEI_FILE_CONNECTING; - *slots -= mei_data2slots(sizeof(struct hbm_client_connect_request)); - if (mei_connect(dev, cl)) { - cl->status = -ENODEV; - cb_pos->information = 0; - list_del(&cb_pos->cb_list); - return -ENODEV; - } else { - list_move_tail(&cb_pos->cb_list, - &dev->ctrl_rd_list.mei_cb.cb_list); - cl->timer_count = MEI_CONNECT_TIMEOUT; - } - return 0; -} + msg_slots = mei_data2slots(sizeof(struct hbm_client_connect_request)); + slots = mei_hbuf_empty_slots(dev); -/** - * _mei_irq_thread_cmpl - processes completed and no-iamthif operation. - * - * @dev: the device structure. - * @slots: free slots. - * @cb_pos: callback block. - * @cl: private data of the file object. - * @cmpl_list: complete list. - * - * returns 0, OK; otherwise, error. - */ -static int _mei_irq_thread_cmpl(struct mei_device *dev, s32 *slots, - struct mei_cl_cb *cb_pos, - struct mei_cl *cl, - struct mei_io_list *cmpl_list) -{ - struct mei_msg_hdr *mei_hdr; + if (mei_cl_is_other_connecting(cl)) + return 0; - if ((*slots * sizeof(u32)) >= (sizeof(struct mei_msg_hdr) + - (cb_pos->request_buffer.size - - cb_pos->information))) { - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = cl->host_client_id; - mei_hdr->me_addr = cl->me_client_id; - mei_hdr->length = cb_pos->request_buffer.size - - cb_pos->information; - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - dev_dbg(&dev->pdev->dev, "cb_pos->request_buffer.size =%d" - "mei_hdr->msg_complete = %d\n", - cb_pos->request_buffer.size, - mei_hdr->msg_complete); - dev_dbg(&dev->pdev->dev, "cb_pos->information =%lu\n", - cb_pos->information); - dev_dbg(&dev->pdev->dev, "mei_hdr->length =%d\n", - mei_hdr->length); - *slots -= mei_data2slots(mei_hdr->length); - if (mei_write_message(dev, mei_hdr, - (unsigned char *) - (cb_pos->request_buffer.data + - cb_pos->information), - mei_hdr->length)) { - cl->status = -ENODEV; - list_move_tail(&cb_pos->cb_list, - &cmpl_list->mei_cb.cb_list); - return -ENODEV; - } else { - if (mei_flow_ctrl_reduce(dev, cl)) - return -ENODEV; - cl->status = 0; - cb_pos->information += mei_hdr->length; - list_move_tail(&cb_pos->cb_list, - &dev->write_waiting_list.mei_cb.cb_list); - } - } else if (*slots == dev->hbuf_depth) { - /* buffer is still empty */ - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = cl->host_client_id; - mei_hdr->me_addr = cl->me_client_id; - mei_hdr->length = - (*slots * sizeof(u32)) - sizeof(struct mei_msg_hdr); - mei_hdr->msg_complete = 0; - mei_hdr->reserved = 0; - *slots -= mei_data2slots(mei_hdr->length); - if (mei_write_message(dev, mei_hdr, - (unsigned char *) - (cb_pos->request_buffer.data + - cb_pos->information), - mei_hdr->length)) { - cl->status = -ENODEV; - list_move_tail(&cb_pos->cb_list, - &cmpl_list->mei_cb.cb_list); - return -ENODEV; - } else { - cb_pos->information += mei_hdr->length; - dev_dbg(&dev->pdev->dev, - "cb_pos->request_buffer.size =%d" - " mei_hdr->msg_complete = %d\n", - cb_pos->request_buffer.size, - mei_hdr->msg_complete); - dev_dbg(&dev->pdev->dev, "cb_pos->information =%lu\n", - cb_pos->information); - dev_dbg(&dev->pdev->dev, "mei_hdr->length =%d\n", - mei_hdr->length); - } + if (slots < msg_slots) return -EMSGSIZE; - } else { - return -EBADMSG; - } - - return 0; -} - -/** - * _mei_irq_thread_cmpl_iamthif - processes completed iamthif operation. - * - * @dev: the device structure. - * @slots: free slots. - * @cb_pos: callback block. - * @cl: private data of the file object. - * @cmpl_list: complete list. - * - * returns 0, OK; otherwise, error. - */ -static int _mei_irq_thread_cmpl_iamthif(struct mei_device *dev, s32 *slots, - struct mei_cl_cb *cb_pos, - struct mei_cl *cl, - struct mei_io_list *cmpl_list) -{ - struct mei_msg_hdr *mei_hdr; - if ((*slots * sizeof(u32)) >= (sizeof(struct mei_msg_hdr) + - dev->iamthif_msg_buf_size - - dev->iamthif_msg_buf_index)) { - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = cl->host_client_id; - mei_hdr->me_addr = cl->me_client_id; - mei_hdr->length = dev->iamthif_msg_buf_size - - dev->iamthif_msg_buf_index; - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; - - *slots -= mei_data2slots(mei_hdr->length); - - if (mei_write_message(dev, mei_hdr, - (dev->iamthif_msg_buf + - dev->iamthif_msg_buf_index), - mei_hdr->length)) { - dev->iamthif_state = MEI_IAMTHIF_IDLE; - cl->status = -ENODEV; - list_del(&cb_pos->cb_list); - return -ENODEV; - } else { - if (mei_flow_ctrl_reduce(dev, cl)) - return -ENODEV; - dev->iamthif_msg_buf_index += mei_hdr->length; - cb_pos->information = dev->iamthif_msg_buf_index; - cl->status = 0; - dev->iamthif_state = MEI_IAMTHIF_FLOW_CONTROL; - dev->iamthif_flow_control_pending = true; - /* save iamthif cb sent to amthi client */ - dev->iamthif_current_cb = cb_pos; - list_move_tail(&cb_pos->cb_list, - &dev->write_waiting_list.mei_cb.cb_list); + cl->state = MEI_FILE_CONNECTING; - } - } else if (*slots == dev->hbuf_depth) { - /* buffer is still empty */ - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = cl->host_client_id; - mei_hdr->me_addr = cl->me_client_id; - mei_hdr->length = - (*slots * sizeof(u32)) - sizeof(struct mei_msg_hdr); - mei_hdr->msg_complete = 0; - mei_hdr->reserved = 0; - - *slots -= mei_data2slots(mei_hdr->length); - - if (mei_write_message(dev, mei_hdr, - (dev->iamthif_msg_buf + - dev->iamthif_msg_buf_index), - mei_hdr->length)) { - cl->status = -ENODEV; - list_del(&cb_pos->cb_list); - } else { - dev->iamthif_msg_buf_index += mei_hdr->length; - } - return -EMSGSIZE; - } else { - return -EBADMSG; + ret = mei_hbm_cl_connect_req(dev, cl); + if (ret) { + cl->status = ret; + cb->buf_idx = 0; + list_del(&cb->list); + return ret; } + list_move_tail(&cb->list, &dev->ctrl_rd_list.list); + cl->timer_count = MEI_CONNECT_TIMEOUT; return 0; } + /** - * mei_irq_thread_read_handler - bottom half read routine after ISR to + * mei_irq_read_handler - bottom half read routine after ISR to * handle the read processing. * - * @cmpl_list: An instance of our list structure * @dev: the device structure + * @cmpl_list: An instance of our list structure * @slots: slots to read. * * returns 0 on success, <0 on failure. */ -static int mei_irq_thread_read_handler(struct mei_io_list *cmpl_list, - struct mei_device *dev, - s32 *slots) +int mei_irq_read_handler(struct mei_device *dev, + struct mei_cl_cb *cmpl_list, s32 *slots) { struct mei_msg_hdr *mei_hdr; - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; - int ret = 0; + struct mei_cl *cl; + int ret; if (!dev->rd_msg_hdr) { - dev->rd_msg_hdr = mei_mecbrw_read(dev); - dev_dbg(&dev->pdev->dev, "slots =%08x.\n", *slots); + dev->rd_msg_hdr = mei_read_hdr(dev); (*slots)--; dev_dbg(&dev->pdev->dev, "slots =%08x.\n", *slots); } mei_hdr = (struct mei_msg_hdr *) &dev->rd_msg_hdr; - dev_dbg(&dev->pdev->dev, "mei_hdr->length =%d\n", mei_hdr->length); + dev_dbg(&dev->pdev->dev, MEI_HDR_FMT, MEI_HDR_PRM(mei_hdr)); if (mei_hdr->reserved || !dev->rd_msg_hdr) { - dev_dbg(&dev->pdev->dev, "corrupted message header.\n"); + dev_err(&dev->pdev->dev, "corrupted message header 0x%08X\n", + dev->rd_msg_hdr); ret = -EBADMSG; goto end; } - if (mei_hdr->host_addr || mei_hdr->me_addr) { - list_for_each_entry_safe(cl_pos, cl_next, - &dev->file_list, link) { - dev_dbg(&dev->pdev->dev, - "list_for_each_entry_safe read host" - " client = %d, ME client = %d\n", - cl_pos->host_client_id, - cl_pos->me_client_id); - if (cl_pos->host_client_id == mei_hdr->host_addr && - cl_pos->me_client_id == mei_hdr->me_addr) - break; - } + if (mei_slots2data(*slots) < mei_hdr->length) { + dev_err(&dev->pdev->dev, "less data available than length=%08x.\n", + *slots); + /* we can't read the message */ + ret = -ENODATA; + goto end; + } - if (&cl_pos->link == &dev->file_list) { - dev_dbg(&dev->pdev->dev, "corrupted message header\n"); - ret = -EBADMSG; + /* HBM message */ + if (mei_hdr->host_addr == 0 && mei_hdr->me_addr == 0) { + ret = mei_hbm_dispatch(dev, mei_hdr); + if (ret) { + dev_dbg(&dev->pdev->dev, "mei_hbm_dispatch failed ret = %d\n", + ret); goto end; } + goto reset_slots; } - if (((*slots) * sizeof(u32)) < mei_hdr->length) { - dev_dbg(&dev->pdev->dev, - "we can't read the message slots =%08x.\n", - *slots); - /* we can't read the message */ - ret = -ERANGE; + + /* find recipient cl */ + list_for_each_entry(cl, &dev->file_list, link) { + if (mei_cl_hbm_equal(cl, mei_hdr)) { + cl_dbg(dev, cl, "got a message\n"); + break; + } + } + + /* if no recipient cl was found we assume corrupted header */ + if (&cl->link == &dev->file_list) { + dev_err(&dev->pdev->dev, "no destination client found 0x%08X\n", + dev->rd_msg_hdr); + ret = -EBADMSG; goto end; } - /* decide where to read the message too */ - if (!mei_hdr->host_addr) { - dev_dbg(&dev->pdev->dev, "call mei_irq_thread_read_bus_message.\n"); - mei_irq_thread_read_bus_message(dev, mei_hdr); - dev_dbg(&dev->pdev->dev, "end mei_irq_thread_read_bus_message.\n"); - } else if (mei_hdr->host_addr == dev->iamthif_cl.host_client_id && - (MEI_FILE_CONNECTED == dev->iamthif_cl.state) && - (dev->iamthif_state == MEI_IAMTHIF_READING)) { - dev_dbg(&dev->pdev->dev, "call mei_irq_thread_read_iamthif_message.\n"); - dev_dbg(&dev->pdev->dev, "mei_hdr->length =%d\n", - mei_hdr->length); - ret = mei_irq_thread_read_amthi_message(cmpl_list, - dev, mei_hdr); - if (ret) - goto end; + if (mei_hdr->host_addr == dev->iamthif_cl.host_client_id && + MEI_FILE_CONNECTED == dev->iamthif_cl.state && + dev->iamthif_state == MEI_IAMTHIF_READING) { + ret = mei_amthif_irq_read_msg(dev, mei_hdr, cmpl_list); + if (ret) { + dev_err(&dev->pdev->dev, "mei_amthif_irq_read_msg failed = %d\n", + ret); + goto end; + } } else { - dev_dbg(&dev->pdev->dev, "call mei_irq_thread_read_client_message.\n"); - ret = mei_irq_thread_read_client_message(cmpl_list, - dev, mei_hdr); - if (ret) + ret = mei_cl_irq_read_msg(dev, mei_hdr, cmpl_list); + if (ret) { + dev_err(&dev->pdev->dev, "mei_cl_irq_read_msg failed = %d\n", + ret); goto end; - + } } +reset_slots: /* reset the number of slots and header */ *slots = mei_count_full_read_slots(dev); dev->rd_msg_hdr = 0; if (*slots == -EOVERFLOW) { /* overflow - reset */ - dev_dbg(&dev->pdev->dev, "resetting due to slots overflow.\n"); + dev_err(&dev->pdev->dev, "resetting due to slots overflow.\n"); /* set the event since message has been read */ ret = -ERANGE; goto end; @@ -1158,59 +415,56 @@ static int mei_irq_thread_read_handler(struct mei_io_list *cmpl_list, end: return ret; } +EXPORT_SYMBOL_GPL(mei_irq_read_handler); /** - * mei_irq_thread_write_handler - bottom half write routine after - * ISR to handle the write processing. + * mei_irq_write_handler - dispatch write requests + * after irq received * - * @cmpl_list: An instance of our list structure * @dev: the device structure - * @slots: slots to write. + * @cmpl_list: An instance of our list structure * * returns 0 on success, <0 on failure. */ -static int mei_irq_thread_write_handler(struct mei_io_list *cmpl_list, - struct mei_device *dev, - s32 *slots) +int mei_irq_write_handler(struct mei_device *dev, struct mei_cl_cb *cmpl_list) { struct mei_cl *cl; - struct mei_cl_cb *pos = NULL, *next = NULL; - struct mei_io_list *list; + struct mei_cl_cb *cb, *next; + struct mei_cl_cb *list; + s32 slots; int ret; - if (!mei_hbuf_is_empty(dev)) { - dev_dbg(&dev->pdev->dev, "host buffer is not empty.\n"); + + if (!mei_hbuf_acquire(dev)) return 0; - } - *slots = mei_hbuf_empty_slots(dev); - if (*slots <= 0) + + slots = mei_hbuf_empty_slots(dev); + if (slots <= 0) return -EMSGSIZE; /* complete all waiting for write CB */ dev_dbg(&dev->pdev->dev, "complete all waiting for write cb.\n"); list = &dev->write_waiting_list; - list_for_each_entry_safe(pos, next, &list->mei_cb.cb_list, cb_list) { - cl = (struct mei_cl *)pos->file_private; + list_for_each_entry_safe(cb, next, &list->list, list) { + cl = cb->cl; if (cl == NULL) continue; cl->status = 0; - list_del(&pos->cb_list); - if (MEI_WRITING == cl->writing_state && - (pos->major_file_operations == MEI_WRITE) && - (cl != &dev->iamthif_cl)) { - dev_dbg(&dev->pdev->dev, "MEI WRITE COMPLETE\n"); + list_del(&cb->list); + if (cb->fop_type == MEI_FOP_WRITE && + cl != &dev->iamthif_cl) { + cl_dbg(dev, cl, "MEI WRITE COMPLETE\n"); cl->writing_state = MEI_WRITE_COMPLETE; - list_add_tail(&pos->cb_list, - &cmpl_list->mei_cb.cb_list); + list_add_tail(&cb->list, &cmpl_list->list); } if (cl == &dev->iamthif_cl) { - dev_dbg(&dev->pdev->dev, "check iamthif flow control.\n"); + cl_dbg(dev, cl, "check iamthif flow control.\n"); if (dev->iamthif_flow_control_pending) { - ret = _mei_irq_thread_iamthif_read(dev, slots); + ret = mei_amthif_irq_read(dev, &slots); if (ret) return ret; } @@ -1219,70 +473,55 @@ static int mei_irq_thread_write_handler(struct mei_io_list *cmpl_list, if (dev->wd_state == MEI_WD_STOPPING) { dev->wd_state = MEI_WD_IDLE; - wake_up_interruptible(&dev->wait_stop_wd); + wake_up(&dev->wait_stop_wd); } - if (dev->extra_write_index) { - dev_dbg(&dev->pdev->dev, "extra_write_index =%d.\n", - dev->extra_write_index); - mei_write_message(dev, - (struct mei_msg_hdr *) &dev->ext_msg_buf[0], - (unsigned char *) &dev->ext_msg_buf[1], - (dev->extra_write_index - 1) * sizeof(u32)); - *slots -= dev->extra_write_index; - dev->extra_write_index = 0; - } - if (dev->dev_state == MEI_DEV_ENABLED) { + if (mei_cl_is_connected(&dev->wd_cl)) { if (dev->wd_pending && - mei_flow_ctrl_creds(dev, &dev->wd_cl) > 0) { - if (mei_wd_send(dev)) - dev_dbg(&dev->pdev->dev, "wd send failed.\n"); - else if (mei_flow_ctrl_reduce(dev, &dev->wd_cl)) - return -ENODEV; - + mei_cl_flow_ctrl_creds(&dev->wd_cl) > 0) { + ret = mei_wd_send(dev); + if (ret) + return ret; dev->wd_pending = false; - - if (dev->wd_state == MEI_WD_RUNNING) - *slots -= mei_data2slots(MEI_WD_START_MSG_SIZE); - else - *slots -= mei_data2slots(MEI_WD_STOP_MSG_SIZE); } } /* complete control write list CB */ dev_dbg(&dev->pdev->dev, "complete control write list cb.\n"); - list_for_each_entry_safe(pos, next, - &dev->ctrl_wr_list.mei_cb.cb_list, cb_list) { - cl = (struct mei_cl *) pos->file_private; + list_for_each_entry_safe(cb, next, &dev->ctrl_wr_list.list, list) { + cl = cb->cl; if (!cl) { - list_del(&pos->cb_list); + list_del(&cb->list); return -ENODEV; } - switch (pos->major_file_operations) { - case MEI_CLOSE: + switch (cb->fop_type) { + case MEI_FOP_CLOSE: /* send disconnect message */ - ret = _mei_irq_thread_close(dev, slots, pos, cl, cmpl_list); + ret = mei_cl_irq_close(cl, cb, cmpl_list); if (ret) return ret; break; - case MEI_READ: + case MEI_FOP_READ: /* send flow control message */ - ret = _mei_irq_thread_read(dev, slots, pos, cl, cmpl_list); + ret = mei_cl_irq_read(cl, cb, cmpl_list); if (ret) return ret; break; - case MEI_IOCTL: + case MEI_FOP_CONNECT: /* connect message */ - if (mei_other_client_is_connecting(dev, cl)) - continue; - ret = _mei_irq_thread_ioctl(dev, slots, pos, cl, cmpl_list); + ret = mei_cl_irq_connect(cl, cb, cmpl_list); if (ret) return ret; break; - + case MEI_FOP_DISCONNECT_RSP: + /* send disconnect resp */ + ret = mei_cl_irq_disconnect_rsp(cl, cb, cmpl_list); + if (ret) + return ret; + break; default: BUG(); } @@ -1290,43 +529,20 @@ static int mei_irq_thread_write_handler(struct mei_io_list *cmpl_list, } /* complete write list CB */ dev_dbg(&dev->pdev->dev, "complete write list cb.\n"); - list_for_each_entry_safe(pos, next, - &dev->write_list.mei_cb.cb_list, cb_list) { - cl = (struct mei_cl *)pos->file_private; + list_for_each_entry_safe(cb, next, &dev->write_list.list, list) { + cl = cb->cl; if (cl == NULL) continue; - - if (cl != &dev->iamthif_cl) { - if (mei_flow_ctrl_creds(dev, cl) <= 0) { - dev_dbg(&dev->pdev->dev, - "No flow control credentials for client %d, not sending.\n", - cl->host_client_id); - continue; - } - ret = _mei_irq_thread_cmpl(dev, slots, pos, - cl, cmpl_list); - if (ret) - return ret; - - } else if (cl == &dev->iamthif_cl) { - /* IAMTHIF IOCTL */ - dev_dbg(&dev->pdev->dev, "complete amthi write cb.\n"); - if (mei_flow_ctrl_creds(dev, cl) <= 0) { - dev_dbg(&dev->pdev->dev, - "No flow control credentials for amthi client %d.\n", - cl->host_client_id); - continue; - } - ret = _mei_irq_thread_cmpl_iamthif(dev, slots, pos, - cl, cmpl_list); - if (ret) - return ret; - - } - + if (cl == &dev->iamthif_cl) + ret = mei_amthif_irq_write(cl, cb, cmpl_list); + else + ret = mei_cl_irq_write(cl, cb, cmpl_list); + if (ret) + return ret; } return 0; } +EXPORT_SYMBOL_GPL(mei_irq_write_handler); @@ -1335,14 +551,11 @@ static int mei_irq_thread_write_handler(struct mei_io_list *cmpl_list, * * @work: pointer to the work_struct structure * - * NOTE: This function is called by timer interrupt work */ void mei_timer(struct work_struct *work) { unsigned long timeout; - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; - struct list_head *amthi_complete_list = NULL; + struct mei_cl *cl; struct mei_cl_cb *cb_pos = NULL; struct mei_cl_cb *cb_next = NULL; @@ -1351,33 +564,42 @@ void mei_timer(struct work_struct *work) mutex_lock(&dev->device_lock); - if (dev->dev_state != MEI_DEV_ENABLED) { - if (dev->dev_state == MEI_DEV_INIT_CLIENTS) { - if (dev->init_clients_timer) { - if (--dev->init_clients_timer == 0) { - dev_dbg(&dev->pdev->dev, "IMEI reset due to init clients timeout ,init clients state = %d.\n", - dev->init_clients_state); - mei_reset(dev, 1); - } + + /* Catch interrupt stalls during HBM init handshake */ + if (dev->dev_state == MEI_DEV_INIT_CLIENTS && + dev->hbm_state != MEI_HBM_IDLE) { + + if (dev->init_clients_timer) { + if (--dev->init_clients_timer == 0) { + dev_err(&dev->pdev->dev, "timer: init clients timeout hbm_state = %d.\n", + dev->hbm_state); + mei_reset(dev); + goto out; } } - goto out; } + + if (dev->dev_state != MEI_DEV_ENABLED) + goto out; + /*** connect/disconnect timeouts ***/ - list_for_each_entry_safe(cl_pos, cl_next, &dev->file_list, link) { - if (cl_pos->timer_count) { - if (--cl_pos->timer_count == 0) { - dev_dbg(&dev->pdev->dev, "HECI reset due to connect/disconnect timeout.\n"); - mei_reset(dev, 1); + list_for_each_entry(cl, &dev->file_list, link) { + if (cl->timer_count) { + if (--cl->timer_count == 0) { + dev_err(&dev->pdev->dev, "timer: connect/disconnect timeout.\n"); + mei_reset(dev); goto out; } } } + if (!mei_cl_is_connected(&dev->iamthif_cl)) + goto out; + if (dev->iamthif_stall_timer) { if (--dev->iamthif_stall_timer == 0) { - dev_dbg(&dev->pdev->dev, "resetting because of hang to amthi.\n"); - mei_reset(dev, 1); + dev_err(&dev->pdev->dev, "timer: amthif hanged.\n"); + mei_reset(dev); dev->iamthif_msg_buf_size = 0; dev->iamthif_msg_buf_index = 0; dev->iamthif_canceled = false; @@ -1385,19 +607,18 @@ void mei_timer(struct work_struct *work) dev->iamthif_state = MEI_IAMTHIF_IDLE; dev->iamthif_timer = 0; - if (dev->iamthif_current_cb) - mei_free_cb_private(dev->iamthif_current_cb); + mei_io_cb_free(dev->iamthif_current_cb); + dev->iamthif_current_cb = NULL; dev->iamthif_file_object = NULL; - dev->iamthif_current_cb = NULL; - mei_run_next_iamthif_cmd(dev); + mei_amthif_run_next_cmd(dev); } } if (dev->iamthif_timer) { timeout = dev->iamthif_timer + - msecs_to_jiffies(IAMTHIF_READ_TIMER); + mei_secs_to_jiffies(MEI_IAMTHIF_READ_TIMER); dev_dbg(&dev->pdev->dev, "dev->iamthif_timer = %ld\n", dev->iamthif_timer); @@ -1411,143 +632,28 @@ void mei_timer(struct work_struct *work) dev_dbg(&dev->pdev->dev, "freeing AMTHI for other requests\n"); - amthi_complete_list = &dev->amthi_read_complete_list. - mei_cb.cb_list; - - list_for_each_entry_safe(cb_pos, cb_next, amthi_complete_list, cb_list) { + list_for_each_entry_safe(cb_pos, cb_next, + &dev->amthif_rd_complete_list.list, list) { - cl_pos = cb_pos->file_object->private_data; + cl = cb_pos->file_object->private_data; /* Finding the AMTHI entry. */ - if (cl_pos == &dev->iamthif_cl) - list_del(&cb_pos->cb_list); + if (cl == &dev->iamthif_cl) + list_del(&cb_pos->list); } - if (dev->iamthif_current_cb) - mei_free_cb_private(dev->iamthif_current_cb); + mei_io_cb_free(dev->iamthif_current_cb); + dev->iamthif_current_cb = NULL; dev->iamthif_file_object->private_data = NULL; dev->iamthif_file_object = NULL; - dev->iamthif_current_cb = NULL; dev->iamthif_timer = 0; - mei_run_next_iamthif_cmd(dev); + mei_amthif_run_next_cmd(dev); } } out: - schedule_delayed_work(&dev->timer_work, 2 * HZ); + if (dev->dev_state != MEI_DEV_DISABLED) + schedule_delayed_work(&dev->timer_work, 2 * HZ); mutex_unlock(&dev->device_lock); } -/** - * mei_interrupt_thread_handler - function called after ISR to handle the interrupt - * processing. - * - * @irq: The irq number - * @dev_id: pointer to the device structure - * - * returns irqreturn_t - * - */ -irqreturn_t mei_interrupt_thread_handler(int irq, void *dev_id) -{ - struct mei_device *dev = (struct mei_device *) dev_id; - struct mei_io_list complete_list; - struct mei_cl_cb *cb_pos = NULL, *cb_next = NULL; - struct mei_cl *cl; - s32 slots; - int rets; - bool bus_message_received; - - - dev_dbg(&dev->pdev->dev, "function called after ISR to handle the interrupt processing.\n"); - /* initialize our complete list */ - mutex_lock(&dev->device_lock); - mei_io_list_init(&complete_list); - dev->host_hw_state = mei_hcsr_read(dev); - - /* Ack the interrupt here - * In case of MSI we don't go through the quick handler */ - if (pci_dev_msi_enabled(dev->pdev)) - mei_reg_write(dev, H_CSR, dev->host_hw_state); - - dev->me_hw_state = mei_mecsr_read(dev); - - /* check if ME wants a reset */ - if ((dev->me_hw_state & ME_RDY_HRA) == 0 && - dev->dev_state != MEI_DEV_RESETING && - dev->dev_state != MEI_DEV_INITIALIZING) { - dev_dbg(&dev->pdev->dev, "FW not ready.\n"); - mei_reset(dev, 1); - mutex_unlock(&dev->device_lock); - return IRQ_HANDLED; - } - - /* check if we need to start the dev */ - if ((dev->host_hw_state & H_RDY) == 0) { - if ((dev->me_hw_state & ME_RDY_HRA) == ME_RDY_HRA) { - dev_dbg(&dev->pdev->dev, "we need to start the dev.\n"); - dev->host_hw_state |= (H_IE | H_IG | H_RDY); - mei_hcsr_set(dev); - dev->dev_state = MEI_DEV_INIT_CLIENTS; - dev_dbg(&dev->pdev->dev, "link is established start sending messages.\n"); - /* link is established - * start sending messages. - */ - mei_host_start_message(dev); - mutex_unlock(&dev->device_lock); - return IRQ_HANDLED; - } else { - dev_dbg(&dev->pdev->dev, "FW not ready.\n"); - mutex_unlock(&dev->device_lock); - return IRQ_HANDLED; - } - } - /* check slots available for reading */ - slots = mei_count_full_read_slots(dev); - dev_dbg(&dev->pdev->dev, "slots =%08x extra_write_index =%08x.\n", - slots, dev->extra_write_index); - while (slots > 0 && !dev->extra_write_index) { - dev_dbg(&dev->pdev->dev, "slots =%08x extra_write_index =%08x.\n", - slots, dev->extra_write_index); - dev_dbg(&dev->pdev->dev, "call mei_irq_thread_read_handler.\n"); - rets = mei_irq_thread_read_handler(&complete_list, dev, &slots); - if (rets) - goto end; - } - rets = mei_irq_thread_write_handler(&complete_list, dev, &slots); -end: - dev_dbg(&dev->pdev->dev, "end of bottom half function.\n"); - dev->host_hw_state = mei_hcsr_read(dev); - dev->mei_host_buffer_is_empty = mei_hbuf_is_empty(dev); - - bus_message_received = false; - if (dev->recvd_msg && waitqueue_active(&dev->wait_recvd_msg)) { - dev_dbg(&dev->pdev->dev, "received waiting bus message\n"); - bus_message_received = true; - } - mutex_unlock(&dev->device_lock); - if (bus_message_received) { - dev_dbg(&dev->pdev->dev, "wake up dev->wait_recvd_msg\n"); - wake_up_interruptible(&dev->wait_recvd_msg); - bus_message_received = false; - } - if (list_empty(&complete_list.mei_cb.cb_list)) - return IRQ_HANDLED; - - - list_for_each_entry_safe(cb_pos, cb_next, - &complete_list.mei_cb.cb_list, cb_list) { - cl = (struct mei_cl *)cb_pos->file_private; - list_del(&cb_pos->cb_list); - if (cl) { - if (cl != &dev->iamthif_cl) { - dev_dbg(&dev->pdev->dev, "completing call back.\n"); - _mei_cmpl(cl, cb_pos); - cb_pos = NULL; - } else if (cl == &dev->iamthif_cl) { - _mei_cmpl_iamthif(dev, cb_pos); - } - } - } - return IRQ_HANDLED; -} diff --git a/drivers/misc/mei/iorw.c b/drivers/misc/mei/iorw.c deleted file mode 100644 index fcba98eb892..00000000000 --- a/drivers/misc/mei/iorw.c +++ /dev/null @@ -1,591 +0,0 @@ -/* - * - * Intel Management Engine Interface (Intel MEI) Linux driver - * Copyright (c) 2003-2012, Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - */ - - -#include <linux/kernel.h> -#include <linux/fs.h> -#include <linux/errno.h> -#include <linux/types.h> -#include <linux/fcntl.h> -#include <linux/aio.h> -#include <linux/pci.h> -#include <linux/init.h> -#include <linux/ioctl.h> -#include <linux/cdev.h> -#include <linux/list.h> -#include <linux/delay.h> -#include <linux/sched.h> -#include <linux/uuid.h> -#include <linux/jiffies.h> -#include <linux/uaccess.h> - - -#include "mei_dev.h" -#include "hw.h" -#include <linux/mei.h> -#include "interface.h" - -/** - * mei_me_cl_by_id return index to me_clients for client_id - * - * @dev: the device structure - * @client_id: me client id - * - * Locking: called under "dev->device_lock" lock - * - * returns index on success, -ENOENT on failure. - */ - -int mei_me_cl_by_id(struct mei_device *dev, u8 client_id) -{ - int i; - for (i = 0; i < dev->me_clients_num; i++) - if (dev->me_clients[i].client_id == client_id) - break; - if (WARN_ON(dev->me_clients[i].client_id != client_id)) - return -ENOENT; - - if (i == dev->me_clients_num) - return -ENOENT; - - return i; -} - -/** - * mei_ioctl_connect_client - the connect to fw client IOCTL function - * - * @dev: the device structure - * @data: IOCTL connect data, input and output parameters - * @file: private data of the file object - * - * Locking: called under "dev->device_lock" lock - * - * returns 0 on success, <0 on failure. - */ -int mei_ioctl_connect_client(struct file *file, - struct mei_connect_client_data *data) -{ - struct mei_device *dev; - struct mei_cl_cb *cb; - struct mei_client *client; - struct mei_cl *cl; - struct mei_cl *cl_pos = NULL; - struct mei_cl *cl_next = NULL; - long timeout = CONNECT_TIMEOUT; - int i; - int err; - int rets; - - cl = file->private_data; - if (WARN_ON(!cl || !cl->dev)) - return -ENODEV; - - dev = cl->dev; - - dev_dbg(&dev->pdev->dev, "mei_ioctl_connect_client() Entry\n"); - - - /* buffered ioctl cb */ - cb = kzalloc(sizeof(struct mei_cl_cb), GFP_KERNEL); - if (!cb) { - rets = -ENOMEM; - goto end; - } - INIT_LIST_HEAD(&cb->cb_list); - - cb->major_file_operations = MEI_IOCTL; - - if (dev->dev_state != MEI_DEV_ENABLED) { - rets = -ENODEV; - goto end; - } - if (cl->state != MEI_FILE_INITIALIZING && - cl->state != MEI_FILE_DISCONNECTED) { - rets = -EBUSY; - goto end; - } - - /* find ME client we're trying to connect to */ - i = mei_me_cl_by_uuid(dev, &data->in_client_uuid); - if (i >= 0 && !dev->me_clients[i].props.fixed_address) { - cl->me_client_id = dev->me_clients[i].client_id; - cl->state = MEI_FILE_CONNECTING; - } - - dev_dbg(&dev->pdev->dev, "Connect to FW Client ID = %d\n", - cl->me_client_id); - dev_dbg(&dev->pdev->dev, "FW Client - Protocol Version = %d\n", - dev->me_clients[i].props.protocol_version); - dev_dbg(&dev->pdev->dev, "FW Client - Max Msg Len = %d\n", - dev->me_clients[i].props.max_msg_length); - - /* if we're connecting to amthi client then we will use the - * existing connection - */ - if (uuid_le_cmp(data->in_client_uuid, mei_amthi_guid) == 0) { - dev_dbg(&dev->pdev->dev, "FW Client is amthi\n"); - if (dev->iamthif_cl.state != MEI_FILE_CONNECTED) { - rets = -ENODEV; - goto end; - } - clear_bit(cl->host_client_id, dev->host_clients_map); - list_for_each_entry_safe(cl_pos, cl_next, - &dev->file_list, link) { - if (mei_cl_cmp_id(cl, cl_pos)) { - dev_dbg(&dev->pdev->dev, - "remove file private data node host" - " client = %d, ME client = %d.\n", - cl_pos->host_client_id, - cl_pos->me_client_id); - list_del(&cl_pos->link); - } - - } - dev_dbg(&dev->pdev->dev, "free file private data memory.\n"); - kfree(cl); - - cl = NULL; - file->private_data = &dev->iamthif_cl; - - client = &data->out_client_properties; - client->max_msg_length = - dev->me_clients[i].props.max_msg_length; - client->protocol_version = - dev->me_clients[i].props.protocol_version; - rets = dev->iamthif_cl.status; - - goto end; - } - - if (cl->state != MEI_FILE_CONNECTING) { - rets = -ENODEV; - goto end; - } - - - /* prepare the output buffer */ - client = &data->out_client_properties; - client->max_msg_length = dev->me_clients[i].props.max_msg_length; - client->protocol_version = dev->me_clients[i].props.protocol_version; - dev_dbg(&dev->pdev->dev, "Can connect?\n"); - if (dev->mei_host_buffer_is_empty - && !mei_other_client_is_connecting(dev, cl)) { - dev_dbg(&dev->pdev->dev, "Sending Connect Message\n"); - dev->mei_host_buffer_is_empty = false; - if (mei_connect(dev, cl)) { - dev_dbg(&dev->pdev->dev, "Sending connect message - failed\n"); - rets = -ENODEV; - goto end; - } else { - dev_dbg(&dev->pdev->dev, "Sending connect message - succeeded\n"); - cl->timer_count = MEI_CONNECT_TIMEOUT; - cb->file_private = cl; - list_add_tail(&cb->cb_list, - &dev->ctrl_rd_list.mei_cb. - cb_list); - } - - - } else { - dev_dbg(&dev->pdev->dev, "Queuing the connect request due to device busy\n"); - cb->file_private = cl; - dev_dbg(&dev->pdev->dev, "add connect cb to control write list.\n"); - list_add_tail(&cb->cb_list, - &dev->ctrl_wr_list.mei_cb.cb_list); - } - mutex_unlock(&dev->device_lock); - err = wait_event_timeout(dev->wait_recvd_msg, - (MEI_FILE_CONNECTED == cl->state || - MEI_FILE_DISCONNECTED == cl->state), - timeout * HZ); - - mutex_lock(&dev->device_lock); - if (MEI_FILE_CONNECTED == cl->state) { - dev_dbg(&dev->pdev->dev, "successfully connected to FW client.\n"); - rets = cl->status; - goto end; - } else { - dev_dbg(&dev->pdev->dev, "failed to connect to FW client.cl->state = %d.\n", - cl->state); - if (!err) { - dev_dbg(&dev->pdev->dev, - "wait_event_interruptible_timeout failed on client" - " connect message fw response message.\n"); - } - rets = -EFAULT; - - mei_io_list_flush(&dev->ctrl_rd_list, cl); - mei_io_list_flush(&dev->ctrl_wr_list, cl); - goto end; - } - rets = 0; -end: - dev_dbg(&dev->pdev->dev, "free connect cb memory."); - kfree(cb); - return rets; -} - -/** - * find_amthi_read_list_entry - finds a amthilist entry for current file - * - * @dev: the device structure - * @file: pointer to file object - * - * returns returned a list entry on success, NULL on failure. - */ -struct mei_cl_cb *find_amthi_read_list_entry( - struct mei_device *dev, - struct file *file) -{ - struct mei_cl *cl_temp; - struct mei_cl_cb *pos = NULL; - struct mei_cl_cb *next = NULL; - - list_for_each_entry_safe(pos, next, - &dev->amthi_read_complete_list.mei_cb.cb_list, cb_list) { - cl_temp = (struct mei_cl *)pos->file_private; - if (cl_temp && cl_temp == &dev->iamthif_cl && - pos->file_object == file) - return pos; - } - return NULL; -} - -/** - * amthi_read - read data from AMTHI client - * - * @dev: the device structure - * @if_num: minor number - * @file: pointer to file object - * @*ubuf: pointer to user data in user space - * @length: data length to read - * @offset: data read offset - * - * Locking: called under "dev->device_lock" lock - * - * returns - * returned data length on success, - * zero if no data to read, - * negative on failure. - */ -int amthi_read(struct mei_device *dev, struct file *file, - char __user *ubuf, size_t length, loff_t *offset) -{ - int rets; - int wait_ret; - struct mei_cl_cb *cb = NULL; - struct mei_cl *cl = file->private_data; - unsigned long timeout; - int i; - - /* Only Posible if we are in timeout */ - if (!cl || cl != &dev->iamthif_cl) { - dev_dbg(&dev->pdev->dev, "bad file ext.\n"); - return -ETIMEDOUT; - } - - i = mei_me_cl_by_id(dev, dev->iamthif_cl.me_client_id); - - if (i < 0) { - dev_dbg(&dev->pdev->dev, "amthi client not found.\n"); - return -ENODEV; - } - dev_dbg(&dev->pdev->dev, "checking amthi data\n"); - cb = find_amthi_read_list_entry(dev, file); - - /* Check for if we can block or not*/ - if (cb == NULL && file->f_flags & O_NONBLOCK) - return -EAGAIN; - - - dev_dbg(&dev->pdev->dev, "waiting for amthi data\n"); - while (cb == NULL) { - /* unlock the Mutex */ - mutex_unlock(&dev->device_lock); - - wait_ret = wait_event_interruptible(dev->iamthif_cl.wait, - (cb = find_amthi_read_list_entry(dev, file))); - - if (wait_ret) - return -ERESTARTSYS; - - dev_dbg(&dev->pdev->dev, "woke up from sleep\n"); - - /* Locking again the Mutex */ - mutex_lock(&dev->device_lock); - } - - - dev_dbg(&dev->pdev->dev, "Got amthi data\n"); - dev->iamthif_timer = 0; - - if (cb) { - timeout = cb->read_time + msecs_to_jiffies(IAMTHIF_READ_TIMER); - dev_dbg(&dev->pdev->dev, "amthi timeout = %lud\n", - timeout); - - if (time_after(jiffies, timeout)) { - dev_dbg(&dev->pdev->dev, "amthi Time out\n"); - /* 15 sec for the message has expired */ - list_del(&cb->cb_list); - rets = -ETIMEDOUT; - goto free; - } - } - /* if the whole message will fit remove it from the list */ - if (cb->information >= *offset && length >= (cb->information - *offset)) - list_del(&cb->cb_list); - else if (cb->information > 0 && cb->information <= *offset) { - /* end of the message has been reached */ - list_del(&cb->cb_list); - rets = 0; - goto free; - } - /* else means that not full buffer will be read and do not - * remove message from deletion list - */ - - dev_dbg(&dev->pdev->dev, "amthi cb->response_buffer size - %d\n", - cb->response_buffer.size); - dev_dbg(&dev->pdev->dev, "amthi cb->information - %lu\n", - cb->information); - - /* length is being turncated to PAGE_SIZE, however, - * the information may be longer */ - length = min_t(size_t, length, (cb->information - *offset)); - - if (copy_to_user(ubuf, cb->response_buffer.data + *offset, length)) - rets = -EFAULT; - else { - rets = length; - if ((*offset + length) < cb->information) { - *offset += length; - goto out; - } - } -free: - dev_dbg(&dev->pdev->dev, "free amthi cb memory.\n"); - *offset = 0; - mei_free_cb_private(cb); -out: - return rets; -} - -/** - * mei_start_read - the start read client message function. - * - * @dev: the device structure - * @if_num: minor number - * @cl: private data of the file object - * - * returns 0 on success, <0 on failure. - */ -int mei_start_read(struct mei_device *dev, struct mei_cl *cl) -{ - struct mei_cl_cb *cb; - int rets = 0; - int i; - - if (cl->state != MEI_FILE_CONNECTED) - return -ENODEV; - - if (dev->dev_state != MEI_DEV_ENABLED) - return -ENODEV; - - dev_dbg(&dev->pdev->dev, "check if read is pending.\n"); - if (cl->read_pending || cl->read_cb) { - dev_dbg(&dev->pdev->dev, "read is pending.\n"); - return -EBUSY; - } - - cb = kzalloc(sizeof(struct mei_cl_cb), GFP_KERNEL); - if (!cb) - return -ENOMEM; - - dev_dbg(&dev->pdev->dev, "allocation call back successful. host client = %d, ME client = %d\n", - cl->host_client_id, cl->me_client_id); - i = mei_me_cl_by_id(dev, cl->me_client_id); - if (i < 0) { - rets = -ENODEV; - goto unlock; - } - - cb->response_buffer.size = dev->me_clients[i].props.max_msg_length; - cb->response_buffer.data = - kmalloc(cb->response_buffer.size, GFP_KERNEL); - if (!cb->response_buffer.data) { - rets = -ENOMEM; - goto unlock; - } - dev_dbg(&dev->pdev->dev, "allocation call back data success.\n"); - cb->major_file_operations = MEI_READ; - /* make sure information is zero before we start */ - cb->information = 0; - cb->file_private = (void *) cl; - cl->read_cb = cb; - if (dev->mei_host_buffer_is_empty) { - dev->mei_host_buffer_is_empty = false; - if (mei_send_flow_control(dev, cl)) { - rets = -ENODEV; - goto unlock; - } - list_add_tail(&cb->cb_list, &dev->read_list.mei_cb.cb_list); - } else { - list_add_tail(&cb->cb_list, &dev->ctrl_wr_list.mei_cb.cb_list); - } - return rets; -unlock: - mei_free_cb_private(cb); - return rets; -} - -/** - * amthi_write - write iamthif data to amthi client - * - * @dev: the device structure - * @cb: mei call back struct - * - * returns 0 on success, <0 on failure. - */ -int amthi_write(struct mei_device *dev, struct mei_cl_cb *cb) -{ - struct mei_msg_hdr mei_hdr; - int ret; - - if (!dev || !cb) - return -ENODEV; - - dev_dbg(&dev->pdev->dev, "write data to amthi client.\n"); - - dev->iamthif_state = MEI_IAMTHIF_WRITING; - dev->iamthif_current_cb = cb; - dev->iamthif_file_object = cb->file_object; - dev->iamthif_canceled = false; - dev->iamthif_ioctl = true; - dev->iamthif_msg_buf_size = cb->request_buffer.size; - memcpy(dev->iamthif_msg_buf, cb->request_buffer.data, - cb->request_buffer.size); - - ret = mei_flow_ctrl_creds(dev, &dev->iamthif_cl); - if (ret < 0) - return ret; - - if (ret && dev->mei_host_buffer_is_empty) { - ret = 0; - dev->mei_host_buffer_is_empty = false; - if (cb->request_buffer.size > mei_hbuf_max_data(dev)) { - mei_hdr.length = mei_hbuf_max_data(dev); - mei_hdr.msg_complete = 0; - } else { - mei_hdr.length = cb->request_buffer.size; - mei_hdr.msg_complete = 1; - } - - mei_hdr.host_addr = dev->iamthif_cl.host_client_id; - mei_hdr.me_addr = dev->iamthif_cl.me_client_id; - mei_hdr.reserved = 0; - dev->iamthif_msg_buf_index += mei_hdr.length; - if (mei_write_message(dev, &mei_hdr, - (unsigned char *)(dev->iamthif_msg_buf), - mei_hdr.length)) - return -ENODEV; - - if (mei_hdr.msg_complete) { - if (mei_flow_ctrl_reduce(dev, &dev->iamthif_cl)) - return -ENODEV; - dev->iamthif_flow_control_pending = true; - dev->iamthif_state = MEI_IAMTHIF_FLOW_CONTROL; - dev_dbg(&dev->pdev->dev, "add amthi cb to write waiting list\n"); - dev->iamthif_current_cb = cb; - dev->iamthif_file_object = cb->file_object; - list_add_tail(&cb->cb_list, - &dev->write_waiting_list.mei_cb.cb_list); - } else { - dev_dbg(&dev->pdev->dev, "message does not complete, " - "so add amthi cb to write list.\n"); - list_add_tail(&cb->cb_list, - &dev->write_list.mei_cb.cb_list); - } - } else { - if (!(dev->mei_host_buffer_is_empty)) - dev_dbg(&dev->pdev->dev, "host buffer is not empty"); - - dev_dbg(&dev->pdev->dev, "No flow control credentials, " - "so add iamthif cb to write list.\n"); - list_add_tail(&cb->cb_list, &dev->write_list.mei_cb.cb_list); - } - return 0; -} - -/** - * iamthif_ioctl_send_msg - send cmd data to amthi client - * - * @dev: the device structure - * - * returns 0 on success, <0 on failure. - */ -void mei_run_next_iamthif_cmd(struct mei_device *dev) -{ - struct mei_cl *cl_tmp; - struct mei_cl_cb *pos = NULL; - struct mei_cl_cb *next = NULL; - int status; - - if (!dev) - return; - - dev->iamthif_msg_buf_size = 0; - dev->iamthif_msg_buf_index = 0; - dev->iamthif_canceled = false; - dev->iamthif_ioctl = true; - dev->iamthif_state = MEI_IAMTHIF_IDLE; - dev->iamthif_timer = 0; - dev->iamthif_file_object = NULL; - - dev_dbg(&dev->pdev->dev, "complete amthi cmd_list cb.\n"); - - list_for_each_entry_safe(pos, next, - &dev->amthi_cmd_list.mei_cb.cb_list, cb_list) { - list_del(&pos->cb_list); - cl_tmp = (struct mei_cl *)pos->file_private; - - if (cl_tmp && cl_tmp == &dev->iamthif_cl) { - status = amthi_write(dev, pos); - if (status) { - dev_dbg(&dev->pdev->dev, - "amthi write failed status = %d\n", - status); - return; - } - break; - } - } -} - -/** - * mei_free_cb_private - free mei_cb_private related memory - * - * @cb: mei callback struct - */ -void mei_free_cb_private(struct mei_cl_cb *cb) -{ - if (cb == NULL) - return; - - kfree(cb->request_buffer.data); - kfree(cb->response_buffer.data); - kfree(cb); -} diff --git a/drivers/misc/mei/main.c b/drivers/misc/mei/main.c index e8b0858132c..66f0a1a0645 100644 --- a/drivers/misc/mei/main.c +++ b/drivers/misc/mei/main.c @@ -13,9 +13,6 @@ * more details. * */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/kernel.h> @@ -37,171 +34,10 @@ #include <linux/interrupt.h> #include <linux/miscdevice.h> -#include "mei_dev.h" #include <linux/mei.h> -#include "interface.h" - -/* AMT device is a singleton on the platform */ -static struct pci_dev *mei_pdev; - -/* mei_pci_tbl - PCI Device ID Table */ -static DEFINE_PCI_DEVICE_TABLE(mei_pci_tbl) = { - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_82946GZ)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_82G35)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_82Q965)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_82G965)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_82GM965)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_82GME965)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_82Q35)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_82G33)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_82Q33)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_82X38)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_3200)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_6)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_7)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_8)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_9)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9_10)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9M_1)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9M_2)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9M_3)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH9M_4)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH10_1)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH10_2)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH10_3)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_ICH10_4)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_IBXPK_1)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_IBXPK_2)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_CPT_1)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_PBG_1)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_PPT_1)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_PPT_2)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_PPT_3)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_LPT)}, - {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MEI_DEV_ID_LPT_LP)}, - - /* required last entry */ - {0, } -}; - -MODULE_DEVICE_TABLE(pci, mei_pci_tbl); - -static DEFINE_MUTEX(mei_mutex); - - -/** - * mei_clear_list - removes all callbacks associated with file - * from mei_cb_list - * - * @dev: device structure. - * @file: file structure - * @mei_cb_list: callbacks list - * - * mei_clear_list is called to clear resources associated with file - * when application calls close function or Ctrl-C was pressed - * - * returns true if callback removed from the list, false otherwise - */ -static bool mei_clear_list(struct mei_device *dev, - struct file *file, struct list_head *mei_cb_list) -{ - struct mei_cl_cb *cb_pos = NULL; - struct mei_cl_cb *cb_next = NULL; - struct file *file_temp; - bool removed = false; - - /* list all list member */ - list_for_each_entry_safe(cb_pos, cb_next, mei_cb_list, cb_list) { - file_temp = (struct file *)cb_pos->file_object; - /* check if list member associated with a file */ - if (file_temp == file) { - /* remove member from the list */ - list_del(&cb_pos->cb_list); - /* check if cb equal to current iamthif cb */ - if (dev->iamthif_current_cb == cb_pos) { - dev->iamthif_current_cb = NULL; - /* send flow control to iamthif client */ - mei_send_flow_control(dev, &dev->iamthif_cl); - } - /* free all allocated buffers */ - mei_free_cb_private(cb_pos); - cb_pos = NULL; - removed = true; - } - } - return removed; -} - -/** - * mei_clear_lists - removes all callbacks associated with file - * - * @dev: device structure - * @file: file structure - * - * mei_clear_lists is called to clear resources associated with file - * when application calls close function or Ctrl-C was pressed - * - * returns true if callback removed from the list, false otherwise - */ -static bool mei_clear_lists(struct mei_device *dev, struct file *file) -{ - bool removed = false; - - /* remove callbacks associated with a file */ - mei_clear_list(dev, file, &dev->amthi_cmd_list.mei_cb.cb_list); - if (mei_clear_list(dev, file, - &dev->amthi_read_complete_list.mei_cb.cb_list)) - removed = true; - - mei_clear_list(dev, file, &dev->ctrl_rd_list.mei_cb.cb_list); - - if (mei_clear_list(dev, file, &dev->ctrl_wr_list.mei_cb.cb_list)) - removed = true; - - if (mei_clear_list(dev, file, &dev->write_waiting_list.mei_cb.cb_list)) - removed = true; - - if (mei_clear_list(dev, file, &dev->write_list.mei_cb.cb_list)) - removed = true; - - /* check if iamthif_current_cb not NULL */ - if (dev->iamthif_current_cb && !removed) { - /* check file and iamthif current cb association */ - if (dev->iamthif_current_cb->file_object == file) { - /* remove cb */ - mei_free_cb_private(dev->iamthif_current_cb); - dev->iamthif_current_cb = NULL; - removed = true; - } - } - return removed; -} -/** - * find_read_list_entry - find read list entry - * - * @dev: device structure - * @file: pointer to file structure - * - * returns cb on success, NULL on error - */ -static struct mei_cl_cb *find_read_list_entry( - struct mei_device *dev, - struct mei_cl *cl) -{ - struct mei_cl_cb *pos = NULL; - struct mei_cl_cb *next = NULL; - dev_dbg(&dev->pdev->dev, "remove read_list CB\n"); - list_for_each_entry_safe(pos, next, - &dev->read_list.mei_cb.cb_list, cb_list) { - struct mei_cl *cl_temp; - cl_temp = (struct mei_cl *)pos->file_private; - - if (mei_cl_cmp_id(cl, cl_temp)) - return pos; - } - return NULL; -} +#include "mei_dev.h" +#include "client.h" /** * mei_open - the open function @@ -213,66 +49,52 @@ static struct mei_cl_cb *find_read_list_entry( */ static int mei_open(struct inode *inode, struct file *file) { + struct miscdevice *misc = file->private_data; + struct pci_dev *pdev; struct mei_cl *cl; struct mei_device *dev; - unsigned long cl_id; + int err; - err = -ENODEV; - if (!mei_pdev) - goto out; + if (!misc->parent) + return -ENODEV; + + pdev = container_of(misc->parent, struct pci_dev, dev); - dev = pci_get_drvdata(mei_pdev); + dev = pci_get_drvdata(pdev); if (!dev) - goto out; + return -ENODEV; mutex_lock(&dev->device_lock); - err = -ENOMEM; - cl = mei_cl_allocate(dev); - if (!cl) - goto out_unlock; + + cl = NULL; err = -ENODEV; if (dev->dev_state != MEI_DEV_ENABLED) { dev_dbg(&dev->pdev->dev, "dev_state != MEI_ENABLED dev_state = %s\n", mei_dev_state_str(dev->dev_state)); - goto out_unlock; + goto err_unlock; } - err = -EMFILE; - if (dev->open_handle_count >= MEI_MAX_OPEN_HANDLE_COUNT) { - dev_err(&dev->pdev->dev, "open_handle_count exceded %d", - MEI_MAX_OPEN_HANDLE_COUNT); - goto out_unlock; - } - - cl_id = find_first_zero_bit(dev->host_clients_map, MEI_CLIENTS_MAX); - if (cl_id >= MEI_CLIENTS_MAX) { - dev_err(&dev->pdev->dev, "client_id exceded %d", - MEI_CLIENTS_MAX) ; - goto out_unlock; - } - - cl->host_client_id = cl_id; - - dev_dbg(&dev->pdev->dev, "client_id = %d\n", cl->host_client_id); - dev->open_handle_count++; - - list_add_tail(&cl->link, &dev->file_list); + err = -ENOMEM; + cl = mei_cl_allocate(dev); + if (!cl) + goto err_unlock; - set_bit(cl->host_client_id, dev->host_clients_map); - cl->state = MEI_FILE_INITIALIZING; - cl->sm_state = 0; + /* open_handle_count check is handled in the mei_cl_link */ + err = mei_cl_link(cl, MEI_HOST_CLIENT_ID_ANY); + if (err) + goto err_unlock; file->private_data = cl; + mutex_unlock(&dev->device_lock); return nonseekable_open(inode, file); -out_unlock: +err_unlock: mutex_unlock(&dev->device_lock); kfree(cl); -out: return err; } @@ -297,67 +119,39 @@ static int mei_release(struct inode *inode, struct file *file) dev = cl->dev; mutex_lock(&dev->device_lock); - if (cl != &dev->iamthif_cl) { - if (cl->state == MEI_FILE_CONNECTED) { - cl->state = MEI_FILE_DISCONNECTING; - dev_dbg(&dev->pdev->dev, - "disconnecting client host client = %d, " - "ME client = %d\n", - cl->host_client_id, - cl->me_client_id); - rets = mei_disconnect_host_client(dev, cl); - } - mei_cl_flush_queues(cl); - dev_dbg(&dev->pdev->dev, "remove client host client = %d, ME client = %d\n", - cl->host_client_id, - cl->me_client_id); - - if (dev->open_handle_count > 0) { - clear_bit(cl->host_client_id, dev->host_clients_map); - dev->open_handle_count--; - } - mei_remove_client_from_file_list(dev, cl->host_client_id); + if (cl == &dev->iamthif_cl) { + rets = mei_amthif_release(dev, file); + goto out; + } + if (cl->state == MEI_FILE_CONNECTED) { + cl->state = MEI_FILE_DISCONNECTING; + cl_dbg(dev, cl, "disconnecting\n"); + rets = mei_cl_disconnect(cl); + } + mei_cl_flush_queues(cl); + cl_dbg(dev, cl, "removing\n"); - /* free read cb */ - cb = NULL; - if (cl->read_cb) { - cb = find_read_list_entry(dev, cl); - /* Remove entry from read list */ - if (cb) - list_del(&cb->cb_list); + mei_cl_unlink(cl); - cb = cl->read_cb; - cl->read_cb = NULL; - } - file->private_data = NULL; + /* free read cb */ + cb = NULL; + if (cl->read_cb) { + cb = mei_cl_find_read_cb(cl); + /* Remove entry from read list */ + if (cb) + list_del(&cb->list); - if (cb) { - mei_free_cb_private(cb); - cb = NULL; - } + cb = cl->read_cb; + cl->read_cb = NULL; + } - kfree(cl); - } else { - if (dev->open_handle_count > 0) - dev->open_handle_count--; - - if (dev->iamthif_file_object == file && - dev->iamthif_state != MEI_IAMTHIF_IDLE) { - - dev_dbg(&dev->pdev->dev, "amthi canceled iamthif state %d\n", - dev->iamthif_state); - dev->iamthif_canceled = true; - if (dev->iamthif_state == MEI_IAMTHIF_READ_COMPLETE) { - dev_dbg(&dev->pdev->dev, "run next amthi iamthif cb\n"); - mei_run_next_iamthif_cmd(dev); - } - } + file->private_data = NULL; - if (mei_clear_lists(dev, file)) - dev->iamthif_state = MEI_IAMTHIF_IDLE; + mei_io_cb_free(cb); - } + kfree(cl); +out: mutex_unlock(&dev->device_lock); return rets; } @@ -380,7 +174,6 @@ static ssize_t mei_read(struct file *file, char __user *ubuf, struct mei_cl_cb *cb_pos = NULL; struct mei_cl_cb *cb = NULL; struct mei_device *dev; - int i; int rets; int err; @@ -390,48 +183,41 @@ static ssize_t mei_read(struct file *file, char __user *ubuf, dev = cl->dev; + mutex_lock(&dev->device_lock); if (dev->dev_state != MEI_DEV_ENABLED) { rets = -ENODEV; goto out; } - if ((cl->sm_state & MEI_WD_STATE_INDEPENDENCE_MSG_SENT) == 0) { - /* Do not allow to read watchdog client */ - i = mei_me_cl_by_uuid(dev, &mei_wd_guid); - if (i >= 0) { - struct mei_me_client *me_client = &dev->me_clients[i]; - if (cl->me_client_id == me_client->client_id) { - rets = -EBADF; - goto out; - } - } - } else { - cl->sm_state &= ~MEI_WD_STATE_INDEPENDENCE_MSG_SENT; + if (length == 0) { + rets = 0; + goto out; } if (cl == &dev->iamthif_cl) { - rets = amthi_read(dev, file, ubuf, length, offset); + rets = mei_amthif_read(dev, file, ubuf, length, offset); goto out; } - if (cl->read_cb && cl->read_cb->information > *offset) { - cb = cl->read_cb; - goto copy_buffer; - } else if (cl->read_cb && cl->read_cb->information > 0 && - cl->read_cb->information <= *offset) { + if (cl->read_cb) { cb = cl->read_cb; - rets = 0; - goto free; - } else if ((!cl->read_cb || !cl->read_cb->information) && - *offset > 0) { - /*Offset needs to be cleaned for contiguous reads*/ + /* read what left */ + if (cb->buf_idx > *offset) + goto copy_buffer; + /* offset is beyond buf_idx we have no more data return 0 */ + if (cb->buf_idx > 0 && cb->buf_idx <= *offset) { + rets = 0; + goto free; + } + /* Offset needs to be cleaned for contiguous reads*/ + if (cb->buf_idx == 0 && *offset > 0) + *offset = 0; + } else if (*offset > 0) { *offset = 0; - rets = 0; - goto out; } - err = mei_start_read(dev, cl); + err = mei_cl_read_start(cl, length); if (err && err != -EBUSY) { dev_dbg(&dev->pdev->dev, "mei start read failure with status = %d\n", err); @@ -449,19 +235,16 @@ static ssize_t mei_read(struct file *file, char __user *ubuf, mutex_unlock(&dev->device_lock); if (wait_event_interruptible(cl->rx_wait, - (MEI_READ_COMPLETE == cl->reading_state || - MEI_FILE_INITIALIZING == cl->state || - MEI_FILE_DISCONNECTED == cl->state || - MEI_FILE_DISCONNECTING == cl->state))) { + MEI_READ_COMPLETE == cl->reading_state || + mei_cl_is_transitioning(cl))) { + if (signal_pending(current)) return -EINTR; return -ERESTARTSYS; } mutex_lock(&dev->device_lock); - if (MEI_FILE_INITIALIZING == cl->state || - MEI_FILE_DISCONNECTED == cl->state || - MEI_FILE_DISCONNECTING == cl->state) { + if (mei_cl_is_transitioning(cl)) { rets = -EBUSY; goto out; } @@ -479,44 +262,41 @@ static ssize_t mei_read(struct file *file, char __user *ubuf, } /* now copy the data to user space */ copy_buffer: - dev_dbg(&dev->pdev->dev, "cb->response_buffer size - %d\n", - cb->response_buffer.size); - dev_dbg(&dev->pdev->dev, "cb->information - %lu\n", - cb->information); - if (length == 0 || ubuf == NULL || *offset > cb->information) { + dev_dbg(&dev->pdev->dev, "buf.size = %d buf.idx= %ld\n", + cb->response_buffer.size, cb->buf_idx); + if (length == 0 || ubuf == NULL || *offset > cb->buf_idx) { rets = -EMSGSIZE; goto free; } - /* length is being truncated to PAGE_SIZE, however, */ - /* information size may be longer */ - length = min_t(size_t, length, (cb->information - *offset)); + /* length is being truncated to PAGE_SIZE, + * however buf_idx may point beyond that */ + length = min_t(size_t, length, cb->buf_idx - *offset); if (copy_to_user(ubuf, cb->response_buffer.data + *offset, length)) { + dev_dbg(&dev->pdev->dev, "failed to copy data to userland\n"); rets = -EFAULT; goto free; } rets = length; *offset += length; - if ((unsigned long)*offset < cb->information) + if ((unsigned long)*offset < cb->buf_idx) goto out; free: - cb_pos = find_read_list_entry(dev, cl); + cb_pos = mei_cl_find_read_cb(cl); /* Remove entry from read list */ if (cb_pos) - list_del(&cb_pos->cb_list); - mei_free_cb_private(cb); + list_del(&cb_pos->list); + mei_io_cb_free(cb); cl->reading_state = MEI_IDLE; cl->read_cb = NULL; - cl->read_pending = 0; out: dev_dbg(&dev->pdev->dev, "end mei read rets= %d\n", rets); mutex_unlock(&dev->device_lock); return rets; } - /** * mei_write - the write function. * @@ -532,11 +312,10 @@ static ssize_t mei_write(struct file *file, const char __user *ubuf, { struct mei_cl *cl = file->private_data; struct mei_cl_cb *write_cb = NULL; - struct mei_msg_hdr mei_hdr; struct mei_device *dev; unsigned long timeout = 0; int rets; - int i; + int id; if (WARN_ON(!cl || !cl->dev)) return -ENODEV; @@ -546,23 +325,45 @@ static ssize_t mei_write(struct file *file, const char __user *ubuf, mutex_lock(&dev->device_lock); if (dev->dev_state != MEI_DEV_ENABLED) { - mutex_unlock(&dev->device_lock); - return -ENODEV; + rets = -ENODEV; + goto out; + } + + id = mei_me_cl_by_id(dev, cl->me_client_id); + if (id < 0) { + rets = -ENOTTY; + goto out; + } + + if (length == 0) { + rets = 0; + goto out; + } + + if (length > dev->me_clients[id].props.max_msg_length) { + rets = -EFBIG; + goto out; } + if (cl->state != MEI_FILE_CONNECTED) { + dev_err(&dev->pdev->dev, "host client = %d, is not connected to ME client = %d", + cl->host_client_id, cl->me_client_id); + rets = -ENODEV; + goto out; + } if (cl == &dev->iamthif_cl) { - write_cb = find_amthi_read_list_entry(dev, file); + write_cb = mei_amthif_find_read_list_entry(dev, file); if (write_cb) { timeout = write_cb->read_time + - msecs_to_jiffies(IAMTHIF_READ_TIMER); + mei_secs_to_jiffies(MEI_IAMTHIF_READ_TIMER); if (time_after(jiffies, timeout) || - cl->reading_state == MEI_READ_COMPLETE) { - *offset = 0; - list_del(&write_cb->cb_list); - mei_free_cb_private(write_cb); - write_cb = NULL; + cl->reading_state == MEI_READ_COMPLETE) { + *offset = 0; + list_del(&write_cb->list); + mei_io_cb_free(write_cb); + write_cb = NULL; } } } @@ -570,181 +371,147 @@ static ssize_t mei_write(struct file *file, const char __user *ubuf, /* free entry used in read */ if (cl->reading_state == MEI_READ_COMPLETE) { *offset = 0; - write_cb = find_read_list_entry(dev, cl); + write_cb = mei_cl_find_read_cb(cl); if (write_cb) { - list_del(&write_cb->cb_list); - mei_free_cb_private(write_cb); + list_del(&write_cb->list); + mei_io_cb_free(write_cb); write_cb = NULL; cl->reading_state = MEI_IDLE; cl->read_cb = NULL; - cl->read_pending = 0; } - } else if (cl->reading_state == MEI_IDLE && !cl->read_pending) + } else if (cl->reading_state == MEI_IDLE) *offset = 0; - write_cb = kzalloc(sizeof(struct mei_cl_cb), GFP_KERNEL); + write_cb = mei_io_cb_init(cl, file); if (!write_cb) { - mutex_unlock(&dev->device_lock); - return -ENOMEM; + dev_err(&dev->pdev->dev, "write cb allocation failed\n"); + rets = -ENOMEM; + goto out; } + rets = mei_io_cb_alloc_req_buf(write_cb, length); + if (rets) + goto out; - write_cb->file_object = file; - write_cb->file_private = cl; - write_cb->request_buffer.data = kmalloc(length, GFP_KERNEL); - rets = -ENOMEM; - if (!write_cb->request_buffer.data) - goto unlock_dev; - - dev_dbg(&dev->pdev->dev, "length =%d\n", (int) length); - - rets = -EFAULT; - if (copy_from_user(write_cb->request_buffer.data, ubuf, length)) - goto unlock_dev; - - cl->sm_state = 0; - if (length == 4 && - ((memcmp(mei_wd_state_independence_msg[0], - write_cb->request_buffer.data, 4) == 0) || - (memcmp(mei_wd_state_independence_msg[1], - write_cb->request_buffer.data, 4) == 0) || - (memcmp(mei_wd_state_independence_msg[2], - write_cb->request_buffer.data, 4) == 0))) - cl->sm_state |= MEI_WD_STATE_INDEPENDENCE_MSG_SENT; - - INIT_LIST_HEAD(&write_cb->cb_list); - if (cl == &dev->iamthif_cl) { - write_cb->response_buffer.data = - kmalloc(dev->iamthif_mtu, GFP_KERNEL); - if (!write_cb->response_buffer.data) { - rets = -ENOMEM; - goto unlock_dev; - } - if (dev->dev_state != MEI_DEV_ENABLED) { - rets = -ENODEV; - goto unlock_dev; - } - i = mei_me_cl_by_id(dev, dev->iamthif_cl.me_client_id); - if (i < 0) { - rets = -ENODEV; - goto unlock_dev; - } - if (length > dev->me_clients[i].props.max_msg_length || - length <= 0) { - rets = -EMSGSIZE; - goto unlock_dev; - } + rets = copy_from_user(write_cb->request_buffer.data, ubuf, length); + if (rets) { + dev_dbg(&dev->pdev->dev, "failed to copy data from userland\n"); + rets = -EFAULT; + goto out; + } - write_cb->response_buffer.size = dev->iamthif_mtu; - write_cb->major_file_operations = MEI_IOCTL; - write_cb->information = 0; - write_cb->request_buffer.size = length; - if (dev->iamthif_cl.state != MEI_FILE_CONNECTED) { - rets = -ENODEV; - goto unlock_dev; - } + if (cl == &dev->iamthif_cl) { + rets = mei_amthif_write(dev, write_cb); - if (!list_empty(&dev->amthi_cmd_list.mei_cb.cb_list) || - dev->iamthif_state != MEI_IAMTHIF_IDLE) { - dev_dbg(&dev->pdev->dev, "amthi_state = %d\n", - (int) dev->iamthif_state); - dev_dbg(&dev->pdev->dev, "add amthi cb to amthi cmd waiting list\n"); - list_add_tail(&write_cb->cb_list, - &dev->amthi_cmd_list.mei_cb.cb_list); - rets = length; - } else { - dev_dbg(&dev->pdev->dev, "call amthi write\n"); - rets = amthi_write(dev, write_cb); - - if (rets) { - dev_dbg(&dev->pdev->dev, "amthi write failed with status = %d\n", - rets); - goto unlock_dev; - } - rets = length; + if (rets) { + dev_err(&dev->pdev->dev, + "amthif write failed with status = %d\n", rets); + goto out; } mutex_unlock(&dev->device_lock); - return rets; + return length; } - write_cb->major_file_operations = MEI_WRITE; - /* make sure information is zero before we start */ + rets = mei_cl_write(cl, write_cb, false); +out: + mutex_unlock(&dev->device_lock); + if (rets < 0) + mei_io_cb_free(write_cb); + return rets; +} - write_cb->information = 0; - write_cb->request_buffer.size = length; +/** + * mei_ioctl_connect_client - the connect to fw client IOCTL function + * + * @dev: the device structure + * @data: IOCTL connect data, input and output parameters + * @file: private data of the file object + * + * Locking: called under "dev->device_lock" lock + * + * returns 0 on success, <0 on failure. + */ +static int mei_ioctl_connect_client(struct file *file, + struct mei_connect_client_data *data) +{ + struct mei_device *dev; + struct mei_client *client; + struct mei_cl *cl; + int i; + int rets; - dev_dbg(&dev->pdev->dev, "host client = %d, ME client = %d\n", - cl->host_client_id, cl->me_client_id); - if (cl->state != MEI_FILE_CONNECTED) { + cl = file->private_data; + if (WARN_ON(!cl || !cl->dev)) + return -ENODEV; + + dev = cl->dev; + + if (dev->dev_state != MEI_DEV_ENABLED) { rets = -ENODEV; - dev_dbg(&dev->pdev->dev, "host client = %d, is not connected to ME client = %d", - cl->host_client_id, - cl->me_client_id); - goto unlock_dev; + goto end; } - i = mei_me_cl_by_id(dev, cl->me_client_id); - if (i < 0) { - rets = -ENODEV; - goto unlock_dev; + + if (cl->state != MEI_FILE_INITIALIZING && + cl->state != MEI_FILE_DISCONNECTED) { + rets = -EBUSY; + goto end; } - if (length > dev->me_clients[i].props.max_msg_length || length <= 0) { - rets = -EINVAL; - goto unlock_dev; + + /* find ME client we're trying to connect to */ + i = mei_me_cl_by_uuid(dev, &data->in_client_uuid); + if (i < 0 || dev->me_clients[i].props.fixed_address) { + dev_dbg(&dev->pdev->dev, "Cannot connect to FW Client UUID = %pUl\n", + &data->in_client_uuid); + rets = -ENOTTY; + goto end; } - write_cb->file_private = cl; - rets = mei_flow_ctrl_creds(dev, cl); - if (rets < 0) - goto unlock_dev; + cl->me_client_id = dev->me_clients[i].client_id; - if (rets && dev->mei_host_buffer_is_empty) { - rets = 0; - dev->mei_host_buffer_is_empty = false; - if (length > mei_hbuf_max_data(dev)) { - mei_hdr.length = mei_hbuf_max_data(dev); - mei_hdr.msg_complete = 0; - } else { - mei_hdr.length = length; - mei_hdr.msg_complete = 1; - } - mei_hdr.host_addr = cl->host_client_id; - mei_hdr.me_addr = cl->me_client_id; - mei_hdr.reserved = 0; - dev_dbg(&dev->pdev->dev, "call mei_write_message header=%08x.\n", - *((u32 *) &mei_hdr)); - if (mei_write_message(dev, &mei_hdr, - (unsigned char *) (write_cb->request_buffer.data), - mei_hdr.length)) { + dev_dbg(&dev->pdev->dev, "Connect to FW Client ID = %d\n", + cl->me_client_id); + dev_dbg(&dev->pdev->dev, "FW Client - Protocol Version = %d\n", + dev->me_clients[i].props.protocol_version); + dev_dbg(&dev->pdev->dev, "FW Client - Max Msg Len = %d\n", + dev->me_clients[i].props.max_msg_length); + + /* if we're connecting to amthif client then we will use the + * existing connection + */ + if (uuid_le_cmp(data->in_client_uuid, mei_amthif_guid) == 0) { + dev_dbg(&dev->pdev->dev, "FW Client is amthi\n"); + if (dev->iamthif_cl.state != MEI_FILE_CONNECTED) { rets = -ENODEV; - goto unlock_dev; - } - cl->writing_state = MEI_WRITING; - write_cb->information = mei_hdr.length; - if (mei_hdr.msg_complete) { - if (mei_flow_ctrl_reduce(dev, cl)) { - rets = -ENODEV; - goto unlock_dev; - } - list_add_tail(&write_cb->cb_list, - &dev->write_waiting_list.mei_cb.cb_list); - } else { - list_add_tail(&write_cb->cb_list, - &dev->write_list.mei_cb.cb_list); + goto end; } + mei_cl_unlink(cl); - } else { + kfree(cl); + cl = NULL; + dev->iamthif_open_count++; + file->private_data = &dev->iamthif_cl; - write_cb->information = 0; - cl->writing_state = MEI_WRITING; - list_add_tail(&write_cb->cb_list, - &dev->write_list.mei_cb.cb_list); + client = &data->out_client_properties; + client->max_msg_length = + dev->me_clients[i].props.max_msg_length; + client->protocol_version = + dev->me_clients[i].props.protocol_version; + rets = dev->iamthif_cl.status; + + goto end; } - mutex_unlock(&dev->device_lock); - return length; -unlock_dev: - mutex_unlock(&dev->device_lock); - mei_free_cb_private(write_cb); + + /* prepare the output buffer */ + client = &data->out_client_properties; + client->max_msg_length = dev->me_clients[i].props.max_msg_length; + client->protocol_version = dev->me_clients[i].props.protocol_version; + dev_dbg(&dev->pdev->dev, "Can connect?\n"); + + + rets = mei_cl_connect(cl, file); + +end: return rets; } @@ -796,6 +563,7 @@ static long mei_ioctl(struct file *file, unsigned int cmd, unsigned long data) rets = -EFAULT; goto out; } + rets = mei_ioctl_connect_client(file, connect_data); /* if all is ok, copying the data back to user. */ @@ -849,34 +617,33 @@ static unsigned int mei_poll(struct file *file, poll_table *wait) unsigned int mask = 0; if (WARN_ON(!cl || !cl->dev)) - return mask; + return POLLERR; dev = cl->dev; mutex_lock(&dev->device_lock); - if (dev->dev_state != MEI_DEV_ENABLED) - goto out; - - - if (cl == &dev->iamthif_cl) { - mutex_unlock(&dev->device_lock); - poll_wait(file, &dev->iamthif_cl.wait, wait); - mutex_lock(&dev->device_lock); - if (dev->iamthif_state == MEI_IAMTHIF_READ_COMPLETE && - dev->iamthif_file_object == file) { - mask |= (POLLIN | POLLRDNORM); - dev_dbg(&dev->pdev->dev, "run next amthi cb\n"); - mei_run_next_iamthif_cmd(dev); - } + if (!mei_cl_is_connected(cl)) { + mask = POLLERR; goto out; } mutex_unlock(&dev->device_lock); + + + if (cl == &dev->iamthif_cl) + return mei_amthif_poll(dev, file, wait); + poll_wait(file, &cl->tx_wait, wait); + mutex_lock(&dev->device_lock); - if (MEI_WRITE_COMPLETE == cl->writing_state) - mask |= (POLLIN | POLLRDNORM); + + if (!mei_cl_is_connected(cl)) { + mask = POLLERR; + goto out; + } + + mask |= (POLLIN | POLLRDNORM); out: mutex_unlock(&dev->device_lock); @@ -900,7 +667,6 @@ static const struct file_operations mei_fops = { .llseek = no_llseek }; - /* * Misc Device Struct */ @@ -910,298 +676,44 @@ static struct miscdevice mei_misc_device = { .minor = MISC_DYNAMIC_MINOR, }; -/** - * mei_quirk_probe - probe for devices that doesn't valid ME interface - * @pdev: PCI device structure - * @ent: entry into pci_device_table - * - * returns true if ME Interface is valid, false otherwise - */ -static bool __devinit mei_quirk_probe(struct pci_dev *pdev, - const struct pci_device_id *ent) -{ - u32 reg; - if (ent->device == MEI_DEV_ID_PBG_1) { - pci_read_config_dword(pdev, 0x48, ®); - /* make sure that bit 9 is up and bit 10 is down */ - if ((reg & 0x600) == 0x200) { - dev_info(&pdev->dev, "Device doesn't have valid ME Interface\n"); - return false; - } - } - return true; -} -/** - * mei_probe - Device Initialization Routine - * - * @pdev: PCI device structure - * @ent: entry in kcs_pci_tbl - * - * returns 0 on success, <0 on failure. - */ -static int __devinit mei_probe(struct pci_dev *pdev, - const struct pci_device_id *ent) -{ - struct mei_device *dev; - int err; - - mutex_lock(&mei_mutex); - - if (!mei_quirk_probe(pdev, ent)) { - err = -ENODEV; - goto end; - } - if (mei_pdev) { - err = -EEXIST; - goto end; - } - /* enable pci dev */ - err = pci_enable_device(pdev); - if (err) { - dev_err(&pdev->dev, "failed to enable pci device.\n"); - goto end; - } - /* set PCI host mastering */ - pci_set_master(pdev); - /* pci request regions for mei driver */ - err = pci_request_regions(pdev, KBUILD_MODNAME); - if (err) { - dev_err(&pdev->dev, "failed to get pci regions.\n"); - goto disable_device; - } - /* allocates and initializes the mei dev structure */ - dev = mei_device_init(pdev); - if (!dev) { - err = -ENOMEM; - goto release_regions; - } - /* mapping IO device memory */ - dev->mem_addr = pci_iomap(pdev, 0, 0); - if (!dev->mem_addr) { - dev_err(&pdev->dev, "mapping I/O device memory failure.\n"); - err = -ENOMEM; - goto free_device; - } - pci_enable_msi(pdev); - - /* request and enable interrupt */ - if (pci_dev_msi_enabled(pdev)) - err = request_threaded_irq(pdev->irq, - NULL, - mei_interrupt_thread_handler, - IRQF_ONESHOT, KBUILD_MODNAME, dev); - else - err = request_threaded_irq(pdev->irq, - mei_interrupt_quick_handler, - mei_interrupt_thread_handler, - IRQF_SHARED, KBUILD_MODNAME, dev); - - if (err) { - dev_err(&pdev->dev, "request_threaded_irq failure. irq = %d\n", - pdev->irq); - goto disable_msi; - } - INIT_DELAYED_WORK(&dev->timer_work, mei_timer); - if (mei_hw_init(dev)) { - dev_err(&pdev->dev, "init hw failure.\n"); - err = -ENODEV; - goto release_irq; - } - - err = misc_register(&mei_misc_device); - if (err) - goto release_irq; - - mei_pdev = pdev; - pci_set_drvdata(pdev, dev); - - - schedule_delayed_work(&dev->timer_work, HZ); - - mutex_unlock(&mei_mutex); +int mei_register(struct mei_device *dev) +{ + int ret; + mei_misc_device.parent = &dev->pdev->dev; + ret = misc_register(&mei_misc_device); + if (ret) + return ret; - pr_debug("initialization successful.\n"); + if (mei_dbgfs_register(dev, mei_misc_device.name)) + dev_err(&dev->pdev->dev, "cannot register debugfs\n"); return 0; - -release_irq: - /* disable interrupts */ - dev->host_hw_state = mei_hcsr_read(dev); - mei_disable_interrupts(dev); - flush_scheduled_work(); - free_irq(pdev->irq, dev); -disable_msi: - pci_disable_msi(pdev); - pci_iounmap(pdev, dev->mem_addr); -free_device: - kfree(dev); -release_regions: - pci_release_regions(pdev); -disable_device: - pci_disable_device(pdev); -end: - mutex_unlock(&mei_mutex); - dev_err(&pdev->dev, "initialization failed.\n"); - return err; } +EXPORT_SYMBOL_GPL(mei_register); -/** - * mei_remove - Device Removal Routine - * - * @pdev: PCI device structure - * - * mei_remove is called by the PCI subsystem to alert the driver - * that it should release a PCI device. - */ -static void __devexit mei_remove(struct pci_dev *pdev) +void mei_deregister(struct mei_device *dev) { - struct mei_device *dev; - - if (mei_pdev != pdev) - return; - - dev = pci_get_drvdata(pdev); - if (!dev) - return; - - mutex_lock(&dev->device_lock); - - cancel_delayed_work(&dev->timer_work); - - mei_wd_stop(dev); - - mei_pdev = NULL; - - if (dev->iamthif_cl.state == MEI_FILE_CONNECTED) { - dev->iamthif_cl.state = MEI_FILE_DISCONNECTING; - mei_disconnect_host_client(dev, &dev->iamthif_cl); - } - if (dev->wd_cl.state == MEI_FILE_CONNECTED) { - dev->wd_cl.state = MEI_FILE_DISCONNECTING; - mei_disconnect_host_client(dev, &dev->wd_cl); - } - - /* Unregistering watchdog device */ - mei_watchdog_unregister(dev); - - /* remove entry if already in list */ - dev_dbg(&pdev->dev, "list del iamthif and wd file list.\n"); - mei_remove_client_from_file_list(dev, dev->wd_cl.host_client_id); - mei_remove_client_from_file_list(dev, dev->iamthif_cl.host_client_id); - - dev->iamthif_current_cb = NULL; - dev->me_clients_num = 0; - - mutex_unlock(&dev->device_lock); - - flush_scheduled_work(); - - /* disable interrupts */ - mei_disable_interrupts(dev); - - free_irq(pdev->irq, dev); - pci_disable_msi(pdev); - pci_set_drvdata(pdev, NULL); - - if (dev->mem_addr) - pci_iounmap(pdev, dev->mem_addr); - - kfree(dev); - - pci_release_regions(pdev); - pci_disable_device(pdev); - + mei_dbgfs_deregister(dev); misc_deregister(&mei_misc_device); + mei_misc_device.parent = NULL; } -#ifdef CONFIG_PM -static int mei_pci_suspend(struct device *device) -{ - struct pci_dev *pdev = to_pci_dev(device); - struct mei_device *dev = pci_get_drvdata(pdev); - int err; +EXPORT_SYMBOL_GPL(mei_deregister); - if (!dev) - return -ENODEV; - mutex_lock(&dev->device_lock); - - cancel_delayed_work(&dev->timer_work); - - /* Stop watchdog if exists */ - err = mei_wd_stop(dev); - /* Set new mei state */ - if (dev->dev_state == MEI_DEV_ENABLED || - dev->dev_state == MEI_DEV_RECOVERING_FROM_RESET) { - dev->dev_state = MEI_DEV_POWER_DOWN; - mei_reset(dev, 0); - } - mutex_unlock(&dev->device_lock); - - free_irq(pdev->irq, dev); - pci_disable_msi(pdev); - - return err; +static int __init mei_init(void) +{ + return mei_cl_bus_init(); } -static int mei_pci_resume(struct device *device) +static void __exit mei_exit(void) { - struct pci_dev *pdev = to_pci_dev(device); - struct mei_device *dev; - int err; - - dev = pci_get_drvdata(pdev); - if (!dev) - return -ENODEV; - - pci_enable_msi(pdev); - - /* request and enable interrupt */ - if (pci_dev_msi_enabled(pdev)) - err = request_threaded_irq(pdev->irq, - NULL, - mei_interrupt_thread_handler, - IRQF_ONESHOT, KBUILD_MODNAME, dev); - else - err = request_threaded_irq(pdev->irq, - mei_interrupt_quick_handler, - mei_interrupt_thread_handler, - IRQF_SHARED, KBUILD_MODNAME, dev); - - if (err) { - dev_err(&pdev->dev, "request_threaded_irq failed: irq = %d.\n", - pdev->irq); - return err; - } - - mutex_lock(&dev->device_lock); - dev->dev_state = MEI_DEV_POWER_UP; - mei_reset(dev, 1); - mutex_unlock(&dev->device_lock); - - /* Start timer if stopped in suspend */ - schedule_delayed_work(&dev->timer_work, HZ); - - return err; + mei_cl_bus_exit(); } -static SIMPLE_DEV_PM_OPS(mei_pm_ops, mei_pci_suspend, mei_pci_resume); -#define MEI_PM_OPS (&mei_pm_ops) -#else -#define MEI_PM_OPS NULL -#endif /* CONFIG_PM */ -/* - * PCI driver structure - */ -static struct pci_driver mei_driver = { - .name = KBUILD_MODNAME, - .id_table = mei_pci_tbl, - .probe = mei_probe, - .remove = __devexit_p(mei_remove), - .shutdown = __devexit_p(mei_remove), - .driver.pm = MEI_PM_OPS, -}; -module_pci_driver(mei_driver); +module_init(mei_init); +module_exit(mei_exit); MODULE_AUTHOR("Intel Corporation"); MODULE_DESCRIPTION("Intel(R) Management Engine Interface"); MODULE_LICENSE("GPL v2"); + diff --git a/drivers/misc/mei/mei_dev.h b/drivers/misc/mei/mei_dev.h index adb35fb9281..5c7e990e2f2 100644 --- a/drivers/misc/mei/mei_dev.h +++ b/drivers/misc/mei/mei_dev.h @@ -19,8 +19,12 @@ #include <linux/types.h> #include <linux/watchdog.h> +#include <linux/poll.h> #include <linux/mei.h> +#include <linux/mei_cl_bus.h> + #include "hw.h" +#include "hbm.h" /* * watch dog definition @@ -43,7 +47,7 @@ /* * AMTHI Client UUID */ -extern const uuid_le mei_amthi_guid; +extern const uuid_le mei_amthif_guid; /* * Watchdog Client UUID @@ -51,25 +55,31 @@ extern const uuid_le mei_amthi_guid; extern const uuid_le mei_wd_guid; /* - * Watchdog independence state message + * Number of Maximum MEI Clients */ -extern const u8 mei_wd_state_independence_msg[3][4]; +#define MEI_CLIENTS_MAX 256 /* - * Number of Maximum MEI Clients + * maximum number of consecutive resets */ -#define MEI_CLIENTS_MAX 256 +#define MEI_MAX_CONSEC_RESET 3 /* * Number of File descriptors/handles * that can be opened to the driver. * - * Limit to 253: 256 Total Clients - * minus internal client for MEI Bus Messags - * minus internal client for AMTHI - * minus internal client for Watchdog + * Limit to 255: 256 Total Clients + * minus internal client for MEI Bus Messages */ -#define MEI_MAX_OPEN_HANDLE_COUNT (MEI_CLIENTS_MAX - 3) +#define MEI_MAX_OPEN_HANDLE_COUNT (MEI_CLIENTS_MAX - 1) + +/* + * Internal Clients Number + */ +#define MEI_HOST_CLIENT_ID_ANY (-1) +#define MEI_HBM_HOST_CLIENT_ID 0 /* not used, just for documentation */ +#define MEI_WD_HOST_CLIENT_ID 1 +#define MEI_IAMTHIF_HOST_CLIENT_ID 2 /* File state */ @@ -86,22 +96,14 @@ enum mei_dev_state { MEI_DEV_INITIALIZING = 0, MEI_DEV_INIT_CLIENTS, MEI_DEV_ENABLED, - MEI_DEV_RESETING, + MEI_DEV_RESETTING, MEI_DEV_DISABLED, - MEI_DEV_RECOVERING_FROM_RESET, MEI_DEV_POWER_DOWN, MEI_DEV_POWER_UP }; const char *mei_dev_state_str(int state); -/* init clients states*/ -enum mei_init_clients_states { - MEI_START_MESSAGE = 0, - MEI_ENUM_CLIENTS_MESSAGE, - MEI_CLIENT_PROPERTIES_MESSAGE -}; - enum iamthif_states { MEI_IAMTHIF_IDLE, MEI_IAMTHIF_WRITING, @@ -125,36 +127,81 @@ enum mei_wd_states { MEI_WD_STOPPING, }; -/* MEI CB */ -enum mei_cb_major_types { - MEI_READ = 0, - MEI_WRITE, - MEI_IOCTL, - MEI_OPEN, - MEI_CLOSE +/** + * enum mei_cb_file_ops - file operation associated with the callback + * @MEI_FOP_READ - read + * @MEI_FOP_WRITE - write + * @MEI_FOP_CONNECT - connect + * @MEI_FOP_DISCONNECT_RSP - disconnect response + * @MEI_FOP_OPEN - open + * @MEI_FOP_CLOSE - close + */ +enum mei_cb_file_ops { + MEI_FOP_READ = 0, + MEI_FOP_WRITE, + MEI_FOP_CONNECT, + MEI_FOP_DISCONNECT_RSP, + MEI_FOP_OPEN, + MEI_FOP_CLOSE }; /* * Intel MEI message data struct */ -struct mei_message_data { +struct mei_msg_data { u32 size; unsigned char *data; }; +/* Maximum number of processed FW status registers */ +#define MEI_FW_STATUS_MAX 2 + +/* + * struct mei_fw_status - storage of FW status data + * + * @count - number of actually available elements in array + * @status - FW status registers + */ +struct mei_fw_status { + int count; + u32 status[MEI_FW_STATUS_MAX]; +}; + +/** + * struct mei_me_client - representation of me (fw) client + * + * @props - client properties + * @client_id - me client id + * @mei_flow_ctrl_creds - flow control credits + */ +struct mei_me_client { + struct mei_client_properties props; + u8 client_id; + u8 mei_flow_ctrl_creds; +}; + + +struct mei_cl; +/** + * struct mei_cl_cb - file operation callback structure + * + * @cl - file client who is running this operation + * @fop_type - file operation type + */ struct mei_cl_cb { - struct list_head cb_list; - enum mei_cb_major_types major_file_operations; - void *file_private; - struct mei_message_data request_buffer; - struct mei_message_data response_buffer; - unsigned long information; + struct list_head list; + struct mei_cl *cl; + enum mei_cb_file_ops fop_type; + struct mei_msg_data request_buffer; + struct mei_msg_data response_buffer; + unsigned long buf_idx; unsigned long read_time; struct file *file_object; + u32 internal:1; }; -/* MEI client instance carried as file->pirvate_data*/ +/* MEI client instance carried as file->private_data*/ struct mei_cl { struct list_head link; struct mei_device *dev; @@ -162,7 +209,6 @@ struct mei_cl { wait_queue_head_t tx_wait; wait_queue_head_t rx_wait; wait_queue_head_t wait; - int read_pending; int status; /* ID of client connected */ u8 host_client_id; @@ -171,97 +217,285 @@ struct mei_cl { u8 timer_count; enum mei_file_transaction_states reading_state; enum mei_file_transaction_states writing_state; - int sm_state; struct mei_cl_cb *read_cb; + + /* MEI CL bus data */ + struct mei_cl_device *device; + struct list_head device_link; + uuid_le device_uuid; +}; + +/** struct mei_hw_ops + * + * @fw_status - read FW status from PCI config space + * @host_is_ready - query for host readiness + + * @hw_is_ready - query if hw is ready + * @hw_reset - reset hw + * @hw_start - start hw after reset + * @hw_config - configure hw + + * @pg_state - power gating state of the device + * @pg_is_enabled - is power gating enabled + + * @intr_clear - clear pending interrupts + * @intr_enable - enable interrupts + * @intr_disable - disable interrupts + + * @hbuf_free_slots - query for write buffer empty slots + * @hbuf_is_ready - query if write buffer is empty + * @hbuf_max_len - query for write buffer max len + + * @write - write a message to FW + + * @rdbuf_full_slots - query how many slots are filled + + * @read_hdr - get first 4 bytes (header) + * @read - read a buffer from the FW + */ +struct mei_hw_ops { + + int (*fw_status)(struct mei_device *dev, + struct mei_fw_status *fw_status); + bool (*host_is_ready)(struct mei_device *dev); + + bool (*hw_is_ready)(struct mei_device *dev); + int (*hw_reset)(struct mei_device *dev, bool enable); + int (*hw_start)(struct mei_device *dev); + void (*hw_config)(struct mei_device *dev); + + enum mei_pg_state (*pg_state)(struct mei_device *dev); + bool (*pg_is_enabled)(struct mei_device *dev); + + void (*intr_clear)(struct mei_device *dev); + void (*intr_enable)(struct mei_device *dev); + void (*intr_disable)(struct mei_device *dev); + + int (*hbuf_free_slots)(struct mei_device *dev); + bool (*hbuf_is_ready)(struct mei_device *dev); + size_t (*hbuf_max_len)(const struct mei_device *dev); + + int (*write)(struct mei_device *dev, + struct mei_msg_hdr *hdr, + unsigned char *buf); + + int (*rdbuf_full_slots)(struct mei_device *dev); + + u32 (*read_hdr)(const struct mei_device *dev); + int (*read)(struct mei_device *dev, + unsigned char *buf, unsigned long len); +}; + +/* MEI bus API*/ + +/** + * struct mei_cl_ops - MEI CL device ops + * This structure allows ME host clients to implement technology + * specific operations. + * + * @enable: Enable an MEI CL device. Some devices require specific + * HECI commands to initialize completely. + * @disable: Disable an MEI CL device. + * @send: Tx hook for the device. This allows ME host clients to trap + * the device driver buffers before actually physically + * pushing it to the ME. + * @recv: Rx hook for the device. This allows ME host clients to trap the + * ME buffers before forwarding them to the device driver. + */ +struct mei_cl_ops { + int (*enable)(struct mei_cl_device *device); + int (*disable)(struct mei_cl_device *device); + int (*send)(struct mei_cl_device *device, u8 *buf, size_t length); + int (*recv)(struct mei_cl_device *device, u8 *buf, size_t length); }; -struct mei_io_list { - struct mei_cl_cb mei_cb; +struct mei_cl_device *mei_cl_add_device(struct mei_device *dev, + uuid_le uuid, char *name, + struct mei_cl_ops *ops); +void mei_cl_remove_device(struct mei_cl_device *device); + +int __mei_cl_async_send(struct mei_cl *cl, u8 *buf, size_t length); +int __mei_cl_send(struct mei_cl *cl, u8 *buf, size_t length); +int __mei_cl_recv(struct mei_cl *cl, u8 *buf, size_t length); +void mei_cl_bus_rx_event(struct mei_cl *cl); +void mei_cl_bus_remove_devices(struct mei_device *dev); +int mei_cl_bus_init(void); +void mei_cl_bus_exit(void); + + +/** + * struct mei_cl_device - MEI device handle + * An mei_cl_device pointer is returned from mei_add_device() + * and links MEI bus clients to their actual ME host client pointer. + * Drivers for MEI devices will get an mei_cl_device pointer + * when being probed and shall use it for doing ME bus I/O. + * + * @dev: linux driver model device pointer + * @uuid: me client uuid + * @cl: mei client + * @ops: ME transport ops + * @event_cb: Drivers register this callback to get asynchronous ME + * events (e.g. Rx buffer pending) notifications. + * @events: Events bitmask sent to the driver. + * @priv_data: client private data + */ +struct mei_cl_device { + struct device dev; + + struct mei_cl *cl; + + const struct mei_cl_ops *ops; + + struct work_struct event_work; + mei_cl_event_cb_t event_cb; + void *event_context; + unsigned long events; + + void *priv_data; +}; + + + /** + * enum mei_pg_event - power gating transition events + * + * @MEI_PG_EVENT_IDLE: the driver is not in power gating transition + * @MEI_PG_EVENT_WAIT: the driver is waiting for a pg event to complete + * @MEI_PG_EVENT_RECEIVED: the driver received pg event + */ +enum mei_pg_event { + MEI_PG_EVENT_IDLE, + MEI_PG_EVENT_WAIT, + MEI_PG_EVENT_RECEIVED, }; /** - * struct mei_deive - MEI private device struct - * @hbuf_depth - depth of host(write) buffer + * enum mei_pg_state - device internal power gating state + * + * @MEI_PG_OFF: device is not power gated - it is active + * @MEI_PG_ON: device is power gated - it is in lower power state + */ +enum mei_pg_state { + MEI_PG_OFF = 0, + MEI_PG_ON = 1, +}; + +/* + * mei_cfg + * + * @fw_status - FW status + * @quirk_probe - device exclusion quirk + */ +struct mei_cfg { + const struct mei_fw_status fw_status; + bool (*quirk_probe)(struct pci_dev *pdev); +}; + + +#define MEI_PCI_DEVICE(dev, cfg) \ + .vendor = PCI_VENDOR_ID_INTEL, .device = (dev), \ + .subvendor = PCI_ANY_ID, .subdevice = PCI_ANY_ID, \ + .driver_data = (kernel_ulong_t)&(cfg) + + +/** + * struct mei_device - MEI private device struct + + * @reset_count - limits the number of consecutive resets + * @hbm_state - state of host bus message protocol + * @pg_event - power gating event + * @mem_addr - mem mapped base register address + + * @hbuf_depth - depth of hardware host/write buffer is slots + * @hbuf_is_ready - query if the host host/write buffer is ready + * @wr_msg - the buffer for hbm control messages + * @cfg - per device generation config and ops */ struct mei_device { struct pci_dev *pdev; /* pointer to pci device struct */ /* * lists of queues */ - /* array of pointers to aio lists */ - struct mei_io_list read_list; /* driver read queue */ - struct mei_io_list write_list; /* driver write queue */ - struct mei_io_list write_waiting_list; /* write waiting queue */ - struct mei_io_list ctrl_wr_list; /* managed write IOCTL list */ - struct mei_io_list ctrl_rd_list; /* managed read IOCTL list */ - struct mei_io_list amthi_cmd_list; /* amthi list for cmd waiting */ - - /* driver managed amthi list for reading completed amthi cmd data */ - struct mei_io_list amthi_read_complete_list; + /* array of pointers to aio lists */ + struct mei_cl_cb read_list; /* driver read queue */ + struct mei_cl_cb write_list; /* driver write queue */ + struct mei_cl_cb write_waiting_list; /* write waiting queue */ + struct mei_cl_cb ctrl_wr_list; /* managed write IOCTL list */ + struct mei_cl_cb ctrl_rd_list; /* managed read IOCTL list */ + /* * list of files */ struct list_head file_list; long open_handle_count; - /* - * memory of device - */ - unsigned int mem_base; - unsigned int mem_length; - void __iomem *mem_addr; + /* * lock for the device */ struct mutex device_lock; /* device lock */ struct delayed_work timer_work; /* MEI timer delayed work (timeouts) */ - bool recvd_msg; - /* - * hw states of host and fw(ME) - */ - u32 host_hw_state; - u32 me_hw_state; - u8 hbuf_depth; + + bool recvd_hw_ready; /* * waiting queue for receive message from FW */ + wait_queue_head_t wait_hw_ready; + wait_queue_head_t wait_pg; wait_queue_head_t wait_recvd_msg; wait_queue_head_t wait_stop_wd; /* * mei device states */ + unsigned long reset_count; enum mei_dev_state dev_state; - enum mei_init_clients_states init_clients_state; + enum mei_hbm_state hbm_state; u16 init_clients_timer; - bool need_reset; - u32 extra_write_index; + /* + * Power Gating support + */ + enum mei_pg_event pg_event; +#ifdef CONFIG_PM_RUNTIME + struct dev_pm_domain pg_domain; +#endif /* CONFIG_PM_RUNTIME */ + unsigned char rd_msg_buf[MEI_RD_MSG_BUF_SIZE]; /* control messages */ - u32 wr_msg_buf[128]; /* used for control messages */ - u32 ext_msg_buf[8]; /* for control responses */ u32 rd_msg_hdr; + /* write buffer */ + u8 hbuf_depth; + bool hbuf_is_ready; + + /* used for control messages */ + struct { + struct mei_msg_hdr hdr; + unsigned char data[128]; + } wr_msg; + struct hbm_version version; struct mei_me_client *me_clients; /* Note: memory has to be allocated */ DECLARE_BITMAP(me_clients_map, MEI_CLIENTS_MAX); DECLARE_BITMAP(host_clients_map, MEI_CLIENTS_MAX); - u8 me_clients_num; - u8 me_client_presentation_num; - u8 me_client_index; - bool mei_host_buffer_is_empty; + unsigned long me_clients_num; + unsigned long me_client_presentation_num; + unsigned long me_client_index; struct mei_cl wd_cl; enum mei_wd_states wd_state; - bool wd_interface_reg; bool wd_pending; u16 wd_timeout; unsigned char wd_data[MEI_WD_START_MSG_SIZE]; + /* amthif list for cmd waiting */ + struct mei_cl_cb amthif_cmd_list; + /* driver managed amthif list for reading completed amthif cmd data */ + struct mei_cl_cb amthif_rd_complete_list; struct file *iamthif_file_object; struct mei_cl iamthif_cl; struct mei_cl_cb *iamthif_current_cb; + long iamthif_open_count; int iamthif_mtu; unsigned long iamthif_timer; u32 iamthif_stall_timer; @@ -272,169 +506,247 @@ struct mei_device { bool iamthif_flow_control_pending; bool iamthif_ioctl; bool iamthif_canceled; -}; + struct work_struct init_work; + struct work_struct reset_work; -/* - * mei init function prototypes - */ -struct mei_device *mei_device_init(struct pci_dev *pdev); -void mei_reset(struct mei_device *dev, int interrupts); -int mei_hw_init(struct mei_device *dev); -int mei_task_initialize_clients(void *data); -int mei_initialize_clients(struct mei_device *dev); -int mei_disconnect_host_client(struct mei_device *dev, struct mei_cl *cl); -void mei_remove_client_from_file_list(struct mei_device *dev, u8 host_client_id); -void mei_host_init_iamthif(struct mei_device *dev); -void mei_allocate_me_clients_storage(struct mei_device *dev); - - -int mei_me_cl_update_filext(struct mei_device *dev, struct mei_cl *cl, - const uuid_le *cguid, u8 host_client_id); -int mei_me_cl_by_uuid(const struct mei_device *dev, const uuid_le *cuuid); -int mei_me_cl_by_id(struct mei_device *dev, u8 client_id); + /* List of bus devices */ + struct list_head device_list; -/* - * MEI IO List Functions - */ -void mei_io_list_init(struct mei_io_list *list); -void mei_io_list_flush(struct mei_io_list *list, struct mei_cl *cl); +#if IS_ENABLED(CONFIG_DEBUG_FS) + struct dentry *dbgfs_dir; +#endif /* CONFIG_DEBUG_FS */ -/* - * MEI ME Client Functions - */ -struct mei_cl *mei_cl_allocate(struct mei_device *dev); -void mei_cl_init(struct mei_cl *cl, struct mei_device *dev); -int mei_cl_flush_queues(struct mei_cl *cl); + const struct mei_hw_ops *ops; + const struct mei_cfg *cfg; + char hw[0] __aligned(sizeof(void *)); +}; + +static inline unsigned long mei_secs_to_jiffies(unsigned long sec) +{ + return msecs_to_jiffies(sec * MSEC_PER_SEC); +} + /** - * mei_cl_cmp_id - tells if file private data have same id - * - * @fe1: private data of 1. file object - * @fe2: private data of 2. file object - * - * returns true - if ids are the same and not NULL + * mei_data2slots - get slots - number of (dwords) from a message length + * + size of the mei header + * @length - size of the messages in bytes + * returns - number of slots */ -static inline bool mei_cl_cmp_id(const struct mei_cl *cl1, - const struct mei_cl *cl2) +static inline u32 mei_data2slots(size_t length) { - return cl1 && cl2 && - (cl1->host_client_id == cl2->host_client_id) && - (cl1->me_client_id == cl2->me_client_id); + return DIV_ROUND_UP(sizeof(struct mei_msg_hdr) + length, 4); } - +/** + * mei_slots2data- get data in slots - bytes from slots + * @slots - number of available slots + * returns - number of bytes in slots + */ +static inline u32 mei_slots2data(int slots) +{ + return slots * 4; +} /* - * MEI Host Client Functions + * mei init function prototypes */ -void mei_host_start_message(struct mei_device *dev); -void mei_host_enum_clients_message(struct mei_device *dev); -int mei_host_client_properties(struct mei_device *dev); +void mei_device_init(struct mei_device *dev, const struct mei_cfg *cfg); +int mei_reset(struct mei_device *dev); +int mei_start(struct mei_device *dev); +int mei_restart(struct mei_device *dev); +void mei_stop(struct mei_device *dev); +void mei_cancel_work(struct mei_device *dev); /* * MEI interrupt functions prototype */ -irqreturn_t mei_interrupt_quick_handler(int irq, void *dev_id); -irqreturn_t mei_interrupt_thread_handler(int irq, void *dev_id); + void mei_timer(struct work_struct *work); +int mei_irq_read_handler(struct mei_device *dev, + struct mei_cl_cb *cmpl_list, s32 *slots); + +int mei_irq_write_handler(struct mei_device *dev, struct mei_cl_cb *cmpl_list); +void mei_irq_compl_handler(struct mei_device *dev, struct mei_cl_cb *cmpl_list); /* - * MEI input output function prototype + * AMTHIF - AMT Host Interface Functions */ -int mei_ioctl_connect_client(struct file *file, - struct mei_connect_client_data *data); +void mei_amthif_reset_params(struct mei_device *dev); + +int mei_amthif_host_init(struct mei_device *dev); -int mei_start_read(struct mei_device *dev, struct mei_cl *cl); +int mei_amthif_write(struct mei_device *dev, struct mei_cl_cb *priv_cb); -int amthi_write(struct mei_device *dev, struct mei_cl_cb *priv_cb); +int mei_amthif_read(struct mei_device *dev, struct file *file, + char __user *ubuf, size_t length, loff_t *offset); -int amthi_read(struct mei_device *dev, struct file *file, - char __user *ubuf, size_t length, loff_t *offset); +unsigned int mei_amthif_poll(struct mei_device *dev, + struct file *file, poll_table *wait); -struct mei_cl_cb *find_amthi_read_list_entry(struct mei_device *dev, +int mei_amthif_release(struct mei_device *dev, struct file *file); + +struct mei_cl_cb *mei_amthif_find_read_list_entry(struct mei_device *dev, struct file *file); -void mei_run_next_iamthif_cmd(struct mei_device *dev); +void mei_amthif_run_next_cmd(struct mei_device *dev); -void mei_free_cb_private(struct mei_cl_cb *priv_cb); +int mei_amthif_irq_write(struct mei_cl *cl, struct mei_cl_cb *cb, + struct mei_cl_cb *cmpl_list); +void mei_amthif_complete(struct mei_device *dev, struct mei_cl_cb *cb); +int mei_amthif_irq_read_msg(struct mei_device *dev, + struct mei_msg_hdr *mei_hdr, + struct mei_cl_cb *complete_list); +int mei_amthif_irq_read(struct mei_device *dev, s32 *slots); /* - * Register Access Function + * NFC functions */ +int mei_nfc_host_init(struct mei_device *dev); +void mei_nfc_host_exit(struct mei_device *dev); -/** - * mei_reg_read - Reads 32bit data from the mei device - * - * @dev: the device structure - * @offset: offset from which to read the data - * - * returns register value (u32) +/* + * NFC Client UUID + */ +extern const uuid_le mei_nfc_guid; + +int mei_wd_send(struct mei_device *dev); +int mei_wd_stop(struct mei_device *dev); +int mei_wd_host_init(struct mei_device *dev); +/* + * mei_watchdog_register - Registering watchdog interface + * once we got connection to the WD Client + * @dev - mei device + */ +int mei_watchdog_register(struct mei_device *dev); +/* + * mei_watchdog_unregister - Unregistering watchdog interface + * @dev - mei device + */ +void mei_watchdog_unregister(struct mei_device *dev); + +/* + * Register Access Function */ -static inline u32 mei_reg_read(const struct mei_device *dev, - unsigned long offset) + + +static inline void mei_hw_config(struct mei_device *dev) { - return ioread32(dev->mem_addr + offset); + dev->ops->hw_config(dev); } -/** - * mei_reg_write - Writes 32bit data to the mei device - * - * @dev: the device structure - * @offset: offset from which to write the data - * @value: register value to write (u32) - */ -static inline void mei_reg_write(const struct mei_device *dev, - unsigned long offset, u32 value) +static inline enum mei_pg_state mei_pg_state(struct mei_device *dev) { - iowrite32(value, dev->mem_addr + offset); + return dev->ops->pg_state(dev); } -/** - * mei_hcsr_read - Reads 32bit data from the host CSR - * - * @dev: the device structure - * - * returns the byte read. - */ -static inline u32 mei_hcsr_read(const struct mei_device *dev) +static inline bool mei_pg_is_enabled(struct mei_device *dev) { - return mei_reg_read(dev, H_CSR); + return dev->ops->pg_is_enabled(dev); } -/** - * mei_mecsr_read - Reads 32bit data from the ME CSR - * - * @dev: the device structure - * - * returns ME_CSR_HA register value (u32) - */ -static inline u32 mei_mecsr_read(const struct mei_device *dev) +static inline int mei_hw_reset(struct mei_device *dev, bool enable) { - return mei_reg_read(dev, ME_CSR_HA); + return dev->ops->hw_reset(dev, enable); } -/** - * get_me_cb_rw - Reads 32bit data from the mei ME_CB_RW register - * - * @dev: the device structure - * - * returns ME_CB_RW register value (u32) - */ -static inline u32 mei_mecbrw_read(const struct mei_device *dev) +static inline int mei_hw_start(struct mei_device *dev) +{ + return dev->ops->hw_start(dev); +} + +static inline void mei_clear_interrupts(struct mei_device *dev) { - return mei_reg_read(dev, ME_CB_RW); + dev->ops->intr_clear(dev); } +static inline void mei_enable_interrupts(struct mei_device *dev) +{ + dev->ops->intr_enable(dev); +} -/* - * mei interface function prototypes - */ -void mei_hcsr_set(struct mei_device *dev); -void mei_csr_clear_his(struct mei_device *dev); +static inline void mei_disable_interrupts(struct mei_device *dev) +{ + dev->ops->intr_disable(dev); +} + +static inline bool mei_host_is_ready(struct mei_device *dev) +{ + return dev->ops->host_is_ready(dev); +} +static inline bool mei_hw_is_ready(struct mei_device *dev) +{ + return dev->ops->hw_is_ready(dev); +} + +static inline bool mei_hbuf_is_ready(struct mei_device *dev) +{ + return dev->ops->hbuf_is_ready(dev); +} + +static inline int mei_hbuf_empty_slots(struct mei_device *dev) +{ + return dev->ops->hbuf_free_slots(dev); +} + +static inline size_t mei_hbuf_max_len(const struct mei_device *dev) +{ + return dev->ops->hbuf_max_len(dev); +} + +static inline int mei_write_message(struct mei_device *dev, + struct mei_msg_hdr *hdr, + unsigned char *buf) +{ + return dev->ops->write(dev, hdr, buf); +} + +static inline u32 mei_read_hdr(const struct mei_device *dev) +{ + return dev->ops->read_hdr(dev); +} + +static inline void mei_read_slots(struct mei_device *dev, + unsigned char *buf, unsigned long len) +{ + dev->ops->read(dev, buf, len); +} + +static inline int mei_count_full_read_slots(struct mei_device *dev) +{ + return dev->ops->rdbuf_full_slots(dev); +} + +int mei_fw_status(struct mei_device *dev, struct mei_fw_status *fw_status); + +#define FW_STS_FMT "%08X %08X" +#define FW_STS_PRM(fw_status) \ + (fw_status).count > 0 ? (fw_status).status[0] : 0xDEADBEEF, \ + (fw_status).count > 1 ? (fw_status).status[1] : 0xDEADBEEF + +bool mei_hbuf_acquire(struct mei_device *dev); + +bool mei_write_is_idle(struct mei_device *dev); + +#if IS_ENABLED(CONFIG_DEBUG_FS) +int mei_dbgfs_register(struct mei_device *dev, const char *name); +void mei_dbgfs_deregister(struct mei_device *dev); +#else +static inline int mei_dbgfs_register(struct mei_device *dev, const char *name) +{ + return 0; +} +static inline void mei_dbgfs_deregister(struct mei_device *dev) {} +#endif /* CONFIG_DEBUG_FS */ + +int mei_register(struct mei_device *dev); +void mei_deregister(struct mei_device *dev); -void mei_enable_interrupts(struct mei_device *dev); -void mei_disable_interrupts(struct mei_device *dev); +#define MEI_HDR_FMT "hdr:host=%02d me=%02d len=%d internal=%1d comp=%1d" +#define MEI_HDR_PRM(hdr) \ + (hdr)->host_addr, (hdr)->me_addr, \ + (hdr)->length, (hdr)->internal, (hdr)->msg_complete #endif diff --git a/drivers/misc/mei/nfc.c b/drivers/misc/mei/nfc.c new file mode 100644 index 00000000000..3095fc514a6 --- /dev/null +++ b/drivers/misc/mei/nfc.c @@ -0,0 +1,558 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2013, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/device.h> +#include <linux/pci.h> +#include <linux/mei_cl_bus.h> + +#include "mei_dev.h" +#include "client.h" + +struct mei_nfc_cmd { + u8 command; + u8 status; + u16 req_id; + u32 reserved; + u16 data_size; + u8 sub_command; + u8 data[]; +} __packed; + +struct mei_nfc_reply { + u8 command; + u8 status; + u16 req_id; + u32 reserved; + u16 data_size; + u8 sub_command; + u8 reply_status; + u8 data[]; +} __packed; + +struct mei_nfc_if_version { + u8 radio_version_sw[3]; + u8 reserved[3]; + u8 radio_version_hw[3]; + u8 i2c_addr; + u8 fw_ivn; + u8 vendor_id; + u8 radio_type; +} __packed; + +struct mei_nfc_connect { + u8 fw_ivn; + u8 vendor_id; +} __packed; + +struct mei_nfc_connect_resp { + u8 fw_ivn; + u8 vendor_id; + u16 me_major; + u16 me_minor; + u16 me_hotfix; + u16 me_build; +} __packed; + +struct mei_nfc_hci_hdr { + u8 cmd; + u8 status; + u16 req_id; + u32 reserved; + u16 data_size; +} __packed; + +#define MEI_NFC_CMD_MAINTENANCE 0x00 +#define MEI_NFC_CMD_HCI_SEND 0x01 +#define MEI_NFC_CMD_HCI_RECV 0x02 + +#define MEI_NFC_SUBCMD_CONNECT 0x00 +#define MEI_NFC_SUBCMD_IF_VERSION 0x01 + +#define MEI_NFC_HEADER_SIZE 10 + +/** mei_nfc_dev - NFC mei device + * + * @cl: NFC host client + * @cl_info: NFC info host client + * @init_work: perform connection to the info client + * @fw_ivn: NFC Interface Version Number + * @vendor_id: NFC manufacturer ID + * @radio_type: NFC radio type + */ +struct mei_nfc_dev { + struct mei_cl *cl; + struct mei_cl *cl_info; + struct work_struct init_work; + wait_queue_head_t send_wq; + u8 fw_ivn; + u8 vendor_id; + u8 radio_type; + char *bus_name; + + u16 req_id; + u16 recv_req_id; +}; + +static struct mei_nfc_dev nfc_dev; + +/* UUIDs for NFC F/W clients */ +const uuid_le mei_nfc_guid = UUID_LE(0x0bb17a78, 0x2a8e, 0x4c50, + 0x94, 0xd4, 0x50, 0x26, + 0x67, 0x23, 0x77, 0x5c); + +static const uuid_le mei_nfc_info_guid = UUID_LE(0xd2de1625, 0x382d, 0x417d, + 0x48, 0xa4, 0xef, 0xab, + 0xba, 0x8a, 0x12, 0x06); + +/* Vendors */ +#define MEI_NFC_VENDOR_INSIDE 0x00 +#define MEI_NFC_VENDOR_NXP 0x01 + +/* Radio types */ +#define MEI_NFC_VENDOR_INSIDE_UREAD 0x00 +#define MEI_NFC_VENDOR_NXP_PN544 0x01 + +static void mei_nfc_free(struct mei_nfc_dev *ndev) +{ + if (ndev->cl) { + list_del(&ndev->cl->device_link); + mei_cl_unlink(ndev->cl); + kfree(ndev->cl); + } + + if (ndev->cl_info) { + list_del(&ndev->cl_info->device_link); + mei_cl_unlink(ndev->cl_info); + kfree(ndev->cl_info); + } + + memset(ndev, 0, sizeof(struct mei_nfc_dev)); +} + +static int mei_nfc_build_bus_name(struct mei_nfc_dev *ndev) +{ + struct mei_device *dev; + + if (!ndev->cl) + return -ENODEV; + + dev = ndev->cl->dev; + + switch (ndev->vendor_id) { + case MEI_NFC_VENDOR_INSIDE: + switch (ndev->radio_type) { + case MEI_NFC_VENDOR_INSIDE_UREAD: + ndev->bus_name = "microread"; + return 0; + + default: + dev_err(&dev->pdev->dev, "Unknown radio type 0x%x\n", + ndev->radio_type); + + return -EINVAL; + } + + case MEI_NFC_VENDOR_NXP: + switch (ndev->radio_type) { + case MEI_NFC_VENDOR_NXP_PN544: + ndev->bus_name = "pn544"; + return 0; + default: + dev_err(&dev->pdev->dev, "Unknown radio type 0x%x\n", + ndev->radio_type); + + return -EINVAL; + } + + default: + dev_err(&dev->pdev->dev, "Unknown vendor ID 0x%x\n", + ndev->vendor_id); + + return -EINVAL; + } + + return 0; +} + +static int mei_nfc_connect(struct mei_nfc_dev *ndev) +{ + struct mei_device *dev; + struct mei_cl *cl; + struct mei_nfc_cmd *cmd, *reply; + struct mei_nfc_connect *connect; + struct mei_nfc_connect_resp *connect_resp; + size_t connect_length, connect_resp_length; + int bytes_recv, ret; + + cl = ndev->cl; + dev = cl->dev; + + connect_length = sizeof(struct mei_nfc_cmd) + + sizeof(struct mei_nfc_connect); + + connect_resp_length = sizeof(struct mei_nfc_cmd) + + sizeof(struct mei_nfc_connect_resp); + + cmd = kzalloc(connect_length, GFP_KERNEL); + if (!cmd) + return -ENOMEM; + connect = (struct mei_nfc_connect *)cmd->data; + + reply = kzalloc(connect_resp_length, GFP_KERNEL); + if (!reply) { + kfree(cmd); + return -ENOMEM; + } + + connect_resp = (struct mei_nfc_connect_resp *)reply->data; + + cmd->command = MEI_NFC_CMD_MAINTENANCE; + cmd->data_size = 3; + cmd->sub_command = MEI_NFC_SUBCMD_CONNECT; + connect->fw_ivn = ndev->fw_ivn; + connect->vendor_id = ndev->vendor_id; + + ret = __mei_cl_send(cl, (u8 *)cmd, connect_length); + if (ret < 0) { + dev_err(&dev->pdev->dev, "Could not send connect cmd\n"); + goto err; + } + + bytes_recv = __mei_cl_recv(cl, (u8 *)reply, connect_resp_length); + if (bytes_recv < 0) { + dev_err(&dev->pdev->dev, "Could not read connect response\n"); + ret = bytes_recv; + goto err; + } + + dev_info(&dev->pdev->dev, "IVN 0x%x Vendor ID 0x%x\n", + connect_resp->fw_ivn, connect_resp->vendor_id); + + dev_info(&dev->pdev->dev, "ME FW %d.%d.%d.%d\n", + connect_resp->me_major, connect_resp->me_minor, + connect_resp->me_hotfix, connect_resp->me_build); + + ret = 0; + +err: + kfree(reply); + kfree(cmd); + + return ret; +} + +static int mei_nfc_if_version(struct mei_nfc_dev *ndev) +{ + struct mei_device *dev; + struct mei_cl *cl; + + struct mei_nfc_cmd cmd; + struct mei_nfc_reply *reply = NULL; + struct mei_nfc_if_version *version; + size_t if_version_length; + int bytes_recv, ret; + + cl = ndev->cl_info; + dev = cl->dev; + + memset(&cmd, 0, sizeof(struct mei_nfc_cmd)); + cmd.command = MEI_NFC_CMD_MAINTENANCE; + cmd.data_size = 1; + cmd.sub_command = MEI_NFC_SUBCMD_IF_VERSION; + + ret = __mei_cl_send(cl, (u8 *)&cmd, sizeof(struct mei_nfc_cmd)); + if (ret < 0) { + dev_err(&dev->pdev->dev, "Could not send IF version cmd\n"); + return ret; + } + + /* to be sure on the stack we alloc memory */ + if_version_length = sizeof(struct mei_nfc_reply) + + sizeof(struct mei_nfc_if_version); + + reply = kzalloc(if_version_length, GFP_KERNEL); + if (!reply) + return -ENOMEM; + + bytes_recv = __mei_cl_recv(cl, (u8 *)reply, if_version_length); + if (bytes_recv < 0 || bytes_recv < sizeof(struct mei_nfc_reply)) { + dev_err(&dev->pdev->dev, "Could not read IF version\n"); + ret = -EIO; + goto err; + } + + version = (struct mei_nfc_if_version *)reply->data; + + ndev->fw_ivn = version->fw_ivn; + ndev->vendor_id = version->vendor_id; + ndev->radio_type = version->radio_type; + +err: + kfree(reply); + return ret; +} + +static int mei_nfc_enable(struct mei_cl_device *cldev) +{ + struct mei_device *dev; + struct mei_nfc_dev *ndev = &nfc_dev; + int ret; + + dev = ndev->cl->dev; + + ret = mei_nfc_connect(ndev); + if (ret < 0) { + dev_err(&dev->pdev->dev, "Could not connect to NFC"); + return ret; + } + + return 0; +} + +static int mei_nfc_disable(struct mei_cl_device *cldev) +{ + return 0; +} + +static int mei_nfc_send(struct mei_cl_device *cldev, u8 *buf, size_t length) +{ + struct mei_device *dev; + struct mei_nfc_dev *ndev; + struct mei_nfc_hci_hdr *hdr; + u8 *mei_buf; + int err; + + ndev = (struct mei_nfc_dev *) cldev->priv_data; + dev = ndev->cl->dev; + + mei_buf = kzalloc(length + MEI_NFC_HEADER_SIZE, GFP_KERNEL); + if (!mei_buf) + return -ENOMEM; + + hdr = (struct mei_nfc_hci_hdr *) mei_buf; + hdr->cmd = MEI_NFC_CMD_HCI_SEND; + hdr->status = 0; + hdr->req_id = ndev->req_id; + hdr->reserved = 0; + hdr->data_size = length; + + memcpy(mei_buf + MEI_NFC_HEADER_SIZE, buf, length); + + err = __mei_cl_send(ndev->cl, mei_buf, length + MEI_NFC_HEADER_SIZE); + if (err < 0) + return err; + + kfree(mei_buf); + + if (!wait_event_interruptible_timeout(ndev->send_wq, + ndev->recv_req_id == ndev->req_id, HZ)) { + dev_err(&dev->pdev->dev, "NFC MEI command timeout\n"); + err = -ETIME; + } else { + ndev->req_id++; + } + + return err; +} + +static int mei_nfc_recv(struct mei_cl_device *cldev, u8 *buf, size_t length) +{ + struct mei_nfc_dev *ndev; + struct mei_nfc_hci_hdr *hci_hdr; + int received_length; + + ndev = (struct mei_nfc_dev *)cldev->priv_data; + + received_length = __mei_cl_recv(ndev->cl, buf, length); + if (received_length < 0) + return received_length; + + hci_hdr = (struct mei_nfc_hci_hdr *) buf; + + if (hci_hdr->cmd == MEI_NFC_CMD_HCI_SEND) { + ndev->recv_req_id = hci_hdr->req_id; + wake_up(&ndev->send_wq); + + return 0; + } + + return received_length; +} + +static struct mei_cl_ops nfc_ops = { + .enable = mei_nfc_enable, + .disable = mei_nfc_disable, + .send = mei_nfc_send, + .recv = mei_nfc_recv, +}; + +static void mei_nfc_init(struct work_struct *work) +{ + struct mei_device *dev; + struct mei_cl_device *cldev; + struct mei_nfc_dev *ndev; + struct mei_cl *cl_info; + + ndev = container_of(work, struct mei_nfc_dev, init_work); + + cl_info = ndev->cl_info; + dev = cl_info->dev; + + mutex_lock(&dev->device_lock); + + if (mei_cl_connect(cl_info, NULL) < 0) { + mutex_unlock(&dev->device_lock); + dev_err(&dev->pdev->dev, + "Could not connect to the NFC INFO ME client"); + + goto err; + } + + mutex_unlock(&dev->device_lock); + + if (mei_nfc_if_version(ndev) < 0) { + dev_err(&dev->pdev->dev, "Could not get the NFC interface version"); + + goto err; + } + + dev_info(&dev->pdev->dev, + "NFC MEI VERSION: IVN 0x%x Vendor ID 0x%x Type 0x%x\n", + ndev->fw_ivn, ndev->vendor_id, ndev->radio_type); + + mutex_lock(&dev->device_lock); + + if (mei_cl_disconnect(cl_info) < 0) { + mutex_unlock(&dev->device_lock); + dev_err(&dev->pdev->dev, + "Could not disconnect the NFC INFO ME client"); + + goto err; + } + + mutex_unlock(&dev->device_lock); + + if (mei_nfc_build_bus_name(ndev) < 0) { + dev_err(&dev->pdev->dev, + "Could not build the bus ID name\n"); + return; + } + + cldev = mei_cl_add_device(dev, mei_nfc_guid, ndev->bus_name, &nfc_ops); + if (!cldev) { + dev_err(&dev->pdev->dev, + "Could not add the NFC device to the MEI bus\n"); + + goto err; + } + + cldev->priv_data = ndev; + + + return; + +err: + mutex_lock(&dev->device_lock); + mei_nfc_free(ndev); + mutex_unlock(&dev->device_lock); + + return; +} + + +int mei_nfc_host_init(struct mei_device *dev) +{ + struct mei_nfc_dev *ndev = &nfc_dev; + struct mei_cl *cl_info, *cl = NULL; + int i, ret; + + /* already initialized */ + if (ndev->cl_info) + return 0; + + ndev->cl_info = mei_cl_allocate(dev); + ndev->cl = mei_cl_allocate(dev); + + cl = ndev->cl; + cl_info = ndev->cl_info; + + if (!cl || !cl_info) { + ret = -ENOMEM; + goto err; + } + + /* check for valid client id */ + i = mei_me_cl_by_uuid(dev, &mei_nfc_info_guid); + if (i < 0) { + dev_info(&dev->pdev->dev, "nfc: failed to find the client\n"); + ret = -ENOTTY; + goto err; + } + + cl_info->me_client_id = dev->me_clients[i].client_id; + + ret = mei_cl_link(cl_info, MEI_HOST_CLIENT_ID_ANY); + if (ret) + goto err; + + cl_info->device_uuid = mei_nfc_info_guid; + + list_add_tail(&cl_info->device_link, &dev->device_list); + + /* check for valid client id */ + i = mei_me_cl_by_uuid(dev, &mei_nfc_guid); + if (i < 0) { + dev_info(&dev->pdev->dev, "nfc: failed to find the client\n"); + ret = -ENOTTY; + goto err; + } + + cl->me_client_id = dev->me_clients[i].client_id; + + ret = mei_cl_link(cl, MEI_HOST_CLIENT_ID_ANY); + if (ret) + goto err; + + cl->device_uuid = mei_nfc_guid; + + + list_add_tail(&cl->device_link, &dev->device_list); + + ndev->req_id = 1; + + INIT_WORK(&ndev->init_work, mei_nfc_init); + init_waitqueue_head(&ndev->send_wq); + schedule_work(&ndev->init_work); + + return 0; + +err: + mei_nfc_free(ndev); + + return ret; +} + +void mei_nfc_host_exit(struct mei_device *dev) +{ + struct mei_nfc_dev *ndev = &nfc_dev; + cancel_work_sync(&ndev->init_work); +} + + diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c new file mode 100644 index 00000000000..1b46c64a649 --- /dev/null +++ b/drivers/misc/mei/pci-me.c @@ -0,0 +1,488 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2003-2012, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/fs.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/fcntl.h> +#include <linux/aio.h> +#include <linux/pci.h> +#include <linux/poll.h> +#include <linux/ioctl.h> +#include <linux/cdev.h> +#include <linux/sched.h> +#include <linux/uuid.h> +#include <linux/compat.h> +#include <linux/jiffies.h> +#include <linux/interrupt.h> +#include <linux/miscdevice.h> + +#include <linux/pm_runtime.h> + +#include <linux/mei.h> + +#include "mei_dev.h" +#include "client.h" +#include "hw-me-regs.h" +#include "hw-me.h" + +/* mei_pci_tbl - PCI Device ID Table */ +static const struct pci_device_id mei_me_pci_tbl[] = { + {MEI_PCI_DEVICE(MEI_DEV_ID_82946GZ, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_82G35, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_82Q965, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_82G965, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_82GM965, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_82GME965, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_82Q35, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_82G33, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_82Q33, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_82X38, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_3200, mei_me_legacy_cfg)}, + + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_6, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_7, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_8, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_9, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9_10, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9M_1, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9M_2, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9M_3, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH9M_4, mei_me_legacy_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH10_1, mei_me_ich_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH10_2, mei_me_ich_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH10_3, mei_me_ich_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_ICH10_4, mei_me_ich_cfg)}, + + {MEI_PCI_DEVICE(MEI_DEV_ID_IBXPK_1, mei_me_pch_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_IBXPK_2, mei_me_pch_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_CPT_1, mei_me_pch_cpt_pbg_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_PBG_1, mei_me_pch_cpt_pbg_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_PPT_1, mei_me_pch_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_PPT_2, mei_me_pch_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_PPT_3, mei_me_pch_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_LPT_H, mei_me_lpt_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_LPT_W, mei_me_lpt_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_LPT_LP, mei_me_pch_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_LPT_HR, mei_me_lpt_cfg)}, + {MEI_PCI_DEVICE(MEI_DEV_ID_WPT_LP, mei_me_pch_cfg)}, + + /* required last entry */ + {0, } +}; + +MODULE_DEVICE_TABLE(pci, mei_me_pci_tbl); + +#ifdef CONFIG_PM_RUNTIME +static inline void mei_me_set_pm_domain(struct mei_device *dev); +static inline void mei_me_unset_pm_domain(struct mei_device *dev); +#else +static inline void mei_me_set_pm_domain(struct mei_device *dev) {} +static inline void mei_me_unset_pm_domain(struct mei_device *dev) {} +#endif /* CONFIG_PM_RUNTIME */ + +/** + * mei_quirk_probe - probe for devices that doesn't valid ME interface + * + * @pdev: PCI device structure + * @cfg: per generation config + * + * returns true if ME Interface is valid, false otherwise + */ +static bool mei_me_quirk_probe(struct pci_dev *pdev, + const struct mei_cfg *cfg) +{ + if (cfg->quirk_probe && cfg->quirk_probe(pdev)) { + dev_info(&pdev->dev, "Device doesn't have valid ME Interface\n"); + return false; + } + + return true; +} + +/** + * mei_probe - Device Initialization Routine + * + * @pdev: PCI device structure + * @ent: entry in kcs_pci_tbl + * + * returns 0 on success, <0 on failure. + */ +static int mei_me_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + const struct mei_cfg *cfg = (struct mei_cfg *)(ent->driver_data); + struct mei_device *dev; + struct mei_me_hw *hw; + int err; + + + if (!mei_me_quirk_probe(pdev, cfg)) + return -ENODEV; + + /* enable pci dev */ + err = pci_enable_device(pdev); + if (err) { + dev_err(&pdev->dev, "failed to enable pci device.\n"); + goto end; + } + /* set PCI host mastering */ + pci_set_master(pdev); + /* pci request regions for mei driver */ + err = pci_request_regions(pdev, KBUILD_MODNAME); + if (err) { + dev_err(&pdev->dev, "failed to get pci regions.\n"); + goto disable_device; + } + + if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(64)) || + dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(64))) { + + err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); + if (err) + err = dma_set_coherent_mask(&pdev->dev, + DMA_BIT_MASK(32)); + } + if (err) { + dev_err(&pdev->dev, "No usable DMA configuration, aborting\n"); + goto release_regions; + } + + + /* allocates and initializes the mei dev structure */ + dev = mei_me_dev_init(pdev, cfg); + if (!dev) { + err = -ENOMEM; + goto release_regions; + } + hw = to_me_hw(dev); + /* mapping IO device memory */ + hw->mem_addr = pci_iomap(pdev, 0, 0); + if (!hw->mem_addr) { + dev_err(&pdev->dev, "mapping I/O device memory failure.\n"); + err = -ENOMEM; + goto free_device; + } + pci_enable_msi(pdev); + + /* request and enable interrupt */ + if (pci_dev_msi_enabled(pdev)) + err = request_threaded_irq(pdev->irq, + NULL, + mei_me_irq_thread_handler, + IRQF_ONESHOT, KBUILD_MODNAME, dev); + else + err = request_threaded_irq(pdev->irq, + mei_me_irq_quick_handler, + mei_me_irq_thread_handler, + IRQF_SHARED, KBUILD_MODNAME, dev); + + if (err) { + dev_err(&pdev->dev, "request_threaded_irq failure. irq = %d\n", + pdev->irq); + goto disable_msi; + } + + if (mei_start(dev)) { + dev_err(&pdev->dev, "init hw failure.\n"); + err = -ENODEV; + goto release_irq; + } + + pm_runtime_set_autosuspend_delay(&pdev->dev, MEI_ME_RPM_TIMEOUT); + pm_runtime_use_autosuspend(&pdev->dev); + + err = mei_register(dev); + if (err) + goto release_irq; + + pci_set_drvdata(pdev, dev); + + schedule_delayed_work(&dev->timer_work, HZ); + + /* + * For not wake-able HW runtime pm framework + * can't be used on pci device level. + * Use domain runtime pm callbacks instead. + */ + if (!pci_dev_run_wake(pdev)) + mei_me_set_pm_domain(dev); + + if (mei_pg_is_enabled(dev)) + pm_runtime_put_noidle(&pdev->dev); + + dev_dbg(&pdev->dev, "initialization successful.\n"); + + return 0; + +release_irq: + mei_cancel_work(dev); + mei_disable_interrupts(dev); + free_irq(pdev->irq, dev); +disable_msi: + pci_disable_msi(pdev); + pci_iounmap(pdev, hw->mem_addr); +free_device: + kfree(dev); +release_regions: + pci_release_regions(pdev); +disable_device: + pci_disable_device(pdev); +end: + dev_err(&pdev->dev, "initialization failed.\n"); + return err; +} + +/** + * mei_remove - Device Removal Routine + * + * @pdev: PCI device structure + * + * mei_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. + */ +static void mei_me_remove(struct pci_dev *pdev) +{ + struct mei_device *dev; + struct mei_me_hw *hw; + + dev = pci_get_drvdata(pdev); + if (!dev) + return; + + if (mei_pg_is_enabled(dev)) + pm_runtime_get_noresume(&pdev->dev); + + hw = to_me_hw(dev); + + + dev_dbg(&pdev->dev, "stop\n"); + mei_stop(dev); + + if (!pci_dev_run_wake(pdev)) + mei_me_unset_pm_domain(dev); + + /* disable interrupts */ + mei_disable_interrupts(dev); + + free_irq(pdev->irq, dev); + pci_disable_msi(pdev); + + if (hw->mem_addr) + pci_iounmap(pdev, hw->mem_addr); + + mei_deregister(dev); + + kfree(dev); + + pci_release_regions(pdev); + pci_disable_device(pdev); + + +} +#ifdef CONFIG_PM_SLEEP +static int mei_me_pci_suspend(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev = pci_get_drvdata(pdev); + + if (!dev) + return -ENODEV; + + dev_dbg(&pdev->dev, "suspend\n"); + + mei_stop(dev); + + mei_disable_interrupts(dev); + + free_irq(pdev->irq, dev); + pci_disable_msi(pdev); + + return 0; +} + +static int mei_me_pci_resume(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + int err; + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + + pci_enable_msi(pdev); + + /* request and enable interrupt */ + if (pci_dev_msi_enabled(pdev)) + err = request_threaded_irq(pdev->irq, + NULL, + mei_me_irq_thread_handler, + IRQF_ONESHOT, KBUILD_MODNAME, dev); + else + err = request_threaded_irq(pdev->irq, + mei_me_irq_quick_handler, + mei_me_irq_thread_handler, + IRQF_SHARED, KBUILD_MODNAME, dev); + + if (err) { + dev_err(&pdev->dev, "request_threaded_irq failed: irq = %d.\n", + pdev->irq); + return err; + } + + err = mei_restart(dev); + if (err) + return err; + + /* Start timer if stopped in suspend */ + schedule_delayed_work(&dev->timer_work, HZ); + + return 0; +} +#endif /* CONFIG_PM_SLEEP */ + +#ifdef CONFIG_PM_RUNTIME +static int mei_me_pm_runtime_idle(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + + dev_dbg(&pdev->dev, "rpm: me: runtime_idle\n"); + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + if (mei_write_is_idle(dev)) + pm_schedule_suspend(device, MEI_ME_RPM_TIMEOUT * 2); + + return -EBUSY; +} + +static int mei_me_pm_runtime_suspend(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + int ret; + + dev_dbg(&pdev->dev, "rpm: me: runtime suspend\n"); + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + + mutex_lock(&dev->device_lock); + + if (mei_write_is_idle(dev)) + ret = mei_me_pg_set_sync(dev); + else + ret = -EAGAIN; + + mutex_unlock(&dev->device_lock); + + dev_dbg(&pdev->dev, "rpm: me: runtime suspend ret=%d\n", ret); + + return ret; +} + +static int mei_me_pm_runtime_resume(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + int ret; + + dev_dbg(&pdev->dev, "rpm: me: runtime resume\n"); + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + + mutex_lock(&dev->device_lock); + + ret = mei_me_pg_unset_sync(dev); + + mutex_unlock(&dev->device_lock); + + dev_dbg(&pdev->dev, "rpm: me: runtime resume ret = %d\n", ret); + + return ret; +} + +/** + * mei_me_set_pm_domain - fill and set pm domian stucture for device + * + * @dev: mei_device + */ +static inline void mei_me_set_pm_domain(struct mei_device *dev) +{ + struct pci_dev *pdev = dev->pdev; + + if (pdev->dev.bus && pdev->dev.bus->pm) { + dev->pg_domain.ops = *pdev->dev.bus->pm; + + dev->pg_domain.ops.runtime_suspend = mei_me_pm_runtime_suspend; + dev->pg_domain.ops.runtime_resume = mei_me_pm_runtime_resume; + dev->pg_domain.ops.runtime_idle = mei_me_pm_runtime_idle; + + pdev->dev.pm_domain = &dev->pg_domain; + } +} + +/** + * mei_me_unset_pm_domain - clean pm domian stucture for device + * + * @dev: mei_device + */ +static inline void mei_me_unset_pm_domain(struct mei_device *dev) +{ + /* stop using pm callbacks if any */ + dev->pdev->dev.pm_domain = NULL; +} +#endif /* CONFIG_PM_RUNTIME */ + +#ifdef CONFIG_PM +static const struct dev_pm_ops mei_me_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(mei_me_pci_suspend, + mei_me_pci_resume) + SET_RUNTIME_PM_OPS( + mei_me_pm_runtime_suspend, + mei_me_pm_runtime_resume, + mei_me_pm_runtime_idle) +}; + +#define MEI_ME_PM_OPS (&mei_me_pm_ops) +#else +#define MEI_ME_PM_OPS NULL +#endif /* CONFIG_PM */ +/* + * PCI driver structure + */ +static struct pci_driver mei_me_driver = { + .name = KBUILD_MODNAME, + .id_table = mei_me_pci_tbl, + .probe = mei_me_probe, + .remove = mei_me_remove, + .shutdown = mei_me_remove, + .driver.pm = MEI_ME_PM_OPS, +}; + +module_pci_driver(mei_me_driver); + +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel(R) Management Engine Interface"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mei/pci-txe.c b/drivers/misc/mei/pci-txe.c new file mode 100644 index 00000000000..2343c6236df --- /dev/null +++ b/drivers/misc/mei/pci-txe.c @@ -0,0 +1,436 @@ +/* + * + * Intel Management Engine Interface (Intel MEI) Linux driver + * Copyright (c) 2013-2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + */ + +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/fs.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/init.h> +#include <linux/sched.h> +#include <linux/uuid.h> +#include <linux/jiffies.h> +#include <linux/interrupt.h> +#include <linux/workqueue.h> +#include <linux/pm_runtime.h> + +#include <linux/mei.h> + + +#include "mei_dev.h" +#include "hw-txe.h" + +static const struct pci_device_id mei_txe_pci_tbl[] = { + {MEI_PCI_DEVICE(0x0F18, mei_txe_cfg)}, /* Baytrail */ + {0, } +}; +MODULE_DEVICE_TABLE(pci, mei_txe_pci_tbl); + +#ifdef CONFIG_PM_RUNTIME +static inline void mei_txe_set_pm_domain(struct mei_device *dev); +static inline void mei_txe_unset_pm_domain(struct mei_device *dev); +#else +static inline void mei_txe_set_pm_domain(struct mei_device *dev) {} +static inline void mei_txe_unset_pm_domain(struct mei_device *dev) {} +#endif /* CONFIG_PM_RUNTIME */ + +static void mei_txe_pci_iounmap(struct pci_dev *pdev, struct mei_txe_hw *hw) +{ + int i; + for (i = SEC_BAR; i < NUM_OF_MEM_BARS; i++) { + if (hw->mem_addr[i]) { + pci_iounmap(pdev, hw->mem_addr[i]); + hw->mem_addr[i] = NULL; + } + } +} +/** + * mei_probe - Device Initialization Routine + * + * @pdev: PCI device structure + * @ent: entry in mei_txe_pci_tbl + * + * returns 0 on success, <0 on failure. + */ +static int mei_txe_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + const struct mei_cfg *cfg = (struct mei_cfg *)(ent->driver_data); + struct mei_device *dev; + struct mei_txe_hw *hw; + int err; + int i; + + /* enable pci dev */ + err = pci_enable_device(pdev); + if (err) { + dev_err(&pdev->dev, "failed to enable pci device.\n"); + goto end; + } + /* set PCI host mastering */ + pci_set_master(pdev); + /* pci request regions for mei driver */ + err = pci_request_regions(pdev, KBUILD_MODNAME); + if (err) { + dev_err(&pdev->dev, "failed to get pci regions.\n"); + goto disable_device; + } + + err = pci_set_dma_mask(pdev, DMA_BIT_MASK(36)); + if (err) { + err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); + if (err) { + dev_err(&pdev->dev, "No suitable DMA available.\n"); + goto release_regions; + } + } + + /* allocates and initializes the mei dev structure */ + dev = mei_txe_dev_init(pdev, cfg); + if (!dev) { + err = -ENOMEM; + goto release_regions; + } + hw = to_txe_hw(dev); + + /* mapping IO device memory */ + for (i = SEC_BAR; i < NUM_OF_MEM_BARS; i++) { + hw->mem_addr[i] = pci_iomap(pdev, i, 0); + if (!hw->mem_addr[i]) { + dev_err(&pdev->dev, "mapping I/O device memory failure.\n"); + err = -ENOMEM; + goto free_device; + } + } + + + pci_enable_msi(pdev); + + /* clear spurious interrupts */ + mei_clear_interrupts(dev); + + /* request and enable interrupt */ + if (pci_dev_msi_enabled(pdev)) + err = request_threaded_irq(pdev->irq, + NULL, + mei_txe_irq_thread_handler, + IRQF_ONESHOT, KBUILD_MODNAME, dev); + else + err = request_threaded_irq(pdev->irq, + mei_txe_irq_quick_handler, + mei_txe_irq_thread_handler, + IRQF_SHARED, KBUILD_MODNAME, dev); + if (err) { + dev_err(&pdev->dev, "mei: request_threaded_irq failure. irq = %d\n", + pdev->irq); + goto free_device; + } + + if (mei_start(dev)) { + dev_err(&pdev->dev, "init hw failure.\n"); + err = -ENODEV; + goto release_irq; + } + + pm_runtime_set_autosuspend_delay(&pdev->dev, MEI_TXI_RPM_TIMEOUT); + pm_runtime_use_autosuspend(&pdev->dev); + + err = mei_register(dev); + if (err) + goto release_irq; + + pci_set_drvdata(pdev, dev); + + /* + * For not wake-able HW runtime pm framework + * can't be used on pci device level. + * Use domain runtime pm callbacks instead. + */ + if (!pci_dev_run_wake(pdev)) + mei_txe_set_pm_domain(dev); + + pm_runtime_put_noidle(&pdev->dev); + + return 0; + +release_irq: + + mei_cancel_work(dev); + + /* disable interrupts */ + mei_disable_interrupts(dev); + + free_irq(pdev->irq, dev); + pci_disable_msi(pdev); + +free_device: + mei_txe_pci_iounmap(pdev, hw); + + kfree(dev); +release_regions: + pci_release_regions(pdev); +disable_device: + pci_disable_device(pdev); +end: + dev_err(&pdev->dev, "initialization failed.\n"); + return err; +} + +/** + * mei_remove - Device Removal Routine + * + * @pdev: PCI device structure + * + * mei_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. + */ +static void mei_txe_remove(struct pci_dev *pdev) +{ + struct mei_device *dev; + struct mei_txe_hw *hw; + + dev = pci_get_drvdata(pdev); + if (!dev) { + dev_err(&pdev->dev, "mei: dev =NULL\n"); + return; + } + + pm_runtime_get_noresume(&pdev->dev); + + hw = to_txe_hw(dev); + + mei_stop(dev); + + if (!pci_dev_run_wake(pdev)) + mei_txe_unset_pm_domain(dev); + + /* disable interrupts */ + mei_disable_interrupts(dev); + free_irq(pdev->irq, dev); + pci_disable_msi(pdev); + + pci_set_drvdata(pdev, NULL); + + mei_txe_pci_iounmap(pdev, hw); + + mei_deregister(dev); + + kfree(dev); + + pci_release_regions(pdev); + pci_disable_device(pdev); +} + + +#ifdef CONFIG_PM_SLEEP +static int mei_txe_pci_suspend(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev = pci_get_drvdata(pdev); + + if (!dev) + return -ENODEV; + + dev_dbg(&pdev->dev, "suspend\n"); + + mei_stop(dev); + + mei_disable_interrupts(dev); + + free_irq(pdev->irq, dev); + pci_disable_msi(pdev); + + return 0; +} + +static int mei_txe_pci_resume(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + int err; + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + + pci_enable_msi(pdev); + + mei_clear_interrupts(dev); + + /* request and enable interrupt */ + if (pci_dev_msi_enabled(pdev)) + err = request_threaded_irq(pdev->irq, + NULL, + mei_txe_irq_thread_handler, + IRQF_ONESHOT, KBUILD_MODNAME, dev); + else + err = request_threaded_irq(pdev->irq, + mei_txe_irq_quick_handler, + mei_txe_irq_thread_handler, + IRQF_SHARED, KBUILD_MODNAME, dev); + if (err) { + dev_err(&pdev->dev, "request_threaded_irq failed: irq = %d.\n", + pdev->irq); + return err; + } + + err = mei_restart(dev); + + return err; +} +#endif /* CONFIG_PM_SLEEP */ + +#ifdef CONFIG_PM_RUNTIME +static int mei_txe_pm_runtime_idle(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + + dev_dbg(&pdev->dev, "rpm: txe: runtime_idle\n"); + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + if (mei_write_is_idle(dev)) + pm_schedule_suspend(device, MEI_TXI_RPM_TIMEOUT * 2); + + return -EBUSY; +} +static int mei_txe_pm_runtime_suspend(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + int ret; + + dev_dbg(&pdev->dev, "rpm: txe: runtime suspend\n"); + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + + mutex_lock(&dev->device_lock); + + if (mei_write_is_idle(dev)) + ret = mei_txe_aliveness_set_sync(dev, 0); + else + ret = -EAGAIN; + + /* + * If everything is okay we're about to enter PCI low + * power state (D3) therefor we need to disable the + * interrupts towards host. + * However if device is not wakeable we do not enter + * D-low state and we need to keep the interrupt kicking + */ + if (!ret && pci_dev_run_wake(pdev)) + mei_disable_interrupts(dev); + + dev_dbg(&pdev->dev, "rpm: txe: runtime suspend ret=%d\n", ret); + + mutex_unlock(&dev->device_lock); + return ret; +} + +static int mei_txe_pm_runtime_resume(struct device *device) +{ + struct pci_dev *pdev = to_pci_dev(device); + struct mei_device *dev; + int ret; + + dev_dbg(&pdev->dev, "rpm: txe: runtime resume\n"); + + dev = pci_get_drvdata(pdev); + if (!dev) + return -ENODEV; + + mutex_lock(&dev->device_lock); + + mei_enable_interrupts(dev); + + ret = mei_txe_aliveness_set_sync(dev, 1); + + mutex_unlock(&dev->device_lock); + + dev_dbg(&pdev->dev, "rpm: txe: runtime resume ret = %d\n", ret); + + return ret; +} + +/** + * mei_txe_set_pm_domain - fill and set pm domian stucture for device + * + * @dev: mei_device + */ +static inline void mei_txe_set_pm_domain(struct mei_device *dev) +{ + struct pci_dev *pdev = dev->pdev; + + if (pdev->dev.bus && pdev->dev.bus->pm) { + dev->pg_domain.ops = *pdev->dev.bus->pm; + + dev->pg_domain.ops.runtime_suspend = mei_txe_pm_runtime_suspend; + dev->pg_domain.ops.runtime_resume = mei_txe_pm_runtime_resume; + dev->pg_domain.ops.runtime_idle = mei_txe_pm_runtime_idle; + + pdev->dev.pm_domain = &dev->pg_domain; + } +} + +/** + * mei_txe_unset_pm_domain - clean pm domian stucture for device + * + * @dev: mei_device + */ +static inline void mei_txe_unset_pm_domain(struct mei_device *dev) +{ + /* stop using pm callbacks if any */ + dev->pdev->dev.pm_domain = NULL; +} +#endif /* CONFIG_PM_RUNTIME */ + +#ifdef CONFIG_PM +static const struct dev_pm_ops mei_txe_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(mei_txe_pci_suspend, + mei_txe_pci_resume) + SET_RUNTIME_PM_OPS( + mei_txe_pm_runtime_suspend, + mei_txe_pm_runtime_resume, + mei_txe_pm_runtime_idle) +}; + +#define MEI_TXE_PM_OPS (&mei_txe_pm_ops) +#else +#define MEI_TXE_PM_OPS NULL +#endif /* CONFIG_PM */ + +/* + * PCI driver structure + */ +static struct pci_driver mei_txe_driver = { + .name = KBUILD_MODNAME, + .id_table = mei_txe_pci_tbl, + .probe = mei_txe_probe, + .remove = mei_txe_remove, + .shutdown = mei_txe_remove, + .driver.pm = MEI_TXE_PM_OPS, +}; + +module_pci_driver(mei_txe_driver); + +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel(R) Trusted Execution Environment Interface"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mei/wd.c b/drivers/misc/mei/wd.c index d96c537f046..a84a664dfcc 100644 --- a/drivers/misc/mei/wd.c +++ b/drivers/misc/mei/wd.c @@ -21,20 +21,15 @@ #include <linux/sched.h> #include <linux/watchdog.h> -#include "mei_dev.h" -#include "hw.h" -#include "interface.h" #include <linux/mei.h> +#include "mei_dev.h" +#include "hbm.h" +#include "client.h" + static const u8 mei_start_wd_params[] = { 0x02, 0x12, 0x13, 0x10 }; static const u8 mei_stop_wd_params[] = { 0x02, 0x02, 0x14, 0x10 }; -const u8 mei_wd_state_independence_msg[3][4] = { - {0x05, 0x02, 0x51, 0x10}, - {0x05, 0x02, 0x52, 0x10}, - {0x07, 0x02, 0x01, 0x10} -}; - /* * AMT Watchdog Device */ @@ -56,38 +51,53 @@ static void mei_wd_set_start_timeout(struct mei_device *dev, u16 timeout) * mei_wd_host_init - connect to the watchdog client * * @dev: the device structure - * returns -ENENT if wd client cannot be found + * + * returns -ENOTTY if wd client cannot be found * -EIO if write has failed * 0 on success */ int mei_wd_host_init(struct mei_device *dev) { - mei_cl_init(&dev->wd_cl, dev); + struct mei_cl *cl = &dev->wd_cl; + int id; + int ret; + + mei_cl_init(cl, dev); - /* look for WD client and connect to it */ - dev->wd_cl.state = MEI_FILE_DISCONNECTED; dev->wd_timeout = MEI_WD_DEFAULT_TIMEOUT; dev->wd_state = MEI_WD_IDLE; - /* find ME WD client */ - mei_me_cl_update_filext(dev, &dev->wd_cl, - &mei_wd_guid, MEI_WD_HOST_CLIENT_ID); - dev_dbg(&dev->pdev->dev, "wd: check client\n"); - if (MEI_FILE_CONNECTING != dev->wd_cl.state) { + /* check for valid client id */ + id = mei_me_cl_by_uuid(dev, &mei_wd_guid); + if (id < 0) { dev_info(&dev->pdev->dev, "wd: failed to find the client\n"); - return -ENOENT; + return -ENOTTY; } - if (mei_connect(dev, &dev->wd_cl)) { - dev_err(&dev->pdev->dev, "wd: failed to connect to the client\n"); - dev->wd_cl.state = MEI_FILE_DISCONNECTED; - dev->wd_cl.host_client_id = 0; - return -EIO; + cl->me_client_id = dev->me_clients[id].client_id; + + ret = mei_cl_link(cl, MEI_WD_HOST_CLIENT_ID); + + if (ret < 0) { + dev_info(&dev->pdev->dev, "wd: failed link client\n"); + return ret; } - dev->wd_cl.timer_count = CONNECT_TIMEOUT; - return 0; + ret = mei_cl_connect(cl, NULL); + + if (ret) { + dev_err(&dev->pdev->dev, "wd: failed to connect = %d\n", ret); + mei_cl_unlink(cl); + return ret; + } + + ret = mei_watchdog_register(dev); + if (ret) { + mei_cl_disconnect(cl); + mei_cl_unlink(cl); + } + return ret; } /** @@ -98,25 +108,42 @@ int mei_wd_host_init(struct mei_device *dev) * returns 0 if success, * -EIO when message send fails * -EINVAL when invalid message is to be sent + * -ENODEV on flow control failure */ int mei_wd_send(struct mei_device *dev) { - struct mei_msg_hdr *mei_hdr; + struct mei_cl *cl = &dev->wd_cl; + struct mei_msg_hdr hdr; + int ret; - mei_hdr = (struct mei_msg_hdr *) &dev->wr_msg_buf[0]; - mei_hdr->host_addr = dev->wd_cl.host_client_id; - mei_hdr->me_addr = dev->wd_cl.me_client_id; - mei_hdr->msg_complete = 1; - mei_hdr->reserved = 0; + hdr.host_addr = cl->host_client_id; + hdr.me_addr = cl->me_client_id; + hdr.msg_complete = 1; + hdr.reserved = 0; + hdr.internal = 0; if (!memcmp(dev->wd_data, mei_start_wd_params, MEI_WD_HDR_SIZE)) - mei_hdr->length = MEI_WD_START_MSG_SIZE; + hdr.length = MEI_WD_START_MSG_SIZE; else if (!memcmp(dev->wd_data, mei_stop_wd_params, MEI_WD_HDR_SIZE)) - mei_hdr->length = MEI_WD_STOP_MSG_SIZE; - else + hdr.length = MEI_WD_STOP_MSG_SIZE; + else { + dev_err(&dev->pdev->dev, "wd: invalid message is to be sent, aborting\n"); return -EINVAL; + } + + ret = mei_write_message(dev, &hdr, dev->wd_data); + if (ret) { + dev_err(&dev->pdev->dev, "wd: write message failed\n"); + return ret; + } + + ret = mei_cl_flow_ctrl_reduce(cl); + if (ret) { + dev_err(&dev->pdev->dev, "wd: flow_ctrl_reduce failed.\n"); + return ret; + } - return mei_write_message(dev, mei_hdr, dev->wd_data, mei_hdr->length); + return 0; } /** @@ -125,9 +152,11 @@ int mei_wd_send(struct mei_device *dev) * @dev: the device structure * @preserve: indicate if to keep the timeout value * - * returns 0 if success, - * -EIO when message send fails + * returns 0 if success + * on error: + * -EIO when message send fails * -EINVAL when invalid message is to be sent + * -ETIME on message timeout */ int mei_wd_stop(struct mei_device *dev) { @@ -141,22 +170,14 @@ int mei_wd_stop(struct mei_device *dev) dev->wd_state = MEI_WD_STOPPING; - ret = mei_flow_ctrl_creds(dev, &dev->wd_cl); + ret = mei_cl_flow_ctrl_creds(&dev->wd_cl); if (ret < 0) - goto out; - - if (ret && dev->mei_host_buffer_is_empty) { - ret = 0; - dev->mei_host_buffer_is_empty = false; - - if (!mei_wd_send(dev)) { - ret = mei_flow_ctrl_reduce(dev, &dev->wd_cl); - if (ret) - goto out; - } else { - dev_err(&dev->pdev->dev, "wd: send stop failed\n"); - } + goto err; + if (ret && mei_hbuf_acquire(dev)) { + ret = mei_wd_send(dev); + if (ret) + goto err; dev->wd_pending = false; } else { dev->wd_pending = true; @@ -164,21 +185,21 @@ int mei_wd_stop(struct mei_device *dev) mutex_unlock(&dev->device_lock); - ret = wait_event_interruptible_timeout(dev->wait_stop_wd, - dev->wd_state == MEI_WD_IDLE, - msecs_to_jiffies(MEI_WD_STOP_TIMEOUT)); + ret = wait_event_timeout(dev->wait_stop_wd, + dev->wd_state == MEI_WD_IDLE, + msecs_to_jiffies(MEI_WD_STOP_TIMEOUT)); mutex_lock(&dev->device_lock); - if (dev->wd_state == MEI_WD_IDLE) { - dev_dbg(&dev->pdev->dev, "wd: stop completed ret=%d.\n", ret); - ret = 0; - } else { - if (!ret) - ret = -ETIMEDOUT; + if (dev->wd_state != MEI_WD_IDLE) { + /* timeout */ + ret = -ETIME; dev_warn(&dev->pdev->dev, "wd: stop failed to complete ret=%d.\n", ret); + goto err; } - -out: + dev_dbg(&dev->pdev->dev, "wd: stop completed after %u msec\n", + MEI_WD_STOP_TIMEOUT - jiffies_to_msecs(ret)); + return 0; +err: return ret; } @@ -252,8 +273,8 @@ static int mei_wd_ops_stop(struct watchdog_device *wd_dev) */ static int mei_wd_ops_ping(struct watchdog_device *wd_dev) { - int ret = 0; struct mei_device *dev; + int ret; dev = watchdog_get_drvdata(wd_dev); if (!dev) @@ -269,26 +290,18 @@ static int mei_wd_ops_ping(struct watchdog_device *wd_dev) dev->wd_state = MEI_WD_RUNNING; + ret = mei_cl_flow_ctrl_creds(&dev->wd_cl); + if (ret < 0) + goto end; /* Check if we can send the ping to HW*/ - if (dev->mei_host_buffer_is_empty && - mei_flow_ctrl_creds(dev, &dev->wd_cl) > 0) { + if (ret && mei_hbuf_acquire(dev)) { - dev->mei_host_buffer_is_empty = false; dev_dbg(&dev->pdev->dev, "wd: sending ping\n"); - if (mei_wd_send(dev)) { - dev_err(&dev->pdev->dev, "wd: send failed.\n"); - ret = -EIO; + ret = mei_wd_send(dev); + if (ret) goto end; - } - - if (mei_flow_ctrl_reduce(dev, &dev->wd_cl)) { - dev_err(&dev->pdev->dev, - "wd: mei_flow_ctrl_reduce() failed.\n"); - ret = -EIO; - goto end; - } - + dev->wd_pending = false; } else { dev->wd_pending = true; } @@ -306,7 +319,8 @@ end: * * returns 0 if success, negative errno code for failure */ -static int mei_wd_ops_set_timeout(struct watchdog_device *wd_dev, unsigned int timeout) +static int mei_wd_ops_set_timeout(struct watchdog_device *wd_dev, + unsigned int timeout) { struct mei_device *dev; @@ -355,28 +369,33 @@ static struct watchdog_device amt_wd_dev = { }; -void mei_watchdog_register(struct mei_device *dev) +int mei_watchdog_register(struct mei_device *dev) { - if (watchdog_register_device(&amt_wd_dev)) { - dev_err(&dev->pdev->dev, - "wd: unable to register watchdog device.\n"); - dev->wd_interface_reg = false; - return; + + int ret; + + /* unlock to perserve correct locking order */ + mutex_unlock(&dev->device_lock); + ret = watchdog_register_device(&amt_wd_dev); + mutex_lock(&dev->device_lock); + if (ret) { + dev_err(&dev->pdev->dev, "wd: unable to register watchdog device = %d.\n", + ret); + return ret; } dev_dbg(&dev->pdev->dev, "wd: successfully register watchdog interface.\n"); - dev->wd_interface_reg = true; watchdog_set_drvdata(&amt_wd_dev, dev); + return 0; } void mei_watchdog_unregister(struct mei_device *dev) { - if (!dev->wd_interface_reg) + if (watchdog_get_drvdata(&amt_wd_dev) == NULL) return; watchdog_set_drvdata(&amt_wd_dev, NULL); watchdog_unregister_device(&amt_wd_dev); - dev->wd_interface_reg = false; } diff --git a/drivers/misc/mic/Kconfig b/drivers/misc/mic/Kconfig new file mode 100644 index 00000000000..462a5b1d865 --- /dev/null +++ b/drivers/misc/mic/Kconfig @@ -0,0 +1,37 @@ +comment "Intel MIC Host Driver" + +config INTEL_MIC_HOST + tristate "Intel MIC Host Driver" + depends on 64BIT && PCI && X86 + select VHOST_RING + help + This enables Host Driver support for the Intel Many Integrated + Core (MIC) family of PCIe form factor coprocessor devices that + run a 64 bit Linux OS. The driver manages card OS state and + enables communication between host and card. Intel MIC X100 + devices are currently supported. + + If you are building a host kernel with an Intel MIC device then + say M (recommended) or Y, else say N. If unsure say N. + + More information about the Intel MIC family as well as the Linux + OS and tools for MIC to use with this driver are available from + <http://software.intel.com/en-us/mic-developer>. + +comment "Intel MIC Card Driver" + +config INTEL_MIC_CARD + tristate "Intel MIC Card Driver" + depends on 64BIT && X86 + select VIRTIO + help + This enables card driver support for the Intel Many Integrated + Core (MIC) device family. The card driver communicates shutdown/ + crash events to the host and allows registration/configuration of + virtio devices. Intel MIC X100 devices are currently supported. + + If you are building a card kernel for an Intel MIC device then + say M (recommended) or Y, else say N. If unsure say N. + + For more information see + <http://software.intel.com/en-us/mic-developer>. diff --git a/drivers/misc/mic/Makefile b/drivers/misc/mic/Makefile new file mode 100644 index 00000000000..05b34d683a5 --- /dev/null +++ b/drivers/misc/mic/Makefile @@ -0,0 +1,6 @@ +# +# Makefile - Intel MIC Linux driver. +# Copyright(c) 2013, Intel Corporation. +# +obj-$(CONFIG_INTEL_MIC_HOST) += host/ +obj-$(CONFIG_INTEL_MIC_CARD) += card/ diff --git a/drivers/misc/mic/card/Makefile b/drivers/misc/mic/card/Makefile new file mode 100644 index 00000000000..69d58bef92c --- /dev/null +++ b/drivers/misc/mic/card/Makefile @@ -0,0 +1,11 @@ +# +# Makefile - Intel MIC Linux driver. +# Copyright(c) 2013, Intel Corporation. +# +ccflags-y += -DINTEL_MIC_CARD + +obj-$(CONFIG_INTEL_MIC_CARD) += mic_card.o +mic_card-y += mic_x100.o +mic_card-y += mic_device.o +mic_card-y += mic_debugfs.o +mic_card-y += mic_virtio.o diff --git a/drivers/misc/mic/card/mic_debugfs.c b/drivers/misc/mic/card/mic_debugfs.c new file mode 100644 index 00000000000..421b3d7911d --- /dev/null +++ b/drivers/misc/mic/card/mic_debugfs.c @@ -0,0 +1,130 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Disclaimer: The codes contained in these modules may be specific to + * the Intel Software Development Platform codenamed: Knights Ferry, and + * the Intel product codenamed: Knights Corner, and are not backward + * compatible with other Intel products. Additionally, Intel will NOT + * support the codes or instruction set in future products. + * + * Intel MIC Card driver. + * + */ +#include <linux/debugfs.h> +#include <linux/delay.h> +#include <linux/seq_file.h> +#include <linux/interrupt.h> +#include <linux/device.h> + +#include "../common/mic_dev.h" +#include "mic_device.h" + +/* Debugfs parent dir */ +static struct dentry *mic_dbg; + +/** + * mic_intr_test - Send interrupts to host. + */ +static int mic_intr_test(struct seq_file *s, void *unused) +{ + struct mic_driver *mdrv = s->private; + struct mic_device *mdev = &mdrv->mdev; + + mic_send_intr(mdev, 0); + msleep(1000); + mic_send_intr(mdev, 1); + msleep(1000); + mic_send_intr(mdev, 2); + msleep(1000); + mic_send_intr(mdev, 3); + msleep(1000); + + return 0; +} + +static int mic_intr_test_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_intr_test, inode->i_private); +} + +static int mic_intr_test_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations intr_test_ops = { + .owner = THIS_MODULE, + .open = mic_intr_test_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_intr_test_release +}; + +/** + * mic_create_card_debug_dir - Initialize MIC debugfs entries. + */ +void __init mic_create_card_debug_dir(struct mic_driver *mdrv) +{ + struct dentry *d; + + if (!mic_dbg) + return; + + mdrv->dbg_dir = debugfs_create_dir(mdrv->name, mic_dbg); + if (!mdrv->dbg_dir) { + dev_err(mdrv->dev, "Cant create dbg_dir %s\n", mdrv->name); + return; + } + + d = debugfs_create_file("intr_test", 0444, mdrv->dbg_dir, + mdrv, &intr_test_ops); + + if (!d) { + dev_err(mdrv->dev, + "Cant create dbg intr_test %s\n", mdrv->name); + return; + } +} + +/** + * mic_delete_card_debug_dir - Uninitialize MIC debugfs entries. + */ +void mic_delete_card_debug_dir(struct mic_driver *mdrv) +{ + if (!mdrv->dbg_dir) + return; + + debugfs_remove_recursive(mdrv->dbg_dir); +} + +/** + * mic_init_card_debugfs - Initialize global debugfs entry. + */ +void __init mic_init_card_debugfs(void) +{ + mic_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); + if (!mic_dbg) + pr_err("can't create debugfs dir\n"); +} + +/** + * mic_exit_card_debugfs - Uninitialize global debugfs entry + */ +void mic_exit_card_debugfs(void) +{ + debugfs_remove(mic_dbg); +} diff --git a/drivers/misc/mic/card/mic_device.c b/drivers/misc/mic/card/mic_device.c new file mode 100644 index 00000000000..d0980ff9683 --- /dev/null +++ b/drivers/misc/mic/card/mic_device.c @@ -0,0 +1,305 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Disclaimer: The codes contained in these modules may be specific to + * the Intel Software Development Platform codenamed: Knights Ferry, and + * the Intel product codenamed: Knights Corner, and are not backward + * compatible with other Intel products. Additionally, Intel will NOT + * support the codes or instruction set in future products. + * + * Intel MIC Card driver. + * + */ +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/interrupt.h> +#include <linux/reboot.h> + +#include <linux/mic_common.h> +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_virtio.h" + +static struct mic_driver *g_drv; +static struct mic_irq *shutdown_cookie; + +static void mic_notify_host(u8 state) +{ + struct mic_driver *mdrv = g_drv; + struct mic_bootparam __iomem *bootparam = mdrv->dp; + + iowrite8(state, &bootparam->shutdown_status); + dev_dbg(mdrv->dev, "%s %d system_state %d\n", + __func__, __LINE__, state); + mic_send_intr(&mdrv->mdev, ioread8(&bootparam->c2h_shutdown_db)); +} + +static int mic_panic_event(struct notifier_block *this, unsigned long event, + void *ptr) +{ + struct mic_driver *mdrv = g_drv; + struct mic_bootparam __iomem *bootparam = mdrv->dp; + + iowrite8(-1, &bootparam->h2c_config_db); + iowrite8(-1, &bootparam->h2c_shutdown_db); + mic_notify_host(MIC_CRASHED); + return NOTIFY_DONE; +} + +static struct notifier_block mic_panic = { + .notifier_call = mic_panic_event, +}; + +static irqreturn_t mic_shutdown_isr(int irq, void *data) +{ + struct mic_driver *mdrv = g_drv; + struct mic_bootparam __iomem *bootparam = mdrv->dp; + + mic_ack_interrupt(&g_drv->mdev); + if (ioread8(&bootparam->shutdown_card)) + orderly_poweroff(true); + return IRQ_HANDLED; +} + +static int mic_shutdown_init(void) +{ + int rc = 0; + struct mic_driver *mdrv = g_drv; + struct mic_bootparam __iomem *bootparam = mdrv->dp; + int shutdown_db; + + shutdown_db = mic_next_card_db(); + shutdown_cookie = mic_request_card_irq(mic_shutdown_isr, + "Shutdown", mdrv, shutdown_db); + if (IS_ERR(shutdown_cookie)) + rc = PTR_ERR(shutdown_cookie); + else + iowrite8(shutdown_db, &bootparam->h2c_shutdown_db); + return rc; +} + +static void mic_shutdown_uninit(void) +{ + struct mic_driver *mdrv = g_drv; + struct mic_bootparam __iomem *bootparam = mdrv->dp; + + iowrite8(-1, &bootparam->h2c_shutdown_db); + mic_free_card_irq(shutdown_cookie, mdrv); +} + +static int __init mic_dp_init(void) +{ + struct mic_driver *mdrv = g_drv; + struct mic_device *mdev = &mdrv->mdev; + struct mic_bootparam __iomem *bootparam; + u64 lo, hi, dp_dma_addr; + u32 magic; + + lo = mic_read_spad(&mdrv->mdev, MIC_DPLO_SPAD); + hi = mic_read_spad(&mdrv->mdev, MIC_DPHI_SPAD); + + dp_dma_addr = lo | (hi << 32); + mdrv->dp = mic_card_map(mdev, dp_dma_addr, MIC_DP_SIZE); + if (!mdrv->dp) { + dev_err(mdrv->dev, "Cannot remap Aperture BAR\n"); + return -ENOMEM; + } + bootparam = mdrv->dp; + magic = ioread32(&bootparam->magic); + if (MIC_MAGIC != magic) { + dev_err(mdrv->dev, "bootparam magic mismatch 0x%x\n", magic); + return -EIO; + } + return 0; +} + +/* Uninitialize the device page */ +static void mic_dp_uninit(void) +{ + mic_card_unmap(&g_drv->mdev, g_drv->dp); +} + +/** + * mic_request_card_irq - request an irq. + * + * @func: The callback function that handles the interrupt. + * @name: The ASCII name of the callee requesting the irq. + * @data: private data that is returned back when calling the + * function handler. + * @index: The doorbell index of the requester. + * + * returns: The cookie that is transparent to the caller. Passed + * back when calling mic_free_irq. An appropriate error code + * is returned on failure. Caller needs to use IS_ERR(return_val) + * to check for failure and PTR_ERR(return_val) to obtained the + * error code. + * + */ +struct mic_irq *mic_request_card_irq(irqreturn_t (*func)(int irq, void *data), + const char *name, void *data, int index) +{ + int rc = 0; + unsigned long cookie; + struct mic_driver *mdrv = g_drv; + + rc = request_irq(mic_db_to_irq(mdrv, index), func, + 0, name, data); + if (rc) { + dev_err(mdrv->dev, "request_irq failed rc = %d\n", rc); + goto err; + } + mdrv->irq_info.irq_usage_count[index]++; + cookie = index; + return (struct mic_irq *)cookie; +err: + return ERR_PTR(rc); +} + +/** + * mic_free_card_irq - free irq. + * + * @cookie: cookie obtained during a successful call to mic_request_irq + * @data: private data specified by the calling function during the + * mic_request_irq + * + * returns: none. + */ +void mic_free_card_irq(struct mic_irq *cookie, void *data) +{ + int index; + struct mic_driver *mdrv = g_drv; + + index = (unsigned long)cookie & 0xFFFFU; + free_irq(mic_db_to_irq(mdrv, index), data); + mdrv->irq_info.irq_usage_count[index]--; +} + +/** + * mic_next_card_db - Get the doorbell with minimum usage count. + * + * Returns the irq index. + */ +int mic_next_card_db(void) +{ + int i; + int index = 0; + struct mic_driver *mdrv = g_drv; + + for (i = 0; i < mdrv->intr_info.num_intr; i++) { + if (mdrv->irq_info.irq_usage_count[i] < + mdrv->irq_info.irq_usage_count[index]) + index = i; + } + + return index; +} + +/** + * mic_init_irq - Initialize irq information. + * + * Returns 0 in success. Appropriate error code on failure. + */ +static int mic_init_irq(void) +{ + struct mic_driver *mdrv = g_drv; + + mdrv->irq_info.irq_usage_count = kzalloc((sizeof(u32) * + mdrv->intr_info.num_intr), + GFP_KERNEL); + if (!mdrv->irq_info.irq_usage_count) + return -ENOMEM; + return 0; +} + +/** + * mic_uninit_irq - Uninitialize irq information. + * + * None. + */ +static void mic_uninit_irq(void) +{ + struct mic_driver *mdrv = g_drv; + + kfree(mdrv->irq_info.irq_usage_count); +} + +/* + * mic_driver_init - MIC driver initialization tasks. + * + * Returns 0 in success. Appropriate error code on failure. + */ +int __init mic_driver_init(struct mic_driver *mdrv) +{ + int rc; + + g_drv = mdrv; + /* + * Unloading the card module is not supported. The MIC card module + * handles fundamental operations like host/card initiated shutdowns + * and informing the host about card crashes and cannot be unloaded. + */ + if (!try_module_get(mdrv->dev->driver->owner)) { + rc = -ENODEV; + goto done; + } + rc = mic_dp_init(); + if (rc) + goto put; + rc = mic_init_irq(); + if (rc) + goto dp_uninit; + rc = mic_shutdown_init(); + if (rc) + goto irq_uninit; + rc = mic_devices_init(mdrv); + if (rc) + goto shutdown_uninit; + mic_create_card_debug_dir(mdrv); + atomic_notifier_chain_register(&panic_notifier_list, &mic_panic); +done: + return rc; +shutdown_uninit: + mic_shutdown_uninit(); +irq_uninit: + mic_uninit_irq(); +dp_uninit: + mic_dp_uninit(); +put: + module_put(mdrv->dev->driver->owner); + return rc; +} + +/* + * mic_driver_uninit - MIC driver uninitialization tasks. + * + * Returns None + */ +void mic_driver_uninit(struct mic_driver *mdrv) +{ + mic_delete_card_debug_dir(mdrv); + mic_devices_uninit(mdrv); + /* + * Inform the host about the shutdown status i.e. poweroff/restart etc. + * The module cannot be unloaded so the only code path to call + * mic_devices_uninit(..) is the shutdown callback. + */ + mic_notify_host(system_state); + mic_shutdown_uninit(); + mic_uninit_irq(); + mic_dp_uninit(); + module_put(mdrv->dev->driver->owner); +} diff --git a/drivers/misc/mic/card/mic_device.h b/drivers/misc/mic/card/mic_device.h new file mode 100644 index 00000000000..306f502be95 --- /dev/null +++ b/drivers/misc/mic/card/mic_device.h @@ -0,0 +1,134 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Disclaimer: The codes contained in these modules may be specific to + * the Intel Software Development Platform codenamed: Knights Ferry, and + * the Intel product codenamed: Knights Corner, and are not backward + * compatible with other Intel products. Additionally, Intel will NOT + * support the codes or instruction set in future products. + * + * Intel MIC Card driver. + * + */ +#ifndef _MIC_CARD_DEVICE_H_ +#define _MIC_CARD_DEVICE_H_ + +#include <linux/workqueue.h> +#include <linux/io.h> +#include <linux/irqreturn.h> + +/** + * struct mic_intr_info - Contains h/w specific interrupt sources info + * + * @num_intr: The number of irqs available + */ +struct mic_intr_info { + u32 num_intr; +}; + +/** + * struct mic_irq_info - OS specific irq information + * + * @irq_usage_count: usage count array tracking the number of sources + * assigned for each irq. + */ +struct mic_irq_info { + int *irq_usage_count; +}; + +/** + * struct mic_device - MIC device information. + * + * @mmio: MMIO bar information. + */ +struct mic_device { + struct mic_mw mmio; +}; + +/** + * struct mic_driver - MIC card driver information. + * + * @name: Name for MIC driver. + * @dbg_dir: debugfs directory of this MIC device. + * @dev: The device backing this MIC. + * @dp: The pointer to the virtio device page. + * @mdev: MIC device information for the host. + * @hotplug_work: Hot plug work for adding/removing virtio devices. + * @irq_info: The OS specific irq information + * @intr_info: H/W specific interrupt information. + */ +struct mic_driver { + char name[20]; + struct dentry *dbg_dir; + struct device *dev; + void __iomem *dp; + struct mic_device mdev; + struct work_struct hotplug_work; + struct mic_irq_info irq_info; + struct mic_intr_info intr_info; +}; + +/** + * struct mic_irq - opaque pointer used as cookie + */ +struct mic_irq; + +/** + * mic_mmio_read - read from an MMIO register. + * @mw: MMIO register base virtual address. + * @offset: register offset. + * + * RETURNS: register value. + */ +static inline u32 mic_mmio_read(struct mic_mw *mw, u32 offset) +{ + return ioread32(mw->va + offset); +} + +/** + * mic_mmio_write - write to an MMIO register. + * @mw: MMIO register base virtual address. + * @val: the data value to put into the register + * @offset: register offset. + * + * RETURNS: none. + */ +static inline void +mic_mmio_write(struct mic_mw *mw, u32 val, u32 offset) +{ + iowrite32(val, mw->va + offset); +} + +int mic_driver_init(struct mic_driver *mdrv); +void mic_driver_uninit(struct mic_driver *mdrv); +int mic_next_card_db(void); +struct mic_irq *mic_request_card_irq(irqreturn_t (*func)(int irq, void *data), + const char *name, void *data, int intr_src); +void mic_free_card_irq(struct mic_irq *cookie, void *data); +u32 mic_read_spad(struct mic_device *mdev, unsigned int idx); +void mic_send_intr(struct mic_device *mdev, int doorbell); +int mic_db_to_irq(struct mic_driver *mdrv, int db); +u32 mic_ack_interrupt(struct mic_device *mdev); +void mic_hw_intr_init(struct mic_driver *mdrv); +void __iomem * +mic_card_map(struct mic_device *mdev, dma_addr_t addr, size_t size); +void mic_card_unmap(struct mic_device *mdev, void __iomem *addr); +void __init mic_create_card_debug_dir(struct mic_driver *mdrv); +void mic_delete_card_debug_dir(struct mic_driver *mdrv); +void __init mic_init_card_debugfs(void); +void mic_exit_card_debugfs(void); +#endif diff --git a/drivers/misc/mic/card/mic_virtio.c b/drivers/misc/mic/card/mic_virtio.c new file mode 100644 index 00000000000..653799b96bf --- /dev/null +++ b/drivers/misc/mic/card/mic_virtio.c @@ -0,0 +1,633 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Disclaimer: The codes contained in these modules may be specific to + * the Intel Software Development Platform codenamed: Knights Ferry, and + * the Intel product codenamed: Knights Corner, and are not backward + * compatible with other Intel products. Additionally, Intel will NOT + * support the codes or instruction set in future products. + * + * Adapted from: + * + * virtio for kvm on s390 + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Christian Borntraeger <borntraeger@de.ibm.com> + * + * Intel MIC Card driver. + * + */ +#include <linux/delay.h> +#include <linux/slab.h> +#include <linux/virtio_config.h> + +#include "../common/mic_dev.h" +#include "mic_virtio.h" + +#define VIRTIO_SUBCODE_64 0x0D00 + +#define MIC_MAX_VRINGS 4 +struct mic_vdev { + struct virtio_device vdev; + struct mic_device_desc __iomem *desc; + struct mic_device_ctrl __iomem *dc; + struct mic_device *mdev; + void __iomem *vr[MIC_MAX_VRINGS]; + int used_size[MIC_MAX_VRINGS]; + struct completion reset_done; + struct mic_irq *virtio_cookie; + int c2h_vdev_db; +}; + +static struct mic_irq *virtio_config_cookie; +#define to_micvdev(vd) container_of(vd, struct mic_vdev, vdev) + +/* Helper API to obtain the parent of the virtio device */ +static inline struct device *mic_dev(struct mic_vdev *mvdev) +{ + return mvdev->vdev.dev.parent; +} + +/* This gets the device's feature bits. */ +static u32 mic_get_features(struct virtio_device *vdev) +{ + unsigned int i, bits; + u32 features = 0; + struct mic_device_desc __iomem *desc = to_micvdev(vdev)->desc; + u8 __iomem *in_features = mic_vq_features(desc); + int feature_len = ioread8(&desc->feature_len); + + bits = min_t(unsigned, feature_len, + sizeof(vdev->features)) * 8; + for (i = 0; i < bits; i++) + if (ioread8(&in_features[i / 8]) & (BIT(i % 8))) + features |= BIT(i); + + return features; +} + +static void mic_finalize_features(struct virtio_device *vdev) +{ + unsigned int i, bits; + struct mic_device_desc __iomem *desc = to_micvdev(vdev)->desc; + u8 feature_len = ioread8(&desc->feature_len); + /* Second half of bitmap is features we accept. */ + u8 __iomem *out_features = + mic_vq_features(desc) + feature_len; + + /* Give virtio_ring a chance to accept features. */ + vring_transport_features(vdev); + + memset_io(out_features, 0, feature_len); + bits = min_t(unsigned, feature_len, + sizeof(vdev->features)) * 8; + for (i = 0; i < bits; i++) { + if (test_bit(i, vdev->features)) + iowrite8(ioread8(&out_features[i / 8]) | (1 << (i % 8)), + &out_features[i / 8]); + } +} + +/* + * Reading and writing elements in config space + */ +static void mic_get(struct virtio_device *vdev, unsigned int offset, + void *buf, unsigned len) +{ + struct mic_device_desc __iomem *desc = to_micvdev(vdev)->desc; + + if (offset + len > ioread8(&desc->config_len)) + return; + memcpy_fromio(buf, mic_vq_configspace(desc) + offset, len); +} + +static void mic_set(struct virtio_device *vdev, unsigned int offset, + const void *buf, unsigned len) +{ + struct mic_device_desc __iomem *desc = to_micvdev(vdev)->desc; + + if (offset + len > ioread8(&desc->config_len)) + return; + memcpy_toio(mic_vq_configspace(desc) + offset, buf, len); +} + +/* + * The operations to get and set the status word just access the status + * field of the device descriptor. set_status also interrupts the host + * to tell about status changes. + */ +static u8 mic_get_status(struct virtio_device *vdev) +{ + return ioread8(&to_micvdev(vdev)->desc->status); +} + +static void mic_set_status(struct virtio_device *vdev, u8 status) +{ + struct mic_vdev *mvdev = to_micvdev(vdev); + if (!status) + return; + iowrite8(status, &mvdev->desc->status); + mic_send_intr(mvdev->mdev, mvdev->c2h_vdev_db); +} + +/* Inform host on a virtio device reset and wait for ack from host */ +static void mic_reset_inform_host(struct virtio_device *vdev) +{ + struct mic_vdev *mvdev = to_micvdev(vdev); + struct mic_device_ctrl __iomem *dc = mvdev->dc; + int retry; + + iowrite8(0, &dc->host_ack); + iowrite8(1, &dc->vdev_reset); + mic_send_intr(mvdev->mdev, mvdev->c2h_vdev_db); + + /* Wait till host completes all card accesses and acks the reset */ + for (retry = 100; retry--;) { + if (ioread8(&dc->host_ack)) + break; + msleep(100); + }; + + dev_dbg(mic_dev(mvdev), "%s: retry: %d\n", __func__, retry); + + /* Reset status to 0 in case we timed out */ + iowrite8(0, &mvdev->desc->status); +} + +static void mic_reset(struct virtio_device *vdev) +{ + struct mic_vdev *mvdev = to_micvdev(vdev); + + dev_dbg(mic_dev(mvdev), "%s: virtio id %d\n", + __func__, vdev->id.device); + + mic_reset_inform_host(vdev); + complete_all(&mvdev->reset_done); +} + +/* + * The virtio_ring code calls this API when it wants to notify the Host. + */ +static bool mic_notify(struct virtqueue *vq) +{ + struct mic_vdev *mvdev = vq->priv; + + mic_send_intr(mvdev->mdev, mvdev->c2h_vdev_db); + return true; +} + +static void mic_del_vq(struct virtqueue *vq, int n) +{ + struct mic_vdev *mvdev = to_micvdev(vq->vdev); + struct vring *vr = (struct vring *)(vq + 1); + + free_pages((unsigned long) vr->used, get_order(mvdev->used_size[n])); + vring_del_virtqueue(vq); + mic_card_unmap(mvdev->mdev, mvdev->vr[n]); + mvdev->vr[n] = NULL; +} + +static void mic_del_vqs(struct virtio_device *vdev) +{ + struct mic_vdev *mvdev = to_micvdev(vdev); + struct virtqueue *vq, *n; + int idx = 0; + + dev_dbg(mic_dev(mvdev), "%s\n", __func__); + + list_for_each_entry_safe(vq, n, &vdev->vqs, list) + mic_del_vq(vq, idx++); +} + +/* + * This routine will assign vring's allocated in host/io memory. Code in + * virtio_ring.c however continues to access this io memory as if it were local + * memory without io accessors. + */ +static struct virtqueue *mic_find_vq(struct virtio_device *vdev, + unsigned index, + void (*callback)(struct virtqueue *vq), + const char *name) +{ + struct mic_vdev *mvdev = to_micvdev(vdev); + struct mic_vqconfig __iomem *vqconfig; + struct mic_vqconfig config; + struct virtqueue *vq; + void __iomem *va; + struct _mic_vring_info __iomem *info; + void *used; + int vr_size, _vr_size, err, magic; + struct vring *vr; + u8 type = ioread8(&mvdev->desc->type); + + if (index >= ioread8(&mvdev->desc->num_vq)) + return ERR_PTR(-ENOENT); + + if (!name) + return ERR_PTR(-ENOENT); + + /* First assign the vring's allocated in host memory */ + vqconfig = mic_vq_config(mvdev->desc) + index; + memcpy_fromio(&config, vqconfig, sizeof(config)); + _vr_size = vring_size(le16_to_cpu(config.num), MIC_VIRTIO_RING_ALIGN); + vr_size = PAGE_ALIGN(_vr_size + sizeof(struct _mic_vring_info)); + va = mic_card_map(mvdev->mdev, le64_to_cpu(config.address), vr_size); + if (!va) + return ERR_PTR(-ENOMEM); + mvdev->vr[index] = va; + memset_io(va, 0x0, _vr_size); + vq = vring_new_virtqueue(index, le16_to_cpu(config.num), + MIC_VIRTIO_RING_ALIGN, vdev, false, + (void __force *)va, mic_notify, callback, + name); + if (!vq) { + err = -ENOMEM; + goto unmap; + } + info = va + _vr_size; + magic = ioread32(&info->magic); + + if (WARN(magic != MIC_MAGIC + type + index, "magic mismatch")) { + err = -EIO; + goto unmap; + } + + /* Allocate and reassign used ring now */ + mvdev->used_size[index] = PAGE_ALIGN(sizeof(__u16) * 3 + + sizeof(struct vring_used_elem) * + le16_to_cpu(config.num)); + used = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, + get_order(mvdev->used_size[index])); + if (!used) { + err = -ENOMEM; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, err); + goto del_vq; + } + iowrite64(virt_to_phys(used), &vqconfig->used_address); + + /* + * To reassign the used ring here we are directly accessing + * struct vring_virtqueue which is a private data structure + * in virtio_ring.c. At the minimum, a BUILD_BUG_ON() in + * vring_new_virtqueue() would ensure that + * (&vq->vring == (struct vring *) (&vq->vq + 1)); + */ + vr = (struct vring *)(vq + 1); + vr->used = used; + + vq->priv = mvdev; + return vq; +del_vq: + vring_del_virtqueue(vq); +unmap: + mic_card_unmap(mvdev->mdev, mvdev->vr[index]); + return ERR_PTR(err); +} + +static int mic_find_vqs(struct virtio_device *vdev, unsigned nvqs, + struct virtqueue *vqs[], + vq_callback_t *callbacks[], + const char *names[]) +{ + struct mic_vdev *mvdev = to_micvdev(vdev); + struct mic_device_ctrl __iomem *dc = mvdev->dc; + int i, err, retry; + + /* We must have this many virtqueues. */ + if (nvqs > ioread8(&mvdev->desc->num_vq)) + return -ENOENT; + + for (i = 0; i < nvqs; ++i) { + dev_dbg(mic_dev(mvdev), "%s: %d: %s\n", + __func__, i, names[i]); + vqs[i] = mic_find_vq(vdev, i, callbacks[i], names[i]); + if (IS_ERR(vqs[i])) { + err = PTR_ERR(vqs[i]); + goto error; + } + } + + iowrite8(1, &dc->used_address_updated); + /* + * Send an interrupt to the host to inform it that used + * rings have been re-assigned. + */ + mic_send_intr(mvdev->mdev, mvdev->c2h_vdev_db); + for (retry = 100; retry--;) { + if (!ioread8(&dc->used_address_updated)) + break; + msleep(100); + }; + + dev_dbg(mic_dev(mvdev), "%s: retry: %d\n", __func__, retry); + if (!retry) { + err = -ENODEV; + goto error; + } + + return 0; +error: + mic_del_vqs(vdev); + return err; +} + +/* + * The config ops structure as defined by virtio config + */ +static struct virtio_config_ops mic_vq_config_ops = { + .get_features = mic_get_features, + .finalize_features = mic_finalize_features, + .get = mic_get, + .set = mic_set, + .get_status = mic_get_status, + .set_status = mic_set_status, + .reset = mic_reset, + .find_vqs = mic_find_vqs, + .del_vqs = mic_del_vqs, +}; + +static irqreturn_t +mic_virtio_intr_handler(int irq, void *data) +{ + struct mic_vdev *mvdev = data; + struct virtqueue *vq; + + mic_ack_interrupt(mvdev->mdev); + list_for_each_entry(vq, &mvdev->vdev.vqs, list) + vring_interrupt(0, vq); + + return IRQ_HANDLED; +} + +static void mic_virtio_release_dev(struct device *_d) +{ + /* + * No need for a release method similar to virtio PCI. + * Provide an empty one to avoid getting a warning from core. + */ +} + +/* + * adds a new device and register it with virtio + * appropriate drivers are loaded by the device model + */ +static int mic_add_device(struct mic_device_desc __iomem *d, + unsigned int offset, struct mic_driver *mdrv) +{ + struct mic_vdev *mvdev; + int ret; + int virtio_db; + u8 type = ioread8(&d->type); + + mvdev = kzalloc(sizeof(*mvdev), GFP_KERNEL); + if (!mvdev) { + dev_err(mdrv->dev, "Cannot allocate mic dev %u type %u\n", + offset, type); + return -ENOMEM; + } + + mvdev->mdev = &mdrv->mdev; + mvdev->vdev.dev.parent = mdrv->dev; + mvdev->vdev.dev.release = mic_virtio_release_dev; + mvdev->vdev.id.device = type; + mvdev->vdev.config = &mic_vq_config_ops; + mvdev->desc = d; + mvdev->dc = (void __iomem *)d + mic_aligned_desc_size(d); + init_completion(&mvdev->reset_done); + + virtio_db = mic_next_card_db(); + mvdev->virtio_cookie = mic_request_card_irq(mic_virtio_intr_handler, + "virtio intr", mvdev, virtio_db); + if (IS_ERR(mvdev->virtio_cookie)) { + ret = PTR_ERR(mvdev->virtio_cookie); + goto kfree; + } + iowrite8((u8)virtio_db, &mvdev->dc->h2c_vdev_db); + mvdev->c2h_vdev_db = ioread8(&mvdev->dc->c2h_vdev_db); + + ret = register_virtio_device(&mvdev->vdev); + if (ret) { + dev_err(mic_dev(mvdev), + "Failed to register mic device %u type %u\n", + offset, type); + goto free_irq; + } + iowrite64((u64)mvdev, &mvdev->dc->vdev); + dev_dbg(mic_dev(mvdev), "%s: registered mic device %u type %u mvdev %p\n", + __func__, offset, type, mvdev); + + return 0; + +free_irq: + mic_free_card_irq(mvdev->virtio_cookie, mvdev); +kfree: + kfree(mvdev); + return ret; +} + +/* + * match for a mic device with a specific desc pointer + */ +static int mic_match_desc(struct device *dev, void *data) +{ + struct virtio_device *vdev = dev_to_virtio(dev); + struct mic_vdev *mvdev = to_micvdev(vdev); + + return mvdev->desc == (void __iomem *)data; +} + +static void mic_handle_config_change(struct mic_device_desc __iomem *d, + unsigned int offset, struct mic_driver *mdrv) +{ + struct mic_device_ctrl __iomem *dc + = (void __iomem *)d + mic_aligned_desc_size(d); + struct mic_vdev *mvdev = (struct mic_vdev *)ioread64(&dc->vdev); + struct virtio_driver *drv; + + if (ioread8(&dc->config_change) != MIC_VIRTIO_PARAM_CONFIG_CHANGED) + return; + + dev_dbg(mdrv->dev, "%s %d\n", __func__, __LINE__); + drv = container_of(mvdev->vdev.dev.driver, + struct virtio_driver, driver); + if (drv->config_changed) + drv->config_changed(&mvdev->vdev); + iowrite8(1, &dc->guest_ack); +} + +/* + * removes a virtio device if a hot remove event has been + * requested by the host. + */ +static int mic_remove_device(struct mic_device_desc __iomem *d, + unsigned int offset, struct mic_driver *mdrv) +{ + struct mic_device_ctrl __iomem *dc + = (void __iomem *)d + mic_aligned_desc_size(d); + struct mic_vdev *mvdev = (struct mic_vdev *)ioread64(&dc->vdev); + u8 status; + int ret = -1; + + if (ioread8(&dc->config_change) == MIC_VIRTIO_PARAM_DEV_REMOVE) { + dev_dbg(mdrv->dev, + "%s %d config_change %d type %d mvdev %p\n", + __func__, __LINE__, + ioread8(&dc->config_change), ioread8(&d->type), mvdev); + + status = ioread8(&d->status); + reinit_completion(&mvdev->reset_done); + unregister_virtio_device(&mvdev->vdev); + mic_free_card_irq(mvdev->virtio_cookie, mvdev); + if (status & VIRTIO_CONFIG_S_DRIVER_OK) + wait_for_completion(&mvdev->reset_done); + kfree(mvdev); + iowrite8(1, &dc->guest_ack); + dev_dbg(mdrv->dev, "%s %d guest_ack %d\n", + __func__, __LINE__, ioread8(&dc->guest_ack)); + ret = 0; + } + + return ret; +} + +#define REMOVE_DEVICES true + +static void mic_scan_devices(struct mic_driver *mdrv, bool remove) +{ + s8 type; + unsigned int i; + struct mic_device_desc __iomem *d; + struct mic_device_ctrl __iomem *dc; + struct device *dev; + int ret; + + for (i = sizeof(struct mic_bootparam); i < MIC_DP_SIZE; + i += mic_total_desc_size(d)) { + d = mdrv->dp + i; + dc = (void __iomem *)d + mic_aligned_desc_size(d); + /* + * This read barrier is paired with the corresponding write + * barrier on the host which is inserted before adding or + * removing a virtio device descriptor, by updating the type. + */ + rmb(); + type = ioread8(&d->type); + + /* end of list */ + if (type == 0) + break; + + if (type == -1) + continue; + + /* device already exists */ + dev = device_find_child(mdrv->dev, (void __force *)d, + mic_match_desc); + if (dev) { + if (remove) + iowrite8(MIC_VIRTIO_PARAM_DEV_REMOVE, + &dc->config_change); + put_device(dev); + mic_handle_config_change(d, i, mdrv); + ret = mic_remove_device(d, i, mdrv); + if (!ret && !remove) + iowrite8(-1, &d->type); + if (remove) { + iowrite8(0, &dc->config_change); + iowrite8(0, &dc->guest_ack); + } + continue; + } + + /* new device */ + dev_dbg(mdrv->dev, "%s %d Adding new virtio device %p\n", + __func__, __LINE__, d); + if (!remove) + mic_add_device(d, i, mdrv); + } +} + +/* + * mic_hotplug_device tries to find changes in the device page. + */ +static void mic_hotplug_devices(struct work_struct *work) +{ + struct mic_driver *mdrv = container_of(work, + struct mic_driver, hotplug_work); + + mic_scan_devices(mdrv, !REMOVE_DEVICES); +} + +/* + * Interrupt handler for hot plug/config changes etc. + */ +static irqreturn_t +mic_extint_handler(int irq, void *data) +{ + struct mic_driver *mdrv = (struct mic_driver *)data; + + dev_dbg(mdrv->dev, "%s %d hotplug work\n", + __func__, __LINE__); + mic_ack_interrupt(&mdrv->mdev); + schedule_work(&mdrv->hotplug_work); + return IRQ_HANDLED; +} + +/* + * Init function for virtio + */ +int mic_devices_init(struct mic_driver *mdrv) +{ + int rc; + struct mic_bootparam __iomem *bootparam; + int config_db; + + INIT_WORK(&mdrv->hotplug_work, mic_hotplug_devices); + mic_scan_devices(mdrv, !REMOVE_DEVICES); + + config_db = mic_next_card_db(); + virtio_config_cookie = mic_request_card_irq(mic_extint_handler, + "virtio_config_intr", mdrv, config_db); + if (IS_ERR(virtio_config_cookie)) { + rc = PTR_ERR(virtio_config_cookie); + goto exit; + } + + bootparam = mdrv->dp; + iowrite8(config_db, &bootparam->h2c_config_db); + return 0; +exit: + return rc; +} + +/* + * Uninit function for virtio + */ +void mic_devices_uninit(struct mic_driver *mdrv) +{ + struct mic_bootparam __iomem *bootparam = mdrv->dp; + iowrite8(-1, &bootparam->h2c_config_db); + mic_free_card_irq(virtio_config_cookie, mdrv); + flush_work(&mdrv->hotplug_work); + mic_scan_devices(mdrv, REMOVE_DEVICES); +} diff --git a/drivers/misc/mic/card/mic_virtio.h b/drivers/misc/mic/card/mic_virtio.h new file mode 100644 index 00000000000..d0407ba53bb --- /dev/null +++ b/drivers/misc/mic/card/mic_virtio.h @@ -0,0 +1,76 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Disclaimer: The codes contained in these modules may be specific to + * the Intel Software Development Platform codenamed: Knights Ferry, and + * the Intel product codenamed: Knights Corner, and are not backward + * compatible with other Intel products. Additionally, Intel will NOT + * support the codes or instruction set in future products. + * + * Intel MIC Card driver. + * + */ +#ifndef __MIC_CARD_VIRTIO_H +#define __MIC_CARD_VIRTIO_H + +#include <linux/mic_common.h> +#include "mic_device.h" + +/* + * 64 bit I/O access + */ +#ifndef ioread64 +#define ioread64 readq +#endif +#ifndef iowrite64 +#define iowrite64 writeq +#endif + +static inline unsigned mic_desc_size(struct mic_device_desc __iomem *desc) +{ + return sizeof(*desc) + + ioread8(&desc->num_vq) * sizeof(struct mic_vqconfig) + + ioread8(&desc->feature_len) * 2 + + ioread8(&desc->config_len); +} + +static inline struct mic_vqconfig __iomem * +mic_vq_config(struct mic_device_desc __iomem *desc) +{ + return (struct mic_vqconfig __iomem *)(desc + 1); +} + +static inline __u8 __iomem * +mic_vq_features(struct mic_device_desc __iomem *desc) +{ + return (__u8 __iomem *)(mic_vq_config(desc) + ioread8(&desc->num_vq)); +} + +static inline __u8 __iomem * +mic_vq_configspace(struct mic_device_desc __iomem *desc) +{ + return mic_vq_features(desc) + ioread8(&desc->feature_len) * 2; +} +static inline unsigned mic_total_desc_size(struct mic_device_desc __iomem *desc) +{ + return mic_aligned_desc_size(desc) + sizeof(struct mic_device_ctrl); +} + +int mic_devices_init(struct mic_driver *mdrv); +void mic_devices_uninit(struct mic_driver *mdrv); + +#endif diff --git a/drivers/misc/mic/card/mic_x100.c b/drivers/misc/mic/card/mic_x100.c new file mode 100644 index 00000000000..2868945c9a4 --- /dev/null +++ b/drivers/misc/mic/card/mic_x100.c @@ -0,0 +1,256 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Disclaimer: The codes contained in these modules may be specific to + * the Intel Software Development Platform codenamed: Knights Ferry, and + * the Intel product codenamed: Knights Corner, and are not backward + * compatible with other Intel products. Additionally, Intel will NOT + * support the codes or instruction set in future products. + * + * Intel MIC Card driver. + * + */ +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/platform_device.h> + +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_x100.h" + +static const char mic_driver_name[] = "mic"; + +static struct mic_driver g_drv; + +/** + * mic_read_spad - read from the scratchpad register + * @mdev: pointer to mic_device instance + * @idx: index to scratchpad register, 0 based + * + * This function allows reading of the 32bit scratchpad register. + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +u32 mic_read_spad(struct mic_device *mdev, unsigned int idx) +{ + return mic_mmio_read(&mdev->mmio, + MIC_X100_SBOX_BASE_ADDRESS + + MIC_X100_SBOX_SPAD0 + idx * 4); +} + +/** + * __mic_send_intr - Send interrupt to Host. + * @mdev: pointer to mic_device instance + * @doorbell: Doorbell number. + */ +void mic_send_intr(struct mic_device *mdev, int doorbell) +{ + struct mic_mw *mw = &mdev->mmio; + + if (doorbell > MIC_X100_MAX_DOORBELL_IDX) + return; + /* Ensure that the interrupt is ordered w.r.t previous stores. */ + wmb(); + mic_mmio_write(mw, MIC_X100_SBOX_SDBIC0_DBREQ_BIT, + MIC_X100_SBOX_BASE_ADDRESS + + (MIC_X100_SBOX_SDBIC0 + (4 * doorbell))); +} + +/** + * mic_ack_interrupt - Device specific interrupt handling. + * @mdev: pointer to mic_device instance + * + * Returns: bitmask of doorbell events triggered. + */ +u32 mic_ack_interrupt(struct mic_device *mdev) +{ + return 0; +} + +static inline int mic_get_sbox_irq(int db) +{ + return MIC_X100_IRQ_BASE + db; +} + +static inline int mic_get_rdmasr_irq(int index) +{ + return MIC_X100_RDMASR_IRQ_BASE + index; +} + +/** + * mic_hw_intr_init - Initialize h/w specific interrupt + * information. + * @mdrv: pointer to mic_driver + */ +void mic_hw_intr_init(struct mic_driver *mdrv) +{ + mdrv->intr_info.num_intr = MIC_X100_NUM_SBOX_IRQ + + MIC_X100_NUM_RDMASR_IRQ; +} + +/** + * mic_db_to_irq - Retrieve irq number corresponding to a doorbell. + * @mdrv: pointer to mic_driver + * @db: The doorbell obtained for which the irq is needed. Doorbell + * may correspond to an sbox doorbell or an rdmasr index. + * + * Returns the irq corresponding to the doorbell. + */ +int mic_db_to_irq(struct mic_driver *mdrv, int db) +{ + int rdmasr_index; + if (db < MIC_X100_NUM_SBOX_IRQ) { + return mic_get_sbox_irq(db); + } else { + rdmasr_index = db - MIC_X100_NUM_SBOX_IRQ + + MIC_X100_RDMASR_IRQ_BASE; + return mic_get_rdmasr_irq(rdmasr_index); + } +} + +/* + * mic_card_map - Allocate virtual address for a remote memory region. + * @mdev: pointer to mic_device instance. + * @addr: Remote DMA address. + * @size: Size of the region. + * + * Returns: Virtual address backing the remote memory region. + */ +void __iomem * +mic_card_map(struct mic_device *mdev, dma_addr_t addr, size_t size) +{ + return ioremap(addr, size); +} + +/* + * mic_card_unmap - Unmap the virtual address for a remote memory region. + * @mdev: pointer to mic_device instance. + * @addr: Virtual address for remote memory region. + * + * Returns: None. + */ +void mic_card_unmap(struct mic_device *mdev, void __iomem *addr) +{ + iounmap(addr); +} + +static int __init mic_probe(struct platform_device *pdev) +{ + struct mic_driver *mdrv = &g_drv; + struct mic_device *mdev = &mdrv->mdev; + int rc = 0; + + mdrv->dev = &pdev->dev; + snprintf(mdrv->name, sizeof(mic_driver_name), mic_driver_name); + + mdev->mmio.pa = MIC_X100_MMIO_BASE; + mdev->mmio.len = MIC_X100_MMIO_LEN; + mdev->mmio.va = ioremap(MIC_X100_MMIO_BASE, MIC_X100_MMIO_LEN); + if (!mdev->mmio.va) { + dev_err(&pdev->dev, "Cannot remap MMIO BAR\n"); + rc = -EIO; + goto done; + } + mic_hw_intr_init(mdrv); + rc = mic_driver_init(mdrv); + if (rc) { + dev_err(&pdev->dev, "mic_driver_init failed rc %d\n", rc); + goto iounmap; + } +done: + return rc; +iounmap: + iounmap(mdev->mmio.va); + return rc; +} + +static int mic_remove(struct platform_device *pdev) +{ + struct mic_driver *mdrv = &g_drv; + struct mic_device *mdev = &mdrv->mdev; + + mic_driver_uninit(mdrv); + iounmap(mdev->mmio.va); + return 0; +} + +static void mic_platform_shutdown(struct platform_device *pdev) +{ + mic_remove(pdev); +} + +static struct platform_device mic_platform_dev = { + .name = mic_driver_name, + .id = 0, + .num_resources = 0, +}; + +static struct platform_driver __refdata mic_platform_driver = { + .probe = mic_probe, + .remove = mic_remove, + .shutdown = mic_platform_shutdown, + .driver = { + .name = mic_driver_name, + .owner = THIS_MODULE, + }, +}; + +static int __init mic_init(void) +{ + int ret; + struct cpuinfo_x86 *c = &cpu_data(0); + + if (!(c->x86 == 11 && c->x86_model == 1)) { + ret = -ENODEV; + pr_err("%s not running on X100 ret %d\n", __func__, ret); + goto done; + } + + mic_init_card_debugfs(); + ret = platform_device_register(&mic_platform_dev); + if (ret) { + pr_err("platform_device_register ret %d\n", ret); + goto cleanup_debugfs; + } + ret = platform_driver_register(&mic_platform_driver); + if (ret) { + pr_err("platform_driver_register ret %d\n", ret); + goto device_unregister; + } + return ret; + +device_unregister: + platform_device_unregister(&mic_platform_dev); +cleanup_debugfs: + mic_exit_card_debugfs(); +done: + return ret; +} + +static void __exit mic_exit(void) +{ + platform_driver_unregister(&mic_platform_driver); + platform_device_unregister(&mic_platform_dev); + mic_exit_card_debugfs(); +} + +module_init(mic_init); +module_exit(mic_exit); + +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel(R) MIC X100 Card driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/card/mic_x100.h b/drivers/misc/mic/card/mic_x100.h new file mode 100644 index 00000000000..d66ea55639c --- /dev/null +++ b/drivers/misc/mic/card/mic_x100.h @@ -0,0 +1,48 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Disclaimer: The codes contained in these modules may be specific to + * the Intel Software Development Platform codenamed: Knights Ferry, and + * the Intel product codenamed: Knights Corner, and are not backward + * compatible with other Intel products. Additionally, Intel will NOT + * support the codes or instruction set in future products. + * + * Intel MIC Card driver. + * + */ +#ifndef _MIC_X100_CARD_H_ +#define _MIC_X100_CARD_H_ + +#define MIC_X100_MMIO_BASE 0x08007C0000ULL +#define MIC_X100_MMIO_LEN 0x00020000ULL +#define MIC_X100_SBOX_BASE_ADDRESS 0x00010000ULL + +#define MIC_X100_SBOX_SPAD0 0x0000AB20 +#define MIC_X100_SBOX_SDBIC0 0x0000CC90 +#define MIC_X100_SBOX_SDBIC0_DBREQ_BIT 0x80000000 +#define MIC_X100_SBOX_RDMASR0 0x0000B180 + +#define MIC_X100_MAX_DOORBELL_IDX 8 + +#define MIC_X100_NUM_SBOX_IRQ 8 +#define MIC_X100_NUM_RDMASR_IRQ 8 +#define MIC_X100_SBOX_IRQ_BASE 0 +#define MIC_X100_RDMASR_IRQ_BASE 17 + +#define MIC_X100_IRQ_BASE 26 + +#endif diff --git a/drivers/misc/mic/common/mic_dev.h b/drivers/misc/mic/common/mic_dev.h new file mode 100644 index 00000000000..92999c2bbf8 --- /dev/null +++ b/drivers/misc/mic/common/mic_dev.h @@ -0,0 +1,51 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC driver. + * + */ +#ifndef __MIC_DEV_H__ +#define __MIC_DEV_H__ + +/** + * struct mic_mw - MIC memory window + * + * @pa: Base physical address. + * @va: Base ioremap'd virtual address. + * @len: Size of the memory window. + */ +struct mic_mw { + phys_addr_t pa; + void __iomem *va; + resource_size_t len; +}; + +/* + * Scratch pad register offsets used by the host to communicate + * device page DMA address to the card. + */ +#define MIC_DPLO_SPAD 14 +#define MIC_DPHI_SPAD 15 + +/* + * These values are supposed to be in the config_change field of the + * device page when the host sends a config change interrupt to the card. + */ +#define MIC_VIRTIO_PARAM_DEV_REMOVE 0x1 +#define MIC_VIRTIO_PARAM_CONFIG_CHANGED 0x2 + +#endif diff --git a/drivers/misc/mic/host/Makefile b/drivers/misc/mic/host/Makefile new file mode 100644 index 00000000000..c2197f99939 --- /dev/null +++ b/drivers/misc/mic/host/Makefile @@ -0,0 +1,14 @@ +# +# Makefile - Intel MIC Linux driver. +# Copyright(c) 2013, Intel Corporation. +# +obj-$(CONFIG_INTEL_MIC_HOST) += mic_host.o +mic_host-objs := mic_main.o +mic_host-objs += mic_x100.o +mic_host-objs += mic_sysfs.o +mic_host-objs += mic_smpt.o +mic_host-objs += mic_intr.o +mic_host-objs += mic_boot.o +mic_host-objs += mic_debugfs.o +mic_host-objs += mic_fops.o +mic_host-objs += mic_virtio.o diff --git a/drivers/misc/mic/host/mic_boot.c b/drivers/misc/mic/host/mic_boot.c new file mode 100644 index 00000000000..b75c6b5cc20 --- /dev/null +++ b/drivers/misc/mic/host/mic_boot.c @@ -0,0 +1,300 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/delay.h> +#include <linux/firmware.h> +#include <linux/pci.h> + +#include <linux/mic_common.h> +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_smpt.h" +#include "mic_virtio.h" + +/** + * mic_reset - Reset the MIC device. + * @mdev: pointer to mic_device instance + */ +static void mic_reset(struct mic_device *mdev) +{ + int i; + +#define MIC_RESET_TO (45) + + reinit_completion(&mdev->reset_wait); + mdev->ops->reset_fw_ready(mdev); + mdev->ops->reset(mdev); + + for (i = 0; i < MIC_RESET_TO; i++) { + if (mdev->ops->is_fw_ready(mdev)) + goto done; + /* + * Resets typically take 10s of seconds to complete. + * Since an MMIO read is required to check if the + * firmware is ready or not, a 1 second delay works nicely. + */ + msleep(1000); + } + mic_set_state(mdev, MIC_RESET_FAILED); +done: + complete_all(&mdev->reset_wait); +} + +/* Initialize the MIC bootparams */ +void mic_bootparam_init(struct mic_device *mdev) +{ + struct mic_bootparam *bootparam = mdev->dp; + + bootparam->magic = cpu_to_le32(MIC_MAGIC); + bootparam->c2h_shutdown_db = mdev->shutdown_db; + bootparam->h2c_shutdown_db = -1; + bootparam->h2c_config_db = -1; + bootparam->shutdown_status = 0; + bootparam->shutdown_card = 0; +} + +/** + * mic_start - Start the MIC. + * @mdev: pointer to mic_device instance + * @buf: buffer containing boot string including firmware/ramdisk path. + * + * This function prepares an MIC for boot and initiates boot. + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +int mic_start(struct mic_device *mdev, const char *buf) +{ + int rc; + mutex_lock(&mdev->mic_mutex); +retry: + if (MIC_OFFLINE != mdev->state) { + rc = -EINVAL; + goto unlock_ret; + } + if (!mdev->ops->is_fw_ready(mdev)) { + mic_reset(mdev); + /* + * The state will either be MIC_OFFLINE if the reset succeeded + * or MIC_RESET_FAILED if the firmware reset failed. + */ + goto retry; + } + rc = mdev->ops->load_mic_fw(mdev, buf); + if (rc) + goto unlock_ret; + mic_smpt_restore(mdev); + mic_intr_restore(mdev); + mdev->intr_ops->enable_interrupts(mdev); + mdev->ops->write_spad(mdev, MIC_DPLO_SPAD, mdev->dp_dma_addr); + mdev->ops->write_spad(mdev, MIC_DPHI_SPAD, mdev->dp_dma_addr >> 32); + mdev->ops->send_firmware_intr(mdev); + mic_set_state(mdev, MIC_ONLINE); +unlock_ret: + mutex_unlock(&mdev->mic_mutex); + return rc; +} + +/** + * mic_stop - Prepare the MIC for reset and trigger reset. + * @mdev: pointer to mic_device instance + * @force: force a MIC to reset even if it is already offline. + * + * RETURNS: None. + */ +void mic_stop(struct mic_device *mdev, bool force) +{ + mutex_lock(&mdev->mic_mutex); + if (MIC_OFFLINE != mdev->state || force) { + mic_virtio_reset_devices(mdev); + mic_bootparam_init(mdev); + mic_reset(mdev); + if (MIC_RESET_FAILED == mdev->state) + goto unlock; + mic_set_shutdown_status(mdev, MIC_NOP); + if (MIC_SUSPENDED != mdev->state) + mic_set_state(mdev, MIC_OFFLINE); + } +unlock: + mutex_unlock(&mdev->mic_mutex); +} + +/** + * mic_shutdown - Initiate MIC shutdown. + * @mdev: pointer to mic_device instance + * + * RETURNS: None. + */ +void mic_shutdown(struct mic_device *mdev) +{ + struct mic_bootparam *bootparam = mdev->dp; + s8 db = bootparam->h2c_shutdown_db; + + mutex_lock(&mdev->mic_mutex); + if (MIC_ONLINE == mdev->state && db != -1) { + bootparam->shutdown_card = 1; + mdev->ops->send_intr(mdev, db); + mic_set_state(mdev, MIC_SHUTTING_DOWN); + } + mutex_unlock(&mdev->mic_mutex); +} + +/** + * mic_shutdown_work - Handle shutdown interrupt from MIC. + * @work: The work structure. + * + * This work is scheduled whenever the host has received a shutdown + * interrupt from the MIC. + */ +void mic_shutdown_work(struct work_struct *work) +{ + struct mic_device *mdev = container_of(work, struct mic_device, + shutdown_work); + struct mic_bootparam *bootparam = mdev->dp; + + mutex_lock(&mdev->mic_mutex); + mic_set_shutdown_status(mdev, bootparam->shutdown_status); + bootparam->shutdown_status = 0; + + /* + * if state is MIC_SUSPENDED, OSPM suspend is in progress. We do not + * change the state here so as to prevent users from booting the card + * during and after the suspend operation. + */ + if (MIC_SHUTTING_DOWN != mdev->state && + MIC_SUSPENDED != mdev->state) + mic_set_state(mdev, MIC_SHUTTING_DOWN); + mutex_unlock(&mdev->mic_mutex); +} + +/** + * mic_reset_trigger_work - Trigger MIC reset. + * @work: The work structure. + * + * This work is scheduled whenever the host wants to reset the MIC. + */ +void mic_reset_trigger_work(struct work_struct *work) +{ + struct mic_device *mdev = container_of(work, struct mic_device, + reset_trigger_work); + + mic_stop(mdev, false); +} + +/** + * mic_complete_resume - Complete MIC Resume after an OSPM suspend/hibernate + * event. + * @mdev: pointer to mic_device instance + * + * RETURNS: None. + */ +void mic_complete_resume(struct mic_device *mdev) +{ + if (mdev->state != MIC_SUSPENDED) { + dev_warn(mdev->sdev->parent, "state %d should be %d\n", + mdev->state, MIC_SUSPENDED); + return; + } + + /* Make sure firmware is ready */ + if (!mdev->ops->is_fw_ready(mdev)) + mic_stop(mdev, true); + + mutex_lock(&mdev->mic_mutex); + mic_set_state(mdev, MIC_OFFLINE); + mutex_unlock(&mdev->mic_mutex); +} + +/** + * mic_prepare_suspend - Handle suspend notification for the MIC device. + * @mdev: pointer to mic_device instance + * + * RETURNS: None. + */ +void mic_prepare_suspend(struct mic_device *mdev) +{ + int rc; + +#define MIC_SUSPEND_TIMEOUT (60 * HZ) + + mutex_lock(&mdev->mic_mutex); + switch (mdev->state) { + case MIC_OFFLINE: + /* + * Card is already offline. Set state to MIC_SUSPENDED + * to prevent users from booting the card. + */ + mic_set_state(mdev, MIC_SUSPENDED); + mutex_unlock(&mdev->mic_mutex); + break; + case MIC_ONLINE: + /* + * Card is online. Set state to MIC_SUSPENDING and notify + * MIC user space daemon which will issue card + * shutdown and reset. + */ + mic_set_state(mdev, MIC_SUSPENDING); + mutex_unlock(&mdev->mic_mutex); + rc = wait_for_completion_timeout(&mdev->reset_wait, + MIC_SUSPEND_TIMEOUT); + /* Force reset the card if the shutdown completion timed out */ + if (!rc) { + mutex_lock(&mdev->mic_mutex); + mic_set_state(mdev, MIC_SUSPENDED); + mutex_unlock(&mdev->mic_mutex); + mic_stop(mdev, true); + } + break; + case MIC_SHUTTING_DOWN: + /* + * Card is shutting down. Set state to MIC_SUSPENDED + * to prevent further boot of the card. + */ + mic_set_state(mdev, MIC_SUSPENDED); + mutex_unlock(&mdev->mic_mutex); + rc = wait_for_completion_timeout(&mdev->reset_wait, + MIC_SUSPEND_TIMEOUT); + /* Force reset the card if the shutdown completion timed out */ + if (!rc) + mic_stop(mdev, true); + break; + default: + mutex_unlock(&mdev->mic_mutex); + break; + } +} + +/** + * mic_suspend - Initiate MIC suspend. Suspend merely issues card shutdown. + * @mdev: pointer to mic_device instance + * + * RETURNS: None. + */ +void mic_suspend(struct mic_device *mdev) +{ + struct mic_bootparam *bootparam = mdev->dp; + s8 db = bootparam->h2c_shutdown_db; + + mutex_lock(&mdev->mic_mutex); + if (MIC_SUSPENDING == mdev->state && db != -1) { + bootparam->shutdown_card = 1; + mdev->ops->send_intr(mdev, db); + mic_set_state(mdev, MIC_SUSPENDED); + } + mutex_unlock(&mdev->mic_mutex); +} diff --git a/drivers/misc/mic/host/mic_debugfs.c b/drivers/misc/mic/host/mic_debugfs.c new file mode 100644 index 00000000000..028ba5d6fd1 --- /dev/null +++ b/drivers/misc/mic/host/mic_debugfs.c @@ -0,0 +1,491 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/debugfs.h> +#include <linux/pci.h> +#include <linux/seq_file.h> + +#include <linux/mic_common.h> +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_smpt.h" +#include "mic_virtio.h" + +/* Debugfs parent dir */ +static struct dentry *mic_dbg; + +/** + * mic_log_buf_show - Display MIC kernel log buffer. + * + * log_buf addr/len is read from System.map by user space + * and populated in sysfs entries. + */ +static int mic_log_buf_show(struct seq_file *s, void *unused) +{ + void __iomem *log_buf_va; + int __iomem *log_buf_len_va; + struct mic_device *mdev = s->private; + void *kva; + int size; + unsigned long aper_offset; + + if (!mdev || !mdev->log_buf_addr || !mdev->log_buf_len) + goto done; + /* + * Card kernel will never be relocated and any kernel text/data mapping + * can be translated to phys address by subtracting __START_KERNEL_map. + */ + aper_offset = (unsigned long)mdev->log_buf_len - __START_KERNEL_map; + log_buf_len_va = mdev->aper.va + aper_offset; + aper_offset = (unsigned long)mdev->log_buf_addr - __START_KERNEL_map; + log_buf_va = mdev->aper.va + aper_offset; + size = ioread32(log_buf_len_va); + + kva = kmalloc(size, GFP_KERNEL); + if (!kva) + goto done; + mutex_lock(&mdev->mic_mutex); + memcpy_fromio(kva, log_buf_va, size); + switch (mdev->state) { + case MIC_ONLINE: + /* Fall through */ + case MIC_SHUTTING_DOWN: + seq_write(s, kva, size); + break; + default: + break; + } + mutex_unlock(&mdev->mic_mutex); + kfree(kva); +done: + return 0; +} + +static int mic_log_buf_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_log_buf_show, inode->i_private); +} + +static int mic_log_buf_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations log_buf_ops = { + .owner = THIS_MODULE, + .open = mic_log_buf_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_log_buf_release +}; + +static int mic_smpt_show(struct seq_file *s, void *pos) +{ + int i; + struct mic_device *mdev = s->private; + unsigned long flags; + + seq_printf(s, "MIC %-2d |%-10s| %-14s %-10s\n", + mdev->id, "SMPT entry", "SW DMA addr", "RefCount"); + seq_puts(s, "====================================================\n"); + + if (mdev->smpt) { + struct mic_smpt_info *smpt_info = mdev->smpt; + spin_lock_irqsave(&smpt_info->smpt_lock, flags); + for (i = 0; i < smpt_info->info.num_reg; i++) { + seq_printf(s, "%9s|%-10d| %-#14llx %-10lld\n", + " ", i, smpt_info->entry[i].dma_addr, + smpt_info->entry[i].ref_count); + } + spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); + } + seq_puts(s, "====================================================\n"); + return 0; +} + +static int mic_smpt_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_smpt_show, inode->i_private); +} + +static int mic_smpt_debug_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations smpt_file_ops = { + .owner = THIS_MODULE, + .open = mic_smpt_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_smpt_debug_release +}; + +static int mic_soft_reset_show(struct seq_file *s, void *pos) +{ + struct mic_device *mdev = s->private; + + mic_stop(mdev, true); + return 0; +} + +static int mic_soft_reset_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_soft_reset_show, inode->i_private); +} + +static int mic_soft_reset_debug_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations soft_reset_ops = { + .owner = THIS_MODULE, + .open = mic_soft_reset_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_soft_reset_debug_release +}; + +static int mic_post_code_show(struct seq_file *s, void *pos) +{ + struct mic_device *mdev = s->private; + u32 reg = mdev->ops->get_postcode(mdev); + + seq_printf(s, "%c%c", reg & 0xff, (reg >> 8) & 0xff); + return 0; +} + +static int mic_post_code_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_post_code_show, inode->i_private); +} + +static int mic_post_code_debug_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations post_code_ops = { + .owner = THIS_MODULE, + .open = mic_post_code_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_post_code_debug_release +}; + +static int mic_dp_show(struct seq_file *s, void *pos) +{ + struct mic_device *mdev = s->private; + struct mic_device_desc *d; + struct mic_device_ctrl *dc; + struct mic_vqconfig *vqconfig; + __u32 *features; + __u8 *config; + struct mic_bootparam *bootparam = mdev->dp; + int i, j; + + seq_printf(s, "Bootparam: magic 0x%x\n", + bootparam->magic); + seq_printf(s, "Bootparam: h2c_shutdown_db %d\n", + bootparam->h2c_shutdown_db); + seq_printf(s, "Bootparam: h2c_config_db %d\n", + bootparam->h2c_config_db); + seq_printf(s, "Bootparam: c2h_shutdown_db %d\n", + bootparam->c2h_shutdown_db); + seq_printf(s, "Bootparam: shutdown_status %d\n", + bootparam->shutdown_status); + seq_printf(s, "Bootparam: shutdown_card %d\n", + bootparam->shutdown_card); + + for (i = sizeof(*bootparam); i < MIC_DP_SIZE; + i += mic_total_desc_size(d)) { + d = mdev->dp + i; + dc = (void *)d + mic_aligned_desc_size(d); + + /* end of list */ + if (d->type == 0) + break; + + if (d->type == -1) + continue; + + seq_printf(s, "Type %d ", d->type); + seq_printf(s, "Num VQ %d ", d->num_vq); + seq_printf(s, "Feature Len %d\n", d->feature_len); + seq_printf(s, "Config Len %d ", d->config_len); + seq_printf(s, "Shutdown Status %d\n", d->status); + + for (j = 0; j < d->num_vq; j++) { + vqconfig = mic_vq_config(d) + j; + seq_printf(s, "vqconfig[%d]: ", j); + seq_printf(s, "address 0x%llx ", vqconfig->address); + seq_printf(s, "num %d ", vqconfig->num); + seq_printf(s, "used address 0x%llx\n", + vqconfig->used_address); + } + + features = (__u32 *)mic_vq_features(d); + seq_printf(s, "Features: Host 0x%x ", features[0]); + seq_printf(s, "Guest 0x%x\n", features[1]); + + config = mic_vq_configspace(d); + for (j = 0; j < d->config_len; j++) + seq_printf(s, "config[%d]=%d\n", j, config[j]); + + seq_puts(s, "Device control:\n"); + seq_printf(s, "Config Change %d ", dc->config_change); + seq_printf(s, "Vdev reset %d\n", dc->vdev_reset); + seq_printf(s, "Guest Ack %d ", dc->guest_ack); + seq_printf(s, "Host ack %d\n", dc->host_ack); + seq_printf(s, "Used address updated %d ", + dc->used_address_updated); + seq_printf(s, "Vdev 0x%llx\n", dc->vdev); + seq_printf(s, "c2h doorbell %d ", dc->c2h_vdev_db); + seq_printf(s, "h2c doorbell %d\n", dc->h2c_vdev_db); + } + + return 0; +} + +static int mic_dp_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_dp_show, inode->i_private); +} + +static int mic_dp_debug_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations dp_ops = { + .owner = THIS_MODULE, + .open = mic_dp_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_dp_debug_release +}; + +static int mic_vdev_info_show(struct seq_file *s, void *unused) +{ + struct mic_device *mdev = s->private; + struct list_head *pos, *tmp; + struct mic_vdev *mvdev; + int i, j; + + mutex_lock(&mdev->mic_mutex); + list_for_each_safe(pos, tmp, &mdev->vdev_list) { + mvdev = list_entry(pos, struct mic_vdev, list); + seq_printf(s, "VDEV type %d state %s in %ld out %ld\n", + mvdev->virtio_id, + mic_vdevup(mvdev) ? "UP" : "DOWN", + mvdev->in_bytes, + mvdev->out_bytes); + for (i = 0; i < MIC_MAX_VRINGS; i++) { + struct vring_desc *desc; + struct vring_avail *avail; + struct vring_used *used; + struct mic_vringh *mvr = &mvdev->mvr[i]; + struct vringh *vrh = &mvr->vrh; + int num = vrh->vring.num; + if (!num) + continue; + desc = vrh->vring.desc; + seq_printf(s, "vring i %d avail_idx %d", + i, mvr->vring.info->avail_idx & (num - 1)); + seq_printf(s, " vring i %d avail_idx %d\n", + i, mvr->vring.info->avail_idx); + seq_printf(s, "vrh i %d weak_barriers %d", + i, vrh->weak_barriers); + seq_printf(s, " last_avail_idx %d last_used_idx %d", + vrh->last_avail_idx, vrh->last_used_idx); + seq_printf(s, " completed %d\n", vrh->completed); + for (j = 0; j < num; j++) { + seq_printf(s, "desc[%d] addr 0x%llx len %d", + j, desc->addr, desc->len); + seq_printf(s, " flags 0x%x next %d\n", + desc->flags, desc->next); + desc++; + } + avail = vrh->vring.avail; + seq_printf(s, "avail flags 0x%x idx %d\n", + avail->flags, avail->idx & (num - 1)); + seq_printf(s, "avail flags 0x%x idx %d\n", + avail->flags, avail->idx); + for (j = 0; j < num; j++) + seq_printf(s, "avail ring[%d] %d\n", + j, avail->ring[j]); + used = vrh->vring.used; + seq_printf(s, "used flags 0x%x idx %d\n", + used->flags, used->idx & (num - 1)); + seq_printf(s, "used flags 0x%x idx %d\n", + used->flags, used->idx); + for (j = 0; j < num; j++) + seq_printf(s, "used ring[%d] id %d len %d\n", + j, used->ring[j].id, + used->ring[j].len); + } + } + mutex_unlock(&mdev->mic_mutex); + + return 0; +} + +static int mic_vdev_info_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_vdev_info_show, inode->i_private); +} + +static int mic_vdev_info_debug_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations vdev_info_ops = { + .owner = THIS_MODULE, + .open = mic_vdev_info_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_vdev_info_debug_release +}; + +static int mic_msi_irq_info_show(struct seq_file *s, void *pos) +{ + struct mic_device *mdev = s->private; + int reg; + int i, j; + u16 entry; + u16 vector; + struct pci_dev *pdev = container_of(mdev->sdev->parent, + struct pci_dev, dev); + + if (pci_dev_msi_enabled(pdev)) { + for (i = 0; i < mdev->irq_info.num_vectors; i++) { + if (pdev->msix_enabled) { + entry = mdev->irq_info.msix_entries[i].entry; + vector = mdev->irq_info.msix_entries[i].vector; + } else { + entry = 0; + vector = pdev->irq; + } + + reg = mdev->intr_ops->read_msi_to_src_map(mdev, entry); + + seq_printf(s, "%s %-10d %s %-10d MXAR[%d]: %08X\n", + "IRQ:", vector, "Entry:", entry, i, reg); + + seq_printf(s, "%-10s", "offset:"); + for (j = (MIC_NUM_OFFSETS - 1); j >= 0; j--) + seq_printf(s, "%4d ", j); + seq_puts(s, "\n"); + + + seq_printf(s, "%-10s", "count:"); + for (j = (MIC_NUM_OFFSETS - 1); j >= 0; j--) + seq_printf(s, "%4d ", + (mdev->irq_info.mic_msi_map[i] & + BIT(j)) ? 1 : 0); + seq_puts(s, "\n\n"); + } + } else { + seq_puts(s, "MSI/MSIx interrupts not enabled\n"); + } + + return 0; +} + +static int mic_msi_irq_info_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, mic_msi_irq_info_show, inode->i_private); +} + +static int +mic_msi_irq_info_debug_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations msi_irq_info_ops = { + .owner = THIS_MODULE, + .open = mic_msi_irq_info_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = mic_msi_irq_info_debug_release +}; + +/** + * mic_create_debug_dir - Initialize MIC debugfs entries. + */ +void mic_create_debug_dir(struct mic_device *mdev) +{ + if (!mic_dbg) + return; + + mdev->dbg_dir = debugfs_create_dir(dev_name(mdev->sdev), mic_dbg); + if (!mdev->dbg_dir) + return; + + debugfs_create_file("log_buf", 0444, mdev->dbg_dir, mdev, &log_buf_ops); + + debugfs_create_file("smpt", 0444, mdev->dbg_dir, mdev, &smpt_file_ops); + + debugfs_create_file("soft_reset", 0444, mdev->dbg_dir, mdev, + &soft_reset_ops); + + debugfs_create_file("post_code", 0444, mdev->dbg_dir, mdev, + &post_code_ops); + + debugfs_create_file("dp", 0444, mdev->dbg_dir, mdev, &dp_ops); + + debugfs_create_file("vdev_info", 0444, mdev->dbg_dir, mdev, + &vdev_info_ops); + + debugfs_create_file("msi_irq_info", 0444, mdev->dbg_dir, mdev, + &msi_irq_info_ops); +} + +/** + * mic_delete_debug_dir - Uninitialize MIC debugfs entries. + */ +void mic_delete_debug_dir(struct mic_device *mdev) +{ + if (!mdev->dbg_dir) + return; + + debugfs_remove_recursive(mdev->dbg_dir); +} + +/** + * mic_init_debugfs - Initialize global debugfs entry. + */ +void __init mic_init_debugfs(void) +{ + mic_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); + if (!mic_dbg) + pr_err("can't create debugfs dir\n"); +} + +/** + * mic_exit_debugfs - Uninitialize global debugfs entry + */ +void mic_exit_debugfs(void) +{ + debugfs_remove(mic_dbg); +} diff --git a/drivers/misc/mic/host/mic_device.h b/drivers/misc/mic/host/mic_device.h new file mode 100644 index 00000000000..0398c696d25 --- /dev/null +++ b/drivers/misc/mic/host/mic_device.h @@ -0,0 +1,207 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#ifndef _MIC_DEVICE_H_ +#define _MIC_DEVICE_H_ + +#include <linux/cdev.h> +#include <linux/idr.h> +#include <linux/notifier.h> +#include <linux/irqreturn.h> + +#include "mic_intr.h" + +/* The maximum number of MIC devices supported in a single host system. */ +#define MIC_MAX_NUM_DEVS 256 + +/** + * enum mic_hw_family - The hardware family to which a device belongs. + */ +enum mic_hw_family { + MIC_FAMILY_X100 = 0, + MIC_FAMILY_UNKNOWN +}; + +/** + * enum mic_stepping - MIC stepping ids. + */ +enum mic_stepping { + MIC_A0_STEP = 0x0, + MIC_B0_STEP = 0x10, + MIC_B1_STEP = 0x11, + MIC_C0_STEP = 0x20, +}; + +/** + * struct mic_device - MIC device information for each card. + * + * @mmio: MMIO bar information. + * @aper: Aperture bar information. + * @family: The MIC family to which this device belongs. + * @ops: MIC HW specific operations. + * @id: The unique device id for this MIC device. + * @stepping: Stepping ID. + * @attr_group: Pointer to list of sysfs attribute groups. + * @sdev: Device for sysfs entries. + * @mic_mutex: Mutex for synchronizing access to mic_device. + * @intr_ops: HW specific interrupt operations. + * @smpt_ops: Hardware specific SMPT operations. + * @smpt: MIC SMPT information. + * @intr_info: H/W specific interrupt information. + * @irq_info: The OS specific irq information + * @dbg_dir: debugfs directory of this MIC device. + * @cmdline: Kernel command line. + * @firmware: Firmware file name. + * @ramdisk: Ramdisk file name. + * @bootmode: Boot mode i.e. "linux" or "elf" for flash updates. + * @bootaddr: MIC boot address. + * @reset_trigger_work: Work for triggering reset requests. + * @shutdown_work: Work for handling shutdown interrupts. + * @state: MIC state. + * @shutdown_status: MIC status reported by card for shutdown/crashes. + * @state_sysfs: Sysfs dirent for notifying ring 3 about MIC state changes. + * @reset_wait: Waitqueue for sleeping while reset completes. + * @log_buf_addr: Log buffer address for MIC. + * @log_buf_len: Log buffer length address for MIC. + * @dp: virtio device page + * @dp_dma_addr: virtio device page DMA address. + * @shutdown_db: shutdown doorbell. + * @shutdown_cookie: shutdown cookie. + * @cdev: Character device for MIC. + * @vdev_list: list of virtio devices. + * @pm_notifier: Handles PM notifications from the OS. + */ +struct mic_device { + struct mic_mw mmio; + struct mic_mw aper; + enum mic_hw_family family; + struct mic_hw_ops *ops; + int id; + enum mic_stepping stepping; + const struct attribute_group **attr_group; + struct device *sdev; + struct mutex mic_mutex; + struct mic_hw_intr_ops *intr_ops; + struct mic_smpt_ops *smpt_ops; + struct mic_smpt_info *smpt; + struct mic_intr_info *intr_info; + struct mic_irq_info irq_info; + struct dentry *dbg_dir; + char *cmdline; + char *firmware; + char *ramdisk; + char *bootmode; + u32 bootaddr; + struct work_struct reset_trigger_work; + struct work_struct shutdown_work; + u8 state; + u8 shutdown_status; + struct kernfs_node *state_sysfs; + struct completion reset_wait; + void *log_buf_addr; + int *log_buf_len; + void *dp; + dma_addr_t dp_dma_addr; + int shutdown_db; + struct mic_irq *shutdown_cookie; + struct cdev cdev; + struct list_head vdev_list; + struct notifier_block pm_notifier; +}; + +/** + * struct mic_hw_ops - MIC HW specific operations. + * @aper_bar: Aperture bar resource number. + * @mmio_bar: MMIO bar resource number. + * @read_spad: Read from scratch pad register. + * @write_spad: Write to scratch pad register. + * @send_intr: Send an interrupt for a particular doorbell on the card. + * @ack_interrupt: Hardware specific operations to ack the h/w on + * receipt of an interrupt. + * @intr_workarounds: Hardware specific workarounds needed after + * handling an interrupt. + * @reset: Reset the remote processor. + * @reset_fw_ready: Reset firmware ready field. + * @is_fw_ready: Check if firmware is ready for OS download. + * @send_firmware_intr: Send an interrupt to the card firmware. + * @load_mic_fw: Load firmware segments required to boot the card + * into card memory. This includes the kernel, command line, ramdisk etc. + * @get_postcode: Get post code status from firmware. + */ +struct mic_hw_ops { + u8 aper_bar; + u8 mmio_bar; + u32 (*read_spad)(struct mic_device *mdev, unsigned int idx); + void (*write_spad)(struct mic_device *mdev, unsigned int idx, u32 val); + void (*send_intr)(struct mic_device *mdev, int doorbell); + u32 (*ack_interrupt)(struct mic_device *mdev); + void (*intr_workarounds)(struct mic_device *mdev); + void (*reset)(struct mic_device *mdev); + void (*reset_fw_ready)(struct mic_device *mdev); + bool (*is_fw_ready)(struct mic_device *mdev); + void (*send_firmware_intr)(struct mic_device *mdev); + int (*load_mic_fw)(struct mic_device *mdev, const char *buf); + u32 (*get_postcode)(struct mic_device *mdev); +}; + +/** + * mic_mmio_read - read from an MMIO register. + * @mw: MMIO register base virtual address. + * @offset: register offset. + * + * RETURNS: register value. + */ +static inline u32 mic_mmio_read(struct mic_mw *mw, u32 offset) +{ + return ioread32(mw->va + offset); +} + +/** + * mic_mmio_write - write to an MMIO register. + * @mw: MMIO register base virtual address. + * @val: the data value to put into the register + * @offset: register offset. + * + * RETURNS: none. + */ +static inline void +mic_mmio_write(struct mic_mw *mw, u32 val, u32 offset) +{ + iowrite32(val, mw->va + offset); +} + +void mic_sysfs_init(struct mic_device *mdev); +int mic_start(struct mic_device *mdev, const char *buf); +void mic_stop(struct mic_device *mdev, bool force); +void mic_shutdown(struct mic_device *mdev); +void mic_reset_delayed_work(struct work_struct *work); +void mic_reset_trigger_work(struct work_struct *work); +void mic_shutdown_work(struct work_struct *work); +void mic_bootparam_init(struct mic_device *mdev); +void mic_set_state(struct mic_device *mdev, u8 state); +void mic_set_shutdown_status(struct mic_device *mdev, u8 status); +void mic_create_debug_dir(struct mic_device *dev); +void mic_delete_debug_dir(struct mic_device *dev); +void __init mic_init_debugfs(void); +void mic_exit_debugfs(void); +void mic_prepare_suspend(struct mic_device *mdev); +void mic_complete_resume(struct mic_device *mdev); +void mic_suspend(struct mic_device *mdev); +#endif diff --git a/drivers/misc/mic/host/mic_fops.c b/drivers/misc/mic/host/mic_fops.c new file mode 100644 index 00000000000..85776d7327f --- /dev/null +++ b/drivers/misc/mic/host/mic_fops.c @@ -0,0 +1,222 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/poll.h> +#include <linux/pci.h> + +#include <linux/mic_common.h> +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_fops.h" +#include "mic_virtio.h" + +int mic_open(struct inode *inode, struct file *f) +{ + struct mic_vdev *mvdev; + struct mic_device *mdev = container_of(inode->i_cdev, + struct mic_device, cdev); + + mvdev = kzalloc(sizeof(*mvdev), GFP_KERNEL); + if (!mvdev) + return -ENOMEM; + + init_waitqueue_head(&mvdev->waitq); + INIT_LIST_HEAD(&mvdev->list); + mvdev->mdev = mdev; + mvdev->virtio_id = -1; + + f->private_data = mvdev; + return 0; +} + +int mic_release(struct inode *inode, struct file *f) +{ + struct mic_vdev *mvdev = (struct mic_vdev *)f->private_data; + + if (-1 != mvdev->virtio_id) + mic_virtio_del_device(mvdev); + f->private_data = NULL; + kfree(mvdev); + return 0; +} + +long mic_ioctl(struct file *f, unsigned int cmd, unsigned long arg) +{ + struct mic_vdev *mvdev = (struct mic_vdev *)f->private_data; + void __user *argp = (void __user *)arg; + int ret; + + switch (cmd) { + case MIC_VIRTIO_ADD_DEVICE: + { + ret = mic_virtio_add_device(mvdev, argp); + if (ret < 0) { + dev_err(mic_dev(mvdev), + "%s %d errno ret %d\n", + __func__, __LINE__, ret); + return ret; + } + break; + } + case MIC_VIRTIO_COPY_DESC: + { + struct mic_copy_desc copy; + + ret = mic_vdev_inited(mvdev); + if (ret) + return ret; + + if (copy_from_user(©, argp, sizeof(copy))) + return -EFAULT; + + dev_dbg(mic_dev(mvdev), + "%s %d === iovcnt 0x%x vr_idx 0x%x update_used %d\n", + __func__, __LINE__, copy.iovcnt, copy.vr_idx, + copy.update_used); + + ret = mic_virtio_copy_desc(mvdev, ©); + if (ret < 0) { + dev_err(mic_dev(mvdev), + "%s %d errno ret %d\n", + __func__, __LINE__, ret); + return ret; + } + if (copy_to_user( + &((struct mic_copy_desc __user *)argp)->out_len, + ©.out_len, sizeof(copy.out_len))) { + dev_err(mic_dev(mvdev), "%s %d errno ret %d\n", + __func__, __LINE__, -EFAULT); + return -EFAULT; + } + break; + } + case MIC_VIRTIO_CONFIG_CHANGE: + { + ret = mic_vdev_inited(mvdev); + if (ret) + return ret; + + ret = mic_virtio_config_change(mvdev, argp); + if (ret < 0) { + dev_err(mic_dev(mvdev), + "%s %d errno ret %d\n", + __func__, __LINE__, ret); + return ret; + } + break; + } + default: + return -ENOIOCTLCMD; + }; + return 0; +} + +/* + * We return POLLIN | POLLOUT from poll when new buffers are enqueued, and + * not when previously enqueued buffers may be available. This means that + * in the card->host (TX) path, when userspace is unblocked by poll it + * must drain all available descriptors or it can stall. + */ +unsigned int mic_poll(struct file *f, poll_table *wait) +{ + struct mic_vdev *mvdev = (struct mic_vdev *)f->private_data; + int mask = 0; + + poll_wait(f, &mvdev->waitq, wait); + + if (mic_vdev_inited(mvdev)) { + mask = POLLERR; + } else if (mvdev->poll_wake) { + mvdev->poll_wake = 0; + mask = POLLIN | POLLOUT; + } + + return mask; +} + +static inline int +mic_query_offset(struct mic_vdev *mvdev, unsigned long offset, + unsigned long *size, unsigned long *pa) +{ + struct mic_device *mdev = mvdev->mdev; + unsigned long start = MIC_DP_SIZE; + int i; + + /* + * MMAP interface is as follows: + * offset region + * 0x0 virtio device_page + * 0x1000 first vring + * 0x1000 + size of 1st vring second vring + * .... + */ + if (!offset) { + *pa = virt_to_phys(mdev->dp); + *size = MIC_DP_SIZE; + return 0; + } + + for (i = 0; i < mvdev->dd->num_vq; i++) { + struct mic_vringh *mvr = &mvdev->mvr[i]; + if (offset == start) { + *pa = virt_to_phys(mvr->vring.va); + *size = mvr->vring.len; + return 0; + } + start += mvr->vring.len; + } + return -1; +} + +/* + * Maps the device page and virtio rings to user space for readonly access. + */ +int +mic_mmap(struct file *f, struct vm_area_struct *vma) +{ + struct mic_vdev *mvdev = (struct mic_vdev *)f->private_data; + unsigned long offset = vma->vm_pgoff << PAGE_SHIFT; + unsigned long pa, size = vma->vm_end - vma->vm_start, size_rem = size; + int i, err; + + err = mic_vdev_inited(mvdev); + if (err) + return err; + + if (vma->vm_flags & VM_WRITE) + return -EACCES; + + while (size_rem) { + i = mic_query_offset(mvdev, offset, &size, &pa); + if (i < 0) + return -EINVAL; + err = remap_pfn_range(vma, vma->vm_start + offset, + pa >> PAGE_SHIFT, size, vma->vm_page_prot); + if (err) + return err; + dev_dbg(mic_dev(mvdev), + "%s %d type %d size 0x%lx off 0x%lx pa 0x%lx vma 0x%lx\n", + __func__, __LINE__, mvdev->virtio_id, size, offset, + pa, vma->vm_start + offset); + size_rem -= size; + offset += size; + } + return 0; +} diff --git a/drivers/misc/mic/host/mic_fops.h b/drivers/misc/mic/host/mic_fops.h new file mode 100644 index 00000000000..dc3893dff66 --- /dev/null +++ b/drivers/misc/mic/host/mic_fops.h @@ -0,0 +1,32 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#ifndef _MIC_FOPS_H_ +#define _MIC_FOPS_H_ + +int mic_open(struct inode *inode, struct file *filp); +int mic_release(struct inode *inode, struct file *filp); +ssize_t mic_read(struct file *filp, char __user *buf, + size_t count, loff_t *pos); +long mic_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); +int mic_mmap(struct file *f, struct vm_area_struct *vma); +unsigned int mic_poll(struct file *f, poll_table *wait); + +#endif diff --git a/drivers/misc/mic/host/mic_intr.c b/drivers/misc/mic/host/mic_intr.c new file mode 100644 index 00000000000..dbc5afde139 --- /dev/null +++ b/drivers/misc/mic/host/mic_intr.c @@ -0,0 +1,630 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/pci.h> +#include <linux/interrupt.h> + +#include "../common/mic_dev.h" +#include "mic_device.h" + +/* + * mic_invoke_callback - Invoke callback functions registered for + * the corresponding source id. + * + * @mdev: pointer to the mic_device instance + * @idx: The interrupt source id. + * + * Returns none. + */ +static inline void mic_invoke_callback(struct mic_device *mdev, int idx) +{ + struct mic_intr_cb *intr_cb; + struct pci_dev *pdev = container_of(mdev->sdev->parent, + struct pci_dev, dev); + + spin_lock(&mdev->irq_info.mic_intr_lock); + list_for_each_entry(intr_cb, &mdev->irq_info.cb_list[idx], list) + if (intr_cb->func) + intr_cb->func(pdev->irq, intr_cb->data); + spin_unlock(&mdev->irq_info.mic_intr_lock); +} + +/** + * mic_interrupt - Generic interrupt handler for + * MSI and INTx based interrupts. + */ +static irqreturn_t mic_interrupt(int irq, void *dev) +{ + struct mic_device *mdev = dev; + struct mic_intr_info *info = mdev->intr_info; + u32 mask; + int i; + + mask = mdev->ops->ack_interrupt(mdev); + if (!mask) + return IRQ_NONE; + + for (i = info->intr_start_idx[MIC_INTR_DB]; + i < info->intr_len[MIC_INTR_DB]; i++) + if (mask & BIT(i)) + mic_invoke_callback(mdev, i); + + return IRQ_HANDLED; +} + +/* Return the interrupt offset from the index. Index is 0 based. */ +static u16 mic_map_src_to_offset(struct mic_device *mdev, + int intr_src, enum mic_intr_type type) +{ + if (type >= MIC_NUM_INTR_TYPES) + return MIC_NUM_OFFSETS; + if (intr_src >= mdev->intr_info->intr_len[type]) + return MIC_NUM_OFFSETS; + + return mdev->intr_info->intr_start_idx[type] + intr_src; +} + +/* Return next available msix_entry. */ +static struct msix_entry *mic_get_available_vector(struct mic_device *mdev) +{ + int i; + struct mic_irq_info *info = &mdev->irq_info; + + for (i = 0; i < info->num_vectors; i++) + if (!info->mic_msi_map[i]) + return &info->msix_entries[i]; + return NULL; +} + +/** + * mic_register_intr_callback - Register a callback handler for the + * given source id. + * + * @mdev: pointer to the mic_device instance + * @idx: The source id to be registered. + * @func: The function to be called when the source id receives + * the interrupt. + * @data: Private data of the requester. + * Return the callback structure that was registered or an + * appropriate error on failure. + */ +static struct mic_intr_cb *mic_register_intr_callback(struct mic_device *mdev, + u8 idx, irqreturn_t (*func) (int irq, void *dev), + void *data) +{ + struct mic_intr_cb *intr_cb; + unsigned long flags; + int rc; + intr_cb = kmalloc(sizeof(*intr_cb), GFP_KERNEL); + + if (!intr_cb) + return ERR_PTR(-ENOMEM); + + intr_cb->func = func; + intr_cb->data = data; + intr_cb->cb_id = ida_simple_get(&mdev->irq_info.cb_ida, + 0, 0, GFP_KERNEL); + if (intr_cb->cb_id < 0) { + rc = intr_cb->cb_id; + goto ida_fail; + } + + spin_lock_irqsave(&mdev->irq_info.mic_intr_lock, flags); + list_add_tail(&intr_cb->list, &mdev->irq_info.cb_list[idx]); + spin_unlock_irqrestore(&mdev->irq_info.mic_intr_lock, flags); + + return intr_cb; +ida_fail: + kfree(intr_cb); + return ERR_PTR(rc); +} + +/** + * mic_unregister_intr_callback - Unregister the callback handler + * identified by its callback id. + * + * @mdev: pointer to the mic_device instance + * @idx: The callback structure id to be unregistered. + * Return the source id that was unregistered or MIC_NUM_OFFSETS if no + * such callback handler was found. + */ +static u8 mic_unregister_intr_callback(struct mic_device *mdev, u32 idx) +{ + struct list_head *pos, *tmp; + struct mic_intr_cb *intr_cb; + unsigned long flags; + int i; + + for (i = 0; i < MIC_NUM_OFFSETS; i++) { + spin_lock_irqsave(&mdev->irq_info.mic_intr_lock, flags); + list_for_each_safe(pos, tmp, &mdev->irq_info.cb_list[i]) { + intr_cb = list_entry(pos, struct mic_intr_cb, list); + if (intr_cb->cb_id == idx) { + list_del(pos); + ida_simple_remove(&mdev->irq_info.cb_ida, + intr_cb->cb_id); + kfree(intr_cb); + spin_unlock_irqrestore( + &mdev->irq_info.mic_intr_lock, flags); + return i; + } + } + spin_unlock_irqrestore(&mdev->irq_info.mic_intr_lock, flags); + } + return MIC_NUM_OFFSETS; +} + +/** + * mic_setup_msix - Initializes MSIx interrupts. + * + * @mdev: pointer to mic_device instance + * + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static int mic_setup_msix(struct mic_device *mdev, struct pci_dev *pdev) +{ + int rc, i; + int entry_size = sizeof(*mdev->irq_info.msix_entries); + + mdev->irq_info.msix_entries = kmalloc_array(MIC_MIN_MSIX, + entry_size, GFP_KERNEL); + if (!mdev->irq_info.msix_entries) { + rc = -ENOMEM; + goto err_nomem1; + } + + for (i = 0; i < MIC_MIN_MSIX; i++) + mdev->irq_info.msix_entries[i].entry = i; + + rc = pci_enable_msix_exact(pdev, mdev->irq_info.msix_entries, + MIC_MIN_MSIX); + if (rc) { + dev_dbg(&pdev->dev, "Error enabling MSIx. rc = %d\n", rc); + goto err_enable_msix; + } + + mdev->irq_info.num_vectors = MIC_MIN_MSIX; + mdev->irq_info.mic_msi_map = kzalloc((sizeof(u32) * + mdev->irq_info.num_vectors), GFP_KERNEL); + + if (!mdev->irq_info.mic_msi_map) { + rc = -ENOMEM; + goto err_nomem2; + } + + dev_dbg(mdev->sdev->parent, + "%d MSIx irqs setup\n", mdev->irq_info.num_vectors); + return 0; +err_nomem2: + pci_disable_msix(pdev); +err_enable_msix: + kfree(mdev->irq_info.msix_entries); +err_nomem1: + mdev->irq_info.num_vectors = 0; + return rc; +} + +/** + * mic_setup_callbacks - Initialize data structures needed + * to handle callbacks. + * + * @mdev: pointer to mic_device instance + */ +static int mic_setup_callbacks(struct mic_device *mdev) +{ + int i; + + mdev->irq_info.cb_list = kmalloc_array(MIC_NUM_OFFSETS, + sizeof(*mdev->irq_info.cb_list), + GFP_KERNEL); + if (!mdev->irq_info.cb_list) + return -ENOMEM; + + for (i = 0; i < MIC_NUM_OFFSETS; i++) + INIT_LIST_HEAD(&mdev->irq_info.cb_list[i]); + ida_init(&mdev->irq_info.cb_ida); + spin_lock_init(&mdev->irq_info.mic_intr_lock); + return 0; +} + +/** + * mic_release_callbacks - Uninitialize data structures needed + * to handle callbacks. + * + * @mdev: pointer to mic_device instance + */ +static void mic_release_callbacks(struct mic_device *mdev) +{ + unsigned long flags; + struct list_head *pos, *tmp; + struct mic_intr_cb *intr_cb; + int i; + + for (i = 0; i < MIC_NUM_OFFSETS; i++) { + spin_lock_irqsave(&mdev->irq_info.mic_intr_lock, flags); + + if (list_empty(&mdev->irq_info.cb_list[i])) { + spin_unlock_irqrestore(&mdev->irq_info.mic_intr_lock, + flags); + break; + } + + list_for_each_safe(pos, tmp, &mdev->irq_info.cb_list[i]) { + intr_cb = list_entry(pos, struct mic_intr_cb, list); + list_del(pos); + ida_simple_remove(&mdev->irq_info.cb_ida, + intr_cb->cb_id); + kfree(intr_cb); + } + spin_unlock_irqrestore(&mdev->irq_info.mic_intr_lock, flags); + } + ida_destroy(&mdev->irq_info.cb_ida); + kfree(mdev->irq_info.cb_list); +} + +/** + * mic_setup_msi - Initializes MSI interrupts. + * + * @mdev: pointer to mic_device instance + * @pdev: PCI device structure + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static int mic_setup_msi(struct mic_device *mdev, struct pci_dev *pdev) +{ + int rc; + + rc = pci_enable_msi(pdev); + if (rc) { + dev_dbg(&pdev->dev, "Error enabling MSI. rc = %d\n", rc); + return rc; + } + + mdev->irq_info.num_vectors = 1; + mdev->irq_info.mic_msi_map = kzalloc((sizeof(u32) * + mdev->irq_info.num_vectors), GFP_KERNEL); + + if (!mdev->irq_info.mic_msi_map) { + rc = -ENOMEM; + goto err_nomem1; + } + + rc = mic_setup_callbacks(mdev); + if (rc) { + dev_err(&pdev->dev, "Error setting up callbacks\n"); + goto err_nomem2; + } + + rc = request_irq(pdev->irq, mic_interrupt, 0 , "mic-msi", mdev); + if (rc) { + dev_err(&pdev->dev, "Error allocating MSI interrupt\n"); + goto err_irq_req_fail; + } + + dev_dbg(&pdev->dev, "%d MSI irqs setup\n", mdev->irq_info.num_vectors); + return 0; +err_irq_req_fail: + mic_release_callbacks(mdev); +err_nomem2: + kfree(mdev->irq_info.mic_msi_map); +err_nomem1: + pci_disable_msi(pdev); + mdev->irq_info.num_vectors = 0; + return rc; +} + +/** + * mic_setup_intx - Initializes legacy interrupts. + * + * @mdev: pointer to mic_device instance + * @pdev: PCI device structure + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static int mic_setup_intx(struct mic_device *mdev, struct pci_dev *pdev) +{ + int rc; + + pci_msi_off(pdev); + + /* Enable intx */ + pci_intx(pdev, 1); + rc = mic_setup_callbacks(mdev); + if (rc) { + dev_err(&pdev->dev, "Error setting up callbacks\n"); + goto err_nomem; + } + + rc = request_irq(pdev->irq, mic_interrupt, + IRQF_SHARED, "mic-intx", mdev); + if (rc) + goto err; + + dev_dbg(&pdev->dev, "intx irq setup\n"); + return 0; +err: + mic_release_callbacks(mdev); +err_nomem: + return rc; +} + +/** + * mic_next_db - Retrieve the next doorbell interrupt source id. + * The id is picked sequentially from the available pool of + * doorlbell ids. + * + * @mdev: pointer to the mic_device instance. + * + * Returns the next doorbell interrupt source. + */ +int mic_next_db(struct mic_device *mdev) +{ + int next_db; + + next_db = mdev->irq_info.next_avail_src % + mdev->intr_info->intr_len[MIC_INTR_DB]; + mdev->irq_info.next_avail_src++; + return next_db; +} + +#define COOKIE_ID_SHIFT 16 +#define GET_ENTRY(cookie) ((cookie) & 0xFFFF) +#define GET_OFFSET(cookie) ((cookie) >> COOKIE_ID_SHIFT) +#define MK_COOKIE(x, y) ((x) | (y) << COOKIE_ID_SHIFT) + +/** + * mic_request_irq - request an irq. mic_mutex needs + * to be held before calling this function. + * + * @mdev: pointer to mic_device instance + * @func: The callback function that handles the interrupt. + * The function needs to call ack_interrupts + * (mdev->ops->ack_interrupt(mdev)) when handling the interrupts. + * @name: The ASCII name of the callee requesting the irq. + * @data: private data that is returned back when calling the + * function handler. + * @intr_src: The source id of the requester. Its the doorbell id + * for Doorbell interrupts and DMA channel id for DMA interrupts. + * @type: The type of interrupt. Values defined in mic_intr_type + * + * returns: The cookie that is transparent to the caller. Passed + * back when calling mic_free_irq. An appropriate error code + * is returned on failure. Caller needs to use IS_ERR(return_val) + * to check for failure and PTR_ERR(return_val) to obtained the + * error code. + * + */ +struct mic_irq *mic_request_irq(struct mic_device *mdev, + irqreturn_t (*func)(int irq, void *dev), + const char *name, void *data, int intr_src, + enum mic_intr_type type) +{ + u16 offset; + int rc = 0; + struct msix_entry *msix = NULL; + unsigned long cookie = 0; + u16 entry; + struct mic_intr_cb *intr_cb; + struct pci_dev *pdev = container_of(mdev->sdev->parent, + struct pci_dev, dev); + + offset = mic_map_src_to_offset(mdev, intr_src, type); + if (offset >= MIC_NUM_OFFSETS) { + dev_err(mdev->sdev->parent, + "Error mapping index %d to a valid source id.\n", + intr_src); + rc = -EINVAL; + goto err; + } + + if (mdev->irq_info.num_vectors > 1) { + msix = mic_get_available_vector(mdev); + if (!msix) { + dev_err(mdev->sdev->parent, + "No MSIx vectors available for use.\n"); + rc = -ENOSPC; + goto err; + } + + rc = request_irq(msix->vector, func, 0, name, data); + if (rc) { + dev_dbg(mdev->sdev->parent, + "request irq failed rc = %d\n", rc); + goto err; + } + entry = msix->entry; + mdev->irq_info.mic_msi_map[entry] |= BIT(offset); + mdev->intr_ops->program_msi_to_src_map(mdev, + entry, offset, true); + cookie = MK_COOKIE(entry, offset); + dev_dbg(mdev->sdev->parent, "irq: %d assigned for src: %d\n", + msix->vector, intr_src); + } else { + intr_cb = mic_register_intr_callback(mdev, + offset, func, data); + if (IS_ERR(intr_cb)) { + dev_err(mdev->sdev->parent, + "No available callback entries for use\n"); + rc = PTR_ERR(intr_cb); + goto err; + } + + entry = 0; + if (pci_dev_msi_enabled(pdev)) { + mdev->irq_info.mic_msi_map[entry] |= (1 << offset); + mdev->intr_ops->program_msi_to_src_map(mdev, + entry, offset, true); + } + cookie = MK_COOKIE(entry, intr_cb->cb_id); + dev_dbg(mdev->sdev->parent, "callback %d registered for src: %d\n", + intr_cb->cb_id, intr_src); + } + return (struct mic_irq *)cookie; +err: + return ERR_PTR(rc); +} + +/** + * mic_free_irq - free irq. mic_mutex + * needs to be held before calling this function. + * + * @mdev: pointer to mic_device instance + * @cookie: cookie obtained during a successful call to mic_request_irq + * @data: private data specified by the calling function during the + * mic_request_irq + * + * returns: none. + */ +void mic_free_irq(struct mic_device *mdev, + struct mic_irq *cookie, void *data) +{ + u32 offset; + u32 entry; + u8 src_id; + unsigned int irq; + struct pci_dev *pdev = container_of(mdev->sdev->parent, + struct pci_dev, dev); + + entry = GET_ENTRY((unsigned long)cookie); + offset = GET_OFFSET((unsigned long)cookie); + if (mdev->irq_info.num_vectors > 1) { + if (entry >= mdev->irq_info.num_vectors) { + dev_warn(mdev->sdev->parent, + "entry %d should be < num_irq %d\n", + entry, mdev->irq_info.num_vectors); + return; + } + irq = mdev->irq_info.msix_entries[entry].vector; + free_irq(irq, data); + mdev->irq_info.mic_msi_map[entry] &= ~(BIT(offset)); + mdev->intr_ops->program_msi_to_src_map(mdev, + entry, offset, false); + + dev_dbg(mdev->sdev->parent, "irq: %d freed\n", irq); + } else { + irq = pdev->irq; + src_id = mic_unregister_intr_callback(mdev, offset); + if (src_id >= MIC_NUM_OFFSETS) { + dev_warn(mdev->sdev->parent, "Error unregistering callback\n"); + return; + } + if (pci_dev_msi_enabled(pdev)) { + mdev->irq_info.mic_msi_map[entry] &= ~(BIT(src_id)); + mdev->intr_ops->program_msi_to_src_map(mdev, + entry, src_id, false); + } + dev_dbg(mdev->sdev->parent, "callback %d unregistered for src: %d\n", + offset, src_id); + } +} + +/** + * mic_setup_interrupts - Initializes interrupts. + * + * @mdev: pointer to mic_device instance + * @pdev: PCI device structure + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +int mic_setup_interrupts(struct mic_device *mdev, struct pci_dev *pdev) +{ + int rc; + + rc = mic_setup_msix(mdev, pdev); + if (!rc) + goto done; + + rc = mic_setup_msi(mdev, pdev); + if (!rc) + goto done; + + rc = mic_setup_intx(mdev, pdev); + if (rc) { + dev_err(mdev->sdev->parent, "no usable interrupts\n"); + return rc; + } +done: + mdev->intr_ops->enable_interrupts(mdev); + return 0; +} + +/** + * mic_free_interrupts - Frees interrupts setup by mic_setup_interrupts + * + * @mdev: pointer to mic_device instance + * @pdev: PCI device structure + * + * returns none. + */ +void mic_free_interrupts(struct mic_device *mdev, struct pci_dev *pdev) +{ + int i; + + mdev->intr_ops->disable_interrupts(mdev); + if (mdev->irq_info.num_vectors > 1) { + for (i = 0; i < mdev->irq_info.num_vectors; i++) { + if (mdev->irq_info.mic_msi_map[i]) + dev_warn(&pdev->dev, "irq %d may still be in use.\n", + mdev->irq_info.msix_entries[i].vector); + } + kfree(mdev->irq_info.mic_msi_map); + kfree(mdev->irq_info.msix_entries); + pci_disable_msix(pdev); + } else { + if (pci_dev_msi_enabled(pdev)) { + free_irq(pdev->irq, mdev); + kfree(mdev->irq_info.mic_msi_map); + pci_disable_msi(pdev); + } else { + free_irq(pdev->irq, mdev); + } + mic_release_callbacks(mdev); + } +} + +/** + * mic_intr_restore - Restore MIC interrupt registers. + * + * @mdev: pointer to mic_device instance. + * + * Restore the interrupt registers to values previously + * stored in the SW data structures. mic_mutex needs to + * be held before calling this function. + * + * returns None. + */ +void mic_intr_restore(struct mic_device *mdev) +{ + int entry, offset; + struct pci_dev *pdev = container_of(mdev->sdev->parent, + struct pci_dev, dev); + + if (!pci_dev_msi_enabled(pdev)) + return; + + for (entry = 0; entry < mdev->irq_info.num_vectors; entry++) { + for (offset = 0; offset < MIC_NUM_OFFSETS; offset++) { + if (mdev->irq_info.mic_msi_map[entry] & BIT(offset)) + mdev->intr_ops->program_msi_to_src_map(mdev, + entry, offset, true); + } + } +} diff --git a/drivers/misc/mic/host/mic_intr.h b/drivers/misc/mic/host/mic_intr.h new file mode 100644 index 00000000000..6091aa97e11 --- /dev/null +++ b/drivers/misc/mic/host/mic_intr.h @@ -0,0 +1,137 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#ifndef _MIC_INTR_H_ +#define _MIC_INTR_H_ + +/* + * The minimum number of msix vectors required for normal operation. + * 3 for virtio network, console and block devices. + * 1 for card shutdown notifications. + */ +#define MIC_MIN_MSIX 4 +#define MIC_NUM_OFFSETS 32 + +/** + * mic_intr_source - The type of source that will generate + * the interrupt.The number of types needs to be in sync with + * MIC_NUM_INTR_TYPES + * + * MIC_INTR_DB: The source is a doorbell + * MIC_INTR_DMA: The source is a DMA channel + * MIC_INTR_ERR: The source is an error interrupt e.g. SBOX ERR + * MIC_NUM_INTR_TYPES: Total number of interrupt sources. + */ +enum mic_intr_type { + MIC_INTR_DB = 0, + MIC_INTR_DMA, + MIC_INTR_ERR, + MIC_NUM_INTR_TYPES +}; + +/** + * struct mic_intr_info - Contains h/w specific interrupt sources + * information. + * + * @intr_start_idx: Contains the starting indexes of the + * interrupt types. + * @intr_len: Contains the length of the interrupt types. + */ +struct mic_intr_info { + u16 intr_start_idx[MIC_NUM_INTR_TYPES]; + u16 intr_len[MIC_NUM_INTR_TYPES]; +}; + +/** + * struct mic_irq_info - OS specific irq information + * + * @next_avail_src: next available doorbell that can be assigned. + * @msix_entries: msix entries allocated while setting up MSI-x + * @mic_msi_map: The MSI/MSI-x mapping information. + * @num_vectors: The number of MSI/MSI-x vectors that have been allocated. + * @cb_ida: callback ID allocator to track the callbacks registered. + * @mic_intr_lock: spinlock to protect the interrupt callback list. + * @cb_list: Array of callback lists one for each source. + */ +struct mic_irq_info { + int next_avail_src; + struct msix_entry *msix_entries; + u32 *mic_msi_map; + u16 num_vectors; + struct ida cb_ida; + spinlock_t mic_intr_lock; + struct list_head *cb_list; +}; + +/** + * struct mic_intr_cb - Interrupt callback structure. + * + * @func: The callback function + * @data: Private data of the requester. + * @cb_id: The callback id. Identifies this callback. + * @list: list head pointing to the next callback structure. + */ +struct mic_intr_cb { + irqreturn_t (*func) (int irq, void *data); + void *data; + int cb_id; + struct list_head list; +}; + +/** + * struct mic_irq - opaque pointer used as cookie + */ +struct mic_irq; + +/* Forward declaration */ +struct mic_device; + +/** + * struct mic_hw_intr_ops: MIC HW specific interrupt operations + * @intr_init: Initialize H/W specific interrupt information. + * @enable_interrupts: Enable interrupts from the hardware. + * @disable_interrupts: Disable interrupts from the hardware. + * @program_msi_to_src_map: Update MSI mapping registers with + * irq information. + * @read_msi_to_src_map: Read MSI mapping registers containing + * irq information. + */ +struct mic_hw_intr_ops { + void (*intr_init)(struct mic_device *mdev); + void (*enable_interrupts)(struct mic_device *mdev); + void (*disable_interrupts)(struct mic_device *mdev); + void (*program_msi_to_src_map) (struct mic_device *mdev, + int idx, int intr_src, bool set); + u32 (*read_msi_to_src_map) (struct mic_device *mdev, + int idx); +}; + +int mic_next_db(struct mic_device *mdev); +struct mic_irq *mic_request_irq(struct mic_device *mdev, + irqreturn_t (*func)(int irq, void *data), + const char *name, void *data, int intr_src, + enum mic_intr_type type); + +void mic_free_irq(struct mic_device *mdev, + struct mic_irq *cookie, void *data); +int mic_setup_interrupts(struct mic_device *mdev, struct pci_dev *pdev); +void mic_free_interrupts(struct mic_device *mdev, struct pci_dev *pdev); +void mic_intr_restore(struct mic_device *mdev); +#endif diff --git a/drivers/misc/mic/host/mic_main.c b/drivers/misc/mic/host/mic_main.c new file mode 100644 index 00000000000..c04a021e20c --- /dev/null +++ b/drivers/misc/mic/host/mic_main.c @@ -0,0 +1,536 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + * Global TODO's across the driver to be added after initial base + * patches are accepted upstream: + * 1) Enable DMA support. + * 2) Enable per vring interrupt support. + */ +#include <linux/fs.h> +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/poll.h> +#include <linux/suspend.h> + +#include <linux/mic_common.h> +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_x100.h" +#include "mic_smpt.h" +#include "mic_fops.h" +#include "mic_virtio.h" + +static const char mic_driver_name[] = "mic"; + +static DEFINE_PCI_DEVICE_TABLE(mic_pci_tbl) = { + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2250)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2251)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2252)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2253)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2254)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2255)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2256)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2257)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2258)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_2259)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225a)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225b)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225c)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225d)}, + {PCI_DEVICE(PCI_VENDOR_ID_INTEL, MIC_X100_PCI_DEVICE_225e)}, + + /* required last entry */ + { 0, } +}; + +MODULE_DEVICE_TABLE(pci, mic_pci_tbl); + +/* ID allocator for MIC devices */ +static struct ida g_mic_ida; +/* Class of MIC devices for sysfs accessibility. */ +static struct class *g_mic_class; +/* Base device node number for MIC devices */ +static dev_t g_mic_devno; + +static const struct file_operations mic_fops = { + .open = mic_open, + .release = mic_release, + .unlocked_ioctl = mic_ioctl, + .poll = mic_poll, + .mmap = mic_mmap, + .owner = THIS_MODULE, +}; + +/* Initialize the device page */ +static int mic_dp_init(struct mic_device *mdev) +{ + mdev->dp = kzalloc(MIC_DP_SIZE, GFP_KERNEL); + if (!mdev->dp) { + dev_err(mdev->sdev->parent, "%s %d err %d\n", + __func__, __LINE__, -ENOMEM); + return -ENOMEM; + } + + mdev->dp_dma_addr = mic_map_single(mdev, + mdev->dp, MIC_DP_SIZE); + if (mic_map_error(mdev->dp_dma_addr)) { + kfree(mdev->dp); + dev_err(mdev->sdev->parent, "%s %d err %d\n", + __func__, __LINE__, -ENOMEM); + return -ENOMEM; + } + mdev->ops->write_spad(mdev, MIC_DPLO_SPAD, mdev->dp_dma_addr); + mdev->ops->write_spad(mdev, MIC_DPHI_SPAD, mdev->dp_dma_addr >> 32); + return 0; +} + +/* Uninitialize the device page */ +static void mic_dp_uninit(struct mic_device *mdev) +{ + mic_unmap_single(mdev, mdev->dp_dma_addr, MIC_DP_SIZE); + kfree(mdev->dp); +} + +/** + * mic_shutdown_db - Shutdown doorbell interrupt handler. + */ +static irqreturn_t mic_shutdown_db(int irq, void *data) +{ + struct mic_device *mdev = data; + struct mic_bootparam *bootparam = mdev->dp; + + mdev->ops->intr_workarounds(mdev); + + switch (bootparam->shutdown_status) { + case MIC_HALTED: + case MIC_POWER_OFF: + case MIC_RESTART: + /* Fall through */ + case MIC_CRASHED: + schedule_work(&mdev->shutdown_work); + break; + default: + break; + }; + return IRQ_HANDLED; +} + +/** + * mic_ops_init: Initialize HW specific operation tables. + * + * @mdev: pointer to mic_device instance + * + * returns none. + */ +static void mic_ops_init(struct mic_device *mdev) +{ + switch (mdev->family) { + case MIC_FAMILY_X100: + mdev->ops = &mic_x100_ops; + mdev->intr_ops = &mic_x100_intr_ops; + mdev->smpt_ops = &mic_x100_smpt_ops; + break; + default: + break; + } +} + +/** + * mic_get_family - Determine hardware family to which this MIC belongs. + * + * @pdev: The pci device structure + * + * returns family. + */ +static enum mic_hw_family mic_get_family(struct pci_dev *pdev) +{ + enum mic_hw_family family; + + switch (pdev->device) { + case MIC_X100_PCI_DEVICE_2250: + case MIC_X100_PCI_DEVICE_2251: + case MIC_X100_PCI_DEVICE_2252: + case MIC_X100_PCI_DEVICE_2253: + case MIC_X100_PCI_DEVICE_2254: + case MIC_X100_PCI_DEVICE_2255: + case MIC_X100_PCI_DEVICE_2256: + case MIC_X100_PCI_DEVICE_2257: + case MIC_X100_PCI_DEVICE_2258: + case MIC_X100_PCI_DEVICE_2259: + case MIC_X100_PCI_DEVICE_225a: + case MIC_X100_PCI_DEVICE_225b: + case MIC_X100_PCI_DEVICE_225c: + case MIC_X100_PCI_DEVICE_225d: + case MIC_X100_PCI_DEVICE_225e: + family = MIC_FAMILY_X100; + break; + default: + family = MIC_FAMILY_UNKNOWN; + break; + } + return family; +} + +/** +* mic_pm_notifier: Notifier callback function that handles +* PM notifications. +* +* @notifier_block: The notifier structure. +* @pm_event: The event for which the driver was notified. +* @unused: Meaningless. Always NULL. +* +* returns NOTIFY_DONE +*/ +static int mic_pm_notifier(struct notifier_block *notifier, + unsigned long pm_event, void *unused) +{ + struct mic_device *mdev = container_of(notifier, + struct mic_device, pm_notifier); + + switch (pm_event) { + case PM_HIBERNATION_PREPARE: + /* Fall through */ + case PM_SUSPEND_PREPARE: + mic_prepare_suspend(mdev); + break; + case PM_POST_HIBERNATION: + /* Fall through */ + case PM_POST_SUSPEND: + /* Fall through */ + case PM_POST_RESTORE: + mic_complete_resume(mdev); + break; + case PM_RESTORE_PREPARE: + break; + default: + break; + } + return NOTIFY_DONE; +} + +/** + * mic_device_init - Allocates and initializes the MIC device structure + * + * @mdev: pointer to mic_device instance + * @pdev: The pci device structure + * + * returns none. + */ +static int +mic_device_init(struct mic_device *mdev, struct pci_dev *pdev) +{ + int rc; + + mdev->family = mic_get_family(pdev); + mdev->stepping = pdev->revision; + mic_ops_init(mdev); + mic_sysfs_init(mdev); + mutex_init(&mdev->mic_mutex); + mdev->irq_info.next_avail_src = 0; + INIT_WORK(&mdev->reset_trigger_work, mic_reset_trigger_work); + INIT_WORK(&mdev->shutdown_work, mic_shutdown_work); + init_completion(&mdev->reset_wait); + INIT_LIST_HEAD(&mdev->vdev_list); + mdev->pm_notifier.notifier_call = mic_pm_notifier; + rc = register_pm_notifier(&mdev->pm_notifier); + if (rc) { + dev_err(&pdev->dev, "register_pm_notifier failed rc %d\n", + rc); + goto register_pm_notifier_fail; + } + return 0; +register_pm_notifier_fail: + flush_work(&mdev->shutdown_work); + flush_work(&mdev->reset_trigger_work); + return rc; +} + +/** + * mic_device_uninit - Frees resources allocated during mic_device_init(..) + * + * @mdev: pointer to mic_device instance + * + * returns none + */ +static void mic_device_uninit(struct mic_device *mdev) +{ + /* The cmdline sysfs entry might have allocated cmdline */ + kfree(mdev->cmdline); + kfree(mdev->firmware); + kfree(mdev->ramdisk); + kfree(mdev->bootmode); + flush_work(&mdev->reset_trigger_work); + flush_work(&mdev->shutdown_work); + unregister_pm_notifier(&mdev->pm_notifier); +} + +/** + * mic_probe - Device Initialization Routine + * + * @pdev: PCI device structure + * @ent: entry in mic_pci_tbl + * + * returns 0 on success, < 0 on failure. + */ +static int mic_probe(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + int rc; + struct mic_device *mdev; + + mdev = kzalloc(sizeof(*mdev), GFP_KERNEL); + if (!mdev) { + rc = -ENOMEM; + dev_err(&pdev->dev, "mdev kmalloc failed rc %d\n", rc); + goto mdev_alloc_fail; + } + mdev->id = ida_simple_get(&g_mic_ida, 0, MIC_MAX_NUM_DEVS, GFP_KERNEL); + if (mdev->id < 0) { + rc = mdev->id; + dev_err(&pdev->dev, "ida_simple_get failed rc %d\n", rc); + goto ida_fail; + } + + rc = mic_device_init(mdev, pdev); + if (rc) { + dev_err(&pdev->dev, "mic_device_init failed rc %d\n", rc); + goto device_init_fail; + } + + rc = pci_enable_device(pdev); + if (rc) { + dev_err(&pdev->dev, "failed to enable pci device.\n"); + goto uninit_device; + } + + pci_set_master(pdev); + + rc = pci_request_regions(pdev, mic_driver_name); + if (rc) { + dev_err(&pdev->dev, "failed to get pci regions.\n"); + goto disable_device; + } + + rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(64)); + if (rc) { + dev_err(&pdev->dev, "Cannot set DMA mask\n"); + goto release_regions; + } + + mdev->mmio.pa = pci_resource_start(pdev, mdev->ops->mmio_bar); + mdev->mmio.len = pci_resource_len(pdev, mdev->ops->mmio_bar); + mdev->mmio.va = pci_ioremap_bar(pdev, mdev->ops->mmio_bar); + if (!mdev->mmio.va) { + dev_err(&pdev->dev, "Cannot remap MMIO BAR\n"); + rc = -EIO; + goto release_regions; + } + + mdev->aper.pa = pci_resource_start(pdev, mdev->ops->aper_bar); + mdev->aper.len = pci_resource_len(pdev, mdev->ops->aper_bar); + mdev->aper.va = ioremap_wc(mdev->aper.pa, mdev->aper.len); + if (!mdev->aper.va) { + dev_err(&pdev->dev, "Cannot remap Aperture BAR\n"); + rc = -EIO; + goto unmap_mmio; + } + + mdev->intr_ops->intr_init(mdev); + rc = mic_setup_interrupts(mdev, pdev); + if (rc) { + dev_err(&pdev->dev, "mic_setup_interrupts failed %d\n", rc); + goto unmap_aper; + } + rc = mic_smpt_init(mdev); + if (rc) { + dev_err(&pdev->dev, "smpt_init failed %d\n", rc); + goto free_interrupts; + } + + pci_set_drvdata(pdev, mdev); + + mdev->sdev = device_create_with_groups(g_mic_class, &pdev->dev, + MKDEV(MAJOR(g_mic_devno), mdev->id), NULL, + mdev->attr_group, "mic%d", mdev->id); + if (IS_ERR(mdev->sdev)) { + rc = PTR_ERR(mdev->sdev); + dev_err(&pdev->dev, + "device_create_with_groups failed rc %d\n", rc); + goto smpt_uninit; + } + mdev->state_sysfs = sysfs_get_dirent(mdev->sdev->kobj.sd, "state"); + if (!mdev->state_sysfs) { + rc = -ENODEV; + dev_err(&pdev->dev, "sysfs_get_dirent failed rc %d\n", rc); + goto destroy_device; + } + + rc = mic_dp_init(mdev); + if (rc) { + dev_err(&pdev->dev, "mic_dp_init failed rc %d\n", rc); + goto sysfs_put; + } + mutex_lock(&mdev->mic_mutex); + + mdev->shutdown_db = mic_next_db(mdev); + mdev->shutdown_cookie = mic_request_irq(mdev, mic_shutdown_db, + "shutdown-interrupt", mdev, mdev->shutdown_db, MIC_INTR_DB); + if (IS_ERR(mdev->shutdown_cookie)) { + rc = PTR_ERR(mdev->shutdown_cookie); + mutex_unlock(&mdev->mic_mutex); + goto dp_uninit; + } + mutex_unlock(&mdev->mic_mutex); + mic_bootparam_init(mdev); + + mic_create_debug_dir(mdev); + cdev_init(&mdev->cdev, &mic_fops); + mdev->cdev.owner = THIS_MODULE; + rc = cdev_add(&mdev->cdev, MKDEV(MAJOR(g_mic_devno), mdev->id), 1); + if (rc) { + dev_err(&pdev->dev, "cdev_add err id %d rc %d\n", mdev->id, rc); + goto cleanup_debug_dir; + } + return 0; +cleanup_debug_dir: + mic_delete_debug_dir(mdev); + mutex_lock(&mdev->mic_mutex); + mic_free_irq(mdev, mdev->shutdown_cookie, mdev); + mutex_unlock(&mdev->mic_mutex); +dp_uninit: + mic_dp_uninit(mdev); +sysfs_put: + sysfs_put(mdev->state_sysfs); +destroy_device: + device_destroy(g_mic_class, MKDEV(MAJOR(g_mic_devno), mdev->id)); +smpt_uninit: + mic_smpt_uninit(mdev); +free_interrupts: + mic_free_interrupts(mdev, pdev); +unmap_aper: + iounmap(mdev->aper.va); +unmap_mmio: + iounmap(mdev->mmio.va); +release_regions: + pci_release_regions(pdev); +disable_device: + pci_disable_device(pdev); +uninit_device: + mic_device_uninit(mdev); +device_init_fail: + ida_simple_remove(&g_mic_ida, mdev->id); +ida_fail: + kfree(mdev); +mdev_alloc_fail: + dev_err(&pdev->dev, "Probe failed rc %d\n", rc); + return rc; +} + +/** + * mic_remove - Device Removal Routine + * mic_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. + * + * @pdev: PCI device structure + */ +static void mic_remove(struct pci_dev *pdev) +{ + struct mic_device *mdev; + + mdev = pci_get_drvdata(pdev); + if (!mdev) + return; + + mic_stop(mdev, false); + cdev_del(&mdev->cdev); + mic_delete_debug_dir(mdev); + mutex_lock(&mdev->mic_mutex); + mic_free_irq(mdev, mdev->shutdown_cookie, mdev); + mutex_unlock(&mdev->mic_mutex); + flush_work(&mdev->shutdown_work); + mic_dp_uninit(mdev); + sysfs_put(mdev->state_sysfs); + device_destroy(g_mic_class, MKDEV(MAJOR(g_mic_devno), mdev->id)); + mic_smpt_uninit(mdev); + mic_free_interrupts(mdev, pdev); + iounmap(mdev->mmio.va); + iounmap(mdev->aper.va); + mic_device_uninit(mdev); + pci_release_regions(pdev); + pci_disable_device(pdev); + ida_simple_remove(&g_mic_ida, mdev->id); + kfree(mdev); +} +static struct pci_driver mic_driver = { + .name = mic_driver_name, + .id_table = mic_pci_tbl, + .probe = mic_probe, + .remove = mic_remove +}; + +static int __init mic_init(void) +{ + int ret; + + ret = alloc_chrdev_region(&g_mic_devno, 0, + MIC_MAX_NUM_DEVS, mic_driver_name); + if (ret) { + pr_err("alloc_chrdev_region failed ret %d\n", ret); + goto error; + } + + g_mic_class = class_create(THIS_MODULE, mic_driver_name); + if (IS_ERR(g_mic_class)) { + ret = PTR_ERR(g_mic_class); + pr_err("class_create failed ret %d\n", ret); + goto cleanup_chrdev; + } + + mic_init_debugfs(); + ida_init(&g_mic_ida); + ret = pci_register_driver(&mic_driver); + if (ret) { + pr_err("pci_register_driver failed ret %d\n", ret); + goto cleanup_debugfs; + } + return ret; +cleanup_debugfs: + mic_exit_debugfs(); + class_destroy(g_mic_class); +cleanup_chrdev: + unregister_chrdev_region(g_mic_devno, MIC_MAX_NUM_DEVS); +error: + return ret; +} + +static void __exit mic_exit(void) +{ + pci_unregister_driver(&mic_driver); + ida_destroy(&g_mic_ida); + mic_exit_debugfs(); + class_destroy(g_mic_class); + unregister_chrdev_region(g_mic_devno, MIC_MAX_NUM_DEVS); +} + +module_init(mic_init); +module_exit(mic_exit); + +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel(R) MIC X100 Host driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/host/mic_smpt.c b/drivers/misc/mic/host/mic_smpt.c new file mode 100644 index 00000000000..fae474c4899 --- /dev/null +++ b/drivers/misc/mic/host/mic_smpt.c @@ -0,0 +1,442 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/pci.h> + +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_smpt.h" + +static inline u64 mic_system_page_mask(struct mic_device *mdev) +{ + return (1ULL << mdev->smpt->info.page_shift) - 1ULL; +} + +static inline u8 mic_sys_addr_to_smpt(struct mic_device *mdev, dma_addr_t pa) +{ + return (pa - mdev->smpt->info.base) >> mdev->smpt->info.page_shift; +} + +static inline u64 mic_smpt_to_pa(struct mic_device *mdev, u8 index) +{ + return mdev->smpt->info.base + (index * mdev->smpt->info.page_size); +} + +static inline u64 mic_smpt_offset(struct mic_device *mdev, dma_addr_t pa) +{ + return pa & mic_system_page_mask(mdev); +} + +static inline u64 mic_smpt_align_low(struct mic_device *mdev, dma_addr_t pa) +{ + return ALIGN(pa - mic_system_page_mask(mdev), + mdev->smpt->info.page_size); +} + +static inline u64 mic_smpt_align_high(struct mic_device *mdev, dma_addr_t pa) +{ + return ALIGN(pa, mdev->smpt->info.page_size); +} + +/* Total Cumulative system memory accessible by MIC across all SMPT entries */ +static inline u64 mic_max_system_memory(struct mic_device *mdev) +{ + return mdev->smpt->info.num_reg * mdev->smpt->info.page_size; +} + +/* Maximum system memory address accessible by MIC */ +static inline u64 mic_max_system_addr(struct mic_device *mdev) +{ + return mdev->smpt->info.base + mic_max_system_memory(mdev) - 1ULL; +} + +/* Check if the DMA address is a MIC system memory address */ +static inline bool +mic_is_system_addr(struct mic_device *mdev, dma_addr_t pa) +{ + return pa >= mdev->smpt->info.base && pa <= mic_max_system_addr(mdev); +} + +/* Populate an SMPT entry and update the reference counts. */ +static void mic_add_smpt_entry(int spt, s64 *ref, u64 addr, + int entries, struct mic_device *mdev) +{ + struct mic_smpt_info *smpt_info = mdev->smpt; + int i; + + for (i = spt; i < spt + entries; i++, + addr += smpt_info->info.page_size) { + if (!smpt_info->entry[i].ref_count && + (smpt_info->entry[i].dma_addr != addr)) { + mdev->smpt_ops->set(mdev, addr, i); + smpt_info->entry[i].dma_addr = addr; + } + smpt_info->entry[i].ref_count += ref[i - spt]; + } +} + +/* + * Find an available MIC address in MIC SMPT address space + * for a given DMA address and size. + */ +static dma_addr_t mic_smpt_op(struct mic_device *mdev, u64 dma_addr, + int entries, s64 *ref, size_t size) +{ + int spt; + int ae = 0; + int i; + unsigned long flags; + dma_addr_t mic_addr = 0; + dma_addr_t addr = dma_addr; + struct mic_smpt_info *smpt_info = mdev->smpt; + + spin_lock_irqsave(&smpt_info->smpt_lock, flags); + + /* find existing entries */ + for (i = 0; i < smpt_info->info.num_reg; i++) { + if (smpt_info->entry[i].dma_addr == addr) { + ae++; + addr += smpt_info->info.page_size; + } else if (ae) /* cannot find contiguous entries */ + goto not_found; + + if (ae == entries) + goto found; + } + + /* find free entry */ + for (ae = 0, i = 0; i < smpt_info->info.num_reg; i++) { + ae = (smpt_info->entry[i].ref_count == 0) ? ae + 1 : 0; + if (ae == entries) + goto found; + } + +not_found: + spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); + return mic_addr; + +found: + spt = i - entries + 1; + mic_addr = mic_smpt_to_pa(mdev, spt); + mic_add_smpt_entry(spt, ref, dma_addr, entries, mdev); + smpt_info->map_count++; + smpt_info->ref_count += (s64)size; + spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); + return mic_addr; +} + +/* + * Returns number of smpt entries needed for dma_addr to dma_addr + size + * also returns the reference count array for each of those entries + * and the starting smpt address + */ +static int mic_get_smpt_ref_count(struct mic_device *mdev, dma_addr_t dma_addr, + size_t size, s64 *ref, u64 *smpt_start) +{ + u64 start = dma_addr; + u64 end = dma_addr + size; + int i = 0; + + while (start < end) { + ref[i++] = min(mic_smpt_align_high(mdev, start + 1), + end) - start; + start = mic_smpt_align_high(mdev, start + 1); + } + + if (smpt_start) + *smpt_start = mic_smpt_align_low(mdev, dma_addr); + + return i; +} + +/* + * mic_to_dma_addr - Converts a MIC address to a DMA address. + * + * @mdev: pointer to mic_device instance. + * @mic_addr: MIC address. + * + * returns a DMA address. + */ +static dma_addr_t +mic_to_dma_addr(struct mic_device *mdev, dma_addr_t mic_addr) +{ + struct mic_smpt_info *smpt_info = mdev->smpt; + int spt; + dma_addr_t dma_addr; + + if (!mic_is_system_addr(mdev, mic_addr)) { + dev_err(mdev->sdev->parent, + "mic_addr is invalid. mic_addr = 0x%llx\n", mic_addr); + return -EINVAL; + } + spt = mic_sys_addr_to_smpt(mdev, mic_addr); + dma_addr = smpt_info->entry[spt].dma_addr + + mic_smpt_offset(mdev, mic_addr); + return dma_addr; +} + +/** + * mic_map - Maps a DMA address to a MIC physical address. + * + * @mdev: pointer to mic_device instance. + * @dma_addr: DMA address. + * @size: Size of the region to be mapped. + * + * This API converts the DMA address provided to a DMA address understood + * by MIC. Caller should check for errors by calling mic_map_error(..). + * + * returns DMA address as required by MIC. + */ +dma_addr_t mic_map(struct mic_device *mdev, dma_addr_t dma_addr, size_t size) +{ + dma_addr_t mic_addr = 0; + int num_entries; + s64 *ref; + u64 smpt_start; + + if (!size || size > mic_max_system_memory(mdev)) + return mic_addr; + + ref = kmalloc(mdev->smpt->info.num_reg * sizeof(s64), GFP_KERNEL); + if (!ref) + return mic_addr; + + num_entries = mic_get_smpt_ref_count(mdev, dma_addr, size, + ref, &smpt_start); + + /* Set the smpt table appropriately and get 16G aligned mic address */ + mic_addr = mic_smpt_op(mdev, smpt_start, num_entries, ref, size); + + kfree(ref); + + /* + * If mic_addr is zero then its an error case + * since mic_addr can never be zero. + * else generate mic_addr by adding the 16G offset in dma_addr + */ + if (!mic_addr && MIC_FAMILY_X100 == mdev->family) { + dev_err(mdev->sdev->parent, + "mic_map failed dma_addr 0x%llx size 0x%lx\n", + dma_addr, size); + return mic_addr; + } else { + return mic_addr + mic_smpt_offset(mdev, dma_addr); + } +} + +/** + * mic_unmap - Unmaps a MIC physical address. + * + * @mdev: pointer to mic_device instance. + * @mic_addr: MIC physical address. + * @size: Size of the region to be unmapped. + * + * This API unmaps the mappings created by mic_map(..). + * + * returns None. + */ +void mic_unmap(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) +{ + struct mic_smpt_info *smpt_info = mdev->smpt; + s64 *ref; + int num_smpt; + int spt; + int i; + unsigned long flags; + + if (!size) + return; + + if (!mic_is_system_addr(mdev, mic_addr)) { + dev_err(mdev->sdev->parent, + "invalid address: 0x%llx\n", mic_addr); + return; + } + + spt = mic_sys_addr_to_smpt(mdev, mic_addr); + ref = kmalloc(mdev->smpt->info.num_reg * sizeof(s64), GFP_KERNEL); + if (!ref) + return; + + /* Get number of smpt entries to be mapped, ref count array */ + num_smpt = mic_get_smpt_ref_count(mdev, mic_addr, size, ref, NULL); + + spin_lock_irqsave(&smpt_info->smpt_lock, flags); + smpt_info->unmap_count++; + smpt_info->ref_count -= (s64)size; + + for (i = spt; i < spt + num_smpt; i++) { + smpt_info->entry[i].ref_count -= ref[i - spt]; + if (smpt_info->entry[i].ref_count < 0) + dev_warn(mdev->sdev->parent, + "ref count for entry %d is negative\n", i); + } + spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); + kfree(ref); +} + +/** + * mic_map_single - Maps a virtual address to a MIC physical address. + * + * @mdev: pointer to mic_device instance. + * @va: Kernel direct mapped virtual address. + * @size: Size of the region to be mapped. + * + * This API calls pci_map_single(..) for the direct mapped virtual address + * and then converts the DMA address provided to a DMA address understood + * by MIC. Caller should check for errors by calling mic_map_error(..). + * + * returns DMA address as required by MIC. + */ +dma_addr_t mic_map_single(struct mic_device *mdev, void *va, size_t size) +{ + dma_addr_t mic_addr = 0; + struct pci_dev *pdev = container_of(mdev->sdev->parent, + struct pci_dev, dev); + dma_addr_t dma_addr = + pci_map_single(pdev, va, size, PCI_DMA_BIDIRECTIONAL); + + if (!pci_dma_mapping_error(pdev, dma_addr)) { + mic_addr = mic_map(mdev, dma_addr, size); + if (!mic_addr) { + dev_err(mdev->sdev->parent, + "mic_map failed dma_addr 0x%llx size 0x%lx\n", + dma_addr, size); + pci_unmap_single(pdev, dma_addr, + size, PCI_DMA_BIDIRECTIONAL); + } + } + return mic_addr; +} + +/** + * mic_unmap_single - Unmaps a MIC physical address. + * + * @mdev: pointer to mic_device instance. + * @mic_addr: MIC physical address. + * @size: Size of the region to be unmapped. + * + * This API unmaps the mappings created by mic_map_single(..). + * + * returns None. + */ +void +mic_unmap_single(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) +{ + struct pci_dev *pdev = container_of(mdev->sdev->parent, + struct pci_dev, dev); + dma_addr_t dma_addr = mic_to_dma_addr(mdev, mic_addr); + mic_unmap(mdev, mic_addr, size); + pci_unmap_single(pdev, dma_addr, size, PCI_DMA_BIDIRECTIONAL); +} + +/** + * mic_smpt_init - Initialize MIC System Memory Page Tables. + * + * @mdev: pointer to mic_device instance. + * + * returns 0 for success and -errno for error. + */ +int mic_smpt_init(struct mic_device *mdev) +{ + int i, err = 0; + dma_addr_t dma_addr; + struct mic_smpt_info *smpt_info; + + mdev->smpt = kmalloc(sizeof(*mdev->smpt), GFP_KERNEL); + if (!mdev->smpt) + return -ENOMEM; + + smpt_info = mdev->smpt; + mdev->smpt_ops->init(mdev); + smpt_info->entry = kmalloc_array(smpt_info->info.num_reg, + sizeof(*smpt_info->entry), GFP_KERNEL); + if (!smpt_info->entry) { + err = -ENOMEM; + goto free_smpt; + } + spin_lock_init(&smpt_info->smpt_lock); + for (i = 0; i < smpt_info->info.num_reg; i++) { + dma_addr = i * smpt_info->info.page_size; + smpt_info->entry[i].dma_addr = dma_addr; + smpt_info->entry[i].ref_count = 0; + mdev->smpt_ops->set(mdev, dma_addr, i); + } + smpt_info->ref_count = 0; + smpt_info->map_count = 0; + smpt_info->unmap_count = 0; + return 0; +free_smpt: + kfree(smpt_info); + return err; +} + +/** + * mic_smpt_uninit - UnInitialize MIC System Memory Page Tables. + * + * @mdev: pointer to mic_device instance. + * + * returns None. + */ +void mic_smpt_uninit(struct mic_device *mdev) +{ + struct mic_smpt_info *smpt_info = mdev->smpt; + int i; + + dev_dbg(mdev->sdev->parent, + "nodeid %d SMPT ref count %lld map %lld unmap %lld\n", + mdev->id, smpt_info->ref_count, + smpt_info->map_count, smpt_info->unmap_count); + + for (i = 0; i < smpt_info->info.num_reg; i++) { + dev_dbg(mdev->sdev->parent, + "SMPT entry[%d] dma_addr = 0x%llx ref_count = %lld\n", + i, smpt_info->entry[i].dma_addr, + smpt_info->entry[i].ref_count); + if (smpt_info->entry[i].ref_count) + dev_warn(mdev->sdev->parent, + "ref count for entry %d is not zero\n", i); + } + kfree(smpt_info->entry); + kfree(smpt_info); +} + +/** + * mic_smpt_restore - Restore MIC System Memory Page Tables. + * + * @mdev: pointer to mic_device instance. + * + * Restore the SMPT registers to values previously stored in the + * SW data structures. Some MIC steppings lose register state + * across resets and this API should be called for performing + * a restore operation if required. + * + * returns None. + */ +void mic_smpt_restore(struct mic_device *mdev) +{ + int i; + dma_addr_t dma_addr; + + for (i = 0; i < mdev->smpt->info.num_reg; i++) { + dma_addr = mdev->smpt->entry[i].dma_addr; + mdev->smpt_ops->set(mdev, dma_addr, i); + } +} diff --git a/drivers/misc/mic/host/mic_smpt.h b/drivers/misc/mic/host/mic_smpt.h new file mode 100644 index 00000000000..51970abfe7d --- /dev/null +++ b/drivers/misc/mic/host/mic_smpt.h @@ -0,0 +1,98 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#ifndef MIC_SMPT_H +#define MIC_SMPT_H +/** + * struct mic_smpt_ops - MIC HW specific SMPT operations. + * @init: Initialize hardware specific SMPT information in mic_smpt_hw_info. + * @set: Set the value for a particular SMPT entry. + */ +struct mic_smpt_ops { + void (*init)(struct mic_device *mdev); + void (*set)(struct mic_device *mdev, dma_addr_t dma_addr, u8 index); +}; + +/** + * struct mic_smpt - MIC SMPT entry information. + * @dma_addr: Base DMA address for this SMPT entry. + * @ref_count: Number of active mappings for this SMPT entry in bytes. + */ +struct mic_smpt { + dma_addr_t dma_addr; + s64 ref_count; +}; + +/** + * struct mic_smpt_hw_info - MIC SMPT hardware specific information. + * @num_reg: Number of SMPT registers. + * @page_shift: System memory page shift. + * @page_size: System memory page size. + * @base: System address base. + */ +struct mic_smpt_hw_info { + u8 num_reg; + u8 page_shift; + u64 page_size; + u64 base; +}; + +/** + * struct mic_smpt_info - MIC SMPT information. + * @entry: Array of SMPT entries. + * @smpt_lock: Spin lock protecting access to SMPT data structures. + * @info: Hardware specific SMPT information. + * @ref_count: Number of active SMPT mappings (for debug). + * @map_count: Number of SMPT mappings created (for debug). + * @unmap_count: Number of SMPT mappings destroyed (for debug). + */ +struct mic_smpt_info { + struct mic_smpt *entry; + spinlock_t smpt_lock; + struct mic_smpt_hw_info info; + s64 ref_count; + s64 map_count; + s64 unmap_count; +}; + +dma_addr_t mic_map_single(struct mic_device *mdev, void *va, size_t size); +void mic_unmap_single(struct mic_device *mdev, + dma_addr_t mic_addr, size_t size); +dma_addr_t mic_map(struct mic_device *mdev, + dma_addr_t dma_addr, size_t size); +void mic_unmap(struct mic_device *mdev, dma_addr_t mic_addr, size_t size); + +/** + * mic_map_error - Check a MIC address for errors. + * + * @mdev: pointer to mic_device instance. + * + * returns Whether there was an error during mic_map..(..) APIs. + */ +static inline bool mic_map_error(dma_addr_t mic_addr) +{ + return !mic_addr; +} + +int mic_smpt_init(struct mic_device *mdev); +void mic_smpt_uninit(struct mic_device *mdev); +void mic_smpt_restore(struct mic_device *mdev); + +#endif diff --git a/drivers/misc/mic/host/mic_sysfs.c b/drivers/misc/mic/host/mic_sysfs.c new file mode 100644 index 00000000000..6dd864e4a61 --- /dev/null +++ b/drivers/misc/mic/host/mic_sysfs.c @@ -0,0 +1,459 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/pci.h> + +#include <linux/mic_common.h> +#include "../common/mic_dev.h" +#include "mic_device.h" + +/* + * A state-to-string lookup table, for exposing a human readable state + * via sysfs. Always keep in sync with enum mic_states + */ +static const char * const mic_state_string[] = { + [MIC_OFFLINE] = "offline", + [MIC_ONLINE] = "online", + [MIC_SHUTTING_DOWN] = "shutting_down", + [MIC_RESET_FAILED] = "reset_failed", + [MIC_SUSPENDING] = "suspending", + [MIC_SUSPENDED] = "suspended", +}; + +/* + * A shutdown-status-to-string lookup table, for exposing a human + * readable state via sysfs. Always keep in sync with enum mic_shutdown_status + */ +static const char * const mic_shutdown_status_string[] = { + [MIC_NOP] = "nop", + [MIC_CRASHED] = "crashed", + [MIC_HALTED] = "halted", + [MIC_POWER_OFF] = "poweroff", + [MIC_RESTART] = "restart", +}; + +void mic_set_shutdown_status(struct mic_device *mdev, u8 shutdown_status) +{ + dev_dbg(mdev->sdev->parent, "Shutdown Status %s -> %s\n", + mic_shutdown_status_string[mdev->shutdown_status], + mic_shutdown_status_string[shutdown_status]); + mdev->shutdown_status = shutdown_status; +} + +void mic_set_state(struct mic_device *mdev, u8 state) +{ + dev_dbg(mdev->sdev->parent, "State %s -> %s\n", + mic_state_string[mdev->state], + mic_state_string[state]); + mdev->state = state; + sysfs_notify_dirent(mdev->state_sysfs); +} + +static ssize_t +family_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + static const char x100[] = "x100"; + static const char unknown[] = "Unknown"; + const char *card = NULL; + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev) + return -EINVAL; + + switch (mdev->family) { + case MIC_FAMILY_X100: + card = x100; + break; + default: + card = unknown; + break; + } + return scnprintf(buf, PAGE_SIZE, "%s\n", card); +} +static DEVICE_ATTR_RO(family); + +static ssize_t +stepping_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + char *string = "??"; + + if (!mdev) + return -EINVAL; + + switch (mdev->stepping) { + case MIC_A0_STEP: + string = "A0"; + break; + case MIC_B0_STEP: + string = "B0"; + break; + case MIC_B1_STEP: + string = "B1"; + break; + case MIC_C0_STEP: + string = "C0"; + break; + default: + break; + } + return scnprintf(buf, PAGE_SIZE, "%s\n", string); +} +static DEVICE_ATTR_RO(stepping); + +static ssize_t +state_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev || mdev->state >= MIC_LAST) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%s\n", + mic_state_string[mdev->state]); +} + +static ssize_t +state_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + int rc = 0; + struct mic_device *mdev = dev_get_drvdata(dev->parent); + if (!mdev) + return -EINVAL; + if (sysfs_streq(buf, "boot")) { + rc = mic_start(mdev, buf); + if (rc) { + dev_err(mdev->sdev->parent, + "mic_boot failed rc %d\n", rc); + count = rc; + } + goto done; + } + + if (sysfs_streq(buf, "reset")) { + schedule_work(&mdev->reset_trigger_work); + goto done; + } + + if (sysfs_streq(buf, "shutdown")) { + mic_shutdown(mdev); + goto done; + } + + if (sysfs_streq(buf, "suspend")) { + mic_suspend(mdev); + goto done; + } + + count = -EINVAL; +done: + return count; +} +static DEVICE_ATTR_RW(state); + +static ssize_t shutdown_status_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev || mdev->shutdown_status >= MIC_STATUS_LAST) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%s\n", + mic_shutdown_status_string[mdev->shutdown_status]); +} +static DEVICE_ATTR_RO(shutdown_status); + +static ssize_t +cmdline_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + char *cmdline; + + if (!mdev) + return -EINVAL; + + cmdline = mdev->cmdline; + + if (cmdline) + return scnprintf(buf, PAGE_SIZE, "%s\n", cmdline); + return 0; +} + +static ssize_t +cmdline_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev) + return -EINVAL; + + mutex_lock(&mdev->mic_mutex); + kfree(mdev->cmdline); + + mdev->cmdline = kmalloc(count + 1, GFP_KERNEL); + if (!mdev->cmdline) { + count = -ENOMEM; + goto unlock; + } + + strncpy(mdev->cmdline, buf, count); + + if (mdev->cmdline[count - 1] == '\n') + mdev->cmdline[count - 1] = '\0'; + else + mdev->cmdline[count] = '\0'; +unlock: + mutex_unlock(&mdev->mic_mutex); + return count; +} +static DEVICE_ATTR_RW(cmdline); + +static ssize_t +firmware_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + char *firmware; + + if (!mdev) + return -EINVAL; + + firmware = mdev->firmware; + + if (firmware) + return scnprintf(buf, PAGE_SIZE, "%s\n", firmware); + return 0; +} + +static ssize_t +firmware_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev) + return -EINVAL; + + mutex_lock(&mdev->mic_mutex); + kfree(mdev->firmware); + + mdev->firmware = kmalloc(count + 1, GFP_KERNEL); + if (!mdev->firmware) { + count = -ENOMEM; + goto unlock; + } + strncpy(mdev->firmware, buf, count); + + if (mdev->firmware[count - 1] == '\n') + mdev->firmware[count - 1] = '\0'; + else + mdev->firmware[count] = '\0'; +unlock: + mutex_unlock(&mdev->mic_mutex); + return count; +} +static DEVICE_ATTR_RW(firmware); + +static ssize_t +ramdisk_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + char *ramdisk; + + if (!mdev) + return -EINVAL; + + ramdisk = mdev->ramdisk; + + if (ramdisk) + return scnprintf(buf, PAGE_SIZE, "%s\n", ramdisk); + return 0; +} + +static ssize_t +ramdisk_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev) + return -EINVAL; + + mutex_lock(&mdev->mic_mutex); + kfree(mdev->ramdisk); + + mdev->ramdisk = kmalloc(count + 1, GFP_KERNEL); + if (!mdev->ramdisk) { + count = -ENOMEM; + goto unlock; + } + + strncpy(mdev->ramdisk, buf, count); + + if (mdev->ramdisk[count - 1] == '\n') + mdev->ramdisk[count - 1] = '\0'; + else + mdev->ramdisk[count] = '\0'; +unlock: + mutex_unlock(&mdev->mic_mutex); + return count; +} +static DEVICE_ATTR_RW(ramdisk); + +static ssize_t +bootmode_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + char *bootmode; + + if (!mdev) + return -EINVAL; + + bootmode = mdev->bootmode; + + if (bootmode) + return scnprintf(buf, PAGE_SIZE, "%s\n", bootmode); + return 0; +} + +static ssize_t +bootmode_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev) + return -EINVAL; + + if (!sysfs_streq(buf, "linux") && !sysfs_streq(buf, "elf")) + return -EINVAL; + + mutex_lock(&mdev->mic_mutex); + kfree(mdev->bootmode); + + mdev->bootmode = kmalloc(count + 1, GFP_KERNEL); + if (!mdev->bootmode) { + count = -ENOMEM; + goto unlock; + } + + strncpy(mdev->bootmode, buf, count); + + if (mdev->bootmode[count - 1] == '\n') + mdev->bootmode[count - 1] = '\0'; + else + mdev->bootmode[count] = '\0'; +unlock: + mutex_unlock(&mdev->mic_mutex); + return count; +} +static DEVICE_ATTR_RW(bootmode); + +static ssize_t +log_buf_addr_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%p\n", mdev->log_buf_addr); +} + +static ssize_t +log_buf_addr_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + int ret; + unsigned long addr; + + if (!mdev) + return -EINVAL; + + ret = kstrtoul(buf, 16, &addr); + if (ret) + goto exit; + + mdev->log_buf_addr = (void *)addr; + ret = count; +exit: + return ret; +} +static DEVICE_ATTR_RW(log_buf_addr); + +static ssize_t +log_buf_len_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + + if (!mdev) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%p\n", mdev->log_buf_len); +} + +static ssize_t +log_buf_len_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct mic_device *mdev = dev_get_drvdata(dev->parent); + int ret; + unsigned long addr; + + if (!mdev) + return -EINVAL; + + ret = kstrtoul(buf, 16, &addr); + if (ret) + goto exit; + + mdev->log_buf_len = (int *)addr; + ret = count; +exit: + return ret; +} +static DEVICE_ATTR_RW(log_buf_len); + +static struct attribute *mic_default_attrs[] = { + &dev_attr_family.attr, + &dev_attr_stepping.attr, + &dev_attr_state.attr, + &dev_attr_shutdown_status.attr, + &dev_attr_cmdline.attr, + &dev_attr_firmware.attr, + &dev_attr_ramdisk.attr, + &dev_attr_bootmode.attr, + &dev_attr_log_buf_addr.attr, + &dev_attr_log_buf_len.attr, + + NULL +}; + +ATTRIBUTE_GROUPS(mic_default); + +void mic_sysfs_init(struct mic_device *mdev) +{ + mdev->attr_group = mic_default_groups; +} diff --git a/drivers/misc/mic/host/mic_virtio.c b/drivers/misc/mic/host/mic_virtio.c new file mode 100644 index 00000000000..7e1ef0ebbb8 --- /dev/null +++ b/drivers/misc/mic/host/mic_virtio.c @@ -0,0 +1,701 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/pci.h> +#include <linux/sched.h> +#include <linux/uaccess.h> + +#include <linux/mic_common.h> +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_smpt.h" +#include "mic_virtio.h" + +/* + * Initiates the copies across the PCIe bus from card memory to + * a user space buffer. + */ +static int mic_virtio_copy_to_user(struct mic_vdev *mvdev, + void __user *ubuf, size_t len, u64 addr) +{ + int err; + void __iomem *dbuf = mvdev->mdev->aper.va + addr; + /* + * We are copying from IO below an should ideally use something + * like copy_to_user_fromio(..) if it existed. + */ + if (copy_to_user(ubuf, (void __force *)dbuf, len)) { + err = -EFAULT; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, err); + goto err; + } + mvdev->in_bytes += len; + err = 0; +err: + return err; +} + +/* + * Initiates copies across the PCIe bus from a user space + * buffer to card memory. + */ +static int mic_virtio_copy_from_user(struct mic_vdev *mvdev, + void __user *ubuf, size_t len, u64 addr) +{ + int err; + void __iomem *dbuf = mvdev->mdev->aper.va + addr; + /* + * We are copying to IO below and should ideally use something + * like copy_from_user_toio(..) if it existed. + */ + if (copy_from_user((void __force *)dbuf, ubuf, len)) { + err = -EFAULT; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, err); + goto err; + } + mvdev->out_bytes += len; + err = 0; +err: + return err; +} + +#define MIC_VRINGH_READ true + +/* The function to call to notify the card about added buffers */ +static void mic_notify(struct vringh *vrh) +{ + struct mic_vringh *mvrh = container_of(vrh, struct mic_vringh, vrh); + struct mic_vdev *mvdev = mvrh->mvdev; + s8 db = mvdev->dc->h2c_vdev_db; + + if (db != -1) + mvdev->mdev->ops->send_intr(mvdev->mdev, db); +} + +/* Determine the total number of bytes consumed in a VRINGH KIOV */ +static inline u32 mic_vringh_iov_consumed(struct vringh_kiov *iov) +{ + int i; + u32 total = iov->consumed; + + for (i = 0; i < iov->i; i++) + total += iov->iov[i].iov_len; + return total; +} + +/* + * Traverse the VRINGH KIOV and issue the APIs to trigger the copies. + * This API is heavily based on the vringh_iov_xfer(..) implementation + * in vringh.c. The reason we cannot reuse vringh_iov_pull_kern(..) + * and vringh_iov_push_kern(..) directly is because there is no + * way to override the VRINGH xfer(..) routines as of v3.10. + */ +static int mic_vringh_copy(struct mic_vdev *mvdev, struct vringh_kiov *iov, + void __user *ubuf, size_t len, bool read, size_t *out_len) +{ + int ret = 0; + size_t partlen, tot_len = 0; + + while (len && iov->i < iov->used) { + partlen = min(iov->iov[iov->i].iov_len, len); + if (read) + ret = mic_virtio_copy_to_user(mvdev, + ubuf, partlen, + (u64)iov->iov[iov->i].iov_base); + else + ret = mic_virtio_copy_from_user(mvdev, + ubuf, partlen, + (u64)iov->iov[iov->i].iov_base); + if (ret) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + break; + } + len -= partlen; + ubuf += partlen; + tot_len += partlen; + iov->consumed += partlen; + iov->iov[iov->i].iov_len -= partlen; + iov->iov[iov->i].iov_base += partlen; + if (!iov->iov[iov->i].iov_len) { + /* Fix up old iov element then increment. */ + iov->iov[iov->i].iov_len = iov->consumed; + iov->iov[iov->i].iov_base -= iov->consumed; + + iov->consumed = 0; + iov->i++; + } + } + *out_len = tot_len; + return ret; +} + +/* + * Use the standard VRINGH infrastructure in the kernel to fetch new + * descriptors, initiate the copies and update the used ring. + */ +static int _mic_virtio_copy(struct mic_vdev *mvdev, + struct mic_copy_desc *copy) +{ + int ret = 0; + u32 iovcnt = copy->iovcnt; + struct iovec iov; + struct iovec __user *u_iov = copy->iov; + void __user *ubuf = NULL; + struct mic_vringh *mvr = &mvdev->mvr[copy->vr_idx]; + struct vringh_kiov *riov = &mvr->riov; + struct vringh_kiov *wiov = &mvr->wiov; + struct vringh *vrh = &mvr->vrh; + u16 *head = &mvr->head; + struct mic_vring *vr = &mvr->vring; + size_t len = 0, out_len; + + copy->out_len = 0; + /* Fetch a new IOVEC if all previous elements have been processed */ + if (riov->i == riov->used && wiov->i == wiov->used) { + ret = vringh_getdesc_kern(vrh, riov, wiov, + head, GFP_KERNEL); + /* Check if there are available descriptors */ + if (ret <= 0) + return ret; + } + while (iovcnt) { + if (!len) { + /* Copy over a new iovec from user space. */ + ret = copy_from_user(&iov, u_iov, sizeof(*u_iov)); + if (ret) { + ret = -EINVAL; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + break; + } + len = iov.iov_len; + ubuf = iov.iov_base; + } + /* Issue all the read descriptors first */ + ret = mic_vringh_copy(mvdev, riov, ubuf, len, + MIC_VRINGH_READ, &out_len); + if (ret) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + break; + } + len -= out_len; + ubuf += out_len; + copy->out_len += out_len; + /* Issue the write descriptors next */ + ret = mic_vringh_copy(mvdev, wiov, ubuf, len, + !MIC_VRINGH_READ, &out_len); + if (ret) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + break; + } + len -= out_len; + ubuf += out_len; + copy->out_len += out_len; + if (!len) { + /* One user space iovec is now completed */ + iovcnt--; + u_iov++; + } + /* Exit loop if all elements in KIOVs have been processed. */ + if (riov->i == riov->used && wiov->i == wiov->used) + break; + } + /* + * Update the used ring if a descriptor was available and some data was + * copied in/out and the user asked for a used ring update. + */ + if (*head != USHRT_MAX && copy->out_len && copy->update_used) { + u32 total = 0; + + /* Determine the total data consumed */ + total += mic_vringh_iov_consumed(riov); + total += mic_vringh_iov_consumed(wiov); + vringh_complete_kern(vrh, *head, total); + *head = USHRT_MAX; + if (vringh_need_notify_kern(vrh) > 0) + vringh_notify(vrh); + vringh_kiov_cleanup(riov); + vringh_kiov_cleanup(wiov); + /* Update avail idx for user space */ + vr->info->avail_idx = vrh->last_avail_idx; + } + return ret; +} + +static inline int mic_verify_copy_args(struct mic_vdev *mvdev, + struct mic_copy_desc *copy) +{ + if (copy->vr_idx >= mvdev->dd->num_vq) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, -EINVAL); + return -EINVAL; + } + return 0; +} + +/* Copy a specified number of virtio descriptors in a chain */ +int mic_virtio_copy_desc(struct mic_vdev *mvdev, + struct mic_copy_desc *copy) +{ + int err; + struct mic_vringh *mvr = &mvdev->mvr[copy->vr_idx]; + + err = mic_verify_copy_args(mvdev, copy); + if (err) + return err; + + mutex_lock(&mvr->vr_mutex); + if (!mic_vdevup(mvdev)) { + err = -ENODEV; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, err); + goto err; + } + err = _mic_virtio_copy(mvdev, copy); + if (err) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, err); + } +err: + mutex_unlock(&mvr->vr_mutex); + return err; +} + +static void mic_virtio_init_post(struct mic_vdev *mvdev) +{ + struct mic_vqconfig *vqconfig = mic_vq_config(mvdev->dd); + int i; + + for (i = 0; i < mvdev->dd->num_vq; i++) { + if (!le64_to_cpu(vqconfig[i].used_address)) { + dev_warn(mic_dev(mvdev), "used_address zero??\n"); + continue; + } + mvdev->mvr[i].vrh.vring.used = + (void __force *)mvdev->mdev->aper.va + + le64_to_cpu(vqconfig[i].used_address); + } + + mvdev->dc->used_address_updated = 0; + + dev_dbg(mic_dev(mvdev), "%s: device type %d LINKUP\n", + __func__, mvdev->virtio_id); +} + +static inline void mic_virtio_device_reset(struct mic_vdev *mvdev) +{ + int i; + + dev_dbg(mic_dev(mvdev), "%s: status %d device type %d RESET\n", + __func__, mvdev->dd->status, mvdev->virtio_id); + + for (i = 0; i < mvdev->dd->num_vq; i++) + /* + * Avoid lockdep false positive. The + 1 is for the mic + * mutex which is held in the reset devices code path. + */ + mutex_lock_nested(&mvdev->mvr[i].vr_mutex, i + 1); + + /* 0 status means "reset" */ + mvdev->dd->status = 0; + mvdev->dc->vdev_reset = 0; + mvdev->dc->host_ack = 1; + + for (i = 0; i < mvdev->dd->num_vq; i++) { + struct vringh *vrh = &mvdev->mvr[i].vrh; + mvdev->mvr[i].vring.info->avail_idx = 0; + vrh->completed = 0; + vrh->last_avail_idx = 0; + vrh->last_used_idx = 0; + } + + for (i = 0; i < mvdev->dd->num_vq; i++) + mutex_unlock(&mvdev->mvr[i].vr_mutex); +} + +void mic_virtio_reset_devices(struct mic_device *mdev) +{ + struct list_head *pos, *tmp; + struct mic_vdev *mvdev; + + dev_dbg(mdev->sdev->parent, "%s\n", __func__); + + list_for_each_safe(pos, tmp, &mdev->vdev_list) { + mvdev = list_entry(pos, struct mic_vdev, list); + mic_virtio_device_reset(mvdev); + mvdev->poll_wake = 1; + wake_up(&mvdev->waitq); + } +} + +void mic_bh_handler(struct work_struct *work) +{ + struct mic_vdev *mvdev = container_of(work, struct mic_vdev, + virtio_bh_work); + + if (mvdev->dc->used_address_updated) + mic_virtio_init_post(mvdev); + + if (mvdev->dc->vdev_reset) + mic_virtio_device_reset(mvdev); + + mvdev->poll_wake = 1; + wake_up(&mvdev->waitq); +} + +static irqreturn_t mic_virtio_intr_handler(int irq, void *data) +{ + struct mic_vdev *mvdev = data; + struct mic_device *mdev = mvdev->mdev; + + mdev->ops->intr_workarounds(mdev); + schedule_work(&mvdev->virtio_bh_work); + return IRQ_HANDLED; +} + +int mic_virtio_config_change(struct mic_vdev *mvdev, + void __user *argp) +{ + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wake); + int ret = 0, retry, i; + struct mic_bootparam *bootparam = mvdev->mdev->dp; + s8 db = bootparam->h2c_config_db; + + mutex_lock(&mvdev->mdev->mic_mutex); + for (i = 0; i < mvdev->dd->num_vq; i++) + mutex_lock_nested(&mvdev->mvr[i].vr_mutex, i + 1); + + if (db == -1 || mvdev->dd->type == -1) { + ret = -EIO; + goto exit; + } + + if (copy_from_user(mic_vq_configspace(mvdev->dd), + argp, mvdev->dd->config_len)) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, -EFAULT); + ret = -EFAULT; + goto exit; + } + mvdev->dc->config_change = MIC_VIRTIO_PARAM_CONFIG_CHANGED; + mvdev->mdev->ops->send_intr(mvdev->mdev, db); + + for (retry = 100; retry--;) { + ret = wait_event_timeout(wake, + mvdev->dc->guest_ack, msecs_to_jiffies(100)); + if (ret) + break; + } + + dev_dbg(mic_dev(mvdev), + "%s %d retry: %d\n", __func__, __LINE__, retry); + mvdev->dc->config_change = 0; + mvdev->dc->guest_ack = 0; +exit: + for (i = 0; i < mvdev->dd->num_vq; i++) + mutex_unlock(&mvdev->mvr[i].vr_mutex); + mutex_unlock(&mvdev->mdev->mic_mutex); + return ret; +} + +static int mic_copy_dp_entry(struct mic_vdev *mvdev, + void __user *argp, + __u8 *type, + struct mic_device_desc **devpage) +{ + struct mic_device *mdev = mvdev->mdev; + struct mic_device_desc dd, *dd_config, *devp; + struct mic_vqconfig *vqconfig; + int ret = 0, i; + bool slot_found = false; + + if (copy_from_user(&dd, argp, sizeof(dd))) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, -EFAULT); + return -EFAULT; + } + + if (mic_aligned_desc_size(&dd) > MIC_MAX_DESC_BLK_SIZE || + dd.num_vq > MIC_MAX_VRINGS) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, -EINVAL); + return -EINVAL; + } + + dd_config = kmalloc(mic_desc_size(&dd), GFP_KERNEL); + if (dd_config == NULL) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, -ENOMEM); + return -ENOMEM; + } + if (copy_from_user(dd_config, argp, mic_desc_size(&dd))) { + ret = -EFAULT; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + goto exit; + } + + vqconfig = mic_vq_config(dd_config); + for (i = 0; i < dd.num_vq; i++) { + if (le16_to_cpu(vqconfig[i].num) > MIC_MAX_VRING_ENTRIES) { + ret = -EINVAL; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + goto exit; + } + } + + /* Find the first free device page entry */ + for (i = sizeof(struct mic_bootparam); + i < MIC_DP_SIZE - mic_total_desc_size(dd_config); + i += mic_total_desc_size(devp)) { + devp = mdev->dp + i; + if (devp->type == 0 || devp->type == -1) { + slot_found = true; + break; + } + } + if (!slot_found) { + ret = -EINVAL; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + goto exit; + } + /* + * Save off the type before doing the memcpy. Type will be set in the + * end after completing all initialization for the new device. + */ + *type = dd_config->type; + dd_config->type = 0; + memcpy(devp, dd_config, mic_desc_size(dd_config)); + + *devpage = devp; +exit: + kfree(dd_config); + return ret; +} + +static void mic_init_device_ctrl(struct mic_vdev *mvdev, + struct mic_device_desc *devpage) +{ + struct mic_device_ctrl *dc; + + dc = (void *)devpage + mic_aligned_desc_size(devpage); + + dc->config_change = 0; + dc->guest_ack = 0; + dc->vdev_reset = 0; + dc->host_ack = 0; + dc->used_address_updated = 0; + dc->c2h_vdev_db = -1; + dc->h2c_vdev_db = -1; + mvdev->dc = dc; +} + +int mic_virtio_add_device(struct mic_vdev *mvdev, + void __user *argp) +{ + struct mic_device *mdev = mvdev->mdev; + struct mic_device_desc *dd = NULL; + struct mic_vqconfig *vqconfig; + int vr_size, i, j, ret; + u8 type = 0; + s8 db; + char irqname[10]; + struct mic_bootparam *bootparam = mdev->dp; + u16 num; + dma_addr_t vr_addr; + + mutex_lock(&mdev->mic_mutex); + + ret = mic_copy_dp_entry(mvdev, argp, &type, &dd); + if (ret) { + mutex_unlock(&mdev->mic_mutex); + return ret; + } + + mic_init_device_ctrl(mvdev, dd); + + mvdev->dd = dd; + mvdev->virtio_id = type; + vqconfig = mic_vq_config(dd); + INIT_WORK(&mvdev->virtio_bh_work, mic_bh_handler); + + for (i = 0; i < dd->num_vq; i++) { + struct mic_vringh *mvr = &mvdev->mvr[i]; + struct mic_vring *vr = &mvdev->mvr[i].vring; + num = le16_to_cpu(vqconfig[i].num); + mutex_init(&mvr->vr_mutex); + vr_size = PAGE_ALIGN(vring_size(num, MIC_VIRTIO_RING_ALIGN) + + sizeof(struct _mic_vring_info)); + vr->va = (void *) + __get_free_pages(GFP_KERNEL | __GFP_ZERO, + get_order(vr_size)); + if (!vr->va) { + ret = -ENOMEM; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + goto err; + } + vr->len = vr_size; + vr->info = vr->va + vring_size(num, MIC_VIRTIO_RING_ALIGN); + vr->info->magic = cpu_to_le32(MIC_MAGIC + mvdev->virtio_id + i); + vr_addr = mic_map_single(mdev, vr->va, vr_size); + if (mic_map_error(vr_addr)) { + free_pages((unsigned long)vr->va, get_order(vr_size)); + ret = -ENOMEM; + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + goto err; + } + vqconfig[i].address = cpu_to_le64(vr_addr); + + vring_init(&vr->vr, num, vr->va, MIC_VIRTIO_RING_ALIGN); + ret = vringh_init_kern(&mvr->vrh, + *(u32 *)mic_vq_features(mvdev->dd), num, false, + vr->vr.desc, vr->vr.avail, vr->vr.used); + if (ret) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, ret); + goto err; + } + vringh_kiov_init(&mvr->riov, NULL, 0); + vringh_kiov_init(&mvr->wiov, NULL, 0); + mvr->head = USHRT_MAX; + mvr->mvdev = mvdev; + mvr->vrh.notify = mic_notify; + dev_dbg(mdev->sdev->parent, + "%s %d index %d va %p info %p vr_size 0x%x\n", + __func__, __LINE__, i, vr->va, vr->info, vr_size); + } + + snprintf(irqname, sizeof(irqname), "mic%dvirtio%d", mdev->id, + mvdev->virtio_id); + mvdev->virtio_db = mic_next_db(mdev); + mvdev->virtio_cookie = mic_request_irq(mdev, mic_virtio_intr_handler, + irqname, mvdev, mvdev->virtio_db, MIC_INTR_DB); + if (IS_ERR(mvdev->virtio_cookie)) { + ret = PTR_ERR(mvdev->virtio_cookie); + dev_dbg(mdev->sdev->parent, "request irq failed\n"); + goto err; + } + + mvdev->dc->c2h_vdev_db = mvdev->virtio_db; + + list_add_tail(&mvdev->list, &mdev->vdev_list); + /* + * Order the type update with previous stores. This write barrier + * is paired with the corresponding read barrier before the uncached + * system memory read of the type, on the card while scanning the + * device page. + */ + smp_wmb(); + dd->type = type; + + dev_dbg(mdev->sdev->parent, "Added virtio device id %d\n", dd->type); + + db = bootparam->h2c_config_db; + if (db != -1) + mdev->ops->send_intr(mdev, db); + mutex_unlock(&mdev->mic_mutex); + return 0; +err: + vqconfig = mic_vq_config(dd); + for (j = 0; j < i; j++) { + struct mic_vringh *mvr = &mvdev->mvr[j]; + mic_unmap_single(mdev, le64_to_cpu(vqconfig[j].address), + mvr->vring.len); + free_pages((unsigned long)mvr->vring.va, + get_order(mvr->vring.len)); + } + mutex_unlock(&mdev->mic_mutex); + return ret; +} + +void mic_virtio_del_device(struct mic_vdev *mvdev) +{ + struct list_head *pos, *tmp; + struct mic_vdev *tmp_mvdev; + struct mic_device *mdev = mvdev->mdev; + DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wake); + int i, ret, retry; + struct mic_vqconfig *vqconfig; + struct mic_bootparam *bootparam = mdev->dp; + s8 db; + + mutex_lock(&mdev->mic_mutex); + db = bootparam->h2c_config_db; + if (db == -1) + goto skip_hot_remove; + dev_dbg(mdev->sdev->parent, + "Requesting hot remove id %d\n", mvdev->virtio_id); + mvdev->dc->config_change = MIC_VIRTIO_PARAM_DEV_REMOVE; + mdev->ops->send_intr(mdev, db); + for (retry = 100; retry--;) { + ret = wait_event_timeout(wake, + mvdev->dc->guest_ack, msecs_to_jiffies(100)); + if (ret) + break; + } + dev_dbg(mdev->sdev->parent, + "Device id %d config_change %d guest_ack %d retry %d\n", + mvdev->virtio_id, mvdev->dc->config_change, + mvdev->dc->guest_ack, retry); + mvdev->dc->config_change = 0; + mvdev->dc->guest_ack = 0; +skip_hot_remove: + mic_free_irq(mdev, mvdev->virtio_cookie, mvdev); + flush_work(&mvdev->virtio_bh_work); + vqconfig = mic_vq_config(mvdev->dd); + for (i = 0; i < mvdev->dd->num_vq; i++) { + struct mic_vringh *mvr = &mvdev->mvr[i]; + vringh_kiov_cleanup(&mvr->riov); + vringh_kiov_cleanup(&mvr->wiov); + mic_unmap_single(mdev, le64_to_cpu(vqconfig[i].address), + mvr->vring.len); + free_pages((unsigned long)mvr->vring.va, + get_order(mvr->vring.len)); + } + + list_for_each_safe(pos, tmp, &mdev->vdev_list) { + tmp_mvdev = list_entry(pos, struct mic_vdev, list); + if (tmp_mvdev == mvdev) { + list_del(pos); + dev_dbg(mdev->sdev->parent, + "Removing virtio device id %d\n", + mvdev->virtio_id); + break; + } + } + /* + * Order the type update with previous stores. This write barrier + * is paired with the corresponding read barrier before the uncached + * system memory read of the type, on the card while scanning the + * device page. + */ + smp_wmb(); + mvdev->dd->type = -1; + mutex_unlock(&mdev->mic_mutex); +} diff --git a/drivers/misc/mic/host/mic_virtio.h b/drivers/misc/mic/host/mic_virtio.h new file mode 100644 index 00000000000..184f3c84805 --- /dev/null +++ b/drivers/misc/mic/host/mic_virtio.h @@ -0,0 +1,138 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#ifndef MIC_VIRTIO_H +#define MIC_VIRTIO_H + +#include <linux/virtio_config.h> +#include <linux/mic_ioctl.h> + +/* + * Note on endianness. + * 1. Host can be both BE or LE + * 2. Guest/card is LE. Host uses le_to_cpu to access desc/avail + * rings and ioreadXX/iowriteXX to access used ring. + * 3. Device page exposed by host to guest contains LE values. Guest + * accesses these using ioreadXX/iowriteXX etc. This way in general we + * obey the virtio spec according to which guest works with native + * endianness and host is aware of guest endianness and does all + * required endianness conversion. + * 4. Data provided from user space to guest (in ADD_DEVICE and + * CONFIG_CHANGE ioctl's) is not interpreted by the driver and should be + * in guest endianness. + */ + +/** + * struct mic_vringh - Virtio ring host information. + * + * @vring: The MIC vring used for setting up user space mappings. + * @vrh: The host VRINGH used for accessing the card vrings. + * @riov: The VRINGH read kernel IOV. + * @wiov: The VRINGH write kernel IOV. + * @head: The VRINGH head index address passed to vringh_getdesc_kern(..). + * @vr_mutex: Mutex for synchronizing access to the VRING. + * @mvdev: Back pointer to MIC virtio device for vringh_notify(..). + */ +struct mic_vringh { + struct mic_vring vring; + struct vringh vrh; + struct vringh_kiov riov; + struct vringh_kiov wiov; + u16 head; + struct mutex vr_mutex; + struct mic_vdev *mvdev; +}; + +/** + * struct mic_vdev - Host information for a card Virtio device. + * + * @virtio_id - Virtio device id. + * @waitq - Waitqueue to allow ring3 apps to poll. + * @mdev - Back pointer to host MIC device. + * @poll_wake - Used for waking up threads blocked in poll. + * @out_bytes - Debug stats for number of bytes copied from host to card. + * @in_bytes - Debug stats for number of bytes copied from card to host. + * @mvr - Store per VRING data structures. + * @virtio_bh_work - Work struct used to schedule virtio bottom half handling. + * @dd - Virtio device descriptor. + * @dc - Virtio device control fields. + * @list - List of Virtio devices. + * @virtio_db - The doorbell used by the card to interrupt the host. + * @virtio_cookie - The cookie returned while requesting interrupts. + */ +struct mic_vdev { + int virtio_id; + wait_queue_head_t waitq; + struct mic_device *mdev; + int poll_wake; + unsigned long out_bytes; + unsigned long in_bytes; + struct mic_vringh mvr[MIC_MAX_VRINGS]; + struct work_struct virtio_bh_work; + struct mic_device_desc *dd; + struct mic_device_ctrl *dc; + struct list_head list; + int virtio_db; + struct mic_irq *virtio_cookie; +}; + +void mic_virtio_uninit(struct mic_device *mdev); +int mic_virtio_add_device(struct mic_vdev *mvdev, + void __user *argp); +void mic_virtio_del_device(struct mic_vdev *mvdev); +int mic_virtio_config_change(struct mic_vdev *mvdev, + void __user *argp); +int mic_virtio_copy_desc(struct mic_vdev *mvdev, + struct mic_copy_desc *request); +void mic_virtio_reset_devices(struct mic_device *mdev); +void mic_bh_handler(struct work_struct *work); + +/* Helper API to obtain the MIC PCIe device */ +static inline struct device *mic_dev(struct mic_vdev *mvdev) +{ + return mvdev->mdev->sdev->parent; +} + +/* Helper API to check if a virtio device is initialized */ +static inline int mic_vdev_inited(struct mic_vdev *mvdev) +{ + /* Device has not been created yet */ + if (!mvdev->dd || !mvdev->dd->type) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, -EINVAL); + return -EINVAL; + } + + /* Device has been removed/deleted */ + if (mvdev->dd->type == -1) { + dev_err(mic_dev(mvdev), "%s %d err %d\n", + __func__, __LINE__, -ENODEV); + return -ENODEV; + } + + return 0; +} + +/* Helper API to check if a virtio device is running */ +static inline bool mic_vdevup(struct mic_vdev *mvdev) +{ + return !!mvdev->dd->status; +} +#endif diff --git a/drivers/misc/mic/host/mic_x100.c b/drivers/misc/mic/host/mic_x100.c new file mode 100644 index 00000000000..5562fdd3ef4 --- /dev/null +++ b/drivers/misc/mic/host/mic_x100.c @@ -0,0 +1,574 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#include <linux/fs.h> +#include <linux/pci.h> +#include <linux/sched.h> +#include <linux/firmware.h> +#include <linux/delay.h> + +#include "../common/mic_dev.h" +#include "mic_device.h" +#include "mic_x100.h" +#include "mic_smpt.h" + +/** + * mic_x100_write_spad - write to the scratchpad register + * @mdev: pointer to mic_device instance + * @idx: index to the scratchpad register, 0 based + * @val: the data value to put into the register + * + * This function allows writing of a 32bit value to the indexed scratchpad + * register. + * + * RETURNS: none. + */ +static void +mic_x100_write_spad(struct mic_device *mdev, unsigned int idx, u32 val) +{ + dev_dbg(mdev->sdev->parent, "Writing 0x%x to scratch pad index %d\n", + val, idx); + mic_mmio_write(&mdev->mmio, val, + MIC_X100_SBOX_BASE_ADDRESS + + MIC_X100_SBOX_SPAD0 + idx * 4); +} + +/** + * mic_x100_read_spad - read from the scratchpad register + * @mdev: pointer to mic_device instance + * @idx: index to scratchpad register, 0 based + * + * This function allows reading of the 32bit scratchpad register. + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static u32 +mic_x100_read_spad(struct mic_device *mdev, unsigned int idx) +{ + u32 val = mic_mmio_read(&mdev->mmio, + MIC_X100_SBOX_BASE_ADDRESS + + MIC_X100_SBOX_SPAD0 + idx * 4); + + dev_dbg(mdev->sdev->parent, + "Reading 0x%x from scratch pad index %d\n", val, idx); + return val; +} + +/** + * mic_x100_enable_interrupts - Enable interrupts. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_enable_interrupts(struct mic_device *mdev) +{ + u32 reg; + struct mic_mw *mw = &mdev->mmio; + u32 sice0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICE0; + u32 siac0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SIAC0; + + reg = mic_mmio_read(mw, sice0); + reg |= MIC_X100_SBOX_DBR_BITS(0xf) | MIC_X100_SBOX_DMA_BITS(0xff); + mic_mmio_write(mw, reg, sice0); + + /* + * Enable auto-clear when enabling interrupts. Applicable only for + * MSI-x. Legacy and MSI mode cannot have auto-clear enabled. + */ + if (mdev->irq_info.num_vectors > 1) { + reg = mic_mmio_read(mw, siac0); + reg |= MIC_X100_SBOX_DBR_BITS(0xf) | + MIC_X100_SBOX_DMA_BITS(0xff); + mic_mmio_write(mw, reg, siac0); + } +} + +/** + * mic_x100_disable_interrupts - Disable interrupts. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_disable_interrupts(struct mic_device *mdev) +{ + u32 reg; + struct mic_mw *mw = &mdev->mmio; + u32 sice0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICE0; + u32 siac0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SIAC0; + u32 sicc0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICC0; + + reg = mic_mmio_read(mw, sice0); + mic_mmio_write(mw, reg, sicc0); + + if (mdev->irq_info.num_vectors > 1) { + reg = mic_mmio_read(mw, siac0); + reg &= ~(MIC_X100_SBOX_DBR_BITS(0xf) | + MIC_X100_SBOX_DMA_BITS(0xff)); + mic_mmio_write(mw, reg, siac0); + } +} + +/** + * mic_x100_send_sbox_intr - Send an MIC_X100_SBOX interrupt to MIC. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_send_sbox_intr(struct mic_device *mdev, + int doorbell) +{ + struct mic_mw *mw = &mdev->mmio; + u64 apic_icr_offset = MIC_X100_SBOX_APICICR0 + doorbell * 8; + u32 apicicr_low = mic_mmio_read(mw, MIC_X100_SBOX_BASE_ADDRESS + + apic_icr_offset); + + /* for MIC we need to make sure we "hit" the send_icr bit (13) */ + apicicr_low = (apicicr_low | (1 << 13)); + + /* Ensure that the interrupt is ordered w.r.t. previous stores. */ + wmb(); + mic_mmio_write(mw, apicicr_low, + MIC_X100_SBOX_BASE_ADDRESS + apic_icr_offset); +} + +/** + * mic_x100_send_rdmasr_intr - Send an RDMASR interrupt to MIC. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_send_rdmasr_intr(struct mic_device *mdev, + int doorbell) +{ + int rdmasr_offset = MIC_X100_SBOX_RDMASR0 + (doorbell << 2); + /* Ensure that the interrupt is ordered w.r.t. previous stores. */ + wmb(); + mic_mmio_write(&mdev->mmio, 0, + MIC_X100_SBOX_BASE_ADDRESS + rdmasr_offset); +} + +/** + * __mic_x100_send_intr - Send interrupt to MIC. + * @mdev: pointer to mic_device instance + * @doorbell: doorbell number. + */ +static void mic_x100_send_intr(struct mic_device *mdev, int doorbell) +{ + int rdmasr_db; + if (doorbell < MIC_X100_NUM_SBOX_IRQ) { + mic_x100_send_sbox_intr(mdev, doorbell); + } else { + rdmasr_db = doorbell - MIC_X100_NUM_SBOX_IRQ + + MIC_X100_RDMASR_IRQ_BASE; + mic_x100_send_rdmasr_intr(mdev, rdmasr_db); + } +} + +/** + * mic_x100_ack_interrupt - Read the interrupt sources register and + * clear it. This function will be called in the MSI/INTx case. + * @mdev: Pointer to mic_device instance. + * + * Returns: bitmask of interrupt sources triggered. + */ +static u32 mic_x100_ack_interrupt(struct mic_device *mdev) +{ + u32 sicr0 = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SICR0; + u32 reg = mic_mmio_read(&mdev->mmio, sicr0); + mic_mmio_write(&mdev->mmio, reg, sicr0); + return reg; +} + +/** + * mic_x100_intr_workarounds - These hardware specific workarounds are + * to be invoked everytime an interrupt is handled. + * @mdev: Pointer to mic_device instance. + * + * Returns: none + */ +static void mic_x100_intr_workarounds(struct mic_device *mdev) +{ + struct mic_mw *mw = &mdev->mmio; + + /* Clear pending bit array. */ + if (MIC_A0_STEP == mdev->stepping) + mic_mmio_write(mw, 1, MIC_X100_SBOX_BASE_ADDRESS + + MIC_X100_SBOX_MSIXPBACR); + + if (mdev->stepping >= MIC_B0_STEP) + mdev->intr_ops->enable_interrupts(mdev); +} + +/** + * mic_x100_hw_intr_init - Initialize h/w specific interrupt + * information. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_hw_intr_init(struct mic_device *mdev) +{ + mdev->intr_info = (struct mic_intr_info *)mic_x100_intr_init; +} + +/** + * mic_x100_read_msi_to_src_map - read from the MSI mapping registers + * @mdev: pointer to mic_device instance + * @idx: index to the mapping register, 0 based + * + * This function allows reading of the 32bit MSI mapping register. + * + * RETURNS: The value in the register. + */ +static u32 +mic_x100_read_msi_to_src_map(struct mic_device *mdev, int idx) +{ + return mic_mmio_read(&mdev->mmio, + MIC_X100_SBOX_BASE_ADDRESS + + MIC_X100_SBOX_MXAR0 + idx * 4); +} + +/** + * mic_x100_program_msi_to_src_map - program the MSI mapping registers + * @mdev: pointer to mic_device instance + * @idx: index to the mapping register, 0 based + * @offset: The bit offset in the register that needs to be updated. + * @set: boolean specifying if the bit in the specified offset needs + * to be set or cleared. + * + * RETURNS: None. + */ +static void +mic_x100_program_msi_to_src_map(struct mic_device *mdev, + int idx, int offset, bool set) +{ + unsigned long reg; + struct mic_mw *mw = &mdev->mmio; + u32 mxar = MIC_X100_SBOX_BASE_ADDRESS + + MIC_X100_SBOX_MXAR0 + idx * 4; + + reg = mic_mmio_read(mw, mxar); + if (set) + __set_bit(offset, ®); + else + __clear_bit(offset, ®); + mic_mmio_write(mw, reg, mxar); +} + +/* + * mic_x100_reset_fw_ready - Reset Firmware ready status field. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_reset_fw_ready(struct mic_device *mdev) +{ + mdev->ops->write_spad(mdev, MIC_X100_DOWNLOAD_INFO, 0); +} + +/* + * mic_x100_is_fw_ready - Check if firmware is ready. + * @mdev: pointer to mic_device instance + */ +static bool mic_x100_is_fw_ready(struct mic_device *mdev) +{ + u32 scratch2 = mdev->ops->read_spad(mdev, MIC_X100_DOWNLOAD_INFO); + return MIC_X100_SPAD2_DOWNLOAD_STATUS(scratch2) ? true : false; +} + +/** + * mic_x100_get_apic_id - Get bootstrap APIC ID. + * @mdev: pointer to mic_device instance + */ +static u32 mic_x100_get_apic_id(struct mic_device *mdev) +{ + u32 scratch2 = 0; + + scratch2 = mdev->ops->read_spad(mdev, MIC_X100_DOWNLOAD_INFO); + return MIC_X100_SPAD2_APIC_ID(scratch2); +} + +/** + * mic_x100_send_firmware_intr - Send an interrupt to the firmware on MIC. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_send_firmware_intr(struct mic_device *mdev) +{ + u32 apicicr_low; + u64 apic_icr_offset = MIC_X100_SBOX_APICICR7; + int vector = MIC_X100_BSP_INTERRUPT_VECTOR; + struct mic_mw *mw = &mdev->mmio; + + /* + * For MIC we need to make sure we "hit" + * the send_icr bit (13). + */ + apicicr_low = (vector | (1 << 13)); + + mic_mmio_write(mw, mic_x100_get_apic_id(mdev), + MIC_X100_SBOX_BASE_ADDRESS + apic_icr_offset + 4); + + /* Ensure that the interrupt is ordered w.r.t. previous stores. */ + wmb(); + mic_mmio_write(mw, apicicr_low, + MIC_X100_SBOX_BASE_ADDRESS + apic_icr_offset); +} + +/** + * mic_x100_hw_reset - Reset the MIC device. + * @mdev: pointer to mic_device instance + */ +static void mic_x100_hw_reset(struct mic_device *mdev) +{ + u32 reset_reg; + u32 rgcr = MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_RGCR; + struct mic_mw *mw = &mdev->mmio; + + /* Ensure that the reset is ordered w.r.t. previous loads and stores */ + mb(); + /* Trigger reset */ + reset_reg = mic_mmio_read(mw, rgcr); + reset_reg |= 0x1; + mic_mmio_write(mw, reset_reg, rgcr); + /* + * It seems we really want to delay at least 1 second + * after touching reset to prevent a lot of problems. + */ + msleep(1000); +} + +/** + * mic_x100_load_command_line - Load command line to MIC. + * @mdev: pointer to mic_device instance + * @fw: the firmware image + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static int +mic_x100_load_command_line(struct mic_device *mdev, const struct firmware *fw) +{ + u32 len = 0; + u32 boot_mem; + char *buf; + void __iomem *cmd_line_va = mdev->aper.va + mdev->bootaddr + fw->size; +#define CMDLINE_SIZE 2048 + + boot_mem = mdev->aper.len >> 20; + buf = kzalloc(CMDLINE_SIZE, GFP_KERNEL); + if (!buf) { + dev_err(mdev->sdev->parent, + "%s %d allocation failed\n", __func__, __LINE__); + return -ENOMEM; + } + len += snprintf(buf, CMDLINE_SIZE - len, + " mem=%dM", boot_mem); + if (mdev->cmdline) + snprintf(buf + len, CMDLINE_SIZE - len, " %s", mdev->cmdline); + memcpy_toio(cmd_line_va, buf, strlen(buf) + 1); + kfree(buf); + return 0; +} + +/** + * mic_x100_load_ramdisk - Load ramdisk to MIC. + * @mdev: pointer to mic_device instance + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static int +mic_x100_load_ramdisk(struct mic_device *mdev) +{ + const struct firmware *fw; + int rc; + struct boot_params __iomem *bp = mdev->aper.va + mdev->bootaddr; + + rc = request_firmware(&fw, + mdev->ramdisk, mdev->sdev->parent); + if (rc < 0) { + dev_err(mdev->sdev->parent, + "ramdisk request_firmware failed: %d %s\n", + rc, mdev->ramdisk); + goto error; + } + /* + * Typically the bootaddr for card OS is 64M + * so copy over the ramdisk @ 128M. + */ + memcpy_toio(mdev->aper.va + (mdev->bootaddr << 1), fw->data, fw->size); + iowrite32(mdev->bootaddr << 1, &bp->hdr.ramdisk_image); + iowrite32(fw->size, &bp->hdr.ramdisk_size); + release_firmware(fw); +error: + return rc; +} + +/** + * mic_x100_get_boot_addr - Get MIC boot address. + * @mdev: pointer to mic_device instance + * + * This function is called during firmware load to determine + * the address at which the OS should be downloaded in card + * memory i.e. GDDR. + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static int +mic_x100_get_boot_addr(struct mic_device *mdev) +{ + u32 scratch2, boot_addr; + int rc = 0; + + scratch2 = mdev->ops->read_spad(mdev, MIC_X100_DOWNLOAD_INFO); + boot_addr = MIC_X100_SPAD2_DOWNLOAD_ADDR(scratch2); + dev_dbg(mdev->sdev->parent, "%s %d boot_addr 0x%x\n", + __func__, __LINE__, boot_addr); + if (boot_addr > (1 << 31)) { + dev_err(mdev->sdev->parent, + "incorrect bootaddr 0x%x\n", + boot_addr); + rc = -EINVAL; + goto error; + } + mdev->bootaddr = boot_addr; +error: + return rc; +} + +/** + * mic_x100_load_firmware - Load firmware to MIC. + * @mdev: pointer to mic_device instance + * @buf: buffer containing boot string including firmware/ramdisk path. + * + * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + */ +static int +mic_x100_load_firmware(struct mic_device *mdev, const char *buf) +{ + int rc; + const struct firmware *fw; + + rc = mic_x100_get_boot_addr(mdev); + if (rc) + goto error; + /* load OS */ + rc = request_firmware(&fw, mdev->firmware, mdev->sdev->parent); + if (rc < 0) { + dev_err(mdev->sdev->parent, + "ramdisk request_firmware failed: %d %s\n", + rc, mdev->firmware); + goto error; + } + if (mdev->bootaddr > mdev->aper.len - fw->size) { + rc = -EINVAL; + dev_err(mdev->sdev->parent, "%s %d rc %d bootaddr 0x%x\n", + __func__, __LINE__, rc, mdev->bootaddr); + release_firmware(fw); + goto error; + } + memcpy_toio(mdev->aper.va + mdev->bootaddr, fw->data, fw->size); + mdev->ops->write_spad(mdev, MIC_X100_FW_SIZE, fw->size); + if (!strcmp(mdev->bootmode, "elf")) + goto done; + /* load command line */ + rc = mic_x100_load_command_line(mdev, fw); + if (rc) { + dev_err(mdev->sdev->parent, "%s %d rc %d\n", + __func__, __LINE__, rc); + goto error; + } + release_firmware(fw); + /* load ramdisk */ + if (mdev->ramdisk) + rc = mic_x100_load_ramdisk(mdev); +error: + dev_dbg(mdev->sdev->parent, "%s %d rc %d\n", __func__, __LINE__, rc); +done: + return rc; +} + +/** + * mic_x100_get_postcode - Get postcode status from firmware. + * @mdev: pointer to mic_device instance + * + * RETURNS: postcode. + */ +static u32 mic_x100_get_postcode(struct mic_device *mdev) +{ + return mic_mmio_read(&mdev->mmio, MIC_X100_POSTCODE); +} + +/** + * mic_x100_smpt_set - Update an SMPT entry with a DMA address. + * @mdev: pointer to mic_device instance + * + * RETURNS: none. + */ +static void +mic_x100_smpt_set(struct mic_device *mdev, dma_addr_t dma_addr, u8 index) +{ +#define SNOOP_ON (0 << 0) +#define SNOOP_OFF (1 << 0) +/* + * Sbox Smpt Reg Bits: + * Bits 31:2 Host address + * Bits 1 RSVD + * Bits 0 No snoop + */ +#define BUILD_SMPT(NO_SNOOP, HOST_ADDR) \ + (u32)(((HOST_ADDR) << 2) | ((NO_SNOOP) & 0x01)) + + uint32_t smpt_reg_val = BUILD_SMPT(SNOOP_ON, + dma_addr >> mdev->smpt->info.page_shift); + mic_mmio_write(&mdev->mmio, smpt_reg_val, + MIC_X100_SBOX_BASE_ADDRESS + + MIC_X100_SBOX_SMPT00 + (4 * index)); +} + +/** + * mic_x100_smpt_hw_init - Initialize SMPT X100 specific fields. + * @mdev: pointer to mic_device instance + * + * RETURNS: none. + */ +static void mic_x100_smpt_hw_init(struct mic_device *mdev) +{ + struct mic_smpt_hw_info *info = &mdev->smpt->info; + + info->num_reg = 32; + info->page_shift = 34; + info->page_size = (1ULL << info->page_shift); + info->base = 0x8000000000ULL; +} + +struct mic_smpt_ops mic_x100_smpt_ops = { + .init = mic_x100_smpt_hw_init, + .set = mic_x100_smpt_set, +}; + +struct mic_hw_ops mic_x100_ops = { + .aper_bar = MIC_X100_APER_BAR, + .mmio_bar = MIC_X100_MMIO_BAR, + .read_spad = mic_x100_read_spad, + .write_spad = mic_x100_write_spad, + .send_intr = mic_x100_send_intr, + .ack_interrupt = mic_x100_ack_interrupt, + .intr_workarounds = mic_x100_intr_workarounds, + .reset = mic_x100_hw_reset, + .reset_fw_ready = mic_x100_reset_fw_ready, + .is_fw_ready = mic_x100_is_fw_ready, + .send_firmware_intr = mic_x100_send_firmware_intr, + .load_mic_fw = mic_x100_load_firmware, + .get_postcode = mic_x100_get_postcode, +}; + +struct mic_hw_intr_ops mic_x100_intr_ops = { + .intr_init = mic_x100_hw_intr_init, + .enable_interrupts = mic_x100_enable_interrupts, + .disable_interrupts = mic_x100_disable_interrupts, + .program_msi_to_src_map = mic_x100_program_msi_to_src_map, + .read_msi_to_src_map = mic_x100_read_msi_to_src_map, +}; diff --git a/drivers/misc/mic/host/mic_x100.h b/drivers/misc/mic/host/mic_x100.h new file mode 100644 index 00000000000..8b7daa182e5 --- /dev/null +++ b/drivers/misc/mic/host/mic_x100.h @@ -0,0 +1,98 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2013 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Host driver. + * + */ +#ifndef _MIC_X100_HW_H_ +#define _MIC_X100_HW_H_ + +#define MIC_X100_PCI_DEVICE_2250 0x2250 +#define MIC_X100_PCI_DEVICE_2251 0x2251 +#define MIC_X100_PCI_DEVICE_2252 0x2252 +#define MIC_X100_PCI_DEVICE_2253 0x2253 +#define MIC_X100_PCI_DEVICE_2254 0x2254 +#define MIC_X100_PCI_DEVICE_2255 0x2255 +#define MIC_X100_PCI_DEVICE_2256 0x2256 +#define MIC_X100_PCI_DEVICE_2257 0x2257 +#define MIC_X100_PCI_DEVICE_2258 0x2258 +#define MIC_X100_PCI_DEVICE_2259 0x2259 +#define MIC_X100_PCI_DEVICE_225a 0x225a +#define MIC_X100_PCI_DEVICE_225b 0x225b +#define MIC_X100_PCI_DEVICE_225c 0x225c +#define MIC_X100_PCI_DEVICE_225d 0x225d +#define MIC_X100_PCI_DEVICE_225e 0x225e + +#define MIC_X100_APER_BAR 0 +#define MIC_X100_MMIO_BAR 4 + +#define MIC_X100_SBOX_BASE_ADDRESS 0x00010000 +#define MIC_X100_SBOX_SPAD0 0x0000AB20 +#define MIC_X100_SBOX_SICR0_DBR(x) ((x) & 0xf) +#define MIC_X100_SBOX_SICR0_DMA(x) (((x) >> 8) & 0xff) +#define MIC_X100_SBOX_SICE0_DBR(x) ((x) & 0xf) +#define MIC_X100_SBOX_DBR_BITS(x) ((x) & 0xf) +#define MIC_X100_SBOX_SICE0_DMA(x) (((x) >> 8) & 0xff) +#define MIC_X100_SBOX_DMA_BITS(x) (((x) & 0xff) << 8) + +#define MIC_X100_SBOX_APICICR0 0x0000A9D0 +#define MIC_X100_SBOX_SICR0 0x00009004 +#define MIC_X100_SBOX_SICE0 0x0000900C +#define MIC_X100_SBOX_SICC0 0x00009010 +#define MIC_X100_SBOX_SIAC0 0x00009014 +#define MIC_X100_SBOX_MSIXPBACR 0x00009084 +#define MIC_X100_SBOX_MXAR0 0x00009044 +#define MIC_X100_SBOX_SMPT00 0x00003100 +#define MIC_X100_SBOX_RDMASR0 0x0000B180 + +#define MIC_X100_DOORBELL_IDX_START 0 +#define MIC_X100_NUM_DOORBELL 4 +#define MIC_X100_DMA_IDX_START 8 +#define MIC_X100_NUM_DMA 8 +#define MIC_X100_ERR_IDX_START 30 +#define MIC_X100_NUM_ERR 1 + +#define MIC_X100_NUM_SBOX_IRQ 8 +#define MIC_X100_NUM_RDMASR_IRQ 8 +#define MIC_X100_RDMASR_IRQ_BASE 17 +#define MIC_X100_SPAD2_DOWNLOAD_STATUS(x) ((x) & 0x1) +#define MIC_X100_SPAD2_APIC_ID(x) (((x) >> 1) & 0x1ff) +#define MIC_X100_SPAD2_DOWNLOAD_ADDR(x) ((x) & 0xfffff000) +#define MIC_X100_SBOX_APICICR7 0x0000AA08 +#define MIC_X100_SBOX_RGCR 0x00004010 +#define MIC_X100_SBOX_SDBIC0 0x0000CC90 +#define MIC_X100_DOWNLOAD_INFO 2 +#define MIC_X100_FW_SIZE 5 +#define MIC_X100_POSTCODE 0x242c + +static const u16 mic_x100_intr_init[] = { + MIC_X100_DOORBELL_IDX_START, + MIC_X100_DMA_IDX_START, + MIC_X100_ERR_IDX_START, + MIC_X100_NUM_DOORBELL, + MIC_X100_NUM_DMA, + MIC_X100_NUM_ERR, +}; + +/* Host->Card(bootstrap) Interrupt Vector */ +#define MIC_X100_BSP_INTERRUPT_VECTOR 229 + +extern struct mic_hw_ops mic_x100_ops; +extern struct mic_smpt_ops mic_x100_smpt_ops; +extern struct mic_hw_intr_ops mic_x100_intr_ops; + +#endif diff --git a/drivers/misc/pch_phub.c b/drivers/misc/pch_phub.c index c9f20dae185..956597321d2 100644 --- a/drivers/misc/pch_phub.c +++ b/drivers/misc/pch_phub.c @@ -633,23 +633,22 @@ static ssize_t show_pch_mac(struct device *dev, struct device_attribute *attr, static ssize_t store_pch_mac(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - u8 mac[6]; + u8 mac[ETH_ALEN]; ssize_t rom_size; struct pch_phub_reg *chip = dev_get_drvdata(dev); + int ret; - if (count != 18) + if (!mac_pton(buf, mac)) return -EINVAL; - sscanf(buf, "%02x:%02x:%02x:%02x:%02x:%02x", - (u32 *)&mac[0], (u32 *)&mac[1], (u32 *)&mac[2], (u32 *)&mac[3], - (u32 *)&mac[4], (u32 *)&mac[5]); - chip->pch_phub_extrom_base_address = pci_map_rom(chip->pdev, &rom_size); if (!chip->pch_phub_extrom_base_address) return -ENOMEM; - pch_phub_write_gbe_mac_addr(chip, mac); + ret = pch_phub_write_gbe_mac_addr(chip, mac); pci_unmap_rom(chip->pdev, chip->pch_phub_extrom_base_address); + if (ret) + return ret; return count; } @@ -666,11 +665,9 @@ static struct bin_attribute pch_bin_attr = { .write = pch_phub_bin_write, }; -static int __devinit pch_phub_probe(struct pci_dev *pdev, +static int pch_phub_probe(struct pci_dev *pdev, const struct pci_device_id *id) { - int retval; - int ret; struct pch_phub_reg *chip; @@ -713,13 +710,13 @@ static int __devinit pch_phub_probe(struct pci_dev *pdev, if (id->driver_data == 1) { /* EG20T PCH */ const char *board_name; - retval = sysfs_create_file(&pdev->dev.kobj, - &dev_attr_pch_mac.attr); - if (retval) + ret = sysfs_create_file(&pdev->dev.kobj, + &dev_attr_pch_mac.attr); + if (ret) goto err_sysfs_create; - retval = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); - if (retval) + ret = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); + if (ret) goto exit_bin_attr; pch_phub_read_modify_write_reg(chip, @@ -743,8 +740,8 @@ static int __devinit pch_phub_probe(struct pci_dev *pdev, chip->pch_opt_rom_start_address = PCH_PHUB_ROM_START_ADDR_EG20T; chip->pch_mac_start_address = PCH_PHUB_MAC_START_ADDR_EG20T; } else if (id->driver_data == 2) { /* ML7213 IOH */ - retval = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); - if (retval) + ret = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); + if (ret) goto err_sysfs_create; /* set the prefech value * Device2(USB OHCI #1/ USB EHCI #1/ USB Device):a @@ -766,12 +763,12 @@ static int __devinit pch_phub_probe(struct pci_dev *pdev, PCH_PHUB_ROM_START_ADDR_ML7223; chip->pch_mac_start_address = PCH_PHUB_MAC_START_ADDR_ML7223; } else if (id->driver_data == 4) { /* ML7223 IOH Bus-n*/ - retval = sysfs_create_file(&pdev->dev.kobj, - &dev_attr_pch_mac.attr); - if (retval) + ret = sysfs_create_file(&pdev->dev.kobj, + &dev_attr_pch_mac.attr); + if (ret) goto err_sysfs_create; - retval = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); - if (retval) + ret = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); + if (ret) goto exit_bin_attr; /* set the prefech value * Device2(USB OHCI #0,1,2,3/ USB EHCI #0):a @@ -783,13 +780,13 @@ static int __devinit pch_phub_probe(struct pci_dev *pdev, PCH_PHUB_ROM_START_ADDR_ML7223; chip->pch_mac_start_address = PCH_PHUB_MAC_START_ADDR_ML7223; } else if (id->driver_data == 5) { /* ML7831 */ - retval = sysfs_create_file(&pdev->dev.kobj, - &dev_attr_pch_mac.attr); - if (retval) + ret = sysfs_create_file(&pdev->dev.kobj, + &dev_attr_pch_mac.attr); + if (ret) goto err_sysfs_create; - retval = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); - if (retval) + ret = sysfs_create_bin_file(&pdev->dev.kobj, &pch_bin_attr); + if (ret) goto exit_bin_attr; /* set the prefech value */ @@ -819,7 +816,7 @@ err_pci_enable_dev: return ret; } -static void __devexit pch_phub_remove(struct pci_dev *pdev) +static void pch_phub_remove(struct pci_dev *pdev) { struct pch_phub_reg *chip = pci_get_drvdata(pdev); @@ -888,7 +885,7 @@ static struct pci_driver pch_phub_driver = { .name = "pch_phub", .id_table = pch_phub_pcidev_id, .probe = pch_phub_probe, - .remove = __devexit_p(pch_phub_remove), + .remove = pch_phub_remove, .suspend = pch_phub_suspend, .resume = pch_phub_resume }; diff --git a/drivers/misc/phantom.c b/drivers/misc/phantom.c index 21b28fc6d91..30754927fd8 100644 --- a/drivers/misc/phantom.c +++ b/drivers/misc/phantom.c @@ -324,7 +324,7 @@ static irqreturn_t phantom_isr(int irq, void *data) * Init and deinit driver */ -static unsigned int __devinit phantom_get_free(void) +static unsigned int phantom_get_free(void) { unsigned int i; @@ -335,7 +335,7 @@ static unsigned int __devinit phantom_get_free(void) return i; } -static int __devinit phantom_probe(struct pci_dev *pdev, +static int phantom_probe(struct pci_dev *pdev, const struct pci_device_id *pci_id) { struct phantom_device *pht; @@ -395,7 +395,7 @@ static int __devinit phantom_probe(struct pci_dev *pdev, iowrite32(0, pht->caddr + PHN_IRQCTL); ioread32(pht->caddr + PHN_IRQCTL); /* PCI posting */ retval = request_irq(pdev->irq, phantom_isr, - IRQF_SHARED | IRQF_DISABLED, "phantom", pht); + IRQF_SHARED, "phantom", pht); if (retval) { dev_err(&pdev->dev, "can't establish ISR\n"); goto err_unmo; @@ -435,7 +435,7 @@ err: return retval; } -static void __devexit phantom_remove(struct pci_dev *pdev) +static void phantom_remove(struct pci_dev *pdev) { struct phantom_device *pht = pci_get_drvdata(pdev); unsigned int minor = MINOR(pht->cdev.dev); @@ -487,7 +487,7 @@ static int phantom_resume(struct pci_dev *pdev) #define phantom_resume NULL #endif -static struct pci_device_id phantom_pci_tbl[] __devinitdata = { +static struct pci_device_id phantom_pci_tbl[] = { { .vendor = PCI_VENDOR_ID_PLX, .device = PCI_DEVICE_ID_PLX_9050, .subvendor = PCI_VENDOR_ID_PLX, .subdevice = PCI_DEVICE_ID_PLX_9050, .class = PCI_CLASS_BRIDGE_OTHER << 8, .class_mask = 0xffff00 }, @@ -499,7 +499,7 @@ static struct pci_driver phantom_pci_driver = { .name = "phantom", .id_table = phantom_pci_tbl, .probe = phantom_probe, - .remove = __devexit_p(phantom_remove), + .remove = phantom_remove, .suspend = phantom_suspend, .resume = phantom_resume }; diff --git a/drivers/misc/pti.c b/drivers/misc/pti.c index 4999b34b7a6..eda38cbe853 100644 --- a/drivers/misc/pti.c +++ b/drivers/misc/pti.c @@ -76,7 +76,7 @@ struct pti_dev { */ static DEFINE_MUTEX(alloclock); -static const struct pci_device_id pci_ids[] __devinitconst = { +static const struct pci_device_id pci_ids[] = { {PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x82B)}, {0} }; @@ -796,7 +796,7 @@ static const struct tty_port_operations tty_port_ops = { * 0 for success * otherwise, error */ -static int __devinit pti_pci_probe(struct pci_dev *pdev, +static int pti_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { unsigned int a; @@ -879,17 +879,19 @@ err: * PCI bus. * @pdev: variable containing pci info of PTI. */ -static void __devexit pti_pci_remove(struct pci_dev *pdev) +static void pti_pci_remove(struct pci_dev *pdev) { struct pti_dev *drv_data = pci_get_drvdata(pdev); + unsigned int a; unregister_console(&pti_console); - tty_unregister_device(pti_tty_driver, 0); - tty_unregister_device(pti_tty_driver, 1); + for (a = 0; a < PTITTY_MINOR_NUM; a++) { + tty_unregister_device(pti_tty_driver, a); + tty_port_destroy(&drv_data->port[a]); + } iounmap(drv_data->pti_ioaddr); - pci_set_drvdata(pdev, NULL); kfree(drv_data); pci_release_region(pdev, 1); pci_disable_device(pdev); @@ -901,7 +903,7 @@ static struct pci_driver pti_pci_driver = { .name = PCINAME, .id_table = pci_ids, .probe = pti_pci_probe, - .remove = __devexit_p(pti_pci_remove), + .remove = pti_pci_remove, }; /** diff --git a/drivers/misc/sgi-gru/grufault.c b/drivers/misc/sgi-gru/grufault.c index c4acac74725..f74fc0ca2ef 100644 --- a/drivers/misc/sgi-gru/grufault.c +++ b/drivers/misc/sgi-gru/grufault.c @@ -876,8 +876,9 @@ int gru_set_context_option(unsigned long arg) switch (req.op) { case sco_blade_chiplet: /* Select blade/chiplet for GRU context */ - if (req.val1 < -1 || req.val1 >= GRU_MAX_BLADES || !gru_base[req.val1] || - req.val0 < -1 || req.val0 >= GRU_CHIPLETS_PER_HUB) { + if (req.val0 < -1 || req.val0 >= GRU_CHIPLETS_PER_HUB || + req.val1 < -1 || req.val1 >= GRU_MAX_BLADES || + (req.val1 >= 0 && !gru_base[req.val1])) { ret = -EINVAL; } else { gts->ts_user_blade_id = req.val1; diff --git a/drivers/misc/sgi-gru/grufile.c b/drivers/misc/sgi-gru/grufile.c index 492c8cac69a..104a05f6b73 100644 --- a/drivers/misc/sgi-gru/grufile.c +++ b/drivers/misc/sgi-gru/grufile.c @@ -6,7 +6,7 @@ * This file supports the user system call for file open, close, mmap, etc. * This also incudes the driver initialization code. * - * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2008-2014 Silicon Graphics, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -58,6 +58,11 @@ static int max_user_cbrs, max_user_dsr_bytes; static struct miscdevice gru_miscdev; +static int gru_supported(void) +{ + return is_uv_system() && + (uv_hub_info->hub_revision < UV3_HUB_REVISION_BASE); +} /* * gru_vma_close @@ -172,6 +177,7 @@ static long gru_get_config_info(unsigned long arg) nodesperblade = 2; else nodesperblade = 1; + memset(&info, 0, sizeof(info)); info.cpus = num_online_cpus(); info.nodes = num_online_nodes(); info.blades = info.nodes / nodesperblade; @@ -517,7 +523,7 @@ static int __init gru_init(void) { int ret; - if (!is_uv_system()) + if (!gru_supported()) return 0; #if defined CONFIG_IA64 @@ -572,7 +578,7 @@ exit0: static void __exit gru_exit(void) { - if (!is_uv_system()) + if (!gru_supported()) return; gru_teardown_tlb_irqs(); diff --git a/drivers/misc/sgi-gru/grukdump.c b/drivers/misc/sgi-gru/grukdump.c index 9b2062d1732..a3700a56b8f 100644 --- a/drivers/misc/sgi-gru/grukdump.c +++ b/drivers/misc/sgi-gru/grukdump.c @@ -139,8 +139,11 @@ static int gru_dump_context(struct gru_state *gru, int ctxnum, ubuf += sizeof(hdr); ubufcch = ubuf; - if (gru_user_copy_handle(&ubuf, cch)) - goto fail; + if (gru_user_copy_handle(&ubuf, cch)) { + if (cch_locked) + unlock_cch_handle(cch); + return -EFAULT; + } if (cch_locked) ubufcch->delresp = 0; bytes = sizeof(hdr) + GRU_CACHE_LINE_BYTES; @@ -175,14 +178,10 @@ static int gru_dump_context(struct gru_state *gru, int ctxnum, hdr.cbrcnt = cbrcnt; hdr.dsrcnt = dsrcnt; hdr.cch_locked = cch_locked; - if (!ret && copy_to_user((void __user *)uhdr, &hdr, sizeof(hdr))) - ret = -EFAULT; - - return ret ? ret : bytes; + if (copy_to_user(uhdr, &hdr, sizeof(hdr))) + return -EFAULT; -fail: - unlock_cch_handle(cch); - return -EFAULT; + return bytes; } int gru_dump_chiplet_request(unsigned long arg) diff --git a/drivers/misc/sgi-gru/gruprocfs.c b/drivers/misc/sgi-gru/gruprocfs.c index 950dbe9ecb3..4f763592239 100644 --- a/drivers/misc/sgi-gru/gruprocfs.c +++ b/drivers/misc/sgi-gru/gruprocfs.c @@ -160,15 +160,11 @@ static int options_show(struct seq_file *s, void *p) static ssize_t options_write(struct file *file, const char __user *userbuf, size_t count, loff_t *data) { - char buf[20]; + int ret; - if (count >= sizeof(buf)) - return -EINVAL; - if (copy_from_user(buf, userbuf, count)) - return -EFAULT; - buf[count] = '\0'; - if (strict_strtoul(buf, 0, &gru_options)) - return -EINVAL; + ret = kstrtoul_from_user(userbuf, count, 0, &gru_options); + if (ret) + return ret; return count; } @@ -355,7 +351,7 @@ static void delete_proc_files(void) for (p = proc_files; p->name; p++) if (p->entry) remove_proc_entry(p->name, proc_gru); - remove_proc_entry("gru", proc_gru->parent); + proc_remove(proc_gru); } } diff --git a/drivers/misc/sgi-gru/grutlbpurge.c b/drivers/misc/sgi-gru/grutlbpurge.c index 240a6d36166..2129274ef7a 100644 --- a/drivers/misc/sgi-gru/grutlbpurge.c +++ b/drivers/misc/sgi-gru/grutlbpurge.c @@ -280,11 +280,10 @@ static struct mmu_notifier *mmu_find_ops(struct mm_struct *mm, const struct mmu_notifier_ops *ops) { struct mmu_notifier *mn, *gru_mn = NULL; - struct hlist_node *n; if (mm->mmu_notifier_mm) { rcu_read_lock(); - hlist_for_each_entry_rcu(mn, n, &mm->mmu_notifier_mm->list, + hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) if (mn->ops == ops) { gru_mn = mn; diff --git a/drivers/misc/sgi-xp/xpc_channel.c b/drivers/misc/sgi-xp/xpc_channel.c index 652593fc486..128d5615c80 100644 --- a/drivers/misc/sgi-xp/xpc_channel.c +++ b/drivers/misc/sgi-xp/xpc_channel.c @@ -828,6 +828,7 @@ enum xp_retval xpc_allocate_msg_wait(struct xpc_channel *ch) { enum xp_retval ret; + DEFINE_WAIT(wait); if (ch->flags & XPC_C_DISCONNECTING) { DBUG_ON(ch->reason == xpInterrupted); @@ -835,7 +836,9 @@ xpc_allocate_msg_wait(struct xpc_channel *ch) } atomic_inc(&ch->n_on_msg_allocate_wq); - ret = interruptible_sleep_on_timeout(&ch->msg_allocate_wq, 1); + prepare_to_wait(&ch->msg_allocate_wq, &wait, TASK_INTERRUPTIBLE); + ret = schedule_timeout(1); + finish_wait(&ch->msg_allocate_wq, &wait); atomic_dec(&ch->n_on_msg_allocate_wq); if (ch->flags & XPC_C_DISCONNECTING) { diff --git a/drivers/misc/sgi-xp/xpc_main.c b/drivers/misc/sgi-xp/xpc_main.c index 8d082b46426..82dc5748f87 100644 --- a/drivers/misc/sgi-xp/xpc_main.c +++ b/drivers/misc/sgi-xp/xpc_main.c @@ -53,6 +53,10 @@ #include <linux/kthread.h> #include "xpc.h" +#ifdef CONFIG_X86_64 +#include <asm/traps.h> +#endif + /* define two XPC debug device structures to be used with dev_dbg() et al */ struct device_driver xpc_dbg_name = { @@ -88,7 +92,7 @@ int xpc_disengage_timelimit = XPC_DISENGAGE_DEFAULT_TIMELIMIT; static int xpc_disengage_min_timelimit; /* = 0 */ static int xpc_disengage_max_timelimit = 120; -static ctl_table xpc_sys_xpc_hb_dir[] = { +static struct ctl_table xpc_sys_xpc_hb_dir[] = { { .procname = "hb_interval", .data = &xpc_hb_interval, @@ -107,7 +111,7 @@ static ctl_table xpc_sys_xpc_hb_dir[] = { .extra2 = &xpc_hb_check_max_interval}, {} }; -static ctl_table xpc_sys_xpc_dir[] = { +static struct ctl_table xpc_sys_xpc_dir[] = { { .procname = "hb", .mode = 0555, @@ -122,7 +126,7 @@ static ctl_table xpc_sys_xpc_dir[] = { .extra2 = &xpc_disengage_max_timelimit}, {} }; -static ctl_table xpc_sys_dir[] = { +static struct ctl_table xpc_sys_dir[] = { { .procname = "xpc", .mode = 0555, @@ -1079,6 +1083,9 @@ xpc_system_reboot(struct notifier_block *nb, unsigned long event, void *unused) return NOTIFY_DONE; } +/* Used to only allow one cpu to complete disconnect */ +static unsigned int xpc_die_disconnecting; + /* * Notify other partitions to deactivate from us by first disengaging from all * references to our memory. @@ -1092,6 +1099,9 @@ xpc_die_deactivate(void) long keep_waiting; long wait_to_print; + if (cmpxchg(&xpc_die_disconnecting, 0, 1)) + return; + /* keep xpc_hb_checker thread from doing anything (just in case) */ xpc_exiting = 1; @@ -1159,7 +1169,7 @@ xpc_die_deactivate(void) * about the lack of a heartbeat. */ static int -xpc_system_die(struct notifier_block *nb, unsigned long event, void *unused) +xpc_system_die(struct notifier_block *nb, unsigned long event, void *_die_args) { #ifdef CONFIG_IA64 /* !!! temporary kludge */ switch (event) { @@ -1191,7 +1201,27 @@ xpc_system_die(struct notifier_block *nb, unsigned long event, void *unused) break; } #else - xpc_die_deactivate(); + struct die_args *die_args = _die_args; + + switch (event) { + case DIE_TRAP: + if (die_args->trapnr == X86_TRAP_DF) + xpc_die_deactivate(); + + if (((die_args->trapnr == X86_TRAP_MF) || + (die_args->trapnr == X86_TRAP_XF)) && + !user_mode_vm(die_args->regs)) + xpc_die_deactivate(); + + break; + case DIE_INT3: + case DIE_DEBUG: + break; + case DIE_OOPS: + case DIE_GPF: + default: + xpc_die_deactivate(); + } #endif return NOTIFY_DONE; diff --git a/drivers/misc/sgi-xp/xpc_uv.c b/drivers/misc/sgi-xp/xpc_uv.c index b9e2000969f..95c894482fd 100644 --- a/drivers/misc/sgi-xp/xpc_uv.c +++ b/drivers/misc/sgi-xp/xpc_uv.c @@ -240,7 +240,7 @@ xpc_create_gru_mq_uv(unsigned int mq_size, int cpu, char *irq_name, nid = cpu_to_node(cpu); page = alloc_pages_exact_node(nid, - GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, + GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE, pg_order); if (page == NULL) { dev_err(xpc_part, "xpc_create_gru_mq_uv() failed to alloc %d " diff --git a/drivers/misc/spear13xx_pcie_gadget.c b/drivers/misc/spear13xx_pcie_gadget.c index 123ed98eec3..2e13614d41e 100644 --- a/drivers/misc/spear13xx_pcie_gadget.c +++ b/drivers/misc/spear13xx_pcie_gadget.c @@ -316,8 +316,12 @@ static ssize_t pcie_gadget_store_no_of_msi( struct spear_pcie_gadget_config *config, const char *buf, size_t count) { - if (strict_strtoul(buf, 0, &config->requested_msi)) - return -EINVAL; + int ret; + + ret = kstrtoul(buf, 0, &config->requested_msi); + if (ret) + return ret; + if (config->requested_msi > 32) config->requested_msi = 32; @@ -330,9 +334,11 @@ static ssize_t pcie_gadget_store_inta( { struct pcie_app_reg __iomem *app_reg = config->va_app_base; ulong en; + int ret; - if (strict_strtoul(buf, 0, &en)) - return -EINVAL; + ret = kstrtoul(buf, 0, &en); + if (ret) + return ret; if (en) writel(readl(&app_reg->app_ctrl_0) | (1 << SYS_INT_ID), @@ -351,9 +357,11 @@ static ssize_t pcie_gadget_store_send_msi( struct pcie_app_reg __iomem *app_reg = config->va_app_base; ulong vector; u32 ven_msi; + int ret; - if (strict_strtoul(buf, 0, &vector)) - return -EINVAL; + ret = kstrtoul(buf, 0, &vector); + if (ret) + return ret; if (!config->configured_msi) return -EINVAL; @@ -395,9 +403,11 @@ static ssize_t pcie_gadget_store_vendor_id( const char *buf, size_t count) { ulong id; + int ret; - if (strict_strtoul(buf, 0, &id)) - return -EINVAL; + ret = kstrtoul(buf, 0, &id); + if (ret) + return ret; spear_dbi_write_reg(config, PCI_VENDOR_ID, 2, id); @@ -420,9 +430,11 @@ static ssize_t pcie_gadget_store_device_id( const char *buf, size_t count) { ulong id; + int ret; - if (strict_strtoul(buf, 0, &id)) - return -EINVAL; + ret = kstrtoul(buf, 0, &id); + if (ret) + return ret; spear_dbi_write_reg(config, PCI_DEVICE_ID, 2, id); @@ -443,9 +455,12 @@ static ssize_t pcie_gadget_store_bar0_size( ulong size; u32 pos, pos1; u32 no_of_bit = 0; + int ret; + + ret = kstrtoul(buf, 0, &size); + if (ret) + return ret; - if (strict_strtoul(buf, 0, &size)) - return -EINVAL; /* min bar size is 256 */ if (size <= 0x100) size = 0x100; @@ -490,9 +505,11 @@ static ssize_t pcie_gadget_store_bar0_address( { struct pcie_app_reg __iomem *app_reg = config->va_app_base; ulong address; + int ret; - if (strict_strtoul(buf, 0, &address)) - return -EINVAL; + ret = kstrtoul(buf, 0, &address); + if (ret) + return ret; address &= ~(config->bar0_size - 1); if (config->va_bar0_address) @@ -518,9 +535,11 @@ static ssize_t pcie_gadget_store_bar0_rw_offset( const char *buf, size_t count) { ulong offset; + int ret; - if (strict_strtoul(buf, 0, &offset)) - return -EINVAL; + ret = kstrtoul(buf, 0, &offset); + if (ret) + return ret; if (offset % 4) return -EINVAL; @@ -549,9 +568,11 @@ static ssize_t pcie_gadget_store_bar0_data( const char *buf, size_t count) { ulong data; + int ret; - if (strict_strtoul(buf, 0, &data)) - return -EINVAL; + ret = kstrtoul(buf, 0, &data); + if (ret) + return ret; if (!config->va_bar0_address) return -ENOMEM; @@ -711,7 +732,7 @@ static void spear13xx_pcie_device_init(struct spear_pcie_gadget_config *config) spear_dbi_write_reg(config, PCI_INTERRUPT_LINE, 1, 1); } -static int __devinit spear_pcie_gadget_probe(struct platform_device *pdev) +static int spear_pcie_gadget_probe(struct platform_device *pdev) { struct resource *res0, *res1; unsigned int status = 0; @@ -776,7 +797,7 @@ static int __devinit spear_pcie_gadget_probe(struct platform_device *pdev) goto err_iounmap_app; } - dev_set_drvdata(&pdev->dev, target); + platform_set_drvdata(pdev, target); irq = platform_get_irq(pdev, 0); if (irq < 0) { @@ -814,9 +835,11 @@ static int __devinit spear_pcie_gadget_probe(struct platform_device *pdev) clk = clk_get_sys("pcie1", NULL); if (IS_ERR(clk)) { pr_err("%s:couldn't get clk for pcie1\n", __func__); + status = PTR_ERR(clk); goto err_irq; } - if (clk_enable(clk)) { + status = clk_enable(clk); + if (status) { pr_err("%s:couldn't enable clk for pcie1\n", __func__); goto err_irq; } @@ -828,9 +851,11 @@ static int __devinit spear_pcie_gadget_probe(struct platform_device *pdev) clk = clk_get_sys("pcie2", NULL); if (IS_ERR(clk)) { pr_err("%s:couldn't get clk for pcie2\n", __func__); + status = PTR_ERR(clk); goto err_irq; } - if (clk_enable(clk)) { + status = clk_enable(clk); + if (status) { pr_err("%s:couldn't enable clk for pcie2\n", __func__); goto err_irq; } @@ -853,7 +878,7 @@ err_rel_res0: return status; } -static int __devexit spear_pcie_gadget_remove(struct platform_device *pdev) +static int spear_pcie_gadget_remove(struct platform_device *pdev) { struct resource *res0, *res1; static struct pcie_gadget_target *target; @@ -863,7 +888,7 @@ static int __devexit spear_pcie_gadget_remove(struct platform_device *pdev) res0 = platform_get_resource(pdev, IORESOURCE_MEM, 0); res1 = platform_get_resource(pdev, IORESOURCE_MEM, 1); irq = platform_get_irq(pdev, 0); - target = dev_get_drvdata(&pdev->dev); + target = platform_get_drvdata(pdev); config = &target->config; free_irq(irq, NULL); diff --git a/drivers/misc/sram.c b/drivers/misc/sram.c new file mode 100644 index 00000000000..21181fa243d --- /dev/null +++ b/drivers/misc/sram.c @@ -0,0 +1,228 @@ +/* + * Generic on-chip SRAM allocation driver + * + * Copyright (C) 2012 Philipp Zabel, Pengutronix + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, + * MA 02110-1301, USA. + */ + +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/clk.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/of.h> +#include <linux/of_address.h> +#include <linux/list.h> +#include <linux/list_sort.h> +#include <linux/platform_device.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/genalloc.h> + +#define SRAM_GRANULARITY 32 + +struct sram_dev { + struct gen_pool *pool; + struct clk *clk; +}; + +struct sram_reserve { + struct list_head list; + u32 start; + u32 size; +}; + +static int sram_reserve_cmp(void *priv, struct list_head *a, + struct list_head *b) +{ + struct sram_reserve *ra = list_entry(a, struct sram_reserve, list); + struct sram_reserve *rb = list_entry(b, struct sram_reserve, list); + + return ra->start - rb->start; +} + +static int sram_probe(struct platform_device *pdev) +{ + void __iomem *virt_base; + struct sram_dev *sram; + struct resource *res; + struct device_node *np = pdev->dev.of_node, *child; + unsigned long size, cur_start, cur_size; + struct sram_reserve *rblocks, *block; + struct list_head reserve_list; + unsigned int nblocks; + int ret; + + INIT_LIST_HEAD(&reserve_list); + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + virt_base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(virt_base)) + return PTR_ERR(virt_base); + + size = resource_size(res); + + sram = devm_kzalloc(&pdev->dev, sizeof(*sram), GFP_KERNEL); + if (!sram) + return -ENOMEM; + + sram->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(sram->clk)) + sram->clk = NULL; + else + clk_prepare_enable(sram->clk); + + sram->pool = devm_gen_pool_create(&pdev->dev, ilog2(SRAM_GRANULARITY), -1); + if (!sram->pool) + return -ENOMEM; + + /* + * We need an additional block to mark the end of the memory region + * after the reserved blocks from the dt are processed. + */ + nblocks = (np) ? of_get_available_child_count(np) + 1 : 1; + rblocks = kmalloc((nblocks) * sizeof(*rblocks), GFP_KERNEL); + if (!rblocks) { + ret = -ENOMEM; + goto err_alloc; + } + + block = &rblocks[0]; + for_each_available_child_of_node(np, child) { + struct resource child_res; + + ret = of_address_to_resource(child, 0, &child_res); + if (ret < 0) { + dev_err(&pdev->dev, + "could not get address for node %s\n", + child->full_name); + goto err_chunks; + } + + if (child_res.start < res->start || child_res.end > res->end) { + dev_err(&pdev->dev, + "reserved block %s outside the sram area\n", + child->full_name); + ret = -EINVAL; + goto err_chunks; + } + + block->start = child_res.start - res->start; + block->size = resource_size(&child_res); + list_add_tail(&block->list, &reserve_list); + + dev_dbg(&pdev->dev, "found reserved block 0x%x-0x%x\n", + block->start, + block->start + block->size); + + block++; + } + + /* the last chunk marks the end of the region */ + rblocks[nblocks - 1].start = size; + rblocks[nblocks - 1].size = 0; + list_add_tail(&rblocks[nblocks - 1].list, &reserve_list); + + list_sort(NULL, &reserve_list, sram_reserve_cmp); + + cur_start = 0; + + list_for_each_entry(block, &reserve_list, list) { + /* can only happen if sections overlap */ + if (block->start < cur_start) { + dev_err(&pdev->dev, + "block at 0x%x starts after current offset 0x%lx\n", + block->start, cur_start); + ret = -EINVAL; + goto err_chunks; + } + + /* current start is in a reserved block, so continue after it */ + if (block->start == cur_start) { + cur_start = block->start + block->size; + continue; + } + + /* + * allocate the space between the current starting + * address and the following reserved block, or the + * end of the region. + */ + cur_size = block->start - cur_start; + + dev_dbg(&pdev->dev, "adding chunk 0x%lx-0x%lx\n", + cur_start, cur_start + cur_size); + ret = gen_pool_add_virt(sram->pool, + (unsigned long)virt_base + cur_start, + res->start + cur_start, cur_size, -1); + if (ret < 0) + goto err_chunks; + + /* next allocation after this reserved block */ + cur_start = block->start + block->size; + } + + kfree(rblocks); + + platform_set_drvdata(pdev, sram); + + dev_dbg(&pdev->dev, "SRAM pool: %ld KiB @ 0x%p\n", size / 1024, virt_base); + + return 0; + +err_chunks: + kfree(rblocks); +err_alloc: + if (sram->clk) + clk_disable_unprepare(sram->clk); + return ret; +} + +static int sram_remove(struct platform_device *pdev) +{ + struct sram_dev *sram = platform_get_drvdata(pdev); + + if (gen_pool_avail(sram->pool) < gen_pool_size(sram->pool)) + dev_dbg(&pdev->dev, "removed while SRAM allocated\n"); + + if (sram->clk) + clk_disable_unprepare(sram->clk); + + return 0; +} + +#ifdef CONFIG_OF +static struct of_device_id sram_dt_ids[] = { + { .compatible = "mmio-sram" }, + {} +}; +#endif + +static struct platform_driver sram_driver = { + .driver = { + .name = "sram", + .of_match_table = of_match_ptr(sram_dt_ids), + }, + .probe = sram_probe, + .remove = sram_remove, +}; + +static int __init sram_init(void) +{ + return platform_driver_register(&sram_driver); +} + +postcore_initcall(sram_init); diff --git a/drivers/misc/ti-st/Kconfig b/drivers/misc/ti-st/Kconfig index abb5de1afce..f34dcc51473 100644 --- a/drivers/misc/ti-st/Kconfig +++ b/drivers/misc/ti-st/Kconfig @@ -5,7 +5,7 @@ menu "Texas Instruments shared transport line discipline" config TI_ST tristate "Shared transport core driver" - depends on NET && GPIOLIB + depends on NET && GPIOLIB && TTY select FW_LOADER help This enables the shared transport core driver for TI diff --git a/drivers/misc/ti-st/st_core.c b/drivers/misc/ti-st/st_core.c index 46937b10726..1972d57aadb 100644 --- a/drivers/misc/ti-st/st_core.c +++ b/drivers/misc/ti-st/st_core.c @@ -22,7 +22,6 @@ #define pr_fmt(fmt) "(stc): " fmt #include <linux/module.h> #include <linux/kernel.h> -#include <linux/init.h> #include <linux/tty.h> #include <linux/seq_file.h> @@ -240,7 +239,8 @@ void st_int_recv(void *disc_data, char *ptr; struct st_proto_s *proto; unsigned short payload_len = 0; - int len = 0, type = 0; + int len = 0; + unsigned char type = 0; unsigned char *plen; struct st_data_s *st_gdata = (struct st_data_s *)disc_data; unsigned long flags; @@ -511,7 +511,6 @@ long st_register(struct st_proto_s *new_proto) unsigned long flags = 0; st_kim_ref(&st_gdata, 0); - pr_info("%s(%d) ", __func__, new_proto->chnl_id); if (st_gdata == NULL || new_proto == NULL || new_proto->recv == NULL || new_proto->reg_complete_cb == NULL) { pr_err("gdata/new_proto/recv or reg_complete_cb not ready"); @@ -562,7 +561,9 @@ long st_register(struct st_proto_s *new_proto) if ((st_gdata->protos_registered != ST_EMPTY) && (test_bit(ST_REG_PENDING, &st_gdata->st_state))) { pr_err(" KIM failure complete callback "); + spin_lock_irqsave(&st_gdata->lock, flags); st_reg_complete(st_gdata, err); + spin_unlock_irqrestore(&st_gdata->lock, flags); clear_bit(ST_REG_PENDING, &st_gdata->st_state); } return -EINVAL; @@ -810,7 +811,7 @@ static void st_tty_flush_buffer(struct tty_struct *tty) kfree_skb(st_gdata->tx_skb); st_gdata->tx_skb = NULL; - tty->ops->flush_buffer(tty); + tty_driver_flush_buffer(tty); return; } diff --git a/drivers/misc/ti-st/st_kim.c b/drivers/misc/ti-st/st_kim.c index 04a819944f6..9d3dbb28734 100644 --- a/drivers/misc/ti-st/st_kim.c +++ b/drivers/misc/ti-st/st_kim.c @@ -218,7 +218,7 @@ static long read_local_version(struct kim_data_s *kim_gdata, char *bts_scr_name) pr_debug("%s", __func__); - INIT_COMPLETION(kim_gdata->kim_rcvd); + reinit_completion(&kim_gdata->kim_rcvd); if (4 != st_int_write(kim_gdata->core_data, read_ver_cmd, 4)) { pr_err("kim: couldn't write 4 bytes"); return -EIO; @@ -229,7 +229,7 @@ static long read_local_version(struct kim_data_s *kim_gdata, char *bts_scr_name) pr_err(" waiting for ver info- timed out "); return -ETIMEDOUT; } - INIT_COMPLETION(kim_gdata->kim_rcvd); + reinit_completion(&kim_gdata->kim_rcvd); /* the positions 12 & 13 in the response buffer provide with the * chip, major & minor numbers */ @@ -362,7 +362,7 @@ static long download_firmware(struct kim_data_s *kim_gdata) /* reinit completion before sending for the * relevant wait */ - INIT_COMPLETION(kim_gdata->kim_rcvd); + reinit_completion(&kim_gdata->kim_rcvd); /* * Free space found in uart buffer, call st_int_write @@ -398,7 +398,7 @@ static long download_firmware(struct kim_data_s *kim_gdata) release_firmware(kim_gdata->fw_entry); return -ETIMEDOUT; } - INIT_COMPLETION(kim_gdata->kim_rcvd); + reinit_completion(&kim_gdata->kim_rcvd); break; case ACTION_DELAY: /* sleep */ pr_info("sleep command in scr"); @@ -468,8 +468,13 @@ long st_kim_start(void *kim_data) if (pdata->chip_enable) pdata->chip_enable(kim_gdata); + /* Configure BT nShutdown to HIGH state */ + gpio_set_value(kim_gdata->nshutdown, GPIO_LOW); + mdelay(5); /* FIXME: a proper toggle */ + gpio_set_value(kim_gdata->nshutdown, GPIO_HIGH); + mdelay(100); /* re-initialize the completion */ - INIT_COMPLETION(kim_gdata->ldisc_installed); + reinit_completion(&kim_gdata->ldisc_installed); /* send notification to UIM */ kim_gdata->ldisc_install = 1; pr_info("ldisc_install = 1"); @@ -509,7 +514,8 @@ long st_kim_start(void *kim_data) * (b) upon failure to either install ldisc or download firmware. * The function is responsible to (a) notify UIM about un-installation, * (b) flush UART if the ldisc was installed. - * (c) invoke platform's chip disabling routine. + * (c) reset BT_EN - pull down nshutdown at the end. + * (d) invoke platform's chip disabling routine. */ long st_kim_stop(void *kim_data) { @@ -519,13 +525,12 @@ long st_kim_stop(void *kim_data) kim_gdata->kim_pdev->dev.platform_data; struct tty_struct *tty = kim_gdata->core_data->tty; - INIT_COMPLETION(kim_gdata->ldisc_installed); + reinit_completion(&kim_gdata->ldisc_installed); if (tty) { /* can be called before ldisc is installed */ /* Flush any pending characters in the driver and discipline. */ tty_ldisc_flush(tty); tty_driver_flush_buffer(tty); - tty->ops->flush_buffer(tty); } /* send uninstall notification to UIM */ @@ -541,6 +546,13 @@ long st_kim_stop(void *kim_data) err = -ETIMEDOUT; } + /* By default configure BT nShutdown to LOW state */ + gpio_set_value(kim_gdata->nshutdown, GPIO_LOW); + mdelay(1); + gpio_set_value(kim_gdata->nshutdown, GPIO_HIGH); + mdelay(1); + gpio_set_value(kim_gdata->nshutdown, GPIO_LOW); + /* platform specific disable */ if (pdata->chip_disable) pdata->chip_disable(kim_gdata); @@ -667,7 +679,7 @@ void st_kim_ref(struct st_data_s **core_data, int id) *core_data = NULL; return; } - kim_gdata = dev_get_drvdata(&pdev->dev); + kim_gdata = platform_get_drvdata(pdev); *core_data = kim_gdata->core_data; } @@ -705,9 +717,9 @@ static const struct file_operations list_debugfs_fops = { static struct dentry *kim_debugfs_dir; static int kim_probe(struct platform_device *pdev) { - long status; struct kim_data_s *kim_gdata; struct ti_st_plat_data *pdata = pdev->dev.platform_data; + int err; if ((pdev->id != -1) && (pdev->id < MAX_ST_DEVICES)) { /* multiple devices could exist */ @@ -722,26 +734,41 @@ static int kim_probe(struct platform_device *pdev) pr_err("no mem to allocate"); return -ENOMEM; } - dev_set_drvdata(&pdev->dev, kim_gdata); + platform_set_drvdata(pdev, kim_gdata); - status = st_core_init(&kim_gdata->core_data); - if (status != 0) { + err = st_core_init(&kim_gdata->core_data); + if (err != 0) { pr_err(" ST core init failed"); - return -EIO; + err = -EIO; + goto err_core_init; } /* refer to itself */ kim_gdata->core_data->kim_data = kim_gdata; + /* Claim the chip enable nShutdown gpio from the system */ + kim_gdata->nshutdown = pdata->nshutdown_gpio; + err = gpio_request(kim_gdata->nshutdown, "kim"); + if (unlikely(err)) { + pr_err(" gpio %ld request failed ", kim_gdata->nshutdown); + return err; + } + + /* Configure nShutdown GPIO as output=0 */ + err = gpio_direction_output(kim_gdata->nshutdown, 0); + if (unlikely(err)) { + pr_err(" unable to configure gpio %ld", kim_gdata->nshutdown); + return err; + } /* get reference of pdev for request_firmware */ kim_gdata->kim_pdev = pdev; init_completion(&kim_gdata->kim_rcvd); init_completion(&kim_gdata->ldisc_installed); - status = sysfs_create_group(&pdev->dev.kobj, &uim_attr_grp); - if (status) { + err = sysfs_create_group(&pdev->dev.kobj, &uim_attr_grp); + if (err) { pr_err("failed to create sysfs entries"); - return status; + goto err_sysfs_group; } /* copying platform data */ @@ -753,8 +780,8 @@ static int kim_probe(struct platform_device *pdev) kim_debugfs_dir = debugfs_create_dir("ti-st", NULL); if (IS_ERR(kim_debugfs_dir)) { pr_err(" debugfs entries creation failed "); - kim_debugfs_dir = NULL; - return -EIO; + err = -EIO; + goto err_debugfs_dir; } debugfs_create_file("version", S_IRUGO, kim_debugfs_dir, @@ -763,13 +790,32 @@ static int kim_probe(struct platform_device *pdev) kim_gdata, &list_debugfs_fops); pr_info(" debugfs entries created "); return 0; + +err_debugfs_dir: + sysfs_remove_group(&pdev->dev.kobj, &uim_attr_grp); + +err_sysfs_group: + st_core_exit(kim_gdata->core_data); + +err_core_init: + kfree(kim_gdata); + + return err; } static int kim_remove(struct platform_device *pdev) { + /* free the GPIOs requested */ + struct ti_st_plat_data *pdata = pdev->dev.platform_data; struct kim_data_s *kim_gdata; - kim_gdata = dev_get_drvdata(&pdev->dev); + kim_gdata = platform_get_drvdata(pdev); + + /* Free the Bluetooth/FM/GPIO + * nShutdown gpio from the system + */ + gpio_free(pdata->nshutdown_gpio); + pr_info("nshutdown GPIO Freed"); debugfs_remove_recursive(kim_debugfs_dir); sysfs_remove_group(&pdev->dev.kobj, &uim_attr_grp); diff --git a/drivers/misc/ti_dac7512.c b/drivers/misc/ti_dac7512.c index 5acbba120de..cb0289b44a1 100644 --- a/drivers/misc/ti_dac7512.c +++ b/drivers/misc/ti_dac7512.c @@ -20,11 +20,8 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <linux/spi/spi.h> - -#define DAC7512_DRV_NAME "dac7512" -#define DRIVER_VERSION "1.0" +#include <linux/of.h> static ssize_t dac7512_store_val(struct device *dev, struct device_attribute *attr, @@ -33,9 +30,11 @@ static ssize_t dac7512_store_val(struct device *dev, struct spi_device *spi = to_spi_device(dev); unsigned char tmp[2]; unsigned long val; + int ret; - if (strict_strtoul(buf, 10, &val) < 0) - return -EINVAL; + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; tmp[0] = val >> 8; tmp[1] = val & 0xff; @@ -54,7 +53,7 @@ static const struct attribute_group dac7512_attr_group = { .attrs = dac7512_attributes, }; -static int __devinit dac7512_probe(struct spi_device *spi) +static int dac7512_probe(struct spi_device *spi) { int ret; @@ -67,19 +66,35 @@ static int __devinit dac7512_probe(struct spi_device *spi) return sysfs_create_group(&spi->dev.kobj, &dac7512_attr_group); } -static int __devexit dac7512_remove(struct spi_device *spi) +static int dac7512_remove(struct spi_device *spi) { sysfs_remove_group(&spi->dev.kobj, &dac7512_attr_group); return 0; } +static const struct spi_device_id dac7512_id_table[] = { + { "dac7512", 0 }, + { } +}; +MODULE_DEVICE_TABLE(spi, dac7512_id_table); + +#ifdef CONFIG_OF +static const struct of_device_id dac7512_of_match[] = { + { .compatible = "ti,dac7512", }, + { } +}; +MODULE_DEVICE_TABLE(of, dac7512_of_match); +#endif + static struct spi_driver dac7512_driver = { .driver = { - .name = DAC7512_DRV_NAME, + .name = "dac7512", .owner = THIS_MODULE, + .of_match_table = of_match_ptr(dac7512_of_match), }, .probe = dac7512_probe, - .remove = __devexit_p(dac7512_remove), + .remove = dac7512_remove, + .id_table = dac7512_id_table, }; module_spi_driver(dac7512_driver); @@ -87,4 +102,3 @@ module_spi_driver(dac7512_driver); MODULE_AUTHOR("Daniel Mack <daniel@caiaq.de>"); MODULE_DESCRIPTION("DAC7512 16-bit DAC"); MODULE_LICENSE("GPL v2"); -MODULE_VERSION(DRIVER_VERSION); diff --git a/drivers/misc/tifm_7xx1.c b/drivers/misc/tifm_7xx1.c index f8d6654391e..a606c8901e1 100644 --- a/drivers/misc/tifm_7xx1.c +++ b/drivers/misc/tifm_7xx1.c @@ -356,8 +356,10 @@ static int tifm_7xx1_probe(struct pci_dev *dev, pci_set_drvdata(dev, fm); fm->addr = pci_ioremap_bar(dev, 0); - if (!fm->addr) + if (!fm->addr) { + rc = -ENODEV; goto err_out_free; + } rc = request_irq(dev->irq, tifm_7xx1_isr, IRQF_SHARED, DRIVER_NAME, fm); if (rc) @@ -378,7 +380,6 @@ err_out_irq: err_out_unmap: iounmap(fm->addr); err_out_free: - pci_set_drvdata(dev, NULL); tifm_free_adapter(fm); err_out_int: pci_intx(dev, 0); @@ -405,8 +406,6 @@ static void tifm_7xx1_remove(struct pci_dev *dev) for (cnt = 0; cnt < fm->num_sockets; cnt++) tifm_7xx1_sock_power_off(tifm_7xx1_sock_addr(fm->addr, cnt)); - pci_set_drvdata(dev, NULL); - iounmap(fm->addr); pci_intx(dev, 0); pci_release_regions(dev); diff --git a/drivers/misc/tifm_core.c b/drivers/misc/tifm_core.c index 0bd5349b042..a511b2a713b 100644 --- a/drivers/misc/tifm_core.c +++ b/drivers/misc/tifm_core.c @@ -145,15 +145,17 @@ static ssize_t type_show(struct device *dev, struct device_attribute *attr, struct tifm_dev *sock = container_of(dev, struct tifm_dev, dev); return sprintf(buf, "%x", sock->type); } +static DEVICE_ATTR_RO(type); -static struct device_attribute tifm_dev_attrs[] = { - __ATTR(type, S_IRUGO, type_show, NULL), - __ATTR_NULL +static struct attribute *tifm_dev_attrs[] = { + &dev_attr_type.attr, + NULL, }; +ATTRIBUTE_GROUPS(tifm_dev); static struct bus_type tifm_bus_type = { .name = "tifm", - .dev_attrs = tifm_dev_attrs, + .dev_groups = tifm_dev_groups, .match = tifm_bus_match, .uevent = tifm_uevent, .probe = tifm_device_probe, @@ -196,13 +198,14 @@ int tifm_add_adapter(struct tifm_adapter *fm) { int rc; - if (!idr_pre_get(&tifm_adapter_idr, GFP_KERNEL)) - return -ENOMEM; - + idr_preload(GFP_KERNEL); spin_lock(&tifm_adapter_lock); - rc = idr_get_new(&tifm_adapter_idr, fm, &fm->id); + rc = idr_alloc(&tifm_adapter_idr, fm, 0, 0, GFP_NOWAIT); + if (rc >= 0) + fm->id = rc; spin_unlock(&tifm_adapter_lock); - if (rc) + idr_preload_end(); + if (rc < 0) return rc; dev_set_name(&fm->dev, "tifm%u", fm->id); diff --git a/drivers/misc/tsl2550.c b/drivers/misc/tsl2550.c index 0beb298a17d..b00335652e5 100644 --- a/drivers/misc/tsl2550.c +++ b/drivers/misc/tsl2550.c @@ -20,7 +20,6 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <linux/slab.h> #include <linux/i2c.h> #include <linux/mutex.h> @@ -204,7 +203,7 @@ static ssize_t tsl2550_store_power_state(struct device *dev, unsigned long val = simple_strtoul(buf, NULL, 10); int ret; - if (val < 0 || val > 1) + if (val > 1) return -EINVAL; mutex_lock(&data->update_lock); @@ -236,7 +235,7 @@ static ssize_t tsl2550_store_operating_mode(struct device *dev, unsigned long val = simple_strtoul(buf, NULL, 10); int ret; - if (val < 0 || val > 1) + if (val > 1) return -EINVAL; if (data->power_state == 0) @@ -347,7 +346,7 @@ static int tsl2550_init_client(struct i2c_client *client) */ static struct i2c_driver tsl2550_driver; -static int __devinit tsl2550_probe(struct i2c_client *client, +static int tsl2550_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent); @@ -405,7 +404,7 @@ exit: return err; } -static int __devexit tsl2550_remove(struct i2c_client *client) +static int tsl2550_remove(struct i2c_client *client) { sysfs_remove_group(&client->dev.kobj, &tsl2550_attr_group); @@ -417,24 +416,26 @@ static int __devexit tsl2550_remove(struct i2c_client *client) return 0; } -#ifdef CONFIG_PM +#ifdef CONFIG_PM_SLEEP -static int tsl2550_suspend(struct i2c_client *client, pm_message_t mesg) +static int tsl2550_suspend(struct device *dev) { - return tsl2550_set_power_state(client, 0); + return tsl2550_set_power_state(to_i2c_client(dev), 0); } -static int tsl2550_resume(struct i2c_client *client) +static int tsl2550_resume(struct device *dev) { - return tsl2550_set_power_state(client, 1); + return tsl2550_set_power_state(to_i2c_client(dev), 1); } +static SIMPLE_DEV_PM_OPS(tsl2550_pm_ops, tsl2550_suspend, tsl2550_resume); +#define TSL2550_PM_OPS (&tsl2550_pm_ops) + #else -#define tsl2550_suspend NULL -#define tsl2550_resume NULL +#define TSL2550_PM_OPS NULL -#endif /* CONFIG_PM */ +#endif /* CONFIG_PM_SLEEP */ static const struct i2c_device_id tsl2550_id[] = { { "tsl2550", 0 }, @@ -446,11 +447,10 @@ static struct i2c_driver tsl2550_driver = { .driver = { .name = TSL2550_DRV_NAME, .owner = THIS_MODULE, + .pm = TSL2550_PM_OPS, }, - .suspend = tsl2550_suspend, - .resume = tsl2550_resume, .probe = tsl2550_probe, - .remove = __devexit_p(tsl2550_remove), + .remove = tsl2550_remove, .id_table = tsl2550_id, }; diff --git a/drivers/misc/vexpress-syscfg.c b/drivers/misc/vexpress-syscfg.c new file mode 100644 index 00000000000..3250fc1df0a --- /dev/null +++ b/drivers/misc/vexpress-syscfg.c @@ -0,0 +1,328 @@ +/* + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Copyright (C) 2014 ARM Limited + */ + +#include <linux/delay.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/of.h> +#include <linux/platform_device.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/syscore_ops.h> +#include <linux/vexpress.h> + + +#define SYS_CFGDATA 0x0 + +#define SYS_CFGCTRL 0x4 +#define SYS_CFGCTRL_START (1 << 31) +#define SYS_CFGCTRL_WRITE (1 << 30) +#define SYS_CFGCTRL_DCC(n) (((n) & 0xf) << 26) +#define SYS_CFGCTRL_FUNC(n) (((n) & 0x3f) << 20) +#define SYS_CFGCTRL_SITE(n) (((n) & 0x3) << 16) +#define SYS_CFGCTRL_POSITION(n) (((n) & 0xf) << 12) +#define SYS_CFGCTRL_DEVICE(n) (((n) & 0xfff) << 0) + +#define SYS_CFGSTAT 0x8 +#define SYS_CFGSTAT_ERR (1 << 1) +#define SYS_CFGSTAT_COMPLETE (1 << 0) + + +struct vexpress_syscfg { + struct device *dev; + void __iomem *base; + struct list_head funcs; +}; + +struct vexpress_syscfg_func { + struct list_head list; + struct vexpress_syscfg *syscfg; + struct regmap *regmap; + int num_templates; + u32 template[0]; /* Keep it last! */ +}; + + +static int vexpress_syscfg_exec(struct vexpress_syscfg_func *func, + int index, bool write, u32 *data) +{ + struct vexpress_syscfg *syscfg = func->syscfg; + u32 command, status; + int tries; + long timeout; + + if (WARN_ON(index > func->num_templates)) + return -EINVAL; + + command = readl(syscfg->base + SYS_CFGCTRL); + if (WARN_ON(command & SYS_CFGCTRL_START)) + return -EBUSY; + + command = func->template[index]; + command |= SYS_CFGCTRL_START; + command |= write ? SYS_CFGCTRL_WRITE : 0; + + /* Use a canary for reads */ + if (!write) + *data = 0xdeadbeef; + + dev_dbg(syscfg->dev, "func %p, command %x, data %x\n", + func, command, *data); + writel(*data, syscfg->base + SYS_CFGDATA); + writel(0, syscfg->base + SYS_CFGSTAT); + writel(command, syscfg->base + SYS_CFGCTRL); + mb(); + + /* The operation can take ages... Go to sleep, 100us initially */ + tries = 100; + timeout = 100; + do { + if (!irqs_disabled()) { + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(usecs_to_jiffies(timeout)); + if (signal_pending(current)) + return -EINTR; + } else { + udelay(timeout); + } + + status = readl(syscfg->base + SYS_CFGSTAT); + if (status & SYS_CFGSTAT_ERR) + return -EFAULT; + + if (timeout > 20) + timeout -= 20; + } while (--tries && !(status & SYS_CFGSTAT_COMPLETE)); + if (WARN_ON_ONCE(!tries)) + return -ETIMEDOUT; + + if (!write) { + *data = readl(syscfg->base + SYS_CFGDATA); + dev_dbg(syscfg->dev, "func %p, read data %x\n", func, *data); + } + + return 0; +} + +static int vexpress_syscfg_read(void *context, unsigned int index, + unsigned int *val) +{ + struct vexpress_syscfg_func *func = context; + + return vexpress_syscfg_exec(func, index, false, val); +} + +static int vexpress_syscfg_write(void *context, unsigned int index, + unsigned int val) +{ + struct vexpress_syscfg_func *func = context; + + return vexpress_syscfg_exec(func, index, true, &val); +} + +struct regmap_config vexpress_syscfg_regmap_config = { + .lock = vexpress_config_lock, + .unlock = vexpress_config_unlock, + .reg_bits = 32, + .val_bits = 32, + .reg_read = vexpress_syscfg_read, + .reg_write = vexpress_syscfg_write, + .reg_format_endian = REGMAP_ENDIAN_LITTLE, + .val_format_endian = REGMAP_ENDIAN_LITTLE, +}; + + +static struct regmap *vexpress_syscfg_regmap_init(struct device *dev, + void *context) +{ + struct platform_device *pdev = to_platform_device(dev); + struct vexpress_syscfg *syscfg = context; + struct vexpress_syscfg_func *func; + struct property *prop; + const __be32 *val = NULL; + __be32 energy_quirk[4]; + int num; + u32 site, position, dcc; + int i; + + if (dev->of_node) { + int err = vexpress_config_get_topo(dev->of_node, &site, + &position, &dcc); + + if (err) + return ERR_PTR(err); + + prop = of_find_property(dev->of_node, + "arm,vexpress-sysreg,func", NULL); + if (!prop) + return ERR_PTR(-EINVAL); + + num = prop->length / sizeof(u32) / 2; + val = prop->value; + } else { + if (pdev->num_resources != 1 || + pdev->resource[0].flags != IORESOURCE_BUS) + return ERR_PTR(-EFAULT); + + site = pdev->resource[0].start; + if (site == VEXPRESS_SITE_MASTER) + site = vexpress_config_get_master(); + position = 0; + dcc = 0; + num = 1; + } + + /* + * "arm,vexpress-energy" function used to be described + * by its first device only, now it requires both + */ + if (num == 1 && of_device_is_compatible(dev->of_node, + "arm,vexpress-energy")) { + num = 2; + energy_quirk[0] = *val; + energy_quirk[2] = *val++; + energy_quirk[1] = *val; + energy_quirk[3] = cpu_to_be32(be32_to_cpup(val) + 1); + val = energy_quirk; + } + + func = kzalloc(sizeof(*func) + sizeof(*func->template) * num, + GFP_KERNEL); + if (!func) + return ERR_PTR(-ENOMEM); + + func->syscfg = syscfg; + func->num_templates = num; + + for (i = 0; i < num; i++) { + u32 function, device; + + if (dev->of_node) { + function = be32_to_cpup(val++); + device = be32_to_cpup(val++); + } else { + function = pdev->resource[0].end; + device = pdev->id; + } + + dev_dbg(dev, "func %p: %u/%u/%u/%u/%u\n", + func, site, position, dcc, + function, device); + + func->template[i] = SYS_CFGCTRL_DCC(dcc); + func->template[i] |= SYS_CFGCTRL_SITE(site); + func->template[i] |= SYS_CFGCTRL_POSITION(position); + func->template[i] |= SYS_CFGCTRL_FUNC(function); + func->template[i] |= SYS_CFGCTRL_DEVICE(device); + } + + vexpress_syscfg_regmap_config.max_register = num - 1; + + func->regmap = regmap_init(dev, NULL, func, + &vexpress_syscfg_regmap_config); + + if (IS_ERR(func->regmap)) { + void *err = func->regmap; + + kfree(func); + return err; + } + + list_add(&func->list, &syscfg->funcs); + + return func->regmap; +} + +static void vexpress_syscfg_regmap_exit(struct regmap *regmap, void *context) +{ + struct vexpress_syscfg *syscfg = context; + struct vexpress_syscfg_func *func, *tmp; + + regmap_exit(regmap); + + list_for_each_entry_safe(func, tmp, &syscfg->funcs, list) { + if (func->regmap == regmap) { + list_del(&syscfg->funcs); + kfree(func); + break; + } + } +} + +static struct vexpress_config_bridge_ops vexpress_syscfg_bridge_ops = { + .regmap_init = vexpress_syscfg_regmap_init, + .regmap_exit = vexpress_syscfg_regmap_exit, +}; + + +/* Non-DT hack, to be gone... */ +static struct device *vexpress_syscfg_bridge; + +int vexpress_syscfg_device_register(struct platform_device *pdev) +{ + pdev->dev.parent = vexpress_syscfg_bridge; + + return platform_device_register(pdev); +} + + +int vexpress_syscfg_probe(struct platform_device *pdev) +{ + struct vexpress_syscfg *syscfg; + struct resource *res; + struct device *bridge; + + syscfg = devm_kzalloc(&pdev->dev, sizeof(*syscfg), GFP_KERNEL); + if (!syscfg) + return -ENOMEM; + syscfg->dev = &pdev->dev; + INIT_LIST_HEAD(&syscfg->funcs); + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!devm_request_mem_region(&pdev->dev, res->start, + resource_size(res), pdev->name)) + return -EBUSY; + + syscfg->base = devm_ioremap(&pdev->dev, res->start, resource_size(res)); + if (!syscfg->base) + return -EFAULT; + + /* Must use dev.parent (MFD), as that's where DT phandle points at... */ + bridge = vexpress_config_bridge_register(pdev->dev.parent, + &vexpress_syscfg_bridge_ops, syscfg); + if (IS_ERR(bridge)) + return PTR_ERR(bridge); + + /* Non-DT case */ + if (!pdev->dev.of_node) + vexpress_syscfg_bridge = bridge; + + return 0; +} + +static const struct platform_device_id vexpress_syscfg_id_table[] = { + { "vexpress-syscfg", }, + {}, +}; + +static struct platform_driver vexpress_syscfg_driver = { + .driver.name = "vexpress-syscfg", + .id_table = vexpress_syscfg_id_table, + .probe = vexpress_syscfg_probe, +}; + +static int __init vexpress_syscfg_init(void) +{ + return platform_driver_register(&vexpress_syscfg_driver); +} +core_initcall(vexpress_syscfg_init); diff --git a/drivers/misc/vmw_balloon.c b/drivers/misc/vmw_balloon.c index cb56e270da1..19161749218 100644 --- a/drivers/misc/vmw_balloon.c +++ b/drivers/misc/vmw_balloon.c @@ -17,7 +17,8 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. * - * Maintained by: Dmitry Torokhov <dtor@vmware.com> + * Maintained by: Xavier Deguillard <xdeguillard@vmware.com> + * Philip Moltmann <moltmann@vmware.com> */ /* @@ -133,7 +134,7 @@ MODULE_LICENSE("GPL"); #define VMWARE_BALLOON_CMD(cmd, data, result) \ ({ \ unsigned long __stat, __dummy1, __dummy2; \ - __asm__ __volatile__ ("inl (%%dx)" : \ + __asm__ __volatile__ ("inl %%dx" : \ "=a"(__stat), \ "=c"(__dummy1), \ "=d"(__dummy2), \ diff --git a/drivers/misc/vmw_vmci/Kconfig b/drivers/misc/vmw_vmci/Kconfig new file mode 100644 index 00000000000..39c2ecadb27 --- /dev/null +++ b/drivers/misc/vmw_vmci/Kconfig @@ -0,0 +1,16 @@ +# +# VMware VMCI device +# + +config VMWARE_VMCI + tristate "VMware VMCI Driver" + depends on X86 && PCI + help + This is VMware's Virtual Machine Communication Interface. It enables + high-speed communication between host and guest in a virtual + environment via the VMCI virtual device. + + If unsure, say N. + + To compile this driver as a module, choose M here: the + module will be called vmw_vmci. diff --git a/drivers/misc/vmw_vmci/Makefile b/drivers/misc/vmw_vmci/Makefile new file mode 100644 index 00000000000..4da9893c394 --- /dev/null +++ b/drivers/misc/vmw_vmci/Makefile @@ -0,0 +1,4 @@ +obj-$(CONFIG_VMWARE_VMCI) += vmw_vmci.o +vmw_vmci-y += vmci_context.o vmci_datagram.o vmci_doorbell.o \ + vmci_driver.o vmci_event.o vmci_guest.o vmci_handle_array.o \ + vmci_host.o vmci_queue_pair.o vmci_resource.o vmci_route.o diff --git a/drivers/misc/vmw_vmci/vmci_context.c b/drivers/misc/vmw_vmci/vmci_context.c new file mode 100644 index 00000000000..f866a4baecb --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_context.c @@ -0,0 +1,1214 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/highmem.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/slab.h> + +#include "vmci_queue_pair.h" +#include "vmci_datagram.h" +#include "vmci_doorbell.h" +#include "vmci_context.h" +#include "vmci_driver.h" +#include "vmci_event.h" + +/* + * List of current VMCI contexts. Contexts can be added by + * vmci_ctx_create() and removed via vmci_ctx_destroy(). + * These, along with context lookup, are protected by the + * list structure's lock. + */ +static struct { + struct list_head head; + spinlock_t lock; /* Spinlock for context list operations */ +} ctx_list = { + .head = LIST_HEAD_INIT(ctx_list.head), + .lock = __SPIN_LOCK_UNLOCKED(ctx_list.lock), +}; + +/* Used by contexts that did not set up notify flag pointers */ +static bool ctx_dummy_notify; + +static void ctx_signal_notify(struct vmci_ctx *context) +{ + *context->notify = true; +} + +static void ctx_clear_notify(struct vmci_ctx *context) +{ + *context->notify = false; +} + +/* + * If nothing requires the attention of the guest, clears both + * notify flag and call. + */ +static void ctx_clear_notify_call(struct vmci_ctx *context) +{ + if (context->pending_datagrams == 0 && + vmci_handle_arr_get_size(context->pending_doorbell_array) == 0) + ctx_clear_notify(context); +} + +/* + * Sets the context's notify flag iff datagrams are pending for this + * context. Called from vmci_setup_notify(). + */ +void vmci_ctx_check_signal_notify(struct vmci_ctx *context) +{ + spin_lock(&context->lock); + if (context->pending_datagrams) + ctx_signal_notify(context); + spin_unlock(&context->lock); +} + +/* + * Allocates and initializes a VMCI context. + */ +struct vmci_ctx *vmci_ctx_create(u32 cid, u32 priv_flags, + uintptr_t event_hnd, + int user_version, + const struct cred *cred) +{ + struct vmci_ctx *context; + int error; + + if (cid == VMCI_INVALID_ID) { + pr_devel("Invalid context ID for VMCI context\n"); + error = -EINVAL; + goto err_out; + } + + if (priv_flags & ~VMCI_PRIVILEGE_ALL_FLAGS) { + pr_devel("Invalid flag (flags=0x%x) for VMCI context\n", + priv_flags); + error = -EINVAL; + goto err_out; + } + + if (user_version == 0) { + pr_devel("Invalid suer_version %d\n", user_version); + error = -EINVAL; + goto err_out; + } + + context = kzalloc(sizeof(*context), GFP_KERNEL); + if (!context) { + pr_warn("Failed to allocate memory for VMCI context\n"); + error = -EINVAL; + goto err_out; + } + + kref_init(&context->kref); + spin_lock_init(&context->lock); + INIT_LIST_HEAD(&context->list_item); + INIT_LIST_HEAD(&context->datagram_queue); + INIT_LIST_HEAD(&context->notifier_list); + + /* Initialize host-specific VMCI context. */ + init_waitqueue_head(&context->host_context.wait_queue); + + context->queue_pair_array = vmci_handle_arr_create(0); + if (!context->queue_pair_array) { + error = -ENOMEM; + goto err_free_ctx; + } + + context->doorbell_array = vmci_handle_arr_create(0); + if (!context->doorbell_array) { + error = -ENOMEM; + goto err_free_qp_array; + } + + context->pending_doorbell_array = vmci_handle_arr_create(0); + if (!context->pending_doorbell_array) { + error = -ENOMEM; + goto err_free_db_array; + } + + context->user_version = user_version; + + context->priv_flags = priv_flags; + + if (cred) + context->cred = get_cred(cred); + + context->notify = &ctx_dummy_notify; + context->notify_page = NULL; + + /* + * If we collide with an existing context we generate a new + * and use it instead. The VMX will determine if regeneration + * is okay. Since there isn't 4B - 16 VMs running on a given + * host, the below loop will terminate. + */ + spin_lock(&ctx_list.lock); + + while (vmci_ctx_exists(cid)) { + /* We reserve the lowest 16 ids for fixed contexts. */ + cid = max(cid, VMCI_RESERVED_CID_LIMIT - 1) + 1; + if (cid == VMCI_INVALID_ID) + cid = VMCI_RESERVED_CID_LIMIT; + } + context->cid = cid; + + list_add_tail_rcu(&context->list_item, &ctx_list.head); + spin_unlock(&ctx_list.lock); + + return context; + + err_free_db_array: + vmci_handle_arr_destroy(context->doorbell_array); + err_free_qp_array: + vmci_handle_arr_destroy(context->queue_pair_array); + err_free_ctx: + kfree(context); + err_out: + return ERR_PTR(error); +} + +/* + * Destroy VMCI context. + */ +void vmci_ctx_destroy(struct vmci_ctx *context) +{ + spin_lock(&ctx_list.lock); + list_del_rcu(&context->list_item); + spin_unlock(&ctx_list.lock); + synchronize_rcu(); + + vmci_ctx_put(context); +} + +/* + * Fire notification for all contexts interested in given cid. + */ +static int ctx_fire_notification(u32 context_id, u32 priv_flags) +{ + u32 i, array_size; + struct vmci_ctx *sub_ctx; + struct vmci_handle_arr *subscriber_array; + struct vmci_handle context_handle = + vmci_make_handle(context_id, VMCI_EVENT_HANDLER); + + /* + * We create an array to hold the subscribers we find when + * scanning through all contexts. + */ + subscriber_array = vmci_handle_arr_create(0); + if (subscriber_array == NULL) + return VMCI_ERROR_NO_MEM; + + /* + * Scan all contexts to find who is interested in being + * notified about given contextID. + */ + rcu_read_lock(); + list_for_each_entry_rcu(sub_ctx, &ctx_list.head, list_item) { + struct vmci_handle_list *node; + + /* + * We only deliver notifications of the removal of + * contexts, if the two contexts are allowed to + * interact. + */ + if (vmci_deny_interaction(priv_flags, sub_ctx->priv_flags)) + continue; + + list_for_each_entry_rcu(node, &sub_ctx->notifier_list, node) { + if (!vmci_handle_is_equal(node->handle, context_handle)) + continue; + + vmci_handle_arr_append_entry(&subscriber_array, + vmci_make_handle(sub_ctx->cid, + VMCI_EVENT_HANDLER)); + } + } + rcu_read_unlock(); + + /* Fire event to all subscribers. */ + array_size = vmci_handle_arr_get_size(subscriber_array); + for (i = 0; i < array_size; i++) { + int result; + struct vmci_event_ctx ev; + + ev.msg.hdr.dst = vmci_handle_arr_get_entry(subscriber_array, i); + ev.msg.hdr.src = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_CONTEXT_RESOURCE_ID); + ev.msg.hdr.payload_size = sizeof(ev) - sizeof(ev.msg.hdr); + ev.msg.event_data.event = VMCI_EVENT_CTX_REMOVED; + ev.payload.context_id = context_id; + + result = vmci_datagram_dispatch(VMCI_HYPERVISOR_CONTEXT_ID, + &ev.msg.hdr, false); + if (result < VMCI_SUCCESS) { + pr_devel("Failed to enqueue event datagram (type=%d) for context (ID=0x%x)\n", + ev.msg.event_data.event, + ev.msg.hdr.dst.context); + /* We continue to enqueue on next subscriber. */ + } + } + vmci_handle_arr_destroy(subscriber_array); + + return VMCI_SUCCESS; +} + +/* + * Returns the current number of pending datagrams. The call may + * also serve as a synchronization point for the datagram queue, + * as no enqueue operations can occur concurrently. + */ +int vmci_ctx_pending_datagrams(u32 cid, u32 *pending) +{ + struct vmci_ctx *context; + + context = vmci_ctx_get(cid); + if (context == NULL) + return VMCI_ERROR_INVALID_ARGS; + + spin_lock(&context->lock); + if (pending) + *pending = context->pending_datagrams; + spin_unlock(&context->lock); + vmci_ctx_put(context); + + return VMCI_SUCCESS; +} + +/* + * Queues a VMCI datagram for the appropriate target VM context. + */ +int vmci_ctx_enqueue_datagram(u32 cid, struct vmci_datagram *dg) +{ + struct vmci_datagram_queue_entry *dq_entry; + struct vmci_ctx *context; + struct vmci_handle dg_src; + size_t vmci_dg_size; + + vmci_dg_size = VMCI_DG_SIZE(dg); + if (vmci_dg_size > VMCI_MAX_DG_SIZE) { + pr_devel("Datagram too large (bytes=%Zu)\n", vmci_dg_size); + return VMCI_ERROR_INVALID_ARGS; + } + + /* Get the target VM's VMCI context. */ + context = vmci_ctx_get(cid); + if (!context) { + pr_devel("Invalid context (ID=0x%x)\n", cid); + return VMCI_ERROR_INVALID_ARGS; + } + + /* Allocate guest call entry and add it to the target VM's queue. */ + dq_entry = kmalloc(sizeof(*dq_entry), GFP_KERNEL); + if (dq_entry == NULL) { + pr_warn("Failed to allocate memory for datagram\n"); + vmci_ctx_put(context); + return VMCI_ERROR_NO_MEM; + } + dq_entry->dg = dg; + dq_entry->dg_size = vmci_dg_size; + dg_src = dg->src; + INIT_LIST_HEAD(&dq_entry->list_item); + + spin_lock(&context->lock); + + /* + * We put a higher limit on datagrams from the hypervisor. If + * the pending datagram is not from hypervisor, then we check + * if enqueueing it would exceed the + * VMCI_MAX_DATAGRAM_QUEUE_SIZE limit on the destination. If + * the pending datagram is from hypervisor, we allow it to be + * queued at the destination side provided we don't reach the + * VMCI_MAX_DATAGRAM_AND_EVENT_QUEUE_SIZE limit. + */ + if (context->datagram_queue_size + vmci_dg_size >= + VMCI_MAX_DATAGRAM_QUEUE_SIZE && + (!vmci_handle_is_equal(dg_src, + vmci_make_handle + (VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_CONTEXT_RESOURCE_ID)) || + context->datagram_queue_size + vmci_dg_size >= + VMCI_MAX_DATAGRAM_AND_EVENT_QUEUE_SIZE)) { + spin_unlock(&context->lock); + vmci_ctx_put(context); + kfree(dq_entry); + pr_devel("Context (ID=0x%x) receive queue is full\n", cid); + return VMCI_ERROR_NO_RESOURCES; + } + + list_add(&dq_entry->list_item, &context->datagram_queue); + context->pending_datagrams++; + context->datagram_queue_size += vmci_dg_size; + ctx_signal_notify(context); + wake_up(&context->host_context.wait_queue); + spin_unlock(&context->lock); + vmci_ctx_put(context); + + return vmci_dg_size; +} + +/* + * Verifies whether a context with the specified context ID exists. + * FIXME: utility is dubious as no decisions can be reliably made + * using this data as context can appear and disappear at any time. + */ +bool vmci_ctx_exists(u32 cid) +{ + struct vmci_ctx *context; + bool exists = false; + + rcu_read_lock(); + + list_for_each_entry_rcu(context, &ctx_list.head, list_item) { + if (context->cid == cid) { + exists = true; + break; + } + } + + rcu_read_unlock(); + return exists; +} + +/* + * Retrieves VMCI context corresponding to the given cid. + */ +struct vmci_ctx *vmci_ctx_get(u32 cid) +{ + struct vmci_ctx *c, *context = NULL; + + if (cid == VMCI_INVALID_ID) + return NULL; + + rcu_read_lock(); + list_for_each_entry_rcu(c, &ctx_list.head, list_item) { + if (c->cid == cid) { + /* + * The context owner drops its own reference to the + * context only after removing it from the list and + * waiting for RCU grace period to expire. This + * means that we are not about to increase the + * reference count of something that is in the + * process of being destroyed. + */ + context = c; + kref_get(&context->kref); + break; + } + } + rcu_read_unlock(); + + return context; +} + +/* + * Deallocates all parts of a context data structure. This + * function doesn't lock the context, because it assumes that + * the caller was holding the last reference to context. + */ +static void ctx_free_ctx(struct kref *kref) +{ + struct vmci_ctx *context = container_of(kref, struct vmci_ctx, kref); + struct vmci_datagram_queue_entry *dq_entry, *dq_entry_tmp; + struct vmci_handle temp_handle; + struct vmci_handle_list *notifier, *tmp; + + /* + * Fire event to all contexts interested in knowing this + * context is dying. + */ + ctx_fire_notification(context->cid, context->priv_flags); + + /* + * Cleanup all queue pair resources attached to context. If + * the VM dies without cleaning up, this code will make sure + * that no resources are leaked. + */ + temp_handle = vmci_handle_arr_get_entry(context->queue_pair_array, 0); + while (!vmci_handle_is_equal(temp_handle, VMCI_INVALID_HANDLE)) { + if (vmci_qp_broker_detach(temp_handle, + context) < VMCI_SUCCESS) { + /* + * When vmci_qp_broker_detach() succeeds it + * removes the handle from the array. If + * detach fails, we must remove the handle + * ourselves. + */ + vmci_handle_arr_remove_entry(context->queue_pair_array, + temp_handle); + } + temp_handle = + vmci_handle_arr_get_entry(context->queue_pair_array, 0); + } + + /* + * It is fine to destroy this without locking the callQueue, as + * this is the only thread having a reference to the context. + */ + list_for_each_entry_safe(dq_entry, dq_entry_tmp, + &context->datagram_queue, list_item) { + WARN_ON(dq_entry->dg_size != VMCI_DG_SIZE(dq_entry->dg)); + list_del(&dq_entry->list_item); + kfree(dq_entry->dg); + kfree(dq_entry); + } + + list_for_each_entry_safe(notifier, tmp, + &context->notifier_list, node) { + list_del(¬ifier->node); + kfree(notifier); + } + + vmci_handle_arr_destroy(context->queue_pair_array); + vmci_handle_arr_destroy(context->doorbell_array); + vmci_handle_arr_destroy(context->pending_doorbell_array); + vmci_ctx_unset_notify(context); + if (context->cred) + put_cred(context->cred); + kfree(context); +} + +/* + * Drops reference to VMCI context. If this is the last reference to + * the context it will be deallocated. A context is created with + * a reference count of one, and on destroy, it is removed from + * the context list before its reference count is decremented. Thus, + * if we reach zero, we are sure that nobody else are about to increment + * it (they need the entry in the context list for that), and so there + * is no need for locking. + */ +void vmci_ctx_put(struct vmci_ctx *context) +{ + kref_put(&context->kref, ctx_free_ctx); +} + +/* + * Dequeues the next datagram and returns it to caller. + * The caller passes in a pointer to the max size datagram + * it can handle and the datagram is only unqueued if the + * size is less than max_size. If larger max_size is set to + * the size of the datagram to give the caller a chance to + * set up a larger buffer for the guestcall. + */ +int vmci_ctx_dequeue_datagram(struct vmci_ctx *context, + size_t *max_size, + struct vmci_datagram **dg) +{ + struct vmci_datagram_queue_entry *dq_entry; + struct list_head *list_item; + int rv; + + /* Dequeue the next datagram entry. */ + spin_lock(&context->lock); + if (context->pending_datagrams == 0) { + ctx_clear_notify_call(context); + spin_unlock(&context->lock); + pr_devel("No datagrams pending\n"); + return VMCI_ERROR_NO_MORE_DATAGRAMS; + } + + list_item = context->datagram_queue.next; + + dq_entry = + list_entry(list_item, struct vmci_datagram_queue_entry, list_item); + + /* Check size of caller's buffer. */ + if (*max_size < dq_entry->dg_size) { + *max_size = dq_entry->dg_size; + spin_unlock(&context->lock); + pr_devel("Caller's buffer should be at least (size=%u bytes)\n", + (u32) *max_size); + return VMCI_ERROR_NO_MEM; + } + + list_del(list_item); + context->pending_datagrams--; + context->datagram_queue_size -= dq_entry->dg_size; + if (context->pending_datagrams == 0) { + ctx_clear_notify_call(context); + rv = VMCI_SUCCESS; + } else { + /* + * Return the size of the next datagram. + */ + struct vmci_datagram_queue_entry *next_entry; + + list_item = context->datagram_queue.next; + next_entry = + list_entry(list_item, struct vmci_datagram_queue_entry, + list_item); + + /* + * The following size_t -> int truncation is fine as + * the maximum size of a (routable) datagram is 68KB. + */ + rv = (int)next_entry->dg_size; + } + spin_unlock(&context->lock); + + /* Caller must free datagram. */ + *dg = dq_entry->dg; + dq_entry->dg = NULL; + kfree(dq_entry); + + return rv; +} + +/* + * Reverts actions set up by vmci_setup_notify(). Unmaps and unlocks the + * page mapped/locked by vmci_setup_notify(). + */ +void vmci_ctx_unset_notify(struct vmci_ctx *context) +{ + struct page *notify_page; + + spin_lock(&context->lock); + + notify_page = context->notify_page; + context->notify = &ctx_dummy_notify; + context->notify_page = NULL; + + spin_unlock(&context->lock); + + if (notify_page) { + kunmap(notify_page); + put_page(notify_page); + } +} + +/* + * Add remote_cid to list of contexts current contexts wants + * notifications from/about. + */ +int vmci_ctx_add_notification(u32 context_id, u32 remote_cid) +{ + struct vmci_ctx *context; + struct vmci_handle_list *notifier, *n; + int result; + bool exists = false; + + context = vmci_ctx_get(context_id); + if (!context) + return VMCI_ERROR_NOT_FOUND; + + if (VMCI_CONTEXT_IS_VM(context_id) && VMCI_CONTEXT_IS_VM(remote_cid)) { + pr_devel("Context removed notifications for other VMs not supported (src=0x%x, remote=0x%x)\n", + context_id, remote_cid); + result = VMCI_ERROR_DST_UNREACHABLE; + goto out; + } + + if (context->priv_flags & VMCI_PRIVILEGE_FLAG_RESTRICTED) { + result = VMCI_ERROR_NO_ACCESS; + goto out; + } + + notifier = kmalloc(sizeof(struct vmci_handle_list), GFP_KERNEL); + if (!notifier) { + result = VMCI_ERROR_NO_MEM; + goto out; + } + + INIT_LIST_HEAD(¬ifier->node); + notifier->handle = vmci_make_handle(remote_cid, VMCI_EVENT_HANDLER); + + spin_lock(&context->lock); + + list_for_each_entry(n, &context->notifier_list, node) { + if (vmci_handle_is_equal(n->handle, notifier->handle)) { + exists = true; + break; + } + } + + if (exists) { + kfree(notifier); + result = VMCI_ERROR_ALREADY_EXISTS; + } else { + list_add_tail_rcu(¬ifier->node, &context->notifier_list); + context->n_notifiers++; + result = VMCI_SUCCESS; + } + + spin_unlock(&context->lock); + + out: + vmci_ctx_put(context); + return result; +} + +/* + * Remove remote_cid from current context's list of contexts it is + * interested in getting notifications from/about. + */ +int vmci_ctx_remove_notification(u32 context_id, u32 remote_cid) +{ + struct vmci_ctx *context; + struct vmci_handle_list *notifier, *tmp; + struct vmci_handle handle; + bool found = false; + + context = vmci_ctx_get(context_id); + if (!context) + return VMCI_ERROR_NOT_FOUND; + + handle = vmci_make_handle(remote_cid, VMCI_EVENT_HANDLER); + + spin_lock(&context->lock); + list_for_each_entry_safe(notifier, tmp, + &context->notifier_list, node) { + if (vmci_handle_is_equal(notifier->handle, handle)) { + list_del_rcu(¬ifier->node); + context->n_notifiers--; + found = true; + break; + } + } + spin_unlock(&context->lock); + + if (found) { + synchronize_rcu(); + kfree(notifier); + } + + vmci_ctx_put(context); + + return found ? VMCI_SUCCESS : VMCI_ERROR_NOT_FOUND; +} + +static int vmci_ctx_get_chkpt_notifiers(struct vmci_ctx *context, + u32 *buf_size, void **pbuf) +{ + u32 *notifiers; + size_t data_size; + struct vmci_handle_list *entry; + int i = 0; + + if (context->n_notifiers == 0) { + *buf_size = 0; + *pbuf = NULL; + return VMCI_SUCCESS; + } + + data_size = context->n_notifiers * sizeof(*notifiers); + if (*buf_size < data_size) { + *buf_size = data_size; + return VMCI_ERROR_MORE_DATA; + } + + notifiers = kmalloc(data_size, GFP_ATOMIC); /* FIXME: want GFP_KERNEL */ + if (!notifiers) + return VMCI_ERROR_NO_MEM; + + list_for_each_entry(entry, &context->notifier_list, node) + notifiers[i++] = entry->handle.context; + + *buf_size = data_size; + *pbuf = notifiers; + return VMCI_SUCCESS; +} + +static int vmci_ctx_get_chkpt_doorbells(struct vmci_ctx *context, + u32 *buf_size, void **pbuf) +{ + struct dbell_cpt_state *dbells; + size_t n_doorbells; + int i; + + n_doorbells = vmci_handle_arr_get_size(context->doorbell_array); + if (n_doorbells > 0) { + size_t data_size = n_doorbells * sizeof(*dbells); + if (*buf_size < data_size) { + *buf_size = data_size; + return VMCI_ERROR_MORE_DATA; + } + + dbells = kmalloc(data_size, GFP_ATOMIC); + if (!dbells) + return VMCI_ERROR_NO_MEM; + + for (i = 0; i < n_doorbells; i++) + dbells[i].handle = vmci_handle_arr_get_entry( + context->doorbell_array, i); + + *buf_size = data_size; + *pbuf = dbells; + } else { + *buf_size = 0; + *pbuf = NULL; + } + + return VMCI_SUCCESS; +} + +/* + * Get current context's checkpoint state of given type. + */ +int vmci_ctx_get_chkpt_state(u32 context_id, + u32 cpt_type, + u32 *buf_size, + void **pbuf) +{ + struct vmci_ctx *context; + int result; + + context = vmci_ctx_get(context_id); + if (!context) + return VMCI_ERROR_NOT_FOUND; + + spin_lock(&context->lock); + + switch (cpt_type) { + case VMCI_NOTIFICATION_CPT_STATE: + result = vmci_ctx_get_chkpt_notifiers(context, buf_size, pbuf); + break; + + case VMCI_WELLKNOWN_CPT_STATE: + /* + * For compatibility with VMX'en with VM to VM communication, we + * always return zero wellknown handles. + */ + + *buf_size = 0; + *pbuf = NULL; + result = VMCI_SUCCESS; + break; + + case VMCI_DOORBELL_CPT_STATE: + result = vmci_ctx_get_chkpt_doorbells(context, buf_size, pbuf); + break; + + default: + pr_devel("Invalid cpt state (type=%d)\n", cpt_type); + result = VMCI_ERROR_INVALID_ARGS; + break; + } + + spin_unlock(&context->lock); + vmci_ctx_put(context); + + return result; +} + +/* + * Set current context's checkpoint state of given type. + */ +int vmci_ctx_set_chkpt_state(u32 context_id, + u32 cpt_type, + u32 buf_size, + void *cpt_buf) +{ + u32 i; + u32 current_id; + int result = VMCI_SUCCESS; + u32 num_ids = buf_size / sizeof(u32); + + if (cpt_type == VMCI_WELLKNOWN_CPT_STATE && num_ids > 0) { + /* + * We would end up here if VMX with VM to VM communication + * attempts to restore a checkpoint with wellknown handles. + */ + pr_warn("Attempt to restore checkpoint with obsolete wellknown handles\n"); + return VMCI_ERROR_OBSOLETE; + } + + if (cpt_type != VMCI_NOTIFICATION_CPT_STATE) { + pr_devel("Invalid cpt state (type=%d)\n", cpt_type); + return VMCI_ERROR_INVALID_ARGS; + } + + for (i = 0; i < num_ids && result == VMCI_SUCCESS; i++) { + current_id = ((u32 *)cpt_buf)[i]; + result = vmci_ctx_add_notification(context_id, current_id); + if (result != VMCI_SUCCESS) + break; + } + if (result != VMCI_SUCCESS) + pr_devel("Failed to set cpt state (type=%d) (error=%d)\n", + cpt_type, result); + + return result; +} + +/* + * Retrieves the specified context's pending notifications in the + * form of a handle array. The handle arrays returned are the + * actual data - not a copy and should not be modified by the + * caller. They must be released using + * vmci_ctx_rcv_notifications_release. + */ +int vmci_ctx_rcv_notifications_get(u32 context_id, + struct vmci_handle_arr **db_handle_array, + struct vmci_handle_arr **qp_handle_array) +{ + struct vmci_ctx *context; + int result = VMCI_SUCCESS; + + context = vmci_ctx_get(context_id); + if (context == NULL) + return VMCI_ERROR_NOT_FOUND; + + spin_lock(&context->lock); + + *db_handle_array = context->pending_doorbell_array; + context->pending_doorbell_array = vmci_handle_arr_create(0); + if (!context->pending_doorbell_array) { + context->pending_doorbell_array = *db_handle_array; + *db_handle_array = NULL; + result = VMCI_ERROR_NO_MEM; + } + *qp_handle_array = NULL; + + spin_unlock(&context->lock); + vmci_ctx_put(context); + + return result; +} + +/* + * Releases handle arrays with pending notifications previously + * retrieved using vmci_ctx_rcv_notifications_get. If the + * notifications were not successfully handed over to the guest, + * success must be false. + */ +void vmci_ctx_rcv_notifications_release(u32 context_id, + struct vmci_handle_arr *db_handle_array, + struct vmci_handle_arr *qp_handle_array, + bool success) +{ + struct vmci_ctx *context = vmci_ctx_get(context_id); + + spin_lock(&context->lock); + if (!success) { + struct vmci_handle handle; + + /* + * New notifications may have been added while we were not + * holding the context lock, so we transfer any new pending + * doorbell notifications to the old array, and reinstate the + * old array. + */ + + handle = vmci_handle_arr_remove_tail( + context->pending_doorbell_array); + while (!vmci_handle_is_invalid(handle)) { + if (!vmci_handle_arr_has_entry(db_handle_array, + handle)) { + vmci_handle_arr_append_entry( + &db_handle_array, handle); + } + handle = vmci_handle_arr_remove_tail( + context->pending_doorbell_array); + } + vmci_handle_arr_destroy(context->pending_doorbell_array); + context->pending_doorbell_array = db_handle_array; + db_handle_array = NULL; + } else { + ctx_clear_notify_call(context); + } + spin_unlock(&context->lock); + vmci_ctx_put(context); + + if (db_handle_array) + vmci_handle_arr_destroy(db_handle_array); + + if (qp_handle_array) + vmci_handle_arr_destroy(qp_handle_array); +} + +/* + * Registers that a new doorbell handle has been allocated by the + * context. Only doorbell handles registered can be notified. + */ +int vmci_ctx_dbell_create(u32 context_id, struct vmci_handle handle) +{ + struct vmci_ctx *context; + int result; + + if (context_id == VMCI_INVALID_ID || vmci_handle_is_invalid(handle)) + return VMCI_ERROR_INVALID_ARGS; + + context = vmci_ctx_get(context_id); + if (context == NULL) + return VMCI_ERROR_NOT_FOUND; + + spin_lock(&context->lock); + if (!vmci_handle_arr_has_entry(context->doorbell_array, handle)) { + vmci_handle_arr_append_entry(&context->doorbell_array, handle); + result = VMCI_SUCCESS; + } else { + result = VMCI_ERROR_DUPLICATE_ENTRY; + } + + spin_unlock(&context->lock); + vmci_ctx_put(context); + + return result; +} + +/* + * Unregisters a doorbell handle that was previously registered + * with vmci_ctx_dbell_create. + */ +int vmci_ctx_dbell_destroy(u32 context_id, struct vmci_handle handle) +{ + struct vmci_ctx *context; + struct vmci_handle removed_handle; + + if (context_id == VMCI_INVALID_ID || vmci_handle_is_invalid(handle)) + return VMCI_ERROR_INVALID_ARGS; + + context = vmci_ctx_get(context_id); + if (context == NULL) + return VMCI_ERROR_NOT_FOUND; + + spin_lock(&context->lock); + removed_handle = + vmci_handle_arr_remove_entry(context->doorbell_array, handle); + vmci_handle_arr_remove_entry(context->pending_doorbell_array, handle); + spin_unlock(&context->lock); + + vmci_ctx_put(context); + + return vmci_handle_is_invalid(removed_handle) ? + VMCI_ERROR_NOT_FOUND : VMCI_SUCCESS; +} + +/* + * Unregisters all doorbell handles that were previously + * registered with vmci_ctx_dbell_create. + */ +int vmci_ctx_dbell_destroy_all(u32 context_id) +{ + struct vmci_ctx *context; + struct vmci_handle handle; + + if (context_id == VMCI_INVALID_ID) + return VMCI_ERROR_INVALID_ARGS; + + context = vmci_ctx_get(context_id); + if (context == NULL) + return VMCI_ERROR_NOT_FOUND; + + spin_lock(&context->lock); + do { + struct vmci_handle_arr *arr = context->doorbell_array; + handle = vmci_handle_arr_remove_tail(arr); + } while (!vmci_handle_is_invalid(handle)); + do { + struct vmci_handle_arr *arr = context->pending_doorbell_array; + handle = vmci_handle_arr_remove_tail(arr); + } while (!vmci_handle_is_invalid(handle)); + spin_unlock(&context->lock); + + vmci_ctx_put(context); + + return VMCI_SUCCESS; +} + +/* + * Registers a notification of a doorbell handle initiated by the + * specified source context. The notification of doorbells are + * subject to the same isolation rules as datagram delivery. To + * allow host side senders of notifications a finer granularity + * of sender rights than those assigned to the sending context + * itself, the host context is required to specify a different + * set of privilege flags that will override the privileges of + * the source context. + */ +int vmci_ctx_notify_dbell(u32 src_cid, + struct vmci_handle handle, + u32 src_priv_flags) +{ + struct vmci_ctx *dst_context; + int result; + + if (vmci_handle_is_invalid(handle)) + return VMCI_ERROR_INVALID_ARGS; + + /* Get the target VM's VMCI context. */ + dst_context = vmci_ctx_get(handle.context); + if (!dst_context) { + pr_devel("Invalid context (ID=0x%x)\n", handle.context); + return VMCI_ERROR_NOT_FOUND; + } + + if (src_cid != handle.context) { + u32 dst_priv_flags; + + if (VMCI_CONTEXT_IS_VM(src_cid) && + VMCI_CONTEXT_IS_VM(handle.context)) { + pr_devel("Doorbell notification from VM to VM not supported (src=0x%x, dst=0x%x)\n", + src_cid, handle.context); + result = VMCI_ERROR_DST_UNREACHABLE; + goto out; + } + + result = vmci_dbell_get_priv_flags(handle, &dst_priv_flags); + if (result < VMCI_SUCCESS) { + pr_warn("Failed to get privilege flags for destination (handle=0x%x:0x%x)\n", + handle.context, handle.resource); + goto out; + } + + if (src_cid != VMCI_HOST_CONTEXT_ID || + src_priv_flags == VMCI_NO_PRIVILEGE_FLAGS) { + src_priv_flags = vmci_context_get_priv_flags(src_cid); + } + + if (vmci_deny_interaction(src_priv_flags, dst_priv_flags)) { + result = VMCI_ERROR_NO_ACCESS; + goto out; + } + } + + if (handle.context == VMCI_HOST_CONTEXT_ID) { + result = vmci_dbell_host_context_notify(src_cid, handle); + } else { + spin_lock(&dst_context->lock); + + if (!vmci_handle_arr_has_entry(dst_context->doorbell_array, + handle)) { + result = VMCI_ERROR_NOT_FOUND; + } else { + if (!vmci_handle_arr_has_entry( + dst_context->pending_doorbell_array, + handle)) { + vmci_handle_arr_append_entry( + &dst_context->pending_doorbell_array, + handle); + + ctx_signal_notify(dst_context); + wake_up(&dst_context->host_context.wait_queue); + + } + result = VMCI_SUCCESS; + } + spin_unlock(&dst_context->lock); + } + + out: + vmci_ctx_put(dst_context); + + return result; +} + +bool vmci_ctx_supports_host_qp(struct vmci_ctx *context) +{ + return context && context->user_version >= VMCI_VERSION_HOSTQP; +} + +/* + * Registers that a new queue pair handle has been allocated by + * the context. + */ +int vmci_ctx_qp_create(struct vmci_ctx *context, struct vmci_handle handle) +{ + int result; + + if (context == NULL || vmci_handle_is_invalid(handle)) + return VMCI_ERROR_INVALID_ARGS; + + if (!vmci_handle_arr_has_entry(context->queue_pair_array, handle)) { + vmci_handle_arr_append_entry(&context->queue_pair_array, + handle); + result = VMCI_SUCCESS; + } else { + result = VMCI_ERROR_DUPLICATE_ENTRY; + } + + return result; +} + +/* + * Unregisters a queue pair handle that was previously registered + * with vmci_ctx_qp_create. + */ +int vmci_ctx_qp_destroy(struct vmci_ctx *context, struct vmci_handle handle) +{ + struct vmci_handle hndl; + + if (context == NULL || vmci_handle_is_invalid(handle)) + return VMCI_ERROR_INVALID_ARGS; + + hndl = vmci_handle_arr_remove_entry(context->queue_pair_array, handle); + + return vmci_handle_is_invalid(hndl) ? + VMCI_ERROR_NOT_FOUND : VMCI_SUCCESS; +} + +/* + * Determines whether a given queue pair handle is registered + * with the given context. + */ +bool vmci_ctx_qp_exists(struct vmci_ctx *context, struct vmci_handle handle) +{ + if (context == NULL || vmci_handle_is_invalid(handle)) + return false; + + return vmci_handle_arr_has_entry(context->queue_pair_array, handle); +} + +/* + * vmci_context_get_priv_flags() - Retrieve privilege flags. + * @context_id: The context ID of the VMCI context. + * + * Retrieves privilege flags of the given VMCI context ID. + */ +u32 vmci_context_get_priv_flags(u32 context_id) +{ + if (vmci_host_code_active()) { + u32 flags; + struct vmci_ctx *context; + + context = vmci_ctx_get(context_id); + if (!context) + return VMCI_LEAST_PRIVILEGE_FLAGS; + + flags = context->priv_flags; + vmci_ctx_put(context); + return flags; + } + return VMCI_NO_PRIVILEGE_FLAGS; +} +EXPORT_SYMBOL_GPL(vmci_context_get_priv_flags); + +/* + * vmci_is_context_owner() - Determimnes if user is the context owner + * @context_id: The context ID of the VMCI context. + * @uid: The host user id (real kernel value). + * + * Determines whether a given UID is the owner of given VMCI context. + */ +bool vmci_is_context_owner(u32 context_id, kuid_t uid) +{ + bool is_owner = false; + + if (vmci_host_code_active()) { + struct vmci_ctx *context = vmci_ctx_get(context_id); + if (context) { + if (context->cred) + is_owner = uid_eq(context->cred->uid, uid); + vmci_ctx_put(context); + } + } + + return is_owner; +} +EXPORT_SYMBOL_GPL(vmci_is_context_owner); diff --git a/drivers/misc/vmw_vmci/vmci_context.h b/drivers/misc/vmw_vmci/vmci_context.h new file mode 100644 index 00000000000..24a88e68a1e --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_context.h @@ -0,0 +1,182 @@ +/* + * VMware VMCI driver (vmciContext.h) + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef _VMCI_CONTEXT_H_ +#define _VMCI_CONTEXT_H_ + +#include <linux/vmw_vmci_defs.h> +#include <linux/atomic.h> +#include <linux/kref.h> +#include <linux/types.h> +#include <linux/wait.h> + +#include "vmci_handle_array.h" +#include "vmci_datagram.h" + +/* Used to determine what checkpoint state to get and set. */ +enum { + VMCI_NOTIFICATION_CPT_STATE = 1, + VMCI_WELLKNOWN_CPT_STATE = 2, + VMCI_DG_OUT_STATE = 3, + VMCI_DG_IN_STATE = 4, + VMCI_DG_IN_SIZE_STATE = 5, + VMCI_DOORBELL_CPT_STATE = 6, +}; + +/* Host specific struct used for signalling */ +struct vmci_host { + wait_queue_head_t wait_queue; +}; + +struct vmci_handle_list { + struct list_head node; + struct vmci_handle handle; +}; + +struct vmci_ctx { + struct list_head list_item; /* For global VMCI list. */ + u32 cid; + struct kref kref; + struct list_head datagram_queue; /* Head of per VM queue. */ + u32 pending_datagrams; + size_t datagram_queue_size; /* Size of datagram queue in bytes. */ + + /* + * Version of the code that created + * this context; e.g., VMX. + */ + int user_version; + spinlock_t lock; /* Locks callQueue and handle_arrays. */ + + /* + * queue_pairs attached to. The array of + * handles for queue pairs is accessed + * from the code for QP API, and there + * it is protected by the QP lock. It + * is also accessed from the context + * clean up path, which does not + * require a lock. VMCILock is not + * used to protect the QP array field. + */ + struct vmci_handle_arr *queue_pair_array; + + /* Doorbells created by context. */ + struct vmci_handle_arr *doorbell_array; + + /* Doorbells pending for context. */ + struct vmci_handle_arr *pending_doorbell_array; + + /* Contexts current context is subscribing to. */ + struct list_head notifier_list; + unsigned int n_notifiers; + + struct vmci_host host_context; + u32 priv_flags; + + const struct cred *cred; + bool *notify; /* Notify flag pointer - hosted only. */ + struct page *notify_page; /* Page backing the notify UVA. */ +}; + +/* VMCINotifyAddRemoveInfo: Used to add/remove remote context notifications. */ +struct vmci_ctx_info { + u32 remote_cid; + int result; +}; + +/* VMCICptBufInfo: Used to set/get current context's checkpoint state. */ +struct vmci_ctx_chkpt_buf_info { + u64 cpt_buf; + u32 cpt_type; + u32 buf_size; + s32 result; + u32 _pad; +}; + +/* + * VMCINotificationReceiveInfo: Used to recieve pending notifications + * for doorbells and queue pairs. + */ +struct vmci_ctx_notify_recv_info { + u64 db_handle_buf_uva; + u64 db_handle_buf_size; + u64 qp_handle_buf_uva; + u64 qp_handle_buf_size; + s32 result; + u32 _pad; +}; + +/* + * Utilility function that checks whether two entities are allowed + * to interact. If one of them is restricted, the other one must + * be trusted. + */ +static inline bool vmci_deny_interaction(u32 part_one, u32 part_two) +{ + return ((part_one & VMCI_PRIVILEGE_FLAG_RESTRICTED) && + !(part_two & VMCI_PRIVILEGE_FLAG_TRUSTED)) || + ((part_two & VMCI_PRIVILEGE_FLAG_RESTRICTED) && + !(part_one & VMCI_PRIVILEGE_FLAG_TRUSTED)); +} + +struct vmci_ctx *vmci_ctx_create(u32 cid, u32 flags, + uintptr_t event_hnd, int version, + const struct cred *cred); +void vmci_ctx_destroy(struct vmci_ctx *context); + +bool vmci_ctx_supports_host_qp(struct vmci_ctx *context); +int vmci_ctx_enqueue_datagram(u32 cid, struct vmci_datagram *dg); +int vmci_ctx_dequeue_datagram(struct vmci_ctx *context, + size_t *max_size, struct vmci_datagram **dg); +int vmci_ctx_pending_datagrams(u32 cid, u32 *pending); +struct vmci_ctx *vmci_ctx_get(u32 cid); +void vmci_ctx_put(struct vmci_ctx *context); +bool vmci_ctx_exists(u32 cid); + +int vmci_ctx_add_notification(u32 context_id, u32 remote_cid); +int vmci_ctx_remove_notification(u32 context_id, u32 remote_cid); +int vmci_ctx_get_chkpt_state(u32 context_id, u32 cpt_type, + u32 *num_cids, void **cpt_buf_ptr); +int vmci_ctx_set_chkpt_state(u32 context_id, u32 cpt_type, + u32 num_cids, void *cpt_buf); + +int vmci_ctx_qp_create(struct vmci_ctx *context, struct vmci_handle handle); +int vmci_ctx_qp_destroy(struct vmci_ctx *context, struct vmci_handle handle); +bool vmci_ctx_qp_exists(struct vmci_ctx *context, struct vmci_handle handle); + +void vmci_ctx_check_signal_notify(struct vmci_ctx *context); +void vmci_ctx_unset_notify(struct vmci_ctx *context); + +int vmci_ctx_dbell_create(u32 context_id, struct vmci_handle handle); +int vmci_ctx_dbell_destroy(u32 context_id, struct vmci_handle handle); +int vmci_ctx_dbell_destroy_all(u32 context_id); +int vmci_ctx_notify_dbell(u32 cid, struct vmci_handle handle, + u32 src_priv_flags); + +int vmci_ctx_rcv_notifications_get(u32 context_id, struct vmci_handle_arr + **db_handle_array, struct vmci_handle_arr + **qp_handle_array); +void vmci_ctx_rcv_notifications_release(u32 context_id, struct vmci_handle_arr + *db_handle_array, struct vmci_handle_arr + *qp_handle_array, bool success); + +static inline u32 vmci_ctx_get_id(struct vmci_ctx *context) +{ + if (!context) + return VMCI_INVALID_ID; + return context->cid; +} + +#endif /* _VMCI_CONTEXT_H_ */ diff --git a/drivers/misc/vmw_vmci/vmci_datagram.c b/drivers/misc/vmw_vmci/vmci_datagram.c new file mode 100644 index 00000000000..f3cdd904fe4 --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_datagram.c @@ -0,0 +1,502 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/bug.h> + +#include "vmci_datagram.h" +#include "vmci_resource.h" +#include "vmci_context.h" +#include "vmci_driver.h" +#include "vmci_event.h" +#include "vmci_route.h" + +/* + * struct datagram_entry describes the datagram entity. It is used for datagram + * entities created only on the host. + */ +struct datagram_entry { + struct vmci_resource resource; + u32 flags; + bool run_delayed; + vmci_datagram_recv_cb recv_cb; + void *client_data; + u32 priv_flags; +}; + +struct delayed_datagram_info { + struct datagram_entry *entry; + struct work_struct work; + bool in_dg_host_queue; + /* msg and msg_payload must be together. */ + struct vmci_datagram msg; + u8 msg_payload[]; +}; + +/* Number of in-flight host->host datagrams */ +static atomic_t delayed_dg_host_queue_size = ATOMIC_INIT(0); + +/* + * Create a datagram entry given a handle pointer. + */ +static int dg_create_handle(u32 resource_id, + u32 flags, + u32 priv_flags, + vmci_datagram_recv_cb recv_cb, + void *client_data, struct vmci_handle *out_handle) +{ + int result; + u32 context_id; + struct vmci_handle handle; + struct datagram_entry *entry; + + if ((flags & VMCI_FLAG_WELLKNOWN_DG_HND) != 0) + return VMCI_ERROR_INVALID_ARGS; + + if ((flags & VMCI_FLAG_ANYCID_DG_HND) != 0) { + context_id = VMCI_INVALID_ID; + } else { + context_id = vmci_get_context_id(); + if (context_id == VMCI_INVALID_ID) + return VMCI_ERROR_NO_RESOURCES; + } + + handle = vmci_make_handle(context_id, resource_id); + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) { + pr_warn("Failed allocating memory for datagram entry\n"); + return VMCI_ERROR_NO_MEM; + } + + entry->run_delayed = (flags & VMCI_FLAG_DG_DELAYED_CB) ? true : false; + entry->flags = flags; + entry->recv_cb = recv_cb; + entry->client_data = client_data; + entry->priv_flags = priv_flags; + + /* Make datagram resource live. */ + result = vmci_resource_add(&entry->resource, + VMCI_RESOURCE_TYPE_DATAGRAM, + handle); + if (result != VMCI_SUCCESS) { + pr_warn("Failed to add new resource (handle=0x%x:0x%x), error: %d\n", + handle.context, handle.resource, result); + kfree(entry); + return result; + } + + *out_handle = vmci_resource_handle(&entry->resource); + return VMCI_SUCCESS; +} + +/* + * Internal utility function with the same purpose as + * vmci_datagram_get_priv_flags that also takes a context_id. + */ +static int vmci_datagram_get_priv_flags(u32 context_id, + struct vmci_handle handle, + u32 *priv_flags) +{ + if (context_id == VMCI_INVALID_ID) + return VMCI_ERROR_INVALID_ARGS; + + if (context_id == VMCI_HOST_CONTEXT_ID) { + struct datagram_entry *src_entry; + struct vmci_resource *resource; + + resource = vmci_resource_by_handle(handle, + VMCI_RESOURCE_TYPE_DATAGRAM); + if (!resource) + return VMCI_ERROR_INVALID_ARGS; + + src_entry = container_of(resource, struct datagram_entry, + resource); + *priv_flags = src_entry->priv_flags; + vmci_resource_put(resource); + } else if (context_id == VMCI_HYPERVISOR_CONTEXT_ID) + *priv_flags = VMCI_MAX_PRIVILEGE_FLAGS; + else + *priv_flags = vmci_context_get_priv_flags(context_id); + + return VMCI_SUCCESS; +} + +/* + * Calls the specified callback in a delayed context. + */ +static void dg_delayed_dispatch(struct work_struct *work) +{ + struct delayed_datagram_info *dg_info = + container_of(work, struct delayed_datagram_info, work); + + dg_info->entry->recv_cb(dg_info->entry->client_data, &dg_info->msg); + + vmci_resource_put(&dg_info->entry->resource); + + if (dg_info->in_dg_host_queue) + atomic_dec(&delayed_dg_host_queue_size); + + kfree(dg_info); +} + +/* + * Dispatch datagram as a host, to the host, or other vm context. This + * function cannot dispatch to hypervisor context handlers. This should + * have been handled before we get here by vmci_datagram_dispatch. + * Returns number of bytes sent on success, error code otherwise. + */ +static int dg_dispatch_as_host(u32 context_id, struct vmci_datagram *dg) +{ + int retval; + size_t dg_size; + u32 src_priv_flags; + + dg_size = VMCI_DG_SIZE(dg); + + /* Host cannot send to the hypervisor. */ + if (dg->dst.context == VMCI_HYPERVISOR_CONTEXT_ID) + return VMCI_ERROR_DST_UNREACHABLE; + + /* Check that source handle matches sending context. */ + if (dg->src.context != context_id) { + pr_devel("Sender context (ID=0x%x) is not owner of src datagram entry (handle=0x%x:0x%x)\n", + context_id, dg->src.context, dg->src.resource); + return VMCI_ERROR_NO_ACCESS; + } + + /* Get hold of privileges of sending endpoint. */ + retval = vmci_datagram_get_priv_flags(context_id, dg->src, + &src_priv_flags); + if (retval != VMCI_SUCCESS) { + pr_warn("Couldn't get privileges (handle=0x%x:0x%x)\n", + dg->src.context, dg->src.resource); + return retval; + } + + /* Determine if we should route to host or guest destination. */ + if (dg->dst.context == VMCI_HOST_CONTEXT_ID) { + /* Route to host datagram entry. */ + struct datagram_entry *dst_entry; + struct vmci_resource *resource; + + if (dg->src.context == VMCI_HYPERVISOR_CONTEXT_ID && + dg->dst.resource == VMCI_EVENT_HANDLER) { + return vmci_event_dispatch(dg); + } + + resource = vmci_resource_by_handle(dg->dst, + VMCI_RESOURCE_TYPE_DATAGRAM); + if (!resource) { + pr_devel("Sending to invalid destination (handle=0x%x:0x%x)\n", + dg->dst.context, dg->dst.resource); + return VMCI_ERROR_INVALID_RESOURCE; + } + dst_entry = container_of(resource, struct datagram_entry, + resource); + if (vmci_deny_interaction(src_priv_flags, + dst_entry->priv_flags)) { + vmci_resource_put(resource); + return VMCI_ERROR_NO_ACCESS; + } + + /* + * If a VMCI datagram destined for the host is also sent by the + * host, we always run it delayed. This ensures that no locks + * are held when the datagram callback runs. + */ + if (dst_entry->run_delayed || + dg->src.context == VMCI_HOST_CONTEXT_ID) { + struct delayed_datagram_info *dg_info; + + if (atomic_add_return(1, &delayed_dg_host_queue_size) + == VMCI_MAX_DELAYED_DG_HOST_QUEUE_SIZE) { + atomic_dec(&delayed_dg_host_queue_size); + vmci_resource_put(resource); + return VMCI_ERROR_NO_MEM; + } + + dg_info = kmalloc(sizeof(*dg_info) + + (size_t) dg->payload_size, GFP_ATOMIC); + if (!dg_info) { + atomic_dec(&delayed_dg_host_queue_size); + vmci_resource_put(resource); + return VMCI_ERROR_NO_MEM; + } + + dg_info->in_dg_host_queue = true; + dg_info->entry = dst_entry; + memcpy(&dg_info->msg, dg, dg_size); + + INIT_WORK(&dg_info->work, dg_delayed_dispatch); + schedule_work(&dg_info->work); + retval = VMCI_SUCCESS; + + } else { + retval = dst_entry->recv_cb(dst_entry->client_data, dg); + vmci_resource_put(resource); + if (retval < VMCI_SUCCESS) + return retval; + } + } else { + /* Route to destination VM context. */ + struct vmci_datagram *new_dg; + + if (context_id != dg->dst.context) { + if (vmci_deny_interaction(src_priv_flags, + vmci_context_get_priv_flags + (dg->dst.context))) { + return VMCI_ERROR_NO_ACCESS; + } else if (VMCI_CONTEXT_IS_VM(context_id)) { + /* + * If the sending context is a VM, it + * cannot reach another VM. + */ + + pr_devel("Datagram communication between VMs not supported (src=0x%x, dst=0x%x)\n", + context_id, dg->dst.context); + return VMCI_ERROR_DST_UNREACHABLE; + } + } + + /* We make a copy to enqueue. */ + new_dg = kmalloc(dg_size, GFP_KERNEL); + if (new_dg == NULL) + return VMCI_ERROR_NO_MEM; + + memcpy(new_dg, dg, dg_size); + retval = vmci_ctx_enqueue_datagram(dg->dst.context, new_dg); + if (retval < VMCI_SUCCESS) { + kfree(new_dg); + return retval; + } + } + + /* + * We currently truncate the size to signed 32 bits. This doesn't + * matter for this handler as it only support 4Kb messages. + */ + return (int)dg_size; +} + +/* + * Dispatch datagram as a guest, down through the VMX and potentially to + * the host. + * Returns number of bytes sent on success, error code otherwise. + */ +static int dg_dispatch_as_guest(struct vmci_datagram *dg) +{ + int retval; + struct vmci_resource *resource; + + resource = vmci_resource_by_handle(dg->src, + VMCI_RESOURCE_TYPE_DATAGRAM); + if (!resource) + return VMCI_ERROR_NO_HANDLE; + + retval = vmci_send_datagram(dg); + vmci_resource_put(resource); + return retval; +} + +/* + * Dispatch datagram. This will determine the routing for the datagram + * and dispatch it accordingly. + * Returns number of bytes sent on success, error code otherwise. + */ +int vmci_datagram_dispatch(u32 context_id, + struct vmci_datagram *dg, bool from_guest) +{ + int retval; + enum vmci_route route; + + BUILD_BUG_ON(sizeof(struct vmci_datagram) != 24); + + if (VMCI_DG_SIZE(dg) > VMCI_MAX_DG_SIZE) { + pr_devel("Payload (size=%llu bytes) too big to send\n", + (unsigned long long)dg->payload_size); + return VMCI_ERROR_INVALID_ARGS; + } + + retval = vmci_route(&dg->src, &dg->dst, from_guest, &route); + if (retval < VMCI_SUCCESS) { + pr_devel("Failed to route datagram (src=0x%x, dst=0x%x, err=%d)\n", + dg->src.context, dg->dst.context, retval); + return retval; + } + + if (VMCI_ROUTE_AS_HOST == route) { + if (VMCI_INVALID_ID == context_id) + context_id = VMCI_HOST_CONTEXT_ID; + return dg_dispatch_as_host(context_id, dg); + } + + if (VMCI_ROUTE_AS_GUEST == route) + return dg_dispatch_as_guest(dg); + + pr_warn("Unknown route (%d) for datagram\n", route); + return VMCI_ERROR_DST_UNREACHABLE; +} + +/* + * Invoke the handler for the given datagram. This is intended to be + * called only when acting as a guest and receiving a datagram from the + * virtual device. + */ +int vmci_datagram_invoke_guest_handler(struct vmci_datagram *dg) +{ + struct vmci_resource *resource; + struct datagram_entry *dst_entry; + + resource = vmci_resource_by_handle(dg->dst, + VMCI_RESOURCE_TYPE_DATAGRAM); + if (!resource) { + pr_devel("destination (handle=0x%x:0x%x) doesn't exist\n", + dg->dst.context, dg->dst.resource); + return VMCI_ERROR_NO_HANDLE; + } + + dst_entry = container_of(resource, struct datagram_entry, resource); + if (dst_entry->run_delayed) { + struct delayed_datagram_info *dg_info; + + dg_info = kmalloc(sizeof(*dg_info) + (size_t)dg->payload_size, + GFP_ATOMIC); + if (!dg_info) { + vmci_resource_put(resource); + return VMCI_ERROR_NO_MEM; + } + + dg_info->in_dg_host_queue = false; + dg_info->entry = dst_entry; + memcpy(&dg_info->msg, dg, VMCI_DG_SIZE(dg)); + + INIT_WORK(&dg_info->work, dg_delayed_dispatch); + schedule_work(&dg_info->work); + } else { + dst_entry->recv_cb(dst_entry->client_data, dg); + vmci_resource_put(resource); + } + + return VMCI_SUCCESS; +} + +/* + * vmci_datagram_create_handle_priv() - Create host context datagram endpoint + * @resource_id: The resource ID. + * @flags: Datagram Flags. + * @priv_flags: Privilege Flags. + * @recv_cb: Callback when receiving datagrams. + * @client_data: Pointer for a datagram_entry struct + * @out_handle: vmci_handle that is populated as a result of this function. + * + * Creates a host context datagram endpoint and returns a handle to it. + */ +int vmci_datagram_create_handle_priv(u32 resource_id, + u32 flags, + u32 priv_flags, + vmci_datagram_recv_cb recv_cb, + void *client_data, + struct vmci_handle *out_handle) +{ + if (out_handle == NULL) + return VMCI_ERROR_INVALID_ARGS; + + if (recv_cb == NULL) { + pr_devel("Client callback needed when creating datagram\n"); + return VMCI_ERROR_INVALID_ARGS; + } + + if (priv_flags & ~VMCI_PRIVILEGE_ALL_FLAGS) + return VMCI_ERROR_INVALID_ARGS; + + return dg_create_handle(resource_id, flags, priv_flags, recv_cb, + client_data, out_handle); +} +EXPORT_SYMBOL_GPL(vmci_datagram_create_handle_priv); + +/* + * vmci_datagram_create_handle() - Create host context datagram endpoint + * @resource_id: Resource ID. + * @flags: Datagram Flags. + * @recv_cb: Callback when receiving datagrams. + * @client_ata: Pointer for a datagram_entry struct + * @out_handle: vmci_handle that is populated as a result of this function. + * + * Creates a host context datagram endpoint and returns a handle to + * it. Same as vmci_datagram_create_handle_priv without the priviledge + * flags argument. + */ +int vmci_datagram_create_handle(u32 resource_id, + u32 flags, + vmci_datagram_recv_cb recv_cb, + void *client_data, + struct vmci_handle *out_handle) +{ + return vmci_datagram_create_handle_priv( + resource_id, flags, + VMCI_DEFAULT_PROC_PRIVILEGE_FLAGS, + recv_cb, client_data, + out_handle); +} +EXPORT_SYMBOL_GPL(vmci_datagram_create_handle); + +/* + * vmci_datagram_destroy_handle() - Destroys datagram handle + * @handle: vmci_handle to be destroyed and reaped. + * + * Use this function to destroy any datagram handles created by + * vmci_datagram_create_handle{,Priv} functions. + */ +int vmci_datagram_destroy_handle(struct vmci_handle handle) +{ + struct datagram_entry *entry; + struct vmci_resource *resource; + + resource = vmci_resource_by_handle(handle, VMCI_RESOURCE_TYPE_DATAGRAM); + if (!resource) { + pr_devel("Failed to destroy datagram (handle=0x%x:0x%x)\n", + handle.context, handle.resource); + return VMCI_ERROR_NOT_FOUND; + } + + entry = container_of(resource, struct datagram_entry, resource); + + vmci_resource_put(&entry->resource); + vmci_resource_remove(&entry->resource); + kfree(entry); + + return VMCI_SUCCESS; +} +EXPORT_SYMBOL_GPL(vmci_datagram_destroy_handle); + +/* + * vmci_datagram_send() - Send a datagram + * @msg: The datagram to send. + * + * Sends the provided datagram on its merry way. + */ +int vmci_datagram_send(struct vmci_datagram *msg) +{ + if (msg == NULL) + return VMCI_ERROR_INVALID_ARGS; + + return vmci_datagram_dispatch(VMCI_INVALID_ID, msg, false); +} +EXPORT_SYMBOL_GPL(vmci_datagram_send); diff --git a/drivers/misc/vmw_vmci/vmci_datagram.h b/drivers/misc/vmw_vmci/vmci_datagram.h new file mode 100644 index 00000000000..eb4aab7f64e --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_datagram.h @@ -0,0 +1,52 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef _VMCI_DATAGRAM_H_ +#define _VMCI_DATAGRAM_H_ + +#include <linux/types.h> +#include <linux/list.h> + +#include "vmci_context.h" + +#define VMCI_MAX_DELAYED_DG_HOST_QUEUE_SIZE 256 + +/* + * The struct vmci_datagram_queue_entry is a queue header for the in-kernel VMCI + * datagram queues. It is allocated in non-paged memory, as the + * content is accessed while holding a spinlock. The pending datagram + * itself may be allocated from paged memory. We shadow the size of + * the datagram in the non-paged queue entry as this size is used + * while holding the same spinlock as above. + */ +struct vmci_datagram_queue_entry { + struct list_head list_item; /* For queuing. */ + size_t dg_size; /* Size of datagram. */ + struct vmci_datagram *dg; /* Pending datagram. */ +}; + +/* VMCIDatagramSendRecvInfo */ +struct vmci_datagram_snd_rcv_info { + u64 addr; + u32 len; + s32 result; +}; + +/* Datagram API for non-public use. */ +int vmci_datagram_dispatch(u32 context_id, struct vmci_datagram *dg, + bool from_guest); +int vmci_datagram_invoke_guest_handler(struct vmci_datagram *dg); + +#endif /* _VMCI_DATAGRAM_H_ */ diff --git a/drivers/misc/vmw_vmci/vmci_doorbell.c b/drivers/misc/vmw_vmci/vmci_doorbell.c new file mode 100644 index 00000000000..a8cee33ae8d --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_doorbell.c @@ -0,0 +1,601 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/completion.h> +#include <linux/hash.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/slab.h> + +#include "vmci_datagram.h" +#include "vmci_doorbell.h" +#include "vmci_resource.h" +#include "vmci_driver.h" +#include "vmci_route.h" + + +#define VMCI_DOORBELL_INDEX_BITS 6 +#define VMCI_DOORBELL_INDEX_TABLE_SIZE (1 << VMCI_DOORBELL_INDEX_BITS) +#define VMCI_DOORBELL_HASH(_idx) hash_32(_idx, VMCI_DOORBELL_INDEX_BITS) + +/* + * DoorbellEntry describes the a doorbell notification handle allocated by the + * host. + */ +struct dbell_entry { + struct vmci_resource resource; + struct hlist_node node; + struct work_struct work; + vmci_callback notify_cb; + void *client_data; + u32 idx; + u32 priv_flags; + bool run_delayed; + atomic_t active; /* Only used by guest personality */ +}; + +/* The VMCI index table keeps track of currently registered doorbells. */ +struct dbell_index_table { + spinlock_t lock; /* Index table lock */ + struct hlist_head entries[VMCI_DOORBELL_INDEX_TABLE_SIZE]; +}; + +static struct dbell_index_table vmci_doorbell_it = { + .lock = __SPIN_LOCK_UNLOCKED(vmci_doorbell_it.lock), +}; + +/* + * The max_notify_idx is one larger than the currently known bitmap index in + * use, and is used to determine how much of the bitmap needs to be scanned. + */ +static u32 max_notify_idx; + +/* + * The notify_idx_count is used for determining whether there are free entries + * within the bitmap (if notify_idx_count + 1 < max_notify_idx). + */ +static u32 notify_idx_count; + +/* + * The last_notify_idx_reserved is used to track the last index handed out - in + * the case where multiple handles share a notification index, we hand out + * indexes round robin based on last_notify_idx_reserved. + */ +static u32 last_notify_idx_reserved; + +/* This is a one entry cache used to by the index allocation. */ +static u32 last_notify_idx_released = PAGE_SIZE; + + +/* + * Utility function that retrieves the privilege flags associated + * with a given doorbell handle. For guest endpoints, the + * privileges are determined by the context ID, but for host + * endpoints privileges are associated with the complete + * handle. Hypervisor endpoints are not yet supported. + */ +int vmci_dbell_get_priv_flags(struct vmci_handle handle, u32 *priv_flags) +{ + if (priv_flags == NULL || handle.context == VMCI_INVALID_ID) + return VMCI_ERROR_INVALID_ARGS; + + if (handle.context == VMCI_HOST_CONTEXT_ID) { + struct dbell_entry *entry; + struct vmci_resource *resource; + + resource = vmci_resource_by_handle(handle, + VMCI_RESOURCE_TYPE_DOORBELL); + if (!resource) + return VMCI_ERROR_NOT_FOUND; + + entry = container_of(resource, struct dbell_entry, resource); + *priv_flags = entry->priv_flags; + vmci_resource_put(resource); + } else if (handle.context == VMCI_HYPERVISOR_CONTEXT_ID) { + /* + * Hypervisor endpoints for notifications are not + * supported (yet). + */ + return VMCI_ERROR_INVALID_ARGS; + } else { + *priv_flags = vmci_context_get_priv_flags(handle.context); + } + + return VMCI_SUCCESS; +} + +/* + * Find doorbell entry by bitmap index. + */ +static struct dbell_entry *dbell_index_table_find(u32 idx) +{ + u32 bucket = VMCI_DOORBELL_HASH(idx); + struct dbell_entry *dbell; + + hlist_for_each_entry(dbell, &vmci_doorbell_it.entries[bucket], + node) { + if (idx == dbell->idx) + return dbell; + } + + return NULL; +} + +/* + * Add the given entry to the index table. This willi take a reference to the + * entry's resource so that the entry is not deleted before it is removed from + * the * table. + */ +static void dbell_index_table_add(struct dbell_entry *entry) +{ + u32 bucket; + u32 new_notify_idx; + + vmci_resource_get(&entry->resource); + + spin_lock_bh(&vmci_doorbell_it.lock); + + /* + * Below we try to allocate an index in the notification + * bitmap with "not too much" sharing between resources. If we + * use less that the full bitmap, we either add to the end if + * there are no unused flags within the currently used area, + * or we search for unused ones. If we use the full bitmap, we + * allocate the index round robin. + */ + if (max_notify_idx < PAGE_SIZE || notify_idx_count < PAGE_SIZE) { + if (last_notify_idx_released < max_notify_idx && + !dbell_index_table_find(last_notify_idx_released)) { + new_notify_idx = last_notify_idx_released; + last_notify_idx_released = PAGE_SIZE; + } else { + bool reused = false; + new_notify_idx = last_notify_idx_reserved; + if (notify_idx_count + 1 < max_notify_idx) { + do { + if (!dbell_index_table_find + (new_notify_idx)) { + reused = true; + break; + } + new_notify_idx = (new_notify_idx + 1) % + max_notify_idx; + } while (new_notify_idx != + last_notify_idx_released); + } + if (!reused) { + new_notify_idx = max_notify_idx; + max_notify_idx++; + } + } + } else { + new_notify_idx = (last_notify_idx_reserved + 1) % PAGE_SIZE; + } + + last_notify_idx_reserved = new_notify_idx; + notify_idx_count++; + + entry->idx = new_notify_idx; + bucket = VMCI_DOORBELL_HASH(entry->idx); + hlist_add_head(&entry->node, &vmci_doorbell_it.entries[bucket]); + + spin_unlock_bh(&vmci_doorbell_it.lock); +} + +/* + * Remove the given entry from the index table. This will release() the + * entry's resource. + */ +static void dbell_index_table_remove(struct dbell_entry *entry) +{ + spin_lock_bh(&vmci_doorbell_it.lock); + + hlist_del_init(&entry->node); + + notify_idx_count--; + if (entry->idx == max_notify_idx - 1) { + /* + * If we delete an entry with the maximum known + * notification index, we take the opportunity to + * prune the current max. As there might be other + * unused indices immediately below, we lower the + * maximum until we hit an index in use. + */ + while (max_notify_idx > 0 && + !dbell_index_table_find(max_notify_idx - 1)) + max_notify_idx--; + } + + last_notify_idx_released = entry->idx; + + spin_unlock_bh(&vmci_doorbell_it.lock); + + vmci_resource_put(&entry->resource); +} + +/* + * Creates a link between the given doorbell handle and the given + * index in the bitmap in the device backend. A notification state + * is created in hypervisor. + */ +static int dbell_link(struct vmci_handle handle, u32 notify_idx) +{ + struct vmci_doorbell_link_msg link_msg; + + link_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_DOORBELL_LINK); + link_msg.hdr.src = VMCI_ANON_SRC_HANDLE; + link_msg.hdr.payload_size = sizeof(link_msg) - VMCI_DG_HEADERSIZE; + link_msg.handle = handle; + link_msg.notify_idx = notify_idx; + + return vmci_send_datagram(&link_msg.hdr); +} + +/* + * Unlinks the given doorbell handle from an index in the bitmap in + * the device backend. The notification state is destroyed in hypervisor. + */ +static int dbell_unlink(struct vmci_handle handle) +{ + struct vmci_doorbell_unlink_msg unlink_msg; + + unlink_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_DOORBELL_UNLINK); + unlink_msg.hdr.src = VMCI_ANON_SRC_HANDLE; + unlink_msg.hdr.payload_size = sizeof(unlink_msg) - VMCI_DG_HEADERSIZE; + unlink_msg.handle = handle; + + return vmci_send_datagram(&unlink_msg.hdr); +} + +/* + * Notify another guest or the host. We send a datagram down to the + * host via the hypervisor with the notification info. + */ +static int dbell_notify_as_guest(struct vmci_handle handle, u32 priv_flags) +{ + struct vmci_doorbell_notify_msg notify_msg; + + notify_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_DOORBELL_NOTIFY); + notify_msg.hdr.src = VMCI_ANON_SRC_HANDLE; + notify_msg.hdr.payload_size = sizeof(notify_msg) - VMCI_DG_HEADERSIZE; + notify_msg.handle = handle; + + return vmci_send_datagram(¬ify_msg.hdr); +} + +/* + * Calls the specified callback in a delayed context. + */ +static void dbell_delayed_dispatch(struct work_struct *work) +{ + struct dbell_entry *entry = container_of(work, + struct dbell_entry, work); + + entry->notify_cb(entry->client_data); + vmci_resource_put(&entry->resource); +} + +/* + * Dispatches a doorbell notification to the host context. + */ +int vmci_dbell_host_context_notify(u32 src_cid, struct vmci_handle handle) +{ + struct dbell_entry *entry; + struct vmci_resource *resource; + + if (vmci_handle_is_invalid(handle)) { + pr_devel("Notifying an invalid doorbell (handle=0x%x:0x%x)\n", + handle.context, handle.resource); + return VMCI_ERROR_INVALID_ARGS; + } + + resource = vmci_resource_by_handle(handle, + VMCI_RESOURCE_TYPE_DOORBELL); + if (!resource) { + pr_devel("Notifying an unknown doorbell (handle=0x%x:0x%x)\n", + handle.context, handle.resource); + return VMCI_ERROR_NOT_FOUND; + } + + entry = container_of(resource, struct dbell_entry, resource); + if (entry->run_delayed) { + schedule_work(&entry->work); + } else { + entry->notify_cb(entry->client_data); + vmci_resource_put(resource); + } + + return VMCI_SUCCESS; +} + +/* + * Register the notification bitmap with the host. + */ +bool vmci_dbell_register_notification_bitmap(u32 bitmap_ppn) +{ + int result; + struct vmci_notify_bm_set_msg bitmap_set_msg; + + bitmap_set_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_SET_NOTIFY_BITMAP); + bitmap_set_msg.hdr.src = VMCI_ANON_SRC_HANDLE; + bitmap_set_msg.hdr.payload_size = sizeof(bitmap_set_msg) - + VMCI_DG_HEADERSIZE; + bitmap_set_msg.bitmap_ppn = bitmap_ppn; + + result = vmci_send_datagram(&bitmap_set_msg.hdr); + if (result != VMCI_SUCCESS) { + pr_devel("Failed to register (PPN=%u) as notification bitmap (error=%d)\n", + bitmap_ppn, result); + return false; + } + return true; +} + +/* + * Executes or schedules the handlers for a given notify index. + */ +static void dbell_fire_entries(u32 notify_idx) +{ + u32 bucket = VMCI_DOORBELL_HASH(notify_idx); + struct dbell_entry *dbell; + + spin_lock_bh(&vmci_doorbell_it.lock); + + hlist_for_each_entry(dbell, &vmci_doorbell_it.entries[bucket], node) { + if (dbell->idx == notify_idx && + atomic_read(&dbell->active) == 1) { + if (dbell->run_delayed) { + vmci_resource_get(&dbell->resource); + schedule_work(&dbell->work); + } else { + dbell->notify_cb(dbell->client_data); + } + } + } + + spin_unlock_bh(&vmci_doorbell_it.lock); +} + +/* + * Scans the notification bitmap, collects pending notifications, + * resets the bitmap and invokes appropriate callbacks. + */ +void vmci_dbell_scan_notification_entries(u8 *bitmap) +{ + u32 idx; + + for (idx = 0; idx < max_notify_idx; idx++) { + if (bitmap[idx] & 0x1) { + bitmap[idx] &= ~1; + dbell_fire_entries(idx); + } + } +} + +/* + * vmci_doorbell_create() - Creates a doorbell + * @handle: A handle used to track the resource. Can be invalid. + * @flags: Flag that determines context of callback. + * @priv_flags: Privileges flags. + * @notify_cb: The callback to be ivoked when the doorbell fires. + * @client_data: A parameter to be passed to the callback. + * + * Creates a doorbell with the given callback. If the handle is + * VMCI_INVALID_HANDLE, a free handle will be assigned, if + * possible. The callback can be run immediately (potentially with + * locks held - the default) or delayed (in a kernel thread) by + * specifying the flag VMCI_FLAG_DELAYED_CB. If delayed execution + * is selected, a given callback may not be run if the kernel is + * unable to allocate memory for the delayed execution (highly + * unlikely). + */ +int vmci_doorbell_create(struct vmci_handle *handle, + u32 flags, + u32 priv_flags, + vmci_callback notify_cb, void *client_data) +{ + struct dbell_entry *entry; + struct vmci_handle new_handle; + int result; + + if (!handle || !notify_cb || flags & ~VMCI_FLAG_DELAYED_CB || + priv_flags & ~VMCI_PRIVILEGE_ALL_FLAGS) + return VMCI_ERROR_INVALID_ARGS; + + entry = kmalloc(sizeof(*entry), GFP_KERNEL); + if (entry == NULL) { + pr_warn("Failed allocating memory for datagram entry\n"); + return VMCI_ERROR_NO_MEM; + } + + if (vmci_handle_is_invalid(*handle)) { + u32 context_id = vmci_get_context_id(); + + /* Let resource code allocate a free ID for us */ + new_handle = vmci_make_handle(context_id, VMCI_INVALID_ID); + } else { + bool valid_context = false; + + /* + * Validate the handle. We must do both of the checks below + * because we can be acting as both a host and a guest at the + * same time. We always allow the host context ID, since the + * host functionality is in practice always there with the + * unified driver. + */ + if (handle->context == VMCI_HOST_CONTEXT_ID || + (vmci_guest_code_active() && + vmci_get_context_id() == handle->context)) { + valid_context = true; + } + + if (!valid_context || handle->resource == VMCI_INVALID_ID) { + pr_devel("Invalid argument (handle=0x%x:0x%x)\n", + handle->context, handle->resource); + result = VMCI_ERROR_INVALID_ARGS; + goto free_mem; + } + + new_handle = *handle; + } + + entry->idx = 0; + INIT_HLIST_NODE(&entry->node); + entry->priv_flags = priv_flags; + INIT_WORK(&entry->work, dbell_delayed_dispatch); + entry->run_delayed = flags & VMCI_FLAG_DELAYED_CB; + entry->notify_cb = notify_cb; + entry->client_data = client_data; + atomic_set(&entry->active, 0); + + result = vmci_resource_add(&entry->resource, + VMCI_RESOURCE_TYPE_DOORBELL, + new_handle); + if (result != VMCI_SUCCESS) { + pr_warn("Failed to add new resource (handle=0x%x:0x%x), error: %d\n", + new_handle.context, new_handle.resource, result); + goto free_mem; + } + + new_handle = vmci_resource_handle(&entry->resource); + if (vmci_guest_code_active()) { + dbell_index_table_add(entry); + result = dbell_link(new_handle, entry->idx); + if (VMCI_SUCCESS != result) + goto destroy_resource; + + atomic_set(&entry->active, 1); + } + + *handle = new_handle; + + return result; + + destroy_resource: + dbell_index_table_remove(entry); + vmci_resource_remove(&entry->resource); + free_mem: + kfree(entry); + return result; +} +EXPORT_SYMBOL_GPL(vmci_doorbell_create); + +/* + * vmci_doorbell_destroy() - Destroy a doorbell. + * @handle: The handle tracking the resource. + * + * Destroys a doorbell previously created with vmcii_doorbell_create. This + * operation may block waiting for a callback to finish. + */ +int vmci_doorbell_destroy(struct vmci_handle handle) +{ + struct dbell_entry *entry; + struct vmci_resource *resource; + + if (vmci_handle_is_invalid(handle)) + return VMCI_ERROR_INVALID_ARGS; + + resource = vmci_resource_by_handle(handle, + VMCI_RESOURCE_TYPE_DOORBELL); + if (!resource) { + pr_devel("Failed to destroy doorbell (handle=0x%x:0x%x)\n", + handle.context, handle.resource); + return VMCI_ERROR_NOT_FOUND; + } + + entry = container_of(resource, struct dbell_entry, resource); + + if (vmci_guest_code_active()) { + int result; + + dbell_index_table_remove(entry); + + result = dbell_unlink(handle); + if (VMCI_SUCCESS != result) { + + /* + * The only reason this should fail would be + * an inconsistency between guest and + * hypervisor state, where the guest believes + * it has an active registration whereas the + * hypervisor doesn't. One case where this may + * happen is if a doorbell is unregistered + * following a hibernation at a time where the + * doorbell state hasn't been restored on the + * hypervisor side yet. Since the handle has + * now been removed in the guest, we just + * print a warning and return success. + */ + pr_devel("Unlink of doorbell (handle=0x%x:0x%x) unknown by hypervisor (error=%d)\n", + handle.context, handle.resource, result); + } + } + + /* + * Now remove the resource from the table. It might still be in use + * after this, in a callback or still on the delayed work queue. + */ + vmci_resource_put(&entry->resource); + vmci_resource_remove(&entry->resource); + + kfree(entry); + + return VMCI_SUCCESS; +} +EXPORT_SYMBOL_GPL(vmci_doorbell_destroy); + +/* + * vmci_doorbell_notify() - Ring the doorbell (and hide in the bushes). + * @dst: The handlle identifying the doorbell resource + * @priv_flags: Priviledge flags. + * + * Generates a notification on the doorbell identified by the + * handle. For host side generation of notifications, the caller + * can specify what the privilege of the calling side is. + */ +int vmci_doorbell_notify(struct vmci_handle dst, u32 priv_flags) +{ + int retval; + enum vmci_route route; + struct vmci_handle src; + + if (vmci_handle_is_invalid(dst) || + (priv_flags & ~VMCI_PRIVILEGE_ALL_FLAGS)) + return VMCI_ERROR_INVALID_ARGS; + + src = VMCI_INVALID_HANDLE; + retval = vmci_route(&src, &dst, false, &route); + if (retval < VMCI_SUCCESS) + return retval; + + if (VMCI_ROUTE_AS_HOST == route) + return vmci_ctx_notify_dbell(VMCI_HOST_CONTEXT_ID, + dst, priv_flags); + + if (VMCI_ROUTE_AS_GUEST == route) + return dbell_notify_as_guest(dst, priv_flags); + + pr_warn("Unknown route (%d) for doorbell\n", route); + return VMCI_ERROR_DST_UNREACHABLE; +} +EXPORT_SYMBOL_GPL(vmci_doorbell_notify); diff --git a/drivers/misc/vmw_vmci/vmci_doorbell.h b/drivers/misc/vmw_vmci/vmci_doorbell.h new file mode 100644 index 00000000000..e4c0b17486a --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_doorbell.h @@ -0,0 +1,51 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef VMCI_DOORBELL_H +#define VMCI_DOORBELL_H + +#include <linux/vmw_vmci_defs.h> +#include <linux/types.h> + +#include "vmci_driver.h" + +/* + * VMCINotifyResourceInfo: Used to create and destroy doorbells, and + * generate a notification for a doorbell or queue pair. + */ +struct vmci_dbell_notify_resource_info { + struct vmci_handle handle; + u16 resource; + u16 action; + s32 result; +}; + +/* + * Structure used for checkpointing the doorbell mappings. It is + * written to the checkpoint as is, so changing this structure will + * break checkpoint compatibility. + */ +struct dbell_cpt_state { + struct vmci_handle handle; + u64 bitmap_idx; +}; + +int vmci_dbell_host_context_notify(u32 src_cid, struct vmci_handle handle); +int vmci_dbell_get_priv_flags(struct vmci_handle handle, u32 *priv_flags); + +bool vmci_dbell_register_notification_bitmap(u32 bitmap_ppn); +void vmci_dbell_scan_notification_entries(u8 *bitmap); + +#endif /* VMCI_DOORBELL_H */ diff --git a/drivers/misc/vmw_vmci/vmci_driver.c b/drivers/misc/vmw_vmci/vmci_driver.c new file mode 100644 index 00000000000..3dee7ae123e --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_driver.c @@ -0,0 +1,117 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/atomic.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> + +#include "vmci_driver.h" +#include "vmci_event.h" + +static bool vmci_disable_host; +module_param_named(disable_host, vmci_disable_host, bool, 0); +MODULE_PARM_DESC(disable_host, + "Disable driver host personality (default=enabled)"); + +static bool vmci_disable_guest; +module_param_named(disable_guest, vmci_disable_guest, bool, 0); +MODULE_PARM_DESC(disable_guest, + "Disable driver guest personality (default=enabled)"); + +static bool vmci_guest_personality_initialized; +static bool vmci_host_personality_initialized; + +/* + * vmci_get_context_id() - Gets the current context ID. + * + * Returns the current context ID. Note that since this is accessed only + * from code running in the host, this always returns the host context ID. + */ +u32 vmci_get_context_id(void) +{ + if (vmci_guest_code_active()) + return vmci_get_vm_context_id(); + else if (vmci_host_code_active()) + return VMCI_HOST_CONTEXT_ID; + + return VMCI_INVALID_ID; +} +EXPORT_SYMBOL_GPL(vmci_get_context_id); + +static int __init vmci_drv_init(void) +{ + int vmci_err; + int error; + + vmci_err = vmci_event_init(); + if (vmci_err < VMCI_SUCCESS) { + pr_err("Failed to initialize VMCIEvent (result=%d)\n", + vmci_err); + return -EINVAL; + } + + if (!vmci_disable_guest) { + error = vmci_guest_init(); + if (error) { + pr_warn("Failed to initialize guest personality (err=%d)\n", + error); + } else { + vmci_guest_personality_initialized = true; + pr_info("Guest personality initialized and is %s\n", + vmci_guest_code_active() ? + "active" : "inactive"); + } + } + + if (!vmci_disable_host) { + error = vmci_host_init(); + if (error) { + pr_warn("Unable to initialize host personality (err=%d)\n", + error); + } else { + vmci_host_personality_initialized = true; + pr_info("Initialized host personality\n"); + } + } + + if (!vmci_guest_personality_initialized && + !vmci_host_personality_initialized) { + vmci_event_exit(); + return -ENODEV; + } + + return 0; +} +module_init(vmci_drv_init); + +static void __exit vmci_drv_exit(void) +{ + if (vmci_guest_personality_initialized) + vmci_guest_exit(); + + if (vmci_host_personality_initialized) + vmci_host_exit(); + + vmci_event_exit(); +} +module_exit(vmci_drv_exit); + +MODULE_AUTHOR("VMware, Inc."); +MODULE_DESCRIPTION("VMware Virtual Machine Communication Interface."); +MODULE_VERSION("1.1.0.0-k"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/vmw_vmci/vmci_driver.h b/drivers/misc/vmw_vmci/vmci_driver.h new file mode 100644 index 00000000000..cee9e977d31 --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_driver.h @@ -0,0 +1,57 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef _VMCI_DRIVER_H_ +#define _VMCI_DRIVER_H_ + +#include <linux/vmw_vmci_defs.h> +#include <linux/wait.h> + +#include "vmci_queue_pair.h" +#include "vmci_context.h" + +enum vmci_obj_type { + VMCIOBJ_VMX_VM = 10, + VMCIOBJ_CONTEXT, + VMCIOBJ_SOCKET, + VMCIOBJ_NOT_SET, +}; + +/* For storing VMCI structures in file handles. */ +struct vmci_obj { + void *ptr; + enum vmci_obj_type type; +}; + +/* + * Needed by other components of this module. It's okay to have one global + * instance of this because there can only ever be one VMCI device. Our + * virtual hardware enforces this. + */ +extern struct pci_dev *vmci_pdev; + +u32 vmci_get_context_id(void); +int vmci_send_datagram(struct vmci_datagram *dg); + +int vmci_host_init(void); +void vmci_host_exit(void); +bool vmci_host_code_active(void); + +int vmci_guest_init(void); +void vmci_guest_exit(void); +bool vmci_guest_code_active(void); +u32 vmci_get_vm_context_id(void); + +#endif /* _VMCI_DRIVER_H_ */ diff --git a/drivers/misc/vmw_vmci/vmci_event.c b/drivers/misc/vmw_vmci/vmci_event.c new file mode 100644 index 00000000000..8449516d6ac --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_event.c @@ -0,0 +1,224 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/slab.h> + +#include "vmci_driver.h" +#include "vmci_event.h" + +#define EVENT_MAGIC 0xEABE0000 +#define VMCI_EVENT_MAX_ATTEMPTS 10 + +struct vmci_subscription { + u32 id; + u32 event; + vmci_event_cb callback; + void *callback_data; + struct list_head node; /* on one of subscriber lists */ +}; + +static struct list_head subscriber_array[VMCI_EVENT_MAX]; +static DEFINE_MUTEX(subscriber_mutex); + +int __init vmci_event_init(void) +{ + int i; + + for (i = 0; i < VMCI_EVENT_MAX; i++) + INIT_LIST_HEAD(&subscriber_array[i]); + + return VMCI_SUCCESS; +} + +void vmci_event_exit(void) +{ + int e; + + /* We free all memory at exit. */ + for (e = 0; e < VMCI_EVENT_MAX; e++) { + struct vmci_subscription *cur, *p2; + list_for_each_entry_safe(cur, p2, &subscriber_array[e], node) { + + /* + * We should never get here because all events + * should have been unregistered before we try + * to unload the driver module. + */ + pr_warn("Unexpected free events occurring\n"); + list_del(&cur->node); + kfree(cur); + } + } +} + +/* + * Find entry. Assumes subscriber_mutex is held. + */ +static struct vmci_subscription *event_find(u32 sub_id) +{ + int e; + + for (e = 0; e < VMCI_EVENT_MAX; e++) { + struct vmci_subscription *cur; + list_for_each_entry(cur, &subscriber_array[e], node) { + if (cur->id == sub_id) + return cur; + } + } + return NULL; +} + +/* + * Actually delivers the events to the subscribers. + * The callback function for each subscriber is invoked. + */ +static void event_deliver(struct vmci_event_msg *event_msg) +{ + struct vmci_subscription *cur; + struct list_head *subscriber_list; + + rcu_read_lock(); + subscriber_list = &subscriber_array[event_msg->event_data.event]; + list_for_each_entry_rcu(cur, subscriber_list, node) { + cur->callback(cur->id, &event_msg->event_data, + cur->callback_data); + } + rcu_read_unlock(); +} + +/* + * Dispatcher for the VMCI_EVENT_RECEIVE datagrams. Calls all + * subscribers for given event. + */ +int vmci_event_dispatch(struct vmci_datagram *msg) +{ + struct vmci_event_msg *event_msg = (struct vmci_event_msg *)msg; + + if (msg->payload_size < sizeof(u32) || + msg->payload_size > sizeof(struct vmci_event_data_max)) + return VMCI_ERROR_INVALID_ARGS; + + if (!VMCI_EVENT_VALID(event_msg->event_data.event)) + return VMCI_ERROR_EVENT_UNKNOWN; + + event_deliver(event_msg); + return VMCI_SUCCESS; +} + +/* + * vmci_event_subscribe() - Subscribe to a given event. + * @event: The event to subscribe to. + * @callback: The callback to invoke upon the event. + * @callback_data: Data to pass to the callback. + * @subscription_id: ID used to track subscription. Used with + * vmci_event_unsubscribe() + * + * Subscribes to the provided event. The callback specified will be + * fired from RCU critical section and therefore must not sleep. + */ +int vmci_event_subscribe(u32 event, + vmci_event_cb callback, + void *callback_data, + u32 *new_subscription_id) +{ + struct vmci_subscription *sub; + int attempts; + int retval; + bool have_new_id = false; + + if (!new_subscription_id) { + pr_devel("%s: Invalid subscription (NULL)\n", __func__); + return VMCI_ERROR_INVALID_ARGS; + } + + if (!VMCI_EVENT_VALID(event) || !callback) { + pr_devel("%s: Failed to subscribe to event (type=%d) (callback=%p) (data=%p)\n", + __func__, event, callback, callback_data); + return VMCI_ERROR_INVALID_ARGS; + } + + sub = kzalloc(sizeof(*sub), GFP_KERNEL); + if (!sub) + return VMCI_ERROR_NO_MEM; + + sub->id = VMCI_EVENT_MAX; + sub->event = event; + sub->callback = callback; + sub->callback_data = callback_data; + INIT_LIST_HEAD(&sub->node); + + mutex_lock(&subscriber_mutex); + + /* Creation of a new event is always allowed. */ + for (attempts = 0; attempts < VMCI_EVENT_MAX_ATTEMPTS; attempts++) { + static u32 subscription_id; + /* + * We try to get an id a couple of time before + * claiming we are out of resources. + */ + + /* Test for duplicate id. */ + if (!event_find(++subscription_id)) { + sub->id = subscription_id; + have_new_id = true; + break; + } + } + + if (have_new_id) { + list_add_rcu(&sub->node, &subscriber_array[event]); + retval = VMCI_SUCCESS; + } else { + retval = VMCI_ERROR_NO_RESOURCES; + } + + mutex_unlock(&subscriber_mutex); + + *new_subscription_id = sub->id; + return retval; +} +EXPORT_SYMBOL_GPL(vmci_event_subscribe); + +/* + * vmci_event_unsubscribe() - unsubscribe from an event. + * @sub_id: A subscription ID as provided by vmci_event_subscribe() + * + * Unsubscribe from given event. Removes it from list and frees it. + * Will return callback_data if requested by caller. + */ +int vmci_event_unsubscribe(u32 sub_id) +{ + struct vmci_subscription *s; + + mutex_lock(&subscriber_mutex); + s = event_find(sub_id); + if (s) + list_del_rcu(&s->node); + mutex_unlock(&subscriber_mutex); + + if (!s) + return VMCI_ERROR_NOT_FOUND; + + synchronize_rcu(); + kfree(s); + + return VMCI_SUCCESS; +} +EXPORT_SYMBOL_GPL(vmci_event_unsubscribe); diff --git a/drivers/misc/vmw_vmci/vmci_event.h b/drivers/misc/vmw_vmci/vmci_event.h new file mode 100644 index 00000000000..7df9b1c0a96 --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_event.h @@ -0,0 +1,25 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef __VMCI_EVENT_H__ +#define __VMCI_EVENT_H__ + +#include <linux/vmw_vmci_api.h> + +int vmci_event_init(void); +void vmci_event_exit(void); +int vmci_event_dispatch(struct vmci_datagram *msg); + +#endif /*__VMCI_EVENT_H__ */ diff --git a/drivers/misc/vmw_vmci/vmci_guest.c b/drivers/misc/vmw_vmci/vmci_guest.c new file mode 100644 index 00000000000..e0d5017785e --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_guest.c @@ -0,0 +1,772 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/moduleparam.h> +#include <linux/interrupt.h> +#include <linux/highmem.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/smp.h> +#include <linux/io.h> +#include <linux/vmalloc.h> + +#include "vmci_datagram.h" +#include "vmci_doorbell.h" +#include "vmci_context.h" +#include "vmci_driver.h" +#include "vmci_event.h" + +#define PCI_VENDOR_ID_VMWARE 0x15AD +#define PCI_DEVICE_ID_VMWARE_VMCI 0x0740 + +#define VMCI_UTIL_NUM_RESOURCES 1 + +static bool vmci_disable_msi; +module_param_named(disable_msi, vmci_disable_msi, bool, 0); +MODULE_PARM_DESC(disable_msi, "Disable MSI use in driver - (default=0)"); + +static bool vmci_disable_msix; +module_param_named(disable_msix, vmci_disable_msix, bool, 0); +MODULE_PARM_DESC(disable_msix, "Disable MSI-X use in driver - (default=0)"); + +static u32 ctx_update_sub_id = VMCI_INVALID_ID; +static u32 vm_context_id = VMCI_INVALID_ID; + +struct vmci_guest_device { + struct device *dev; /* PCI device we are attached to */ + void __iomem *iobase; + + unsigned int irq; + unsigned int intr_type; + bool exclusive_vectors; + struct msix_entry msix_entries[VMCI_MAX_INTRS]; + + struct tasklet_struct datagram_tasklet; + struct tasklet_struct bm_tasklet; + + void *data_buffer; + void *notification_bitmap; + dma_addr_t notification_base; +}; + +/* vmci_dev singleton device and supporting data*/ +struct pci_dev *vmci_pdev; +static struct vmci_guest_device *vmci_dev_g; +static DEFINE_SPINLOCK(vmci_dev_spinlock); + +static atomic_t vmci_num_guest_devices = ATOMIC_INIT(0); + +bool vmci_guest_code_active(void) +{ + return atomic_read(&vmci_num_guest_devices) != 0; +} + +u32 vmci_get_vm_context_id(void) +{ + if (vm_context_id == VMCI_INVALID_ID) { + struct vmci_datagram get_cid_msg; + get_cid_msg.dst = + vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_GET_CONTEXT_ID); + get_cid_msg.src = VMCI_ANON_SRC_HANDLE; + get_cid_msg.payload_size = 0; + vm_context_id = vmci_send_datagram(&get_cid_msg); + } + return vm_context_id; +} + +/* + * VM to hypervisor call mechanism. We use the standard VMware naming + * convention since shared code is calling this function as well. + */ +int vmci_send_datagram(struct vmci_datagram *dg) +{ + unsigned long flags; + int result; + + /* Check args. */ + if (dg == NULL) + return VMCI_ERROR_INVALID_ARGS; + + /* + * Need to acquire spinlock on the device because the datagram + * data may be spread over multiple pages and the monitor may + * interleave device user rpc calls from multiple + * VCPUs. Acquiring the spinlock precludes that + * possibility. Disabling interrupts to avoid incoming + * datagrams during a "rep out" and possibly landing up in + * this function. + */ + spin_lock_irqsave(&vmci_dev_spinlock, flags); + + if (vmci_dev_g) { + iowrite8_rep(vmci_dev_g->iobase + VMCI_DATA_OUT_ADDR, + dg, VMCI_DG_SIZE(dg)); + result = ioread32(vmci_dev_g->iobase + VMCI_RESULT_LOW_ADDR); + } else { + result = VMCI_ERROR_UNAVAILABLE; + } + + spin_unlock_irqrestore(&vmci_dev_spinlock, flags); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_send_datagram); + +/* + * Gets called with the new context id if updated or resumed. + * Context id. + */ +static void vmci_guest_cid_update(u32 sub_id, + const struct vmci_event_data *event_data, + void *client_data) +{ + const struct vmci_event_payld_ctx *ev_payload = + vmci_event_data_const_payload(event_data); + + if (sub_id != ctx_update_sub_id) { + pr_devel("Invalid subscriber (ID=0x%x)\n", sub_id); + return; + } + + if (!event_data || ev_payload->context_id == VMCI_INVALID_ID) { + pr_devel("Invalid event data\n"); + return; + } + + pr_devel("Updating context from (ID=0x%x) to (ID=0x%x) on event (type=%d)\n", + vm_context_id, ev_payload->context_id, event_data->event); + + vm_context_id = ev_payload->context_id; +} + +/* + * Verify that the host supports the hypercalls we need. If it does not, + * try to find fallback hypercalls and use those instead. Returns + * true if required hypercalls (or fallback hypercalls) are + * supported by the host, false otherwise. + */ +static int vmci_check_host_caps(struct pci_dev *pdev) +{ + bool result; + struct vmci_resource_query_msg *msg; + u32 msg_size = sizeof(struct vmci_resource_query_hdr) + + VMCI_UTIL_NUM_RESOURCES * sizeof(u32); + struct vmci_datagram *check_msg; + + check_msg = kmalloc(msg_size, GFP_KERNEL); + if (!check_msg) { + dev_err(&pdev->dev, "%s: Insufficient memory\n", __func__); + return -ENOMEM; + } + + check_msg->dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_RESOURCES_QUERY); + check_msg->src = VMCI_ANON_SRC_HANDLE; + check_msg->payload_size = msg_size - VMCI_DG_HEADERSIZE; + msg = (struct vmci_resource_query_msg *)VMCI_DG_PAYLOAD(check_msg); + + msg->num_resources = VMCI_UTIL_NUM_RESOURCES; + msg->resources[0] = VMCI_GET_CONTEXT_ID; + + /* Checks that hyper calls are supported */ + result = vmci_send_datagram(check_msg) == 0x01; + kfree(check_msg); + + dev_dbg(&pdev->dev, "%s: Host capability check: %s\n", + __func__, result ? "PASSED" : "FAILED"); + + /* We need the vector. There are no fallbacks. */ + return result ? 0 : -ENXIO; +} + +/* + * Reads datagrams from the data in port and dispatches them. We + * always start reading datagrams into only the first page of the + * datagram buffer. If the datagrams don't fit into one page, we + * use the maximum datagram buffer size for the remainder of the + * invocation. This is a simple heuristic for not penalizing + * small datagrams. + * + * This function assumes that it has exclusive access to the data + * in port for the duration of the call. + */ +static void vmci_dispatch_dgs(unsigned long data) +{ + struct vmci_guest_device *vmci_dev = (struct vmci_guest_device *)data; + u8 *dg_in_buffer = vmci_dev->data_buffer; + struct vmci_datagram *dg; + size_t dg_in_buffer_size = VMCI_MAX_DG_SIZE; + size_t current_dg_in_buffer_size = PAGE_SIZE; + size_t remaining_bytes; + + BUILD_BUG_ON(VMCI_MAX_DG_SIZE < PAGE_SIZE); + + ioread8_rep(vmci_dev->iobase + VMCI_DATA_IN_ADDR, + vmci_dev->data_buffer, current_dg_in_buffer_size); + dg = (struct vmci_datagram *)dg_in_buffer; + remaining_bytes = current_dg_in_buffer_size; + + while (dg->dst.resource != VMCI_INVALID_ID || + remaining_bytes > PAGE_SIZE) { + unsigned dg_in_size; + + /* + * When the input buffer spans multiple pages, a datagram can + * start on any page boundary in the buffer. + */ + if (dg->dst.resource == VMCI_INVALID_ID) { + dg = (struct vmci_datagram *)roundup( + (uintptr_t)dg + 1, PAGE_SIZE); + remaining_bytes = + (size_t)(dg_in_buffer + + current_dg_in_buffer_size - + (u8 *)dg); + continue; + } + + dg_in_size = VMCI_DG_SIZE_ALIGNED(dg); + + if (dg_in_size <= dg_in_buffer_size) { + int result; + + /* + * If the remaining bytes in the datagram + * buffer doesn't contain the complete + * datagram, we first make sure we have enough + * room for it and then we read the reminder + * of the datagram and possibly any following + * datagrams. + */ + if (dg_in_size > remaining_bytes) { + if (remaining_bytes != + current_dg_in_buffer_size) { + + /* + * We move the partial + * datagram to the front and + * read the reminder of the + * datagram and possibly + * following calls into the + * following bytes. + */ + memmove(dg_in_buffer, dg_in_buffer + + current_dg_in_buffer_size - + remaining_bytes, + remaining_bytes); + dg = (struct vmci_datagram *) + dg_in_buffer; + } + + if (current_dg_in_buffer_size != + dg_in_buffer_size) + current_dg_in_buffer_size = + dg_in_buffer_size; + + ioread8_rep(vmci_dev->iobase + + VMCI_DATA_IN_ADDR, + vmci_dev->data_buffer + + remaining_bytes, + current_dg_in_buffer_size - + remaining_bytes); + } + + /* + * We special case event datagrams from the + * hypervisor. + */ + if (dg->src.context == VMCI_HYPERVISOR_CONTEXT_ID && + dg->dst.resource == VMCI_EVENT_HANDLER) { + result = vmci_event_dispatch(dg); + } else { + result = vmci_datagram_invoke_guest_handler(dg); + } + if (result < VMCI_SUCCESS) + dev_dbg(vmci_dev->dev, + "Datagram with resource (ID=0x%x) failed (err=%d)\n", + dg->dst.resource, result); + + /* On to the next datagram. */ + dg = (struct vmci_datagram *)((u8 *)dg + + dg_in_size); + } else { + size_t bytes_to_skip; + + /* + * Datagram doesn't fit in datagram buffer of maximal + * size. We drop it. + */ + dev_dbg(vmci_dev->dev, + "Failed to receive datagram (size=%u bytes)\n", + dg_in_size); + + bytes_to_skip = dg_in_size - remaining_bytes; + if (current_dg_in_buffer_size != dg_in_buffer_size) + current_dg_in_buffer_size = dg_in_buffer_size; + + for (;;) { + ioread8_rep(vmci_dev->iobase + + VMCI_DATA_IN_ADDR, + vmci_dev->data_buffer, + current_dg_in_buffer_size); + if (bytes_to_skip <= current_dg_in_buffer_size) + break; + + bytes_to_skip -= current_dg_in_buffer_size; + } + dg = (struct vmci_datagram *)(dg_in_buffer + + bytes_to_skip); + } + + remaining_bytes = + (size_t) (dg_in_buffer + current_dg_in_buffer_size - + (u8 *)dg); + + if (remaining_bytes < VMCI_DG_HEADERSIZE) { + /* Get the next batch of datagrams. */ + + ioread8_rep(vmci_dev->iobase + VMCI_DATA_IN_ADDR, + vmci_dev->data_buffer, + current_dg_in_buffer_size); + dg = (struct vmci_datagram *)dg_in_buffer; + remaining_bytes = current_dg_in_buffer_size; + } + } +} + +/* + * Scans the notification bitmap for raised flags, clears them + * and handles the notifications. + */ +static void vmci_process_bitmap(unsigned long data) +{ + struct vmci_guest_device *dev = (struct vmci_guest_device *)data; + + if (!dev->notification_bitmap) { + dev_dbg(dev->dev, "No bitmap present in %s\n", __func__); + return; + } + + vmci_dbell_scan_notification_entries(dev->notification_bitmap); +} + +/* + * Enable MSI-X. Try exclusive vectors first, then shared vectors. + */ +static int vmci_enable_msix(struct pci_dev *pdev, + struct vmci_guest_device *vmci_dev) +{ + int i; + int result; + + for (i = 0; i < VMCI_MAX_INTRS; ++i) { + vmci_dev->msix_entries[i].entry = i; + vmci_dev->msix_entries[i].vector = i; + } + + result = pci_enable_msix_exact(pdev, + vmci_dev->msix_entries, VMCI_MAX_INTRS); + if (result == 0) + vmci_dev->exclusive_vectors = true; + else if (result == -ENOSPC) + result = pci_enable_msix_exact(pdev, vmci_dev->msix_entries, 1); + + return result; +} + +/* + * Interrupt handler for legacy or MSI interrupt, or for first MSI-X + * interrupt (vector VMCI_INTR_DATAGRAM). + */ +static irqreturn_t vmci_interrupt(int irq, void *_dev) +{ + struct vmci_guest_device *dev = _dev; + + /* + * If we are using MSI-X with exclusive vectors then we simply schedule + * the datagram tasklet, since we know the interrupt was meant for us. + * Otherwise we must read the ICR to determine what to do. + */ + + if (dev->intr_type == VMCI_INTR_TYPE_MSIX && dev->exclusive_vectors) { + tasklet_schedule(&dev->datagram_tasklet); + } else { + unsigned int icr; + + /* Acknowledge interrupt and determine what needs doing. */ + icr = ioread32(dev->iobase + VMCI_ICR_ADDR); + if (icr == 0 || icr == ~0) + return IRQ_NONE; + + if (icr & VMCI_ICR_DATAGRAM) { + tasklet_schedule(&dev->datagram_tasklet); + icr &= ~VMCI_ICR_DATAGRAM; + } + + if (icr & VMCI_ICR_NOTIFICATION) { + tasklet_schedule(&dev->bm_tasklet); + icr &= ~VMCI_ICR_NOTIFICATION; + } + + if (icr != 0) + dev_warn(dev->dev, + "Ignoring unknown interrupt cause (%d)\n", + icr); + } + + return IRQ_HANDLED; +} + +/* + * Interrupt handler for MSI-X interrupt vector VMCI_INTR_NOTIFICATION, + * which is for the notification bitmap. Will only get called if we are + * using MSI-X with exclusive vectors. + */ +static irqreturn_t vmci_interrupt_bm(int irq, void *_dev) +{ + struct vmci_guest_device *dev = _dev; + + /* For MSI-X we can just assume it was meant for us. */ + tasklet_schedule(&dev->bm_tasklet); + + return IRQ_HANDLED; +} + +/* + * Most of the initialization at module load time is done here. + */ +static int vmci_guest_probe_device(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + struct vmci_guest_device *vmci_dev; + void __iomem *iobase; + unsigned int capabilities; + unsigned long cmd; + int vmci_err; + int error; + + dev_dbg(&pdev->dev, "Probing for vmci/PCI guest device\n"); + + error = pcim_enable_device(pdev); + if (error) { + dev_err(&pdev->dev, + "Failed to enable VMCI device: %d\n", error); + return error; + } + + error = pcim_iomap_regions(pdev, 1 << 0, KBUILD_MODNAME); + if (error) { + dev_err(&pdev->dev, "Failed to reserve/map IO regions\n"); + return error; + } + + iobase = pcim_iomap_table(pdev)[0]; + + dev_info(&pdev->dev, "Found VMCI PCI device at %#lx, irq %u\n", + (unsigned long)iobase, pdev->irq); + + vmci_dev = devm_kzalloc(&pdev->dev, sizeof(*vmci_dev), GFP_KERNEL); + if (!vmci_dev) { + dev_err(&pdev->dev, + "Can't allocate memory for VMCI device\n"); + return -ENOMEM; + } + + vmci_dev->dev = &pdev->dev; + vmci_dev->intr_type = VMCI_INTR_TYPE_INTX; + vmci_dev->exclusive_vectors = false; + vmci_dev->iobase = iobase; + + tasklet_init(&vmci_dev->datagram_tasklet, + vmci_dispatch_dgs, (unsigned long)vmci_dev); + tasklet_init(&vmci_dev->bm_tasklet, + vmci_process_bitmap, (unsigned long)vmci_dev); + + vmci_dev->data_buffer = vmalloc(VMCI_MAX_DG_SIZE); + if (!vmci_dev->data_buffer) { + dev_err(&pdev->dev, + "Can't allocate memory for datagram buffer\n"); + return -ENOMEM; + } + + pci_set_master(pdev); /* To enable queue_pair functionality. */ + + /* + * Verify that the VMCI Device supports the capabilities that + * we need. If the device is missing capabilities that we would + * like to use, check for fallback capabilities and use those + * instead (so we can run a new VM on old hosts). Fail the load if + * a required capability is missing and there is no fallback. + * + * Right now, we need datagrams. There are no fallbacks. + */ + capabilities = ioread32(vmci_dev->iobase + VMCI_CAPS_ADDR); + if (!(capabilities & VMCI_CAPS_DATAGRAM)) { + dev_err(&pdev->dev, "Device does not support datagrams\n"); + error = -ENXIO; + goto err_free_data_buffer; + } + + /* + * If the hardware supports notifications, we will use that as + * well. + */ + if (capabilities & VMCI_CAPS_NOTIFICATIONS) { + vmci_dev->notification_bitmap = dma_alloc_coherent( + &pdev->dev, PAGE_SIZE, &vmci_dev->notification_base, + GFP_KERNEL); + if (!vmci_dev->notification_bitmap) { + dev_warn(&pdev->dev, + "Unable to allocate notification bitmap\n"); + } else { + memset(vmci_dev->notification_bitmap, 0, PAGE_SIZE); + capabilities |= VMCI_CAPS_NOTIFICATIONS; + } + } + + dev_info(&pdev->dev, "Using capabilities 0x%x\n", capabilities); + + /* Let the host know which capabilities we intend to use. */ + iowrite32(capabilities, vmci_dev->iobase + VMCI_CAPS_ADDR); + + /* Set up global device so that we can start sending datagrams */ + spin_lock_irq(&vmci_dev_spinlock); + vmci_dev_g = vmci_dev; + vmci_pdev = pdev; + spin_unlock_irq(&vmci_dev_spinlock); + + /* + * Register notification bitmap with device if that capability is + * used. + */ + if (capabilities & VMCI_CAPS_NOTIFICATIONS) { + unsigned long bitmap_ppn = + vmci_dev->notification_base >> PAGE_SHIFT; + if (!vmci_dbell_register_notification_bitmap(bitmap_ppn)) { + dev_warn(&pdev->dev, + "VMCI device unable to register notification bitmap with PPN 0x%x\n", + (u32) bitmap_ppn); + error = -ENXIO; + goto err_remove_vmci_dev_g; + } + } + + /* Check host capabilities. */ + error = vmci_check_host_caps(pdev); + if (error) + goto err_remove_bitmap; + + /* Enable device. */ + + /* + * We subscribe to the VMCI_EVENT_CTX_ID_UPDATE here so we can + * update the internal context id when needed. + */ + vmci_err = vmci_event_subscribe(VMCI_EVENT_CTX_ID_UPDATE, + vmci_guest_cid_update, NULL, + &ctx_update_sub_id); + if (vmci_err < VMCI_SUCCESS) + dev_warn(&pdev->dev, + "Failed to subscribe to event (type=%d): %d\n", + VMCI_EVENT_CTX_ID_UPDATE, vmci_err); + + /* + * Enable interrupts. Try MSI-X first, then MSI, and then fallback on + * legacy interrupts. + */ + if (!vmci_disable_msix && !vmci_enable_msix(pdev, vmci_dev)) { + vmci_dev->intr_type = VMCI_INTR_TYPE_MSIX; + vmci_dev->irq = vmci_dev->msix_entries[0].vector; + } else if (!vmci_disable_msi && !pci_enable_msi(pdev)) { + vmci_dev->intr_type = VMCI_INTR_TYPE_MSI; + vmci_dev->irq = pdev->irq; + } else { + vmci_dev->intr_type = VMCI_INTR_TYPE_INTX; + vmci_dev->irq = pdev->irq; + } + + /* + * Request IRQ for legacy or MSI interrupts, or for first + * MSI-X vector. + */ + error = request_irq(vmci_dev->irq, vmci_interrupt, IRQF_SHARED, + KBUILD_MODNAME, vmci_dev); + if (error) { + dev_err(&pdev->dev, "Irq %u in use: %d\n", + vmci_dev->irq, error); + goto err_disable_msi; + } + + /* + * For MSI-X with exclusive vectors we need to request an + * interrupt for each vector so that we get a separate + * interrupt handler routine. This allows us to distinguish + * between the vectors. + */ + if (vmci_dev->exclusive_vectors) { + error = request_irq(vmci_dev->msix_entries[1].vector, + vmci_interrupt_bm, 0, KBUILD_MODNAME, + vmci_dev); + if (error) { + dev_err(&pdev->dev, + "Failed to allocate irq %u: %d\n", + vmci_dev->msix_entries[1].vector, error); + goto err_free_irq; + } + } + + dev_dbg(&pdev->dev, "Registered device\n"); + + atomic_inc(&vmci_num_guest_devices); + + /* Enable specific interrupt bits. */ + cmd = VMCI_IMR_DATAGRAM; + if (capabilities & VMCI_CAPS_NOTIFICATIONS) + cmd |= VMCI_IMR_NOTIFICATION; + iowrite32(cmd, vmci_dev->iobase + VMCI_IMR_ADDR); + + /* Enable interrupts. */ + iowrite32(VMCI_CONTROL_INT_ENABLE, + vmci_dev->iobase + VMCI_CONTROL_ADDR); + + pci_set_drvdata(pdev, vmci_dev); + return 0; + +err_free_irq: + free_irq(vmci_dev->irq, vmci_dev); + tasklet_kill(&vmci_dev->datagram_tasklet); + tasklet_kill(&vmci_dev->bm_tasklet); + +err_disable_msi: + if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSIX) + pci_disable_msix(pdev); + else if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSI) + pci_disable_msi(pdev); + + vmci_err = vmci_event_unsubscribe(ctx_update_sub_id); + if (vmci_err < VMCI_SUCCESS) + dev_warn(&pdev->dev, + "Failed to unsubscribe from event (type=%d) with subscriber (ID=0x%x): %d\n", + VMCI_EVENT_CTX_ID_UPDATE, ctx_update_sub_id, vmci_err); + +err_remove_bitmap: + if (vmci_dev->notification_bitmap) { + iowrite32(VMCI_CONTROL_RESET, + vmci_dev->iobase + VMCI_CONTROL_ADDR); + dma_free_coherent(&pdev->dev, PAGE_SIZE, + vmci_dev->notification_bitmap, + vmci_dev->notification_base); + } + +err_remove_vmci_dev_g: + spin_lock_irq(&vmci_dev_spinlock); + vmci_pdev = NULL; + vmci_dev_g = NULL; + spin_unlock_irq(&vmci_dev_spinlock); + +err_free_data_buffer: + vfree(vmci_dev->data_buffer); + + /* The rest are managed resources and will be freed by PCI core */ + return error; +} + +static void vmci_guest_remove_device(struct pci_dev *pdev) +{ + struct vmci_guest_device *vmci_dev = pci_get_drvdata(pdev); + int vmci_err; + + dev_dbg(&pdev->dev, "Removing device\n"); + + atomic_dec(&vmci_num_guest_devices); + + vmci_qp_guest_endpoints_exit(); + + vmci_err = vmci_event_unsubscribe(ctx_update_sub_id); + if (vmci_err < VMCI_SUCCESS) + dev_warn(&pdev->dev, + "Failed to unsubscribe from event (type=%d) with subscriber (ID=0x%x): %d\n", + VMCI_EVENT_CTX_ID_UPDATE, ctx_update_sub_id, vmci_err); + + spin_lock_irq(&vmci_dev_spinlock); + vmci_dev_g = NULL; + vmci_pdev = NULL; + spin_unlock_irq(&vmci_dev_spinlock); + + dev_dbg(&pdev->dev, "Resetting vmci device\n"); + iowrite32(VMCI_CONTROL_RESET, vmci_dev->iobase + VMCI_CONTROL_ADDR); + + /* + * Free IRQ and then disable MSI/MSI-X as appropriate. For + * MSI-X, we might have multiple vectors, each with their own + * IRQ, which we must free too. + */ + free_irq(vmci_dev->irq, vmci_dev); + if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSIX) { + if (vmci_dev->exclusive_vectors) + free_irq(vmci_dev->msix_entries[1].vector, vmci_dev); + pci_disable_msix(pdev); + } else if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSI) { + pci_disable_msi(pdev); + } + + tasklet_kill(&vmci_dev->datagram_tasklet); + tasklet_kill(&vmci_dev->bm_tasklet); + + if (vmci_dev->notification_bitmap) { + /* + * The device reset above cleared the bitmap state of the + * device, so we can safely free it here. + */ + + dma_free_coherent(&pdev->dev, PAGE_SIZE, + vmci_dev->notification_bitmap, + vmci_dev->notification_base); + } + + vfree(vmci_dev->data_buffer); + + /* The rest are managed resources and will be freed by PCI core */ +} + +static DEFINE_PCI_DEVICE_TABLE(vmci_ids) = { + { PCI_DEVICE(PCI_VENDOR_ID_VMWARE, PCI_DEVICE_ID_VMWARE_VMCI), }, + { 0 }, +}; +MODULE_DEVICE_TABLE(pci, vmci_ids); + +static struct pci_driver vmci_guest_driver = { + .name = KBUILD_MODNAME, + .id_table = vmci_ids, + .probe = vmci_guest_probe_device, + .remove = vmci_guest_remove_device, +}; + +int __init vmci_guest_init(void) +{ + return pci_register_driver(&vmci_guest_driver); +} + +void __exit vmci_guest_exit(void) +{ + pci_unregister_driver(&vmci_guest_driver); +} diff --git a/drivers/misc/vmw_vmci/vmci_handle_array.c b/drivers/misc/vmw_vmci/vmci_handle_array.c new file mode 100644 index 00000000000..344973a0fb0 --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_handle_array.c @@ -0,0 +1,142 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/slab.h> +#include "vmci_handle_array.h" + +static size_t handle_arr_calc_size(size_t capacity) +{ + return sizeof(struct vmci_handle_arr) + + capacity * sizeof(struct vmci_handle); +} + +struct vmci_handle_arr *vmci_handle_arr_create(size_t capacity) +{ + struct vmci_handle_arr *array; + + if (capacity == 0) + capacity = VMCI_HANDLE_ARRAY_DEFAULT_SIZE; + + array = kmalloc(handle_arr_calc_size(capacity), GFP_ATOMIC); + if (!array) + return NULL; + + array->capacity = capacity; + array->size = 0; + + return array; +} + +void vmci_handle_arr_destroy(struct vmci_handle_arr *array) +{ + kfree(array); +} + +void vmci_handle_arr_append_entry(struct vmci_handle_arr **array_ptr, + struct vmci_handle handle) +{ + struct vmci_handle_arr *array = *array_ptr; + + if (unlikely(array->size >= array->capacity)) { + /* reallocate. */ + struct vmci_handle_arr *new_array; + size_t new_capacity = array->capacity * VMCI_ARR_CAP_MULT; + size_t new_size = handle_arr_calc_size(new_capacity); + + new_array = krealloc(array, new_size, GFP_ATOMIC); + if (!new_array) + return; + + new_array->capacity = new_capacity; + *array_ptr = array = new_array; + } + + array->entries[array->size] = handle; + array->size++; +} + +/* + * Handle that was removed, VMCI_INVALID_HANDLE if entry not found. + */ +struct vmci_handle vmci_handle_arr_remove_entry(struct vmci_handle_arr *array, + struct vmci_handle entry_handle) +{ + struct vmci_handle handle = VMCI_INVALID_HANDLE; + size_t i; + + for (i = 0; i < array->size; i++) { + if (vmci_handle_is_equal(array->entries[i], entry_handle)) { + handle = array->entries[i]; + array->size--; + array->entries[i] = array->entries[array->size]; + array->entries[array->size] = VMCI_INVALID_HANDLE; + break; + } + } + + return handle; +} + +/* + * Handle that was removed, VMCI_INVALID_HANDLE if array was empty. + */ +struct vmci_handle vmci_handle_arr_remove_tail(struct vmci_handle_arr *array) +{ + struct vmci_handle handle = VMCI_INVALID_HANDLE; + + if (array->size) { + array->size--; + handle = array->entries[array->size]; + array->entries[array->size] = VMCI_INVALID_HANDLE; + } + + return handle; +} + +/* + * Handle at given index, VMCI_INVALID_HANDLE if invalid index. + */ +struct vmci_handle +vmci_handle_arr_get_entry(const struct vmci_handle_arr *array, size_t index) +{ + if (unlikely(index >= array->size)) + return VMCI_INVALID_HANDLE; + + return array->entries[index]; +} + +bool vmci_handle_arr_has_entry(const struct vmci_handle_arr *array, + struct vmci_handle entry_handle) +{ + size_t i; + + for (i = 0; i < array->size; i++) + if (vmci_handle_is_equal(array->entries[i], entry_handle)) + return true; + + return false; +} + +/* + * NULL if the array is empty. Otherwise, a pointer to the array + * of VMCI handles in the handle array. + */ +struct vmci_handle *vmci_handle_arr_get_handles(struct vmci_handle_arr *array) +{ + if (array->size) + return array->entries; + + return NULL; +} diff --git a/drivers/misc/vmw_vmci/vmci_handle_array.h b/drivers/misc/vmw_vmci/vmci_handle_array.h new file mode 100644 index 00000000000..b5f3a7f98cf --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_handle_array.h @@ -0,0 +1,52 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef _VMCI_HANDLE_ARRAY_H_ +#define _VMCI_HANDLE_ARRAY_H_ + +#include <linux/vmw_vmci_defs.h> +#include <linux/types.h> + +#define VMCI_HANDLE_ARRAY_DEFAULT_SIZE 4 +#define VMCI_ARR_CAP_MULT 2 /* Array capacity multiplier */ + +struct vmci_handle_arr { + size_t capacity; + size_t size; + struct vmci_handle entries[]; +}; + +struct vmci_handle_arr *vmci_handle_arr_create(size_t capacity); +void vmci_handle_arr_destroy(struct vmci_handle_arr *array); +void vmci_handle_arr_append_entry(struct vmci_handle_arr **array_ptr, + struct vmci_handle handle); +struct vmci_handle vmci_handle_arr_remove_entry(struct vmci_handle_arr *array, + struct vmci_handle + entry_handle); +struct vmci_handle vmci_handle_arr_remove_tail(struct vmci_handle_arr *array); +struct vmci_handle +vmci_handle_arr_get_entry(const struct vmci_handle_arr *array, size_t index); +bool vmci_handle_arr_has_entry(const struct vmci_handle_arr *array, + struct vmci_handle entry_handle); +struct vmci_handle *vmci_handle_arr_get_handles(struct vmci_handle_arr *array); + +static inline size_t vmci_handle_arr_get_size( + const struct vmci_handle_arr *array) +{ + return array->size; +} + + +#endif /* _VMCI_HANDLE_ARRAY_H_ */ diff --git a/drivers/misc/vmw_vmci/vmci_host.c b/drivers/misc/vmw_vmci/vmci_host.c new file mode 100644 index 00000000000..1723a6e4f2e --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_host.c @@ -0,0 +1,1039 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/moduleparam.h> +#include <linux/miscdevice.h> +#include <linux/interrupt.h> +#include <linux/highmem.h> +#include <linux/atomic.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/file.h> +#include <linux/init.h> +#include <linux/poll.h> +#include <linux/pci.h> +#include <linux/smp.h> +#include <linux/fs.h> +#include <linux/io.h> + +#include "vmci_handle_array.h" +#include "vmci_queue_pair.h" +#include "vmci_datagram.h" +#include "vmci_doorbell.h" +#include "vmci_resource.h" +#include "vmci_context.h" +#include "vmci_driver.h" +#include "vmci_event.h" + +#define VMCI_UTIL_NUM_RESOURCES 1 + +enum { + VMCI_NOTIFY_RESOURCE_QUEUE_PAIR = 0, + VMCI_NOTIFY_RESOURCE_DOOR_BELL = 1, +}; + +enum { + VMCI_NOTIFY_RESOURCE_ACTION_NOTIFY = 0, + VMCI_NOTIFY_RESOURCE_ACTION_CREATE = 1, + VMCI_NOTIFY_RESOURCE_ACTION_DESTROY = 2, +}; + +/* + * VMCI driver initialization. This block can also be used to + * pass initial group membership etc. + */ +struct vmci_init_blk { + u32 cid; + u32 flags; +}; + +/* VMCIqueue_pairAllocInfo_VMToVM */ +struct vmci_qp_alloc_info_vmvm { + struct vmci_handle handle; + u32 peer; + u32 flags; + u64 produce_size; + u64 consume_size; + u64 produce_page_file; /* User VA. */ + u64 consume_page_file; /* User VA. */ + u64 produce_page_file_size; /* Size of the file name array. */ + u64 consume_page_file_size; /* Size of the file name array. */ + s32 result; + u32 _pad; +}; + +/* VMCISetNotifyInfo: Used to pass notify flag's address to the host driver. */ +struct vmci_set_notify_info { + u64 notify_uva; + s32 result; + u32 _pad; +}; + +/* + * Per-instance host state + */ +struct vmci_host_dev { + struct vmci_ctx *context; + int user_version; + enum vmci_obj_type ct_type; + struct mutex lock; /* Mutex lock for vmci context access */ +}; + +static struct vmci_ctx *host_context; +static bool vmci_host_device_initialized; +static atomic_t vmci_host_active_users = ATOMIC_INIT(0); + +/* + * Determines whether the VMCI host personality is + * available. Since the core functionality of the host driver is + * always present, all guests could possibly use the host + * personality. However, to minimize the deviation from the + * pre-unified driver state of affairs, we only consider the host + * device active if there is no active guest device or if there + * are VMX'en with active VMCI contexts using the host device. + */ +bool vmci_host_code_active(void) +{ + return vmci_host_device_initialized && + (!vmci_guest_code_active() || + atomic_read(&vmci_host_active_users) > 0); +} + +/* + * Called on open of /dev/vmci. + */ +static int vmci_host_open(struct inode *inode, struct file *filp) +{ + struct vmci_host_dev *vmci_host_dev; + + vmci_host_dev = kzalloc(sizeof(struct vmci_host_dev), GFP_KERNEL); + if (vmci_host_dev == NULL) + return -ENOMEM; + + vmci_host_dev->ct_type = VMCIOBJ_NOT_SET; + mutex_init(&vmci_host_dev->lock); + filp->private_data = vmci_host_dev; + + return 0; +} + +/* + * Called on close of /dev/vmci, most often when the process + * exits. + */ +static int vmci_host_close(struct inode *inode, struct file *filp) +{ + struct vmci_host_dev *vmci_host_dev = filp->private_data; + + if (vmci_host_dev->ct_type == VMCIOBJ_CONTEXT) { + vmci_ctx_destroy(vmci_host_dev->context); + vmci_host_dev->context = NULL; + + /* + * The number of active contexts is used to track whether any + * VMX'en are using the host personality. It is incremented when + * a context is created through the IOCTL_VMCI_INIT_CONTEXT + * ioctl. + */ + atomic_dec(&vmci_host_active_users); + } + vmci_host_dev->ct_type = VMCIOBJ_NOT_SET; + + kfree(vmci_host_dev); + filp->private_data = NULL; + return 0; +} + +/* + * This is used to wake up the VMX when a VMCI call arrives, or + * to wake up select() or poll() at the next clock tick. + */ +static unsigned int vmci_host_poll(struct file *filp, poll_table *wait) +{ + struct vmci_host_dev *vmci_host_dev = filp->private_data; + struct vmci_ctx *context = vmci_host_dev->context; + unsigned int mask = 0; + + if (vmci_host_dev->ct_type == VMCIOBJ_CONTEXT) { + /* Check for VMCI calls to this VM context. */ + if (wait) + poll_wait(filp, &context->host_context.wait_queue, + wait); + + spin_lock(&context->lock); + if (context->pending_datagrams > 0 || + vmci_handle_arr_get_size( + context->pending_doorbell_array) > 0) { + mask = POLLIN; + } + spin_unlock(&context->lock); + } + return mask; +} + +/* + * Copies the handles of a handle array into a user buffer, and + * returns the new length in userBufferSize. If the copy to the + * user buffer fails, the functions still returns VMCI_SUCCESS, + * but retval != 0. + */ +static int drv_cp_harray_to_user(void __user *user_buf_uva, + u64 *user_buf_size, + struct vmci_handle_arr *handle_array, + int *retval) +{ + u32 array_size = 0; + struct vmci_handle *handles; + + if (handle_array) + array_size = vmci_handle_arr_get_size(handle_array); + + if (array_size * sizeof(*handles) > *user_buf_size) + return VMCI_ERROR_MORE_DATA; + + *user_buf_size = array_size * sizeof(*handles); + if (*user_buf_size) + *retval = copy_to_user(user_buf_uva, + vmci_handle_arr_get_handles + (handle_array), *user_buf_size); + + return VMCI_SUCCESS; +} + +/* + * Sets up a given context for notify to work. Calls drv_map_bool_ptr() + * which maps the notify boolean in user VA in kernel space. + */ +static int vmci_host_setup_notify(struct vmci_ctx *context, + unsigned long uva) +{ + struct page *page; + int retval; + + if (context->notify_page) { + pr_devel("%s: Notify mechanism is already set up\n", __func__); + return VMCI_ERROR_DUPLICATE_ENTRY; + } + + /* + * We are using 'bool' internally, but let's make sure we explicit + * about the size. + */ + BUILD_BUG_ON(sizeof(bool) != sizeof(u8)); + if (!access_ok(VERIFY_WRITE, (void __user *)uva, sizeof(u8))) + return VMCI_ERROR_GENERIC; + + /* + * Lock physical page backing a given user VA. + */ + retval = get_user_pages_fast(PAGE_ALIGN(uva), 1, 1, &page); + if (retval != 1) + return VMCI_ERROR_GENERIC; + + /* + * Map the locked page and set up notify pointer. + */ + context->notify = kmap(page) + (uva & (PAGE_SIZE - 1)); + vmci_ctx_check_signal_notify(context); + + return VMCI_SUCCESS; +} + +static int vmci_host_get_version(struct vmci_host_dev *vmci_host_dev, + unsigned int cmd, void __user *uptr) +{ + if (cmd == IOCTL_VMCI_VERSION2) { + int __user *vptr = uptr; + if (get_user(vmci_host_dev->user_version, vptr)) + return -EFAULT; + } + + /* + * The basic logic here is: + * + * If the user sends in a version of 0 tell it our version. + * If the user didn't send in a version, tell it our version. + * If the user sent in an old version, tell it -its- version. + * If the user sent in an newer version, tell it our version. + * + * The rationale behind telling the caller its version is that + * Workstation 6.5 required that VMX and VMCI kernel module were + * version sync'd. All new VMX users will be programmed to + * handle the VMCI kernel module version. + */ + + if (vmci_host_dev->user_version > 0 && + vmci_host_dev->user_version < VMCI_VERSION_HOSTQP) { + return vmci_host_dev->user_version; + } + + return VMCI_VERSION; +} + +#define vmci_ioctl_err(fmt, ...) \ + pr_devel("%s: " fmt, ioctl_name, ##__VA_ARGS__) + +static int vmci_host_do_init_context(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_init_blk init_block; + const struct cred *cred; + int retval; + + if (copy_from_user(&init_block, uptr, sizeof(init_block))) { + vmci_ioctl_err("error reading init block\n"); + return -EFAULT; + } + + mutex_lock(&vmci_host_dev->lock); + + if (vmci_host_dev->ct_type != VMCIOBJ_NOT_SET) { + vmci_ioctl_err("received VMCI init on initialized handle\n"); + retval = -EINVAL; + goto out; + } + + if (init_block.flags & ~VMCI_PRIVILEGE_FLAG_RESTRICTED) { + vmci_ioctl_err("unsupported VMCI restriction flag\n"); + retval = -EINVAL; + goto out; + } + + cred = get_current_cred(); + vmci_host_dev->context = vmci_ctx_create(init_block.cid, + init_block.flags, 0, + vmci_host_dev->user_version, + cred); + put_cred(cred); + if (IS_ERR(vmci_host_dev->context)) { + retval = PTR_ERR(vmci_host_dev->context); + vmci_ioctl_err("error initializing context\n"); + goto out; + } + + /* + * Copy cid to userlevel, we do this to allow the VMX + * to enforce its policy on cid generation. + */ + init_block.cid = vmci_ctx_get_id(vmci_host_dev->context); + if (copy_to_user(uptr, &init_block, sizeof(init_block))) { + vmci_ctx_destroy(vmci_host_dev->context); + vmci_host_dev->context = NULL; + vmci_ioctl_err("error writing init block\n"); + retval = -EFAULT; + goto out; + } + + vmci_host_dev->ct_type = VMCIOBJ_CONTEXT; + atomic_inc(&vmci_host_active_users); + + retval = 0; + +out: + mutex_unlock(&vmci_host_dev->lock); + return retval; +} + +static int vmci_host_do_send_datagram(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_datagram_snd_rcv_info send_info; + struct vmci_datagram *dg = NULL; + u32 cid; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&send_info, uptr, sizeof(send_info))) + return -EFAULT; + + if (send_info.len > VMCI_MAX_DG_SIZE) { + vmci_ioctl_err("datagram is too big (size=%d)\n", + send_info.len); + return -EINVAL; + } + + if (send_info.len < sizeof(*dg)) { + vmci_ioctl_err("datagram is too small (size=%d)\n", + send_info.len); + return -EINVAL; + } + + dg = kmalloc(send_info.len, GFP_KERNEL); + if (!dg) { + vmci_ioctl_err( + "cannot allocate memory to dispatch datagram\n"); + return -ENOMEM; + } + + if (copy_from_user(dg, (void __user *)(uintptr_t)send_info.addr, + send_info.len)) { + vmci_ioctl_err("error getting datagram\n"); + kfree(dg); + return -EFAULT; + } + + pr_devel("Datagram dst (handle=0x%x:0x%x) src (handle=0x%x:0x%x), payload (size=%llu bytes)\n", + dg->dst.context, dg->dst.resource, + dg->src.context, dg->src.resource, + (unsigned long long)dg->payload_size); + + /* Get source context id. */ + cid = vmci_ctx_get_id(vmci_host_dev->context); + send_info.result = vmci_datagram_dispatch(cid, dg, true); + kfree(dg); + + return copy_to_user(uptr, &send_info, sizeof(send_info)) ? -EFAULT : 0; +} + +static int vmci_host_do_receive_datagram(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_datagram_snd_rcv_info recv_info; + struct vmci_datagram *dg = NULL; + int retval; + size_t size; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&recv_info, uptr, sizeof(recv_info))) + return -EFAULT; + + size = recv_info.len; + recv_info.result = vmci_ctx_dequeue_datagram(vmci_host_dev->context, + &size, &dg); + + if (recv_info.result >= VMCI_SUCCESS) { + void __user *ubuf = (void __user *)(uintptr_t)recv_info.addr; + retval = copy_to_user(ubuf, dg, VMCI_DG_SIZE(dg)); + kfree(dg); + if (retval != 0) + return -EFAULT; + } + + return copy_to_user(uptr, &recv_info, sizeof(recv_info)) ? -EFAULT : 0; +} + +static int vmci_host_do_alloc_queuepair(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_handle handle; + int vmci_status; + int __user *retptr; + u32 cid; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + cid = vmci_ctx_get_id(vmci_host_dev->context); + + if (vmci_host_dev->user_version < VMCI_VERSION_NOVMVM) { + struct vmci_qp_alloc_info_vmvm alloc_info; + struct vmci_qp_alloc_info_vmvm __user *info = uptr; + + if (copy_from_user(&alloc_info, uptr, sizeof(alloc_info))) + return -EFAULT; + + handle = alloc_info.handle; + retptr = &info->result; + + vmci_status = vmci_qp_broker_alloc(alloc_info.handle, + alloc_info.peer, + alloc_info.flags, + VMCI_NO_PRIVILEGE_FLAGS, + alloc_info.produce_size, + alloc_info.consume_size, + NULL, + vmci_host_dev->context); + + if (vmci_status == VMCI_SUCCESS) + vmci_status = VMCI_SUCCESS_QUEUEPAIR_CREATE; + } else { + struct vmci_qp_alloc_info alloc_info; + struct vmci_qp_alloc_info __user *info = uptr; + struct vmci_qp_page_store page_store; + + if (copy_from_user(&alloc_info, uptr, sizeof(alloc_info))) + return -EFAULT; + + handle = alloc_info.handle; + retptr = &info->result; + + page_store.pages = alloc_info.ppn_va; + page_store.len = alloc_info.num_ppns; + + vmci_status = vmci_qp_broker_alloc(alloc_info.handle, + alloc_info.peer, + alloc_info.flags, + VMCI_NO_PRIVILEGE_FLAGS, + alloc_info.produce_size, + alloc_info.consume_size, + &page_store, + vmci_host_dev->context); + } + + if (put_user(vmci_status, retptr)) { + if (vmci_status >= VMCI_SUCCESS) { + vmci_status = vmci_qp_broker_detach(handle, + vmci_host_dev->context); + } + return -EFAULT; + } + + return 0; +} + +static int vmci_host_do_queuepair_setva(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_qp_set_va_info set_va_info; + struct vmci_qp_set_va_info __user *info = uptr; + s32 result; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (vmci_host_dev->user_version < VMCI_VERSION_NOVMVM) { + vmci_ioctl_err("is not allowed\n"); + return -EINVAL; + } + + if (copy_from_user(&set_va_info, uptr, sizeof(set_va_info))) + return -EFAULT; + + if (set_va_info.va) { + /* + * VMX is passing down a new VA for the queue + * pair mapping. + */ + result = vmci_qp_broker_map(set_va_info.handle, + vmci_host_dev->context, + set_va_info.va); + } else { + /* + * The queue pair is about to be unmapped by + * the VMX. + */ + result = vmci_qp_broker_unmap(set_va_info.handle, + vmci_host_dev->context, 0); + } + + return put_user(result, &info->result) ? -EFAULT : 0; +} + +static int vmci_host_do_queuepair_setpf(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_qp_page_file_info page_file_info; + struct vmci_qp_page_file_info __user *info = uptr; + s32 result; + + if (vmci_host_dev->user_version < VMCI_VERSION_HOSTQP || + vmci_host_dev->user_version >= VMCI_VERSION_NOVMVM) { + vmci_ioctl_err("not supported on this VMX (version=%d)\n", + vmci_host_dev->user_version); + return -EINVAL; + } + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&page_file_info, uptr, sizeof(*info))) + return -EFAULT; + + /* + * Communicate success pre-emptively to the caller. Note that the + * basic premise is that it is incumbent upon the caller not to look at + * the info.result field until after the ioctl() returns. And then, + * only if the ioctl() result indicates no error. We send up the + * SUCCESS status before calling SetPageStore() store because failing + * to copy up the result code means unwinding the SetPageStore(). + * + * It turns out the logic to unwind a SetPageStore() opens a can of + * worms. For example, if a host had created the queue_pair and a + * guest attaches and SetPageStore() is successful but writing success + * fails, then ... the host has to be stopped from writing (anymore) + * data into the queue_pair. That means an additional test in the + * VMCI_Enqueue() code path. Ugh. + */ + + if (put_user(VMCI_SUCCESS, &info->result)) { + /* + * In this case, we can't write a result field of the + * caller's info block. So, we don't even try to + * SetPageStore(). + */ + return -EFAULT; + } + + result = vmci_qp_broker_set_page_store(page_file_info.handle, + page_file_info.produce_va, + page_file_info.consume_va, + vmci_host_dev->context); + if (result < VMCI_SUCCESS) { + if (put_user(result, &info->result)) { + /* + * Note that in this case the SetPageStore() + * call failed but we were unable to + * communicate that to the caller (because the + * copy_to_user() call failed). So, if we + * simply return an error (in this case + * -EFAULT) then the caller will know that the + * SetPageStore failed even though we couldn't + * put the result code in the result field and + * indicate exactly why it failed. + * + * That says nothing about the issue where we + * were once able to write to the caller's info + * memory and now can't. Something more + * serious is probably going on than the fact + * that SetPageStore() didn't work. + */ + return -EFAULT; + } + } + + return 0; +} + +static int vmci_host_do_qp_detach(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_qp_dtch_info detach_info; + struct vmci_qp_dtch_info __user *info = uptr; + s32 result; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&detach_info, uptr, sizeof(detach_info))) + return -EFAULT; + + result = vmci_qp_broker_detach(detach_info.handle, + vmci_host_dev->context); + if (result == VMCI_SUCCESS && + vmci_host_dev->user_version < VMCI_VERSION_NOVMVM) { + result = VMCI_SUCCESS_LAST_DETACH; + } + + return put_user(result, &info->result) ? -EFAULT : 0; +} + +static int vmci_host_do_ctx_add_notify(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_ctx_info ar_info; + struct vmci_ctx_info __user *info = uptr; + s32 result; + u32 cid; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&ar_info, uptr, sizeof(ar_info))) + return -EFAULT; + + cid = vmci_ctx_get_id(vmci_host_dev->context); + result = vmci_ctx_add_notification(cid, ar_info.remote_cid); + + return put_user(result, &info->result) ? -EFAULT : 0; +} + +static int vmci_host_do_ctx_remove_notify(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_ctx_info ar_info; + struct vmci_ctx_info __user *info = uptr; + u32 cid; + int result; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&ar_info, uptr, sizeof(ar_info))) + return -EFAULT; + + cid = vmci_ctx_get_id(vmci_host_dev->context); + result = vmci_ctx_remove_notification(cid, + ar_info.remote_cid); + + return put_user(result, &info->result) ? -EFAULT : 0; +} + +static int vmci_host_do_ctx_get_cpt_state(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_ctx_chkpt_buf_info get_info; + u32 cid; + void *cpt_buf; + int retval; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&get_info, uptr, sizeof(get_info))) + return -EFAULT; + + cid = vmci_ctx_get_id(vmci_host_dev->context); + get_info.result = vmci_ctx_get_chkpt_state(cid, get_info.cpt_type, + &get_info.buf_size, &cpt_buf); + if (get_info.result == VMCI_SUCCESS && get_info.buf_size) { + void __user *ubuf = (void __user *)(uintptr_t)get_info.cpt_buf; + retval = copy_to_user(ubuf, cpt_buf, get_info.buf_size); + kfree(cpt_buf); + + if (retval) + return -EFAULT; + } + + return copy_to_user(uptr, &get_info, sizeof(get_info)) ? -EFAULT : 0; +} + +static int vmci_host_do_ctx_set_cpt_state(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_ctx_chkpt_buf_info set_info; + u32 cid; + void *cpt_buf; + int retval; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&set_info, uptr, sizeof(set_info))) + return -EFAULT; + + cpt_buf = kmalloc(set_info.buf_size, GFP_KERNEL); + if (!cpt_buf) { + vmci_ioctl_err( + "cannot allocate memory to set cpt state (type=%d)\n", + set_info.cpt_type); + return -ENOMEM; + } + + if (copy_from_user(cpt_buf, (void __user *)(uintptr_t)set_info.cpt_buf, + set_info.buf_size)) { + retval = -EFAULT; + goto out; + } + + cid = vmci_ctx_get_id(vmci_host_dev->context); + set_info.result = vmci_ctx_set_chkpt_state(cid, set_info.cpt_type, + set_info.buf_size, cpt_buf); + + retval = copy_to_user(uptr, &set_info, sizeof(set_info)) ? -EFAULT : 0; + +out: + kfree(cpt_buf); + return retval; +} + +static int vmci_host_do_get_context_id(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + u32 __user *u32ptr = uptr; + + return put_user(VMCI_HOST_CONTEXT_ID, u32ptr) ? -EFAULT : 0; +} + +static int vmci_host_do_set_notify(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_set_notify_info notify_info; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(¬ify_info, uptr, sizeof(notify_info))) + return -EFAULT; + + if (notify_info.notify_uva) { + notify_info.result = + vmci_host_setup_notify(vmci_host_dev->context, + notify_info.notify_uva); + } else { + vmci_ctx_unset_notify(vmci_host_dev->context); + notify_info.result = VMCI_SUCCESS; + } + + return copy_to_user(uptr, ¬ify_info, sizeof(notify_info)) ? + -EFAULT : 0; +} + +static int vmci_host_do_notify_resource(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_dbell_notify_resource_info info; + u32 cid; + + if (vmci_host_dev->user_version < VMCI_VERSION_NOTIFY) { + vmci_ioctl_err("invalid for current VMX versions\n"); + return -EINVAL; + } + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (copy_from_user(&info, uptr, sizeof(info))) + return -EFAULT; + + cid = vmci_ctx_get_id(vmci_host_dev->context); + + switch (info.action) { + case VMCI_NOTIFY_RESOURCE_ACTION_NOTIFY: + if (info.resource == VMCI_NOTIFY_RESOURCE_DOOR_BELL) { + u32 flags = VMCI_NO_PRIVILEGE_FLAGS; + info.result = vmci_ctx_notify_dbell(cid, info.handle, + flags); + } else { + info.result = VMCI_ERROR_UNAVAILABLE; + } + break; + + case VMCI_NOTIFY_RESOURCE_ACTION_CREATE: + info.result = vmci_ctx_dbell_create(cid, info.handle); + break; + + case VMCI_NOTIFY_RESOURCE_ACTION_DESTROY: + info.result = vmci_ctx_dbell_destroy(cid, info.handle); + break; + + default: + vmci_ioctl_err("got unknown action (action=%d)\n", + info.action); + info.result = VMCI_ERROR_INVALID_ARGS; + } + + return copy_to_user(uptr, &info, sizeof(info)) ? -EFAULT : 0; +} + +static int vmci_host_do_recv_notifications(struct vmci_host_dev *vmci_host_dev, + const char *ioctl_name, + void __user *uptr) +{ + struct vmci_ctx_notify_recv_info info; + struct vmci_handle_arr *db_handle_array; + struct vmci_handle_arr *qp_handle_array; + void __user *ubuf; + u32 cid; + int retval = 0; + + if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) { + vmci_ioctl_err("only valid for contexts\n"); + return -EINVAL; + } + + if (vmci_host_dev->user_version < VMCI_VERSION_NOTIFY) { + vmci_ioctl_err("not supported for the current vmx version\n"); + return -EINVAL; + } + + if (copy_from_user(&info, uptr, sizeof(info))) + return -EFAULT; + + if ((info.db_handle_buf_size && !info.db_handle_buf_uva) || + (info.qp_handle_buf_size && !info.qp_handle_buf_uva)) { + return -EINVAL; + } + + cid = vmci_ctx_get_id(vmci_host_dev->context); + + info.result = vmci_ctx_rcv_notifications_get(cid, + &db_handle_array, &qp_handle_array); + if (info.result != VMCI_SUCCESS) + return copy_to_user(uptr, &info, sizeof(info)) ? -EFAULT : 0; + + ubuf = (void __user *)(uintptr_t)info.db_handle_buf_uva; + info.result = drv_cp_harray_to_user(ubuf, &info.db_handle_buf_size, + db_handle_array, &retval); + if (info.result == VMCI_SUCCESS && !retval) { + ubuf = (void __user *)(uintptr_t)info.qp_handle_buf_uva; + info.result = drv_cp_harray_to_user(ubuf, + &info.qp_handle_buf_size, + qp_handle_array, &retval); + } + + if (!retval && copy_to_user(uptr, &info, sizeof(info))) + retval = -EFAULT; + + vmci_ctx_rcv_notifications_release(cid, + db_handle_array, qp_handle_array, + info.result == VMCI_SUCCESS && !retval); + + return retval; +} + +static long vmci_host_unlocked_ioctl(struct file *filp, + unsigned int iocmd, unsigned long ioarg) +{ +#define VMCI_DO_IOCTL(ioctl_name, ioctl_fn) do { \ + char *name = __stringify(IOCTL_VMCI_ ## ioctl_name); \ + return vmci_host_do_ ## ioctl_fn( \ + vmci_host_dev, name, uptr); \ + } while (0) + + struct vmci_host_dev *vmci_host_dev = filp->private_data; + void __user *uptr = (void __user *)ioarg; + + switch (iocmd) { + case IOCTL_VMCI_INIT_CONTEXT: + VMCI_DO_IOCTL(INIT_CONTEXT, init_context); + case IOCTL_VMCI_DATAGRAM_SEND: + VMCI_DO_IOCTL(DATAGRAM_SEND, send_datagram); + case IOCTL_VMCI_DATAGRAM_RECEIVE: + VMCI_DO_IOCTL(DATAGRAM_RECEIVE, receive_datagram); + case IOCTL_VMCI_QUEUEPAIR_ALLOC: + VMCI_DO_IOCTL(QUEUEPAIR_ALLOC, alloc_queuepair); + case IOCTL_VMCI_QUEUEPAIR_SETVA: + VMCI_DO_IOCTL(QUEUEPAIR_SETVA, queuepair_setva); + case IOCTL_VMCI_QUEUEPAIR_SETPAGEFILE: + VMCI_DO_IOCTL(QUEUEPAIR_SETPAGEFILE, queuepair_setpf); + case IOCTL_VMCI_QUEUEPAIR_DETACH: + VMCI_DO_IOCTL(QUEUEPAIR_DETACH, qp_detach); + case IOCTL_VMCI_CTX_ADD_NOTIFICATION: + VMCI_DO_IOCTL(CTX_ADD_NOTIFICATION, ctx_add_notify); + case IOCTL_VMCI_CTX_REMOVE_NOTIFICATION: + VMCI_DO_IOCTL(CTX_REMOVE_NOTIFICATION, ctx_remove_notify); + case IOCTL_VMCI_CTX_GET_CPT_STATE: + VMCI_DO_IOCTL(CTX_GET_CPT_STATE, ctx_get_cpt_state); + case IOCTL_VMCI_CTX_SET_CPT_STATE: + VMCI_DO_IOCTL(CTX_SET_CPT_STATE, ctx_set_cpt_state); + case IOCTL_VMCI_GET_CONTEXT_ID: + VMCI_DO_IOCTL(GET_CONTEXT_ID, get_context_id); + case IOCTL_VMCI_SET_NOTIFY: + VMCI_DO_IOCTL(SET_NOTIFY, set_notify); + case IOCTL_VMCI_NOTIFY_RESOURCE: + VMCI_DO_IOCTL(NOTIFY_RESOURCE, notify_resource); + case IOCTL_VMCI_NOTIFICATIONS_RECEIVE: + VMCI_DO_IOCTL(NOTIFICATIONS_RECEIVE, recv_notifications); + + case IOCTL_VMCI_VERSION: + case IOCTL_VMCI_VERSION2: + return vmci_host_get_version(vmci_host_dev, iocmd, uptr); + + default: + pr_devel("%s: Unknown ioctl (iocmd=%d)\n", __func__, iocmd); + return -EINVAL; + } + +#undef VMCI_DO_IOCTL +} + +static const struct file_operations vmuser_fops = { + .owner = THIS_MODULE, + .open = vmci_host_open, + .release = vmci_host_close, + .poll = vmci_host_poll, + .unlocked_ioctl = vmci_host_unlocked_ioctl, + .compat_ioctl = vmci_host_unlocked_ioctl, +}; + +static struct miscdevice vmci_host_miscdev = { + .name = "vmci", + .minor = MISC_DYNAMIC_MINOR, + .fops = &vmuser_fops, +}; + +int __init vmci_host_init(void) +{ + int error; + + host_context = vmci_ctx_create(VMCI_HOST_CONTEXT_ID, + VMCI_DEFAULT_PROC_PRIVILEGE_FLAGS, + -1, VMCI_VERSION, NULL); + if (IS_ERR(host_context)) { + error = PTR_ERR(host_context); + pr_warn("Failed to initialize VMCIContext (error%d)\n", + error); + return error; + } + + error = misc_register(&vmci_host_miscdev); + if (error) { + pr_warn("Module registration error (name=%s, major=%d, minor=%d, err=%d)\n", + vmci_host_miscdev.name, + MISC_MAJOR, vmci_host_miscdev.minor, + error); + pr_warn("Unable to initialize host personality\n"); + vmci_ctx_destroy(host_context); + return error; + } + + pr_info("VMCI host device registered (name=%s, major=%d, minor=%d)\n", + vmci_host_miscdev.name, MISC_MAJOR, vmci_host_miscdev.minor); + + vmci_host_device_initialized = true; + return 0; +} + +void __exit vmci_host_exit(void) +{ + int error; + + vmci_host_device_initialized = false; + + error = misc_deregister(&vmci_host_miscdev); + if (error) + pr_warn("Error unregistering character device: %d\n", error); + + vmci_ctx_destroy(host_context); + vmci_qp_broker_exit(); + + pr_debug("VMCI host driver module unloaded\n"); +} diff --git a/drivers/misc/vmw_vmci/vmci_queue_pair.c b/drivers/misc/vmw_vmci/vmci_queue_pair.c new file mode 100644 index 00000000000..1b7b303085d --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_queue_pair.c @@ -0,0 +1,3339 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> +#include <linux/highmem.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/mutex.h> +#include <linux/pagemap.h> +#include <linux/pci.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/uio.h> +#include <linux/wait.h> +#include <linux/vmalloc.h> + +#include "vmci_handle_array.h" +#include "vmci_queue_pair.h" +#include "vmci_datagram.h" +#include "vmci_resource.h" +#include "vmci_context.h" +#include "vmci_driver.h" +#include "vmci_event.h" +#include "vmci_route.h" + +/* + * In the following, we will distinguish between two kinds of VMX processes - + * the ones with versions lower than VMCI_VERSION_NOVMVM that use specialized + * VMCI page files in the VMX and supporting VM to VM communication and the + * newer ones that use the guest memory directly. We will in the following + * refer to the older VMX versions as old-style VMX'en, and the newer ones as + * new-style VMX'en. + * + * The state transition datagram is as follows (the VMCIQPB_ prefix has been + * removed for readability) - see below for more details on the transtions: + * + * -------------- NEW ------------- + * | | + * \_/ \_/ + * CREATED_NO_MEM <-----------------> CREATED_MEM + * | | | + * | o-----------------------o | + * | | | + * \_/ \_/ \_/ + * ATTACHED_NO_MEM <----------------> ATTACHED_MEM + * | | | + * | o----------------------o | + * | | | + * \_/ \_/ \_/ + * SHUTDOWN_NO_MEM <----------------> SHUTDOWN_MEM + * | | + * | | + * -------------> gone <------------- + * + * In more detail. When a VMCI queue pair is first created, it will be in the + * VMCIQPB_NEW state. It will then move into one of the following states: + * + * - VMCIQPB_CREATED_NO_MEM: this state indicates that either: + * + * - the created was performed by a host endpoint, in which case there is + * no backing memory yet. + * + * - the create was initiated by an old-style VMX, that uses + * vmci_qp_broker_set_page_store to specify the UVAs of the queue pair at + * a later point in time. This state can be distinguished from the one + * above by the context ID of the creator. A host side is not allowed to + * attach until the page store has been set. + * + * - VMCIQPB_CREATED_MEM: this state is the result when the queue pair + * is created by a VMX using the queue pair device backend that + * sets the UVAs of the queue pair immediately and stores the + * information for later attachers. At this point, it is ready for + * the host side to attach to it. + * + * Once the queue pair is in one of the created states (with the exception of + * the case mentioned for older VMX'en above), it is possible to attach to the + * queue pair. Again we have two new states possible: + * + * - VMCIQPB_ATTACHED_MEM: this state can be reached through the following + * paths: + * + * - from VMCIQPB_CREATED_NO_MEM when a new-style VMX allocates a queue + * pair, and attaches to a queue pair previously created by the host side. + * + * - from VMCIQPB_CREATED_MEM when the host side attaches to a queue pair + * already created by a guest. + * + * - from VMCIQPB_ATTACHED_NO_MEM, when an old-style VMX calls + * vmci_qp_broker_set_page_store (see below). + * + * - VMCIQPB_ATTACHED_NO_MEM: If the queue pair already was in the + * VMCIQPB_CREATED_NO_MEM due to a host side create, an old-style VMX will + * bring the queue pair into this state. Once vmci_qp_broker_set_page_store + * is called to register the user memory, the VMCIQPB_ATTACH_MEM state + * will be entered. + * + * From the attached queue pair, the queue pair can enter the shutdown states + * when either side of the queue pair detaches. If the guest side detaches + * first, the queue pair will enter the VMCIQPB_SHUTDOWN_NO_MEM state, where + * the content of the queue pair will no longer be available. If the host + * side detaches first, the queue pair will either enter the + * VMCIQPB_SHUTDOWN_MEM, if the guest memory is currently mapped, or + * VMCIQPB_SHUTDOWN_NO_MEM, if the guest memory is not mapped + * (e.g., the host detaches while a guest is stunned). + * + * New-style VMX'en will also unmap guest memory, if the guest is + * quiesced, e.g., during a snapshot operation. In that case, the guest + * memory will no longer be available, and the queue pair will transition from + * *_MEM state to a *_NO_MEM state. The VMX may later map the memory once more, + * in which case the queue pair will transition from the *_NO_MEM state at that + * point back to the *_MEM state. Note that the *_NO_MEM state may have changed, + * since the peer may have either attached or detached in the meantime. The + * values are laid out such that ++ on a state will move from a *_NO_MEM to a + * *_MEM state, and vice versa. + */ + +/* + * VMCIMemcpy{To,From}QueueFunc() prototypes. Functions of these + * types are passed around to enqueue and dequeue routines. Note that + * often the functions passed are simply wrappers around memcpy + * itself. + * + * Note: In order for the memcpy typedefs to be compatible with the VMKernel, + * there's an unused last parameter for the hosted side. In + * ESX, that parameter holds a buffer type. + */ +typedef int vmci_memcpy_to_queue_func(struct vmci_queue *queue, + u64 queue_offset, const void *src, + size_t src_offset, size_t size); +typedef int vmci_memcpy_from_queue_func(void *dest, size_t dest_offset, + const struct vmci_queue *queue, + u64 queue_offset, size_t size); + +/* The Kernel specific component of the struct vmci_queue structure. */ +struct vmci_queue_kern_if { + struct mutex __mutex; /* Protects the queue. */ + struct mutex *mutex; /* Shared by producer and consumer queues. */ + size_t num_pages; /* Number of pages incl. header. */ + bool host; /* Host or guest? */ + union { + struct { + dma_addr_t *pas; + void **vas; + } g; /* Used by the guest. */ + struct { + struct page **page; + struct page **header_page; + } h; /* Used by the host. */ + } u; +}; + +/* + * This structure is opaque to the clients. + */ +struct vmci_qp { + struct vmci_handle handle; + struct vmci_queue *produce_q; + struct vmci_queue *consume_q; + u64 produce_q_size; + u64 consume_q_size; + u32 peer; + u32 flags; + u32 priv_flags; + bool guest_endpoint; + unsigned int blocked; + unsigned int generation; + wait_queue_head_t event; +}; + +enum qp_broker_state { + VMCIQPB_NEW, + VMCIQPB_CREATED_NO_MEM, + VMCIQPB_CREATED_MEM, + VMCIQPB_ATTACHED_NO_MEM, + VMCIQPB_ATTACHED_MEM, + VMCIQPB_SHUTDOWN_NO_MEM, + VMCIQPB_SHUTDOWN_MEM, + VMCIQPB_GONE +}; + +#define QPBROKERSTATE_HAS_MEM(_qpb) (_qpb->state == VMCIQPB_CREATED_MEM || \ + _qpb->state == VMCIQPB_ATTACHED_MEM || \ + _qpb->state == VMCIQPB_SHUTDOWN_MEM) + +/* + * In the queue pair broker, we always use the guest point of view for + * the produce and consume queue values and references, e.g., the + * produce queue size stored is the guests produce queue size. The + * host endpoint will need to swap these around. The only exception is + * the local queue pairs on the host, in which case the host endpoint + * that creates the queue pair will have the right orientation, and + * the attaching host endpoint will need to swap. + */ +struct qp_entry { + struct list_head list_item; + struct vmci_handle handle; + u32 peer; + u32 flags; + u64 produce_size; + u64 consume_size; + u32 ref_count; +}; + +struct qp_broker_entry { + struct vmci_resource resource; + struct qp_entry qp; + u32 create_id; + u32 attach_id; + enum qp_broker_state state; + bool require_trusted_attach; + bool created_by_trusted; + bool vmci_page_files; /* Created by VMX using VMCI page files */ + struct vmci_queue *produce_q; + struct vmci_queue *consume_q; + struct vmci_queue_header saved_produce_q; + struct vmci_queue_header saved_consume_q; + vmci_event_release_cb wakeup_cb; + void *client_data; + void *local_mem; /* Kernel memory for local queue pair */ +}; + +struct qp_guest_endpoint { + struct vmci_resource resource; + struct qp_entry qp; + u64 num_ppns; + void *produce_q; + void *consume_q; + struct ppn_set ppn_set; +}; + +struct qp_list { + struct list_head head; + struct mutex mutex; /* Protect queue list. */ +}; + +static struct qp_list qp_broker_list = { + .head = LIST_HEAD_INIT(qp_broker_list.head), + .mutex = __MUTEX_INITIALIZER(qp_broker_list.mutex), +}; + +static struct qp_list qp_guest_endpoints = { + .head = LIST_HEAD_INIT(qp_guest_endpoints.head), + .mutex = __MUTEX_INITIALIZER(qp_guest_endpoints.mutex), +}; + +#define INVALID_VMCI_GUEST_MEM_ID 0 +#define QPE_NUM_PAGES(_QPE) ((u32) \ + (DIV_ROUND_UP(_QPE.produce_size, PAGE_SIZE) + \ + DIV_ROUND_UP(_QPE.consume_size, PAGE_SIZE) + 2)) + + +/* + * Frees kernel VA space for a given queue and its queue header, and + * frees physical data pages. + */ +static void qp_free_queue(void *q, u64 size) +{ + struct vmci_queue *queue = q; + + if (queue) { + u64 i; + + /* Given size does not include header, so add in a page here. */ + for (i = 0; i < DIV_ROUND_UP(size, PAGE_SIZE) + 1; i++) { + dma_free_coherent(&vmci_pdev->dev, PAGE_SIZE, + queue->kernel_if->u.g.vas[i], + queue->kernel_if->u.g.pas[i]); + } + + vfree(queue); + } +} + +/* + * Allocates kernel queue pages of specified size with IOMMU mappings, + * plus space for the queue structure/kernel interface and the queue + * header. + */ +static void *qp_alloc_queue(u64 size, u32 flags) +{ + u64 i; + struct vmci_queue *queue; + const size_t num_pages = DIV_ROUND_UP(size, PAGE_SIZE) + 1; + const size_t pas_size = num_pages * sizeof(*queue->kernel_if->u.g.pas); + const size_t vas_size = num_pages * sizeof(*queue->kernel_if->u.g.vas); + const size_t queue_size = + sizeof(*queue) + sizeof(*queue->kernel_if) + + pas_size + vas_size; + + queue = vmalloc(queue_size); + if (!queue) + return NULL; + + queue->q_header = NULL; + queue->saved_header = NULL; + queue->kernel_if = (struct vmci_queue_kern_if *)(queue + 1); + queue->kernel_if->mutex = NULL; + queue->kernel_if->num_pages = num_pages; + queue->kernel_if->u.g.pas = (dma_addr_t *)(queue->kernel_if + 1); + queue->kernel_if->u.g.vas = + (void **)((u8 *)queue->kernel_if->u.g.pas + pas_size); + queue->kernel_if->host = false; + + for (i = 0; i < num_pages; i++) { + queue->kernel_if->u.g.vas[i] = + dma_alloc_coherent(&vmci_pdev->dev, PAGE_SIZE, + &queue->kernel_if->u.g.pas[i], + GFP_KERNEL); + if (!queue->kernel_if->u.g.vas[i]) { + /* Size excl. the header. */ + qp_free_queue(queue, i * PAGE_SIZE); + return NULL; + } + } + + /* Queue header is the first page. */ + queue->q_header = queue->kernel_if->u.g.vas[0]; + + return queue; +} + +/* + * Copies from a given buffer or iovector to a VMCI Queue. Uses + * kmap()/kunmap() to dynamically map/unmap required portions of the queue + * by traversing the offset -> page translation structure for the queue. + * Assumes that offset + size does not wrap around in the queue. + */ +static int __qp_memcpy_to_queue(struct vmci_queue *queue, + u64 queue_offset, + const void *src, + size_t size, + bool is_iovec) +{ + struct vmci_queue_kern_if *kernel_if = queue->kernel_if; + size_t bytes_copied = 0; + + while (bytes_copied < size) { + const u64 page_index = + (queue_offset + bytes_copied) / PAGE_SIZE; + const size_t page_offset = + (queue_offset + bytes_copied) & (PAGE_SIZE - 1); + void *va; + size_t to_copy; + + if (kernel_if->host) + va = kmap(kernel_if->u.h.page[page_index]); + else + va = kernel_if->u.g.vas[page_index + 1]; + /* Skip header. */ + + if (size - bytes_copied > PAGE_SIZE - page_offset) + /* Enough payload to fill up from this page. */ + to_copy = PAGE_SIZE - page_offset; + else + to_copy = size - bytes_copied; + + if (is_iovec) { + struct iovec *iov = (struct iovec *)src; + int err; + + /* The iovec will track bytes_copied internally. */ + err = memcpy_fromiovec((u8 *)va + page_offset, + iov, to_copy); + if (err != 0) { + if (kernel_if->host) + kunmap(kernel_if->u.h.page[page_index]); + return VMCI_ERROR_INVALID_ARGS; + } + } else { + memcpy((u8 *)va + page_offset, + (u8 *)src + bytes_copied, to_copy); + } + + bytes_copied += to_copy; + if (kernel_if->host) + kunmap(kernel_if->u.h.page[page_index]); + } + + return VMCI_SUCCESS; +} + +/* + * Copies to a given buffer or iovector from a VMCI Queue. Uses + * kmap()/kunmap() to dynamically map/unmap required portions of the queue + * by traversing the offset -> page translation structure for the queue. + * Assumes that offset + size does not wrap around in the queue. + */ +static int __qp_memcpy_from_queue(void *dest, + const struct vmci_queue *queue, + u64 queue_offset, + size_t size, + bool is_iovec) +{ + struct vmci_queue_kern_if *kernel_if = queue->kernel_if; + size_t bytes_copied = 0; + + while (bytes_copied < size) { + const u64 page_index = + (queue_offset + bytes_copied) / PAGE_SIZE; + const size_t page_offset = + (queue_offset + bytes_copied) & (PAGE_SIZE - 1); + void *va; + size_t to_copy; + + if (kernel_if->host) + va = kmap(kernel_if->u.h.page[page_index]); + else + va = kernel_if->u.g.vas[page_index + 1]; + /* Skip header. */ + + if (size - bytes_copied > PAGE_SIZE - page_offset) + /* Enough payload to fill up this page. */ + to_copy = PAGE_SIZE - page_offset; + else + to_copy = size - bytes_copied; + + if (is_iovec) { + struct iovec *iov = (struct iovec *)dest; + int err; + + /* The iovec will track bytes_copied internally. */ + err = memcpy_toiovec(iov, (u8 *)va + page_offset, + to_copy); + if (err != 0) { + if (kernel_if->host) + kunmap(kernel_if->u.h.page[page_index]); + return VMCI_ERROR_INVALID_ARGS; + } + } else { + memcpy((u8 *)dest + bytes_copied, + (u8 *)va + page_offset, to_copy); + } + + bytes_copied += to_copy; + if (kernel_if->host) + kunmap(kernel_if->u.h.page[page_index]); + } + + return VMCI_SUCCESS; +} + +/* + * Allocates two list of PPNs --- one for the pages in the produce queue, + * and the other for the pages in the consume queue. Intializes the list + * of PPNs with the page frame numbers of the KVA for the two queues (and + * the queue headers). + */ +static int qp_alloc_ppn_set(void *prod_q, + u64 num_produce_pages, + void *cons_q, + u64 num_consume_pages, struct ppn_set *ppn_set) +{ + u32 *produce_ppns; + u32 *consume_ppns; + struct vmci_queue *produce_q = prod_q; + struct vmci_queue *consume_q = cons_q; + u64 i; + + if (!produce_q || !num_produce_pages || !consume_q || + !num_consume_pages || !ppn_set) + return VMCI_ERROR_INVALID_ARGS; + + if (ppn_set->initialized) + return VMCI_ERROR_ALREADY_EXISTS; + + produce_ppns = + kmalloc(num_produce_pages * sizeof(*produce_ppns), GFP_KERNEL); + if (!produce_ppns) + return VMCI_ERROR_NO_MEM; + + consume_ppns = + kmalloc(num_consume_pages * sizeof(*consume_ppns), GFP_KERNEL); + if (!consume_ppns) { + kfree(produce_ppns); + return VMCI_ERROR_NO_MEM; + } + + for (i = 0; i < num_produce_pages; i++) { + unsigned long pfn; + + produce_ppns[i] = + produce_q->kernel_if->u.g.pas[i] >> PAGE_SHIFT; + pfn = produce_ppns[i]; + + /* Fail allocation if PFN isn't supported by hypervisor. */ + if (sizeof(pfn) > sizeof(*produce_ppns) + && pfn != produce_ppns[i]) + goto ppn_error; + } + + for (i = 0; i < num_consume_pages; i++) { + unsigned long pfn; + + consume_ppns[i] = + consume_q->kernel_if->u.g.pas[i] >> PAGE_SHIFT; + pfn = consume_ppns[i]; + + /* Fail allocation if PFN isn't supported by hypervisor. */ + if (sizeof(pfn) > sizeof(*consume_ppns) + && pfn != consume_ppns[i]) + goto ppn_error; + } + + ppn_set->num_produce_pages = num_produce_pages; + ppn_set->num_consume_pages = num_consume_pages; + ppn_set->produce_ppns = produce_ppns; + ppn_set->consume_ppns = consume_ppns; + ppn_set->initialized = true; + return VMCI_SUCCESS; + + ppn_error: + kfree(produce_ppns); + kfree(consume_ppns); + return VMCI_ERROR_INVALID_ARGS; +} + +/* + * Frees the two list of PPNs for a queue pair. + */ +static void qp_free_ppn_set(struct ppn_set *ppn_set) +{ + if (ppn_set->initialized) { + /* Do not call these functions on NULL inputs. */ + kfree(ppn_set->produce_ppns); + kfree(ppn_set->consume_ppns); + } + memset(ppn_set, 0, sizeof(*ppn_set)); +} + +/* + * Populates the list of PPNs in the hypercall structure with the PPNS + * of the produce queue and the consume queue. + */ +static int qp_populate_ppn_set(u8 *call_buf, const struct ppn_set *ppn_set) +{ + memcpy(call_buf, ppn_set->produce_ppns, + ppn_set->num_produce_pages * sizeof(*ppn_set->produce_ppns)); + memcpy(call_buf + + ppn_set->num_produce_pages * sizeof(*ppn_set->produce_ppns), + ppn_set->consume_ppns, + ppn_set->num_consume_pages * sizeof(*ppn_set->consume_ppns)); + + return VMCI_SUCCESS; +} + +static int qp_memcpy_to_queue(struct vmci_queue *queue, + u64 queue_offset, + const void *src, size_t src_offset, size_t size) +{ + return __qp_memcpy_to_queue(queue, queue_offset, + (u8 *)src + src_offset, size, false); +} + +static int qp_memcpy_from_queue(void *dest, + size_t dest_offset, + const struct vmci_queue *queue, + u64 queue_offset, size_t size) +{ + return __qp_memcpy_from_queue((u8 *)dest + dest_offset, + queue, queue_offset, size, false); +} + +/* + * Copies from a given iovec from a VMCI Queue. + */ +static int qp_memcpy_to_queue_iov(struct vmci_queue *queue, + u64 queue_offset, + const void *src, + size_t src_offset, size_t size) +{ + + /* + * We ignore src_offset because src is really a struct iovec * and will + * maintain offset internally. + */ + return __qp_memcpy_to_queue(queue, queue_offset, src, size, true); +} + +/* + * Copies to a given iovec from a VMCI Queue. + */ +static int qp_memcpy_from_queue_iov(void *dest, + size_t dest_offset, + const struct vmci_queue *queue, + u64 queue_offset, size_t size) +{ + /* + * We ignore dest_offset because dest is really a struct iovec * and + * will maintain offset internally. + */ + return __qp_memcpy_from_queue(dest, queue, queue_offset, size, true); +} + +/* + * Allocates kernel VA space of specified size plus space for the queue + * and kernel interface. This is different from the guest queue allocator, + * because we do not allocate our own queue header/data pages here but + * share those of the guest. + */ +static struct vmci_queue *qp_host_alloc_queue(u64 size) +{ + struct vmci_queue *queue; + const size_t num_pages = DIV_ROUND_UP(size, PAGE_SIZE) + 1; + const size_t queue_size = sizeof(*queue) + sizeof(*(queue->kernel_if)); + const size_t queue_page_size = + num_pages * sizeof(*queue->kernel_if->u.h.page); + + queue = kzalloc(queue_size + queue_page_size, GFP_KERNEL); + if (queue) { + queue->q_header = NULL; + queue->saved_header = NULL; + queue->kernel_if = (struct vmci_queue_kern_if *)(queue + 1); + queue->kernel_if->host = true; + queue->kernel_if->mutex = NULL; + queue->kernel_if->num_pages = num_pages; + queue->kernel_if->u.h.header_page = + (struct page **)((u8 *)queue + queue_size); + queue->kernel_if->u.h.page = + &queue->kernel_if->u.h.header_page[1]; + } + + return queue; +} + +/* + * Frees kernel memory for a given queue (header plus translation + * structure). + */ +static void qp_host_free_queue(struct vmci_queue *queue, u64 queue_size) +{ + kfree(queue); +} + +/* + * Initialize the mutex for the pair of queues. This mutex is used to + * protect the q_header and the buffer from changing out from under any + * users of either queue. Of course, it's only any good if the mutexes + * are actually acquired. Queue structure must lie on non-paged memory + * or we cannot guarantee access to the mutex. + */ +static void qp_init_queue_mutex(struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + /* + * Only the host queue has shared state - the guest queues do not + * need to synchronize access using a queue mutex. + */ + + if (produce_q->kernel_if->host) { + produce_q->kernel_if->mutex = &produce_q->kernel_if->__mutex; + consume_q->kernel_if->mutex = &produce_q->kernel_if->__mutex; + mutex_init(produce_q->kernel_if->mutex); + } +} + +/* + * Cleans up the mutex for the pair of queues. + */ +static void qp_cleanup_queue_mutex(struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + if (produce_q->kernel_if->host) { + produce_q->kernel_if->mutex = NULL; + consume_q->kernel_if->mutex = NULL; + } +} + +/* + * Acquire the mutex for the queue. Note that the produce_q and + * the consume_q share a mutex. So, only one of the two need to + * be passed in to this routine. Either will work just fine. + */ +static void qp_acquire_queue_mutex(struct vmci_queue *queue) +{ + if (queue->kernel_if->host) + mutex_lock(queue->kernel_if->mutex); +} + +/* + * Release the mutex for the queue. Note that the produce_q and + * the consume_q share a mutex. So, only one of the two need to + * be passed in to this routine. Either will work just fine. + */ +static void qp_release_queue_mutex(struct vmci_queue *queue) +{ + if (queue->kernel_if->host) + mutex_unlock(queue->kernel_if->mutex); +} + +/* + * Helper function to release pages in the PageStoreAttachInfo + * previously obtained using get_user_pages. + */ +static void qp_release_pages(struct page **pages, + u64 num_pages, bool dirty) +{ + int i; + + for (i = 0; i < num_pages; i++) { + if (dirty) + set_page_dirty(pages[i]); + + page_cache_release(pages[i]); + pages[i] = NULL; + } +} + +/* + * Lock the user pages referenced by the {produce,consume}Buffer + * struct into memory and populate the {produce,consume}Pages + * arrays in the attach structure with them. + */ +static int qp_host_get_user_memory(u64 produce_uva, + u64 consume_uva, + struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + int retval; + int err = VMCI_SUCCESS; + + retval = get_user_pages_fast((uintptr_t) produce_uva, + produce_q->kernel_if->num_pages, 1, + produce_q->kernel_if->u.h.header_page); + if (retval < produce_q->kernel_if->num_pages) { + pr_warn("get_user_pages(produce) failed (retval=%d)", retval); + qp_release_pages(produce_q->kernel_if->u.h.header_page, + retval, false); + err = VMCI_ERROR_NO_MEM; + goto out; + } + + retval = get_user_pages_fast((uintptr_t) consume_uva, + consume_q->kernel_if->num_pages, 1, + consume_q->kernel_if->u.h.header_page); + if (retval < consume_q->kernel_if->num_pages) { + pr_warn("get_user_pages(consume) failed (retval=%d)", retval); + qp_release_pages(consume_q->kernel_if->u.h.header_page, + retval, false); + qp_release_pages(produce_q->kernel_if->u.h.header_page, + produce_q->kernel_if->num_pages, false); + err = VMCI_ERROR_NO_MEM; + } + + out: + return err; +} + +/* + * Registers the specification of the user pages used for backing a queue + * pair. Enough information to map in pages is stored in the OS specific + * part of the struct vmci_queue structure. + */ +static int qp_host_register_user_memory(struct vmci_qp_page_store *page_store, + struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + u64 produce_uva; + u64 consume_uva; + + /* + * The new style and the old style mapping only differs in + * that we either get a single or two UVAs, so we split the + * single UVA range at the appropriate spot. + */ + produce_uva = page_store->pages; + consume_uva = page_store->pages + + produce_q->kernel_if->num_pages * PAGE_SIZE; + return qp_host_get_user_memory(produce_uva, consume_uva, produce_q, + consume_q); +} + +/* + * Releases and removes the references to user pages stored in the attach + * struct. Pages are released from the page cache and may become + * swappable again. + */ +static void qp_host_unregister_user_memory(struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + qp_release_pages(produce_q->kernel_if->u.h.header_page, + produce_q->kernel_if->num_pages, true); + memset(produce_q->kernel_if->u.h.header_page, 0, + sizeof(*produce_q->kernel_if->u.h.header_page) * + produce_q->kernel_if->num_pages); + qp_release_pages(consume_q->kernel_if->u.h.header_page, + consume_q->kernel_if->num_pages, true); + memset(consume_q->kernel_if->u.h.header_page, 0, + sizeof(*consume_q->kernel_if->u.h.header_page) * + consume_q->kernel_if->num_pages); +} + +/* + * Once qp_host_register_user_memory has been performed on a + * queue, the queue pair headers can be mapped into the + * kernel. Once mapped, they must be unmapped with + * qp_host_unmap_queues prior to calling + * qp_host_unregister_user_memory. + * Pages are pinned. + */ +static int qp_host_map_queues(struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + int result; + + if (!produce_q->q_header || !consume_q->q_header) { + struct page *headers[2]; + + if (produce_q->q_header != consume_q->q_header) + return VMCI_ERROR_QUEUEPAIR_MISMATCH; + + if (produce_q->kernel_if->u.h.header_page == NULL || + *produce_q->kernel_if->u.h.header_page == NULL) + return VMCI_ERROR_UNAVAILABLE; + + headers[0] = *produce_q->kernel_if->u.h.header_page; + headers[1] = *consume_q->kernel_if->u.h.header_page; + + produce_q->q_header = vmap(headers, 2, VM_MAP, PAGE_KERNEL); + if (produce_q->q_header != NULL) { + consume_q->q_header = + (struct vmci_queue_header *)((u8 *) + produce_q->q_header + + PAGE_SIZE); + result = VMCI_SUCCESS; + } else { + pr_warn("vmap failed\n"); + result = VMCI_ERROR_NO_MEM; + } + } else { + result = VMCI_SUCCESS; + } + + return result; +} + +/* + * Unmaps previously mapped queue pair headers from the kernel. + * Pages are unpinned. + */ +static int qp_host_unmap_queues(u32 gid, + struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + if (produce_q->q_header) { + if (produce_q->q_header < consume_q->q_header) + vunmap(produce_q->q_header); + else + vunmap(consume_q->q_header); + + produce_q->q_header = NULL; + consume_q->q_header = NULL; + } + + return VMCI_SUCCESS; +} + +/* + * Finds the entry in the list corresponding to a given handle. Assumes + * that the list is locked. + */ +static struct qp_entry *qp_list_find(struct qp_list *qp_list, + struct vmci_handle handle) +{ + struct qp_entry *entry; + + if (vmci_handle_is_invalid(handle)) + return NULL; + + list_for_each_entry(entry, &qp_list->head, list_item) { + if (vmci_handle_is_equal(entry->handle, handle)) + return entry; + } + + return NULL; +} + +/* + * Finds the entry in the list corresponding to a given handle. + */ +static struct qp_guest_endpoint * +qp_guest_handle_to_entry(struct vmci_handle handle) +{ + struct qp_guest_endpoint *entry; + struct qp_entry *qp = qp_list_find(&qp_guest_endpoints, handle); + + entry = qp ? container_of( + qp, struct qp_guest_endpoint, qp) : NULL; + return entry; +} + +/* + * Finds the entry in the list corresponding to a given handle. + */ +static struct qp_broker_entry * +qp_broker_handle_to_entry(struct vmci_handle handle) +{ + struct qp_broker_entry *entry; + struct qp_entry *qp = qp_list_find(&qp_broker_list, handle); + + entry = qp ? container_of( + qp, struct qp_broker_entry, qp) : NULL; + return entry; +} + +/* + * Dispatches a queue pair event message directly into the local event + * queue. + */ +static int qp_notify_peer_local(bool attach, struct vmci_handle handle) +{ + u32 context_id = vmci_get_context_id(); + struct vmci_event_qp ev; + + ev.msg.hdr.dst = vmci_make_handle(context_id, VMCI_EVENT_HANDLER); + ev.msg.hdr.src = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_CONTEXT_RESOURCE_ID); + ev.msg.hdr.payload_size = sizeof(ev) - sizeof(ev.msg.hdr); + ev.msg.event_data.event = + attach ? VMCI_EVENT_QP_PEER_ATTACH : VMCI_EVENT_QP_PEER_DETACH; + ev.payload.peer_id = context_id; + ev.payload.handle = handle; + + return vmci_event_dispatch(&ev.msg.hdr); +} + +/* + * Allocates and initializes a qp_guest_endpoint structure. + * Allocates a queue_pair rid (and handle) iff the given entry has + * an invalid handle. 0 through VMCI_RESERVED_RESOURCE_ID_MAX + * are reserved handles. Assumes that the QP list mutex is held + * by the caller. + */ +static struct qp_guest_endpoint * +qp_guest_endpoint_create(struct vmci_handle handle, + u32 peer, + u32 flags, + u64 produce_size, + u64 consume_size, + void *produce_q, + void *consume_q) +{ + int result; + struct qp_guest_endpoint *entry; + /* One page each for the queue headers. */ + const u64 num_ppns = DIV_ROUND_UP(produce_size, PAGE_SIZE) + + DIV_ROUND_UP(consume_size, PAGE_SIZE) + 2; + + if (vmci_handle_is_invalid(handle)) { + u32 context_id = vmci_get_context_id(); + + handle = vmci_make_handle(context_id, VMCI_INVALID_ID); + } + + entry = kzalloc(sizeof(*entry), GFP_KERNEL); + if (entry) { + entry->qp.peer = peer; + entry->qp.flags = flags; + entry->qp.produce_size = produce_size; + entry->qp.consume_size = consume_size; + entry->qp.ref_count = 0; + entry->num_ppns = num_ppns; + entry->produce_q = produce_q; + entry->consume_q = consume_q; + INIT_LIST_HEAD(&entry->qp.list_item); + + /* Add resource obj */ + result = vmci_resource_add(&entry->resource, + VMCI_RESOURCE_TYPE_QPAIR_GUEST, + handle); + entry->qp.handle = vmci_resource_handle(&entry->resource); + if ((result != VMCI_SUCCESS) || + qp_list_find(&qp_guest_endpoints, entry->qp.handle)) { + pr_warn("Failed to add new resource (handle=0x%x:0x%x), error: %d", + handle.context, handle.resource, result); + kfree(entry); + entry = NULL; + } + } + return entry; +} + +/* + * Frees a qp_guest_endpoint structure. + */ +static void qp_guest_endpoint_destroy(struct qp_guest_endpoint *entry) +{ + qp_free_ppn_set(&entry->ppn_set); + qp_cleanup_queue_mutex(entry->produce_q, entry->consume_q); + qp_free_queue(entry->produce_q, entry->qp.produce_size); + qp_free_queue(entry->consume_q, entry->qp.consume_size); + /* Unlink from resource hash table and free callback */ + vmci_resource_remove(&entry->resource); + + kfree(entry); +} + +/* + * Helper to make a queue_pairAlloc hypercall when the driver is + * supporting a guest device. + */ +static int qp_alloc_hypercall(const struct qp_guest_endpoint *entry) +{ + struct vmci_qp_alloc_msg *alloc_msg; + size_t msg_size; + int result; + + if (!entry || entry->num_ppns <= 2) + return VMCI_ERROR_INVALID_ARGS; + + msg_size = sizeof(*alloc_msg) + + (size_t) entry->num_ppns * sizeof(u32); + alloc_msg = kmalloc(msg_size, GFP_KERNEL); + if (!alloc_msg) + return VMCI_ERROR_NO_MEM; + + alloc_msg->hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_QUEUEPAIR_ALLOC); + alloc_msg->hdr.src = VMCI_ANON_SRC_HANDLE; + alloc_msg->hdr.payload_size = msg_size - VMCI_DG_HEADERSIZE; + alloc_msg->handle = entry->qp.handle; + alloc_msg->peer = entry->qp.peer; + alloc_msg->flags = entry->qp.flags; + alloc_msg->produce_size = entry->qp.produce_size; + alloc_msg->consume_size = entry->qp.consume_size; + alloc_msg->num_ppns = entry->num_ppns; + + result = qp_populate_ppn_set((u8 *)alloc_msg + sizeof(*alloc_msg), + &entry->ppn_set); + if (result == VMCI_SUCCESS) + result = vmci_send_datagram(&alloc_msg->hdr); + + kfree(alloc_msg); + + return result; +} + +/* + * Helper to make a queue_pairDetach hypercall when the driver is + * supporting a guest device. + */ +static int qp_detatch_hypercall(struct vmci_handle handle) +{ + struct vmci_qp_detach_msg detach_msg; + + detach_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_QUEUEPAIR_DETACH); + detach_msg.hdr.src = VMCI_ANON_SRC_HANDLE; + detach_msg.hdr.payload_size = sizeof(handle); + detach_msg.handle = handle; + + return vmci_send_datagram(&detach_msg.hdr); +} + +/* + * Adds the given entry to the list. Assumes that the list is locked. + */ +static void qp_list_add_entry(struct qp_list *qp_list, struct qp_entry *entry) +{ + if (entry) + list_add(&entry->list_item, &qp_list->head); +} + +/* + * Removes the given entry from the list. Assumes that the list is locked. + */ +static void qp_list_remove_entry(struct qp_list *qp_list, + struct qp_entry *entry) +{ + if (entry) + list_del(&entry->list_item); +} + +/* + * Helper for VMCI queue_pair detach interface. Frees the physical + * pages for the queue pair. + */ +static int qp_detatch_guest_work(struct vmci_handle handle) +{ + int result; + struct qp_guest_endpoint *entry; + u32 ref_count = ~0; /* To avoid compiler warning below */ + + mutex_lock(&qp_guest_endpoints.mutex); + + entry = qp_guest_handle_to_entry(handle); + if (!entry) { + mutex_unlock(&qp_guest_endpoints.mutex); + return VMCI_ERROR_NOT_FOUND; + } + + if (entry->qp.flags & VMCI_QPFLAG_LOCAL) { + result = VMCI_SUCCESS; + + if (entry->qp.ref_count > 1) { + result = qp_notify_peer_local(false, handle); + /* + * We can fail to notify a local queuepair + * because we can't allocate. We still want + * to release the entry if that happens, so + * don't bail out yet. + */ + } + } else { + result = qp_detatch_hypercall(handle); + if (result < VMCI_SUCCESS) { + /* + * We failed to notify a non-local queuepair. + * That other queuepair might still be + * accessing the shared memory, so don't + * release the entry yet. It will get cleaned + * up by VMCIqueue_pair_Exit() if necessary + * (assuming we are going away, otherwise why + * did this fail?). + */ + + mutex_unlock(&qp_guest_endpoints.mutex); + return result; + } + } + + /* + * If we get here then we either failed to notify a local queuepair, or + * we succeeded in all cases. Release the entry if required. + */ + + entry->qp.ref_count--; + if (entry->qp.ref_count == 0) + qp_list_remove_entry(&qp_guest_endpoints, &entry->qp); + + /* If we didn't remove the entry, this could change once we unlock. */ + if (entry) + ref_count = entry->qp.ref_count; + + mutex_unlock(&qp_guest_endpoints.mutex); + + if (ref_count == 0) + qp_guest_endpoint_destroy(entry); + + return result; +} + +/* + * This functions handles the actual allocation of a VMCI queue + * pair guest endpoint. Allocates physical pages for the queue + * pair. It makes OS dependent calls through generic wrappers. + */ +static int qp_alloc_guest_work(struct vmci_handle *handle, + struct vmci_queue **produce_q, + u64 produce_size, + struct vmci_queue **consume_q, + u64 consume_size, + u32 peer, + u32 flags, + u32 priv_flags) +{ + const u64 num_produce_pages = + DIV_ROUND_UP(produce_size, PAGE_SIZE) + 1; + const u64 num_consume_pages = + DIV_ROUND_UP(consume_size, PAGE_SIZE) + 1; + void *my_produce_q = NULL; + void *my_consume_q = NULL; + int result; + struct qp_guest_endpoint *queue_pair_entry = NULL; + + if (priv_flags != VMCI_NO_PRIVILEGE_FLAGS) + return VMCI_ERROR_NO_ACCESS; + + mutex_lock(&qp_guest_endpoints.mutex); + + queue_pair_entry = qp_guest_handle_to_entry(*handle); + if (queue_pair_entry) { + if (queue_pair_entry->qp.flags & VMCI_QPFLAG_LOCAL) { + /* Local attach case. */ + if (queue_pair_entry->qp.ref_count > 1) { + pr_devel("Error attempting to attach more than once\n"); + result = VMCI_ERROR_UNAVAILABLE; + goto error_keep_entry; + } + + if (queue_pair_entry->qp.produce_size != consume_size || + queue_pair_entry->qp.consume_size != + produce_size || + queue_pair_entry->qp.flags != + (flags & ~VMCI_QPFLAG_ATTACH_ONLY)) { + pr_devel("Error mismatched queue pair in local attach\n"); + result = VMCI_ERROR_QUEUEPAIR_MISMATCH; + goto error_keep_entry; + } + + /* + * Do a local attach. We swap the consume and + * produce queues for the attacher and deliver + * an attach event. + */ + result = qp_notify_peer_local(true, *handle); + if (result < VMCI_SUCCESS) + goto error_keep_entry; + + my_produce_q = queue_pair_entry->consume_q; + my_consume_q = queue_pair_entry->produce_q; + goto out; + } + + result = VMCI_ERROR_ALREADY_EXISTS; + goto error_keep_entry; + } + + my_produce_q = qp_alloc_queue(produce_size, flags); + if (!my_produce_q) { + pr_warn("Error allocating pages for produce queue\n"); + result = VMCI_ERROR_NO_MEM; + goto error; + } + + my_consume_q = qp_alloc_queue(consume_size, flags); + if (!my_consume_q) { + pr_warn("Error allocating pages for consume queue\n"); + result = VMCI_ERROR_NO_MEM; + goto error; + } + + queue_pair_entry = qp_guest_endpoint_create(*handle, peer, flags, + produce_size, consume_size, + my_produce_q, my_consume_q); + if (!queue_pair_entry) { + pr_warn("Error allocating memory in %s\n", __func__); + result = VMCI_ERROR_NO_MEM; + goto error; + } + + result = qp_alloc_ppn_set(my_produce_q, num_produce_pages, my_consume_q, + num_consume_pages, + &queue_pair_entry->ppn_set); + if (result < VMCI_SUCCESS) { + pr_warn("qp_alloc_ppn_set failed\n"); + goto error; + } + + /* + * It's only necessary to notify the host if this queue pair will be + * attached to from another context. + */ + if (queue_pair_entry->qp.flags & VMCI_QPFLAG_LOCAL) { + /* Local create case. */ + u32 context_id = vmci_get_context_id(); + + /* + * Enforce similar checks on local queue pairs as we + * do for regular ones. The handle's context must + * match the creator or attacher context id (here they + * are both the current context id) and the + * attach-only flag cannot exist during create. We + * also ensure specified peer is this context or an + * invalid one. + */ + if (queue_pair_entry->qp.handle.context != context_id || + (queue_pair_entry->qp.peer != VMCI_INVALID_ID && + queue_pair_entry->qp.peer != context_id)) { + result = VMCI_ERROR_NO_ACCESS; + goto error; + } + + if (queue_pair_entry->qp.flags & VMCI_QPFLAG_ATTACH_ONLY) { + result = VMCI_ERROR_NOT_FOUND; + goto error; + } + } else { + result = qp_alloc_hypercall(queue_pair_entry); + if (result < VMCI_SUCCESS) { + pr_warn("qp_alloc_hypercall result = %d\n", result); + goto error; + } + } + + qp_init_queue_mutex((struct vmci_queue *)my_produce_q, + (struct vmci_queue *)my_consume_q); + + qp_list_add_entry(&qp_guest_endpoints, &queue_pair_entry->qp); + + out: + queue_pair_entry->qp.ref_count++; + *handle = queue_pair_entry->qp.handle; + *produce_q = (struct vmci_queue *)my_produce_q; + *consume_q = (struct vmci_queue *)my_consume_q; + + /* + * We should initialize the queue pair header pages on a local + * queue pair create. For non-local queue pairs, the + * hypervisor initializes the header pages in the create step. + */ + if ((queue_pair_entry->qp.flags & VMCI_QPFLAG_LOCAL) && + queue_pair_entry->qp.ref_count == 1) { + vmci_q_header_init((*produce_q)->q_header, *handle); + vmci_q_header_init((*consume_q)->q_header, *handle); + } + + mutex_unlock(&qp_guest_endpoints.mutex); + + return VMCI_SUCCESS; + + error: + mutex_unlock(&qp_guest_endpoints.mutex); + if (queue_pair_entry) { + /* The queues will be freed inside the destroy routine. */ + qp_guest_endpoint_destroy(queue_pair_entry); + } else { + qp_free_queue(my_produce_q, produce_size); + qp_free_queue(my_consume_q, consume_size); + } + return result; + + error_keep_entry: + /* This path should only be used when an existing entry was found. */ + mutex_unlock(&qp_guest_endpoints.mutex); + return result; +} + +/* + * The first endpoint issuing a queue pair allocation will create the state + * of the queue pair in the queue pair broker. + * + * If the creator is a guest, it will associate a VMX virtual address range + * with the queue pair as specified by the page_store. For compatibility with + * older VMX'en, that would use a separate step to set the VMX virtual + * address range, the virtual address range can be registered later using + * vmci_qp_broker_set_page_store. In that case, a page_store of NULL should be + * used. + * + * If the creator is the host, a page_store of NULL should be used as well, + * since the host is not able to supply a page store for the queue pair. + * + * For older VMX and host callers, the queue pair will be created in the + * VMCIQPB_CREATED_NO_MEM state, and for current VMX callers, it will be + * created in VMCOQPB_CREATED_MEM state. + */ +static int qp_broker_create(struct vmci_handle handle, + u32 peer, + u32 flags, + u32 priv_flags, + u64 produce_size, + u64 consume_size, + struct vmci_qp_page_store *page_store, + struct vmci_ctx *context, + vmci_event_release_cb wakeup_cb, + void *client_data, struct qp_broker_entry **ent) +{ + struct qp_broker_entry *entry = NULL; + const u32 context_id = vmci_ctx_get_id(context); + bool is_local = flags & VMCI_QPFLAG_LOCAL; + int result; + u64 guest_produce_size; + u64 guest_consume_size; + + /* Do not create if the caller asked not to. */ + if (flags & VMCI_QPFLAG_ATTACH_ONLY) + return VMCI_ERROR_NOT_FOUND; + + /* + * Creator's context ID should match handle's context ID or the creator + * must allow the context in handle's context ID as the "peer". + */ + if (handle.context != context_id && handle.context != peer) + return VMCI_ERROR_NO_ACCESS; + + if (VMCI_CONTEXT_IS_VM(context_id) && VMCI_CONTEXT_IS_VM(peer)) + return VMCI_ERROR_DST_UNREACHABLE; + + /* + * Creator's context ID for local queue pairs should match the + * peer, if a peer is specified. + */ + if (is_local && peer != VMCI_INVALID_ID && context_id != peer) + return VMCI_ERROR_NO_ACCESS; + + entry = kzalloc(sizeof(*entry), GFP_ATOMIC); + if (!entry) + return VMCI_ERROR_NO_MEM; + + if (vmci_ctx_get_id(context) == VMCI_HOST_CONTEXT_ID && !is_local) { + /* + * The queue pair broker entry stores values from the guest + * point of view, so a creating host side endpoint should swap + * produce and consume values -- unless it is a local queue + * pair, in which case no swapping is necessary, since the local + * attacher will swap queues. + */ + + guest_produce_size = consume_size; + guest_consume_size = produce_size; + } else { + guest_produce_size = produce_size; + guest_consume_size = consume_size; + } + + entry->qp.handle = handle; + entry->qp.peer = peer; + entry->qp.flags = flags; + entry->qp.produce_size = guest_produce_size; + entry->qp.consume_size = guest_consume_size; + entry->qp.ref_count = 1; + entry->create_id = context_id; + entry->attach_id = VMCI_INVALID_ID; + entry->state = VMCIQPB_NEW; + entry->require_trusted_attach = + !!(context->priv_flags & VMCI_PRIVILEGE_FLAG_RESTRICTED); + entry->created_by_trusted = + !!(priv_flags & VMCI_PRIVILEGE_FLAG_TRUSTED); + entry->vmci_page_files = false; + entry->wakeup_cb = wakeup_cb; + entry->client_data = client_data; + entry->produce_q = qp_host_alloc_queue(guest_produce_size); + if (entry->produce_q == NULL) { + result = VMCI_ERROR_NO_MEM; + goto error; + } + entry->consume_q = qp_host_alloc_queue(guest_consume_size); + if (entry->consume_q == NULL) { + result = VMCI_ERROR_NO_MEM; + goto error; + } + + qp_init_queue_mutex(entry->produce_q, entry->consume_q); + + INIT_LIST_HEAD(&entry->qp.list_item); + + if (is_local) { + u8 *tmp; + + entry->local_mem = kcalloc(QPE_NUM_PAGES(entry->qp), + PAGE_SIZE, GFP_KERNEL); + if (entry->local_mem == NULL) { + result = VMCI_ERROR_NO_MEM; + goto error; + } + entry->state = VMCIQPB_CREATED_MEM; + entry->produce_q->q_header = entry->local_mem; + tmp = (u8 *)entry->local_mem + PAGE_SIZE * + (DIV_ROUND_UP(entry->qp.produce_size, PAGE_SIZE) + 1); + entry->consume_q->q_header = (struct vmci_queue_header *)tmp; + } else if (page_store) { + /* + * The VMX already initialized the queue pair headers, so no + * need for the kernel side to do that. + */ + result = qp_host_register_user_memory(page_store, + entry->produce_q, + entry->consume_q); + if (result < VMCI_SUCCESS) + goto error; + + entry->state = VMCIQPB_CREATED_MEM; + } else { + /* + * A create without a page_store may be either a host + * side create (in which case we are waiting for the + * guest side to supply the memory) or an old style + * queue pair create (in which case we will expect a + * set page store call as the next step). + */ + entry->state = VMCIQPB_CREATED_NO_MEM; + } + + qp_list_add_entry(&qp_broker_list, &entry->qp); + if (ent != NULL) + *ent = entry; + + /* Add to resource obj */ + result = vmci_resource_add(&entry->resource, + VMCI_RESOURCE_TYPE_QPAIR_HOST, + handle); + if (result != VMCI_SUCCESS) { + pr_warn("Failed to add new resource (handle=0x%x:0x%x), error: %d", + handle.context, handle.resource, result); + goto error; + } + + entry->qp.handle = vmci_resource_handle(&entry->resource); + if (is_local) { + vmci_q_header_init(entry->produce_q->q_header, + entry->qp.handle); + vmci_q_header_init(entry->consume_q->q_header, + entry->qp.handle); + } + + vmci_ctx_qp_create(context, entry->qp.handle); + + return VMCI_SUCCESS; + + error: + if (entry != NULL) { + qp_host_free_queue(entry->produce_q, guest_produce_size); + qp_host_free_queue(entry->consume_q, guest_consume_size); + kfree(entry); + } + + return result; +} + +/* + * Enqueues an event datagram to notify the peer VM attached to + * the given queue pair handle about attach/detach event by the + * given VM. Returns Payload size of datagram enqueued on + * success, error code otherwise. + */ +static int qp_notify_peer(bool attach, + struct vmci_handle handle, + u32 my_id, + u32 peer_id) +{ + int rv; + struct vmci_event_qp ev; + + if (vmci_handle_is_invalid(handle) || my_id == VMCI_INVALID_ID || + peer_id == VMCI_INVALID_ID) + return VMCI_ERROR_INVALID_ARGS; + + /* + * In vmci_ctx_enqueue_datagram() we enforce the upper limit on + * number of pending events from the hypervisor to a given VM + * otherwise a rogue VM could do an arbitrary number of attach + * and detach operations causing memory pressure in the host + * kernel. + */ + + ev.msg.hdr.dst = vmci_make_handle(peer_id, VMCI_EVENT_HANDLER); + ev.msg.hdr.src = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID, + VMCI_CONTEXT_RESOURCE_ID); + ev.msg.hdr.payload_size = sizeof(ev) - sizeof(ev.msg.hdr); + ev.msg.event_data.event = attach ? + VMCI_EVENT_QP_PEER_ATTACH : VMCI_EVENT_QP_PEER_DETACH; + ev.payload.handle = handle; + ev.payload.peer_id = my_id; + + rv = vmci_datagram_dispatch(VMCI_HYPERVISOR_CONTEXT_ID, + &ev.msg.hdr, false); + if (rv < VMCI_SUCCESS) + pr_warn("Failed to enqueue queue_pair %s event datagram for context (ID=0x%x)\n", + attach ? "ATTACH" : "DETACH", peer_id); + + return rv; +} + +/* + * The second endpoint issuing a queue pair allocation will attach to + * the queue pair registered with the queue pair broker. + * + * If the attacher is a guest, it will associate a VMX virtual address + * range with the queue pair as specified by the page_store. At this + * point, the already attach host endpoint may start using the queue + * pair, and an attach event is sent to it. For compatibility with + * older VMX'en, that used a separate step to set the VMX virtual + * address range, the virtual address range can be registered later + * using vmci_qp_broker_set_page_store. In that case, a page_store of + * NULL should be used, and the attach event will be generated once + * the actual page store has been set. + * + * If the attacher is the host, a page_store of NULL should be used as + * well, since the page store information is already set by the guest. + * + * For new VMX and host callers, the queue pair will be moved to the + * VMCIQPB_ATTACHED_MEM state, and for older VMX callers, it will be + * moved to the VMCOQPB_ATTACHED_NO_MEM state. + */ +static int qp_broker_attach(struct qp_broker_entry *entry, + u32 peer, + u32 flags, + u32 priv_flags, + u64 produce_size, + u64 consume_size, + struct vmci_qp_page_store *page_store, + struct vmci_ctx *context, + vmci_event_release_cb wakeup_cb, + void *client_data, + struct qp_broker_entry **ent) +{ + const u32 context_id = vmci_ctx_get_id(context); + bool is_local = flags & VMCI_QPFLAG_LOCAL; + int result; + + if (entry->state != VMCIQPB_CREATED_NO_MEM && + entry->state != VMCIQPB_CREATED_MEM) + return VMCI_ERROR_UNAVAILABLE; + + if (is_local) { + if (!(entry->qp.flags & VMCI_QPFLAG_LOCAL) || + context_id != entry->create_id) { + return VMCI_ERROR_INVALID_ARGS; + } + } else if (context_id == entry->create_id || + context_id == entry->attach_id) { + return VMCI_ERROR_ALREADY_EXISTS; + } + + if (VMCI_CONTEXT_IS_VM(context_id) && + VMCI_CONTEXT_IS_VM(entry->create_id)) + return VMCI_ERROR_DST_UNREACHABLE; + + /* + * If we are attaching from a restricted context then the queuepair + * must have been created by a trusted endpoint. + */ + if ((context->priv_flags & VMCI_PRIVILEGE_FLAG_RESTRICTED) && + !entry->created_by_trusted) + return VMCI_ERROR_NO_ACCESS; + + /* + * If we are attaching to a queuepair that was created by a restricted + * context then we must be trusted. + */ + if (entry->require_trusted_attach && + (!(priv_flags & VMCI_PRIVILEGE_FLAG_TRUSTED))) + return VMCI_ERROR_NO_ACCESS; + + /* + * If the creator specifies VMCI_INVALID_ID in "peer" field, access + * control check is not performed. + */ + if (entry->qp.peer != VMCI_INVALID_ID && entry->qp.peer != context_id) + return VMCI_ERROR_NO_ACCESS; + + if (entry->create_id == VMCI_HOST_CONTEXT_ID) { + /* + * Do not attach if the caller doesn't support Host Queue Pairs + * and a host created this queue pair. + */ + + if (!vmci_ctx_supports_host_qp(context)) + return VMCI_ERROR_INVALID_RESOURCE; + + } else if (context_id == VMCI_HOST_CONTEXT_ID) { + struct vmci_ctx *create_context; + bool supports_host_qp; + + /* + * Do not attach a host to a user created queue pair if that + * user doesn't support host queue pair end points. + */ + + create_context = vmci_ctx_get(entry->create_id); + supports_host_qp = vmci_ctx_supports_host_qp(create_context); + vmci_ctx_put(create_context); + + if (!supports_host_qp) + return VMCI_ERROR_INVALID_RESOURCE; + } + + if ((entry->qp.flags & ~VMCI_QP_ASYMM) != (flags & ~VMCI_QP_ASYMM_PEER)) + return VMCI_ERROR_QUEUEPAIR_MISMATCH; + + if (context_id != VMCI_HOST_CONTEXT_ID) { + /* + * The queue pair broker entry stores values from the guest + * point of view, so an attaching guest should match the values + * stored in the entry. + */ + + if (entry->qp.produce_size != produce_size || + entry->qp.consume_size != consume_size) { + return VMCI_ERROR_QUEUEPAIR_MISMATCH; + } + } else if (entry->qp.produce_size != consume_size || + entry->qp.consume_size != produce_size) { + return VMCI_ERROR_QUEUEPAIR_MISMATCH; + } + + if (context_id != VMCI_HOST_CONTEXT_ID) { + /* + * If a guest attached to a queue pair, it will supply + * the backing memory. If this is a pre NOVMVM vmx, + * the backing memory will be supplied by calling + * vmci_qp_broker_set_page_store() following the + * return of the vmci_qp_broker_alloc() call. If it is + * a vmx of version NOVMVM or later, the page store + * must be supplied as part of the + * vmci_qp_broker_alloc call. Under all circumstances + * must the initially created queue pair not have any + * memory associated with it already. + */ + + if (entry->state != VMCIQPB_CREATED_NO_MEM) + return VMCI_ERROR_INVALID_ARGS; + + if (page_store != NULL) { + /* + * Patch up host state to point to guest + * supplied memory. The VMX already + * initialized the queue pair headers, so no + * need for the kernel side to do that. + */ + + result = qp_host_register_user_memory(page_store, + entry->produce_q, + entry->consume_q); + if (result < VMCI_SUCCESS) + return result; + + entry->state = VMCIQPB_ATTACHED_MEM; + } else { + entry->state = VMCIQPB_ATTACHED_NO_MEM; + } + } else if (entry->state == VMCIQPB_CREATED_NO_MEM) { + /* + * The host side is attempting to attach to a queue + * pair that doesn't have any memory associated with + * it. This must be a pre NOVMVM vmx that hasn't set + * the page store information yet, or a quiesced VM. + */ + + return VMCI_ERROR_UNAVAILABLE; + } else { + /* The host side has successfully attached to a queue pair. */ + entry->state = VMCIQPB_ATTACHED_MEM; + } + + if (entry->state == VMCIQPB_ATTACHED_MEM) { + result = + qp_notify_peer(true, entry->qp.handle, context_id, + entry->create_id); + if (result < VMCI_SUCCESS) + pr_warn("Failed to notify peer (ID=0x%x) of attach to queue pair (handle=0x%x:0x%x)\n", + entry->create_id, entry->qp.handle.context, + entry->qp.handle.resource); + } + + entry->attach_id = context_id; + entry->qp.ref_count++; + if (wakeup_cb) { + entry->wakeup_cb = wakeup_cb; + entry->client_data = client_data; + } + + /* + * When attaching to local queue pairs, the context already has + * an entry tracking the queue pair, so don't add another one. + */ + if (!is_local) + vmci_ctx_qp_create(context, entry->qp.handle); + + if (ent != NULL) + *ent = entry; + + return VMCI_SUCCESS; +} + +/* + * queue_pair_Alloc for use when setting up queue pair endpoints + * on the host. + */ +static int qp_broker_alloc(struct vmci_handle handle, + u32 peer, + u32 flags, + u32 priv_flags, + u64 produce_size, + u64 consume_size, + struct vmci_qp_page_store *page_store, + struct vmci_ctx *context, + vmci_event_release_cb wakeup_cb, + void *client_data, + struct qp_broker_entry **ent, + bool *swap) +{ + const u32 context_id = vmci_ctx_get_id(context); + bool create; + struct qp_broker_entry *entry = NULL; + bool is_local = flags & VMCI_QPFLAG_LOCAL; + int result; + + if (vmci_handle_is_invalid(handle) || + (flags & ~VMCI_QP_ALL_FLAGS) || is_local || + !(produce_size || consume_size) || + !context || context_id == VMCI_INVALID_ID || + handle.context == VMCI_INVALID_ID) { + return VMCI_ERROR_INVALID_ARGS; + } + + if (page_store && !VMCI_QP_PAGESTORE_IS_WELLFORMED(page_store)) + return VMCI_ERROR_INVALID_ARGS; + + /* + * In the initial argument check, we ensure that non-vmkernel hosts + * are not allowed to create local queue pairs. + */ + + mutex_lock(&qp_broker_list.mutex); + + if (!is_local && vmci_ctx_qp_exists(context, handle)) { + pr_devel("Context (ID=0x%x) already attached to queue pair (handle=0x%x:0x%x)\n", + context_id, handle.context, handle.resource); + mutex_unlock(&qp_broker_list.mutex); + return VMCI_ERROR_ALREADY_EXISTS; + } + + if (handle.resource != VMCI_INVALID_ID) + entry = qp_broker_handle_to_entry(handle); + + if (!entry) { + create = true; + result = + qp_broker_create(handle, peer, flags, priv_flags, + produce_size, consume_size, page_store, + context, wakeup_cb, client_data, ent); + } else { + create = false; + result = + qp_broker_attach(entry, peer, flags, priv_flags, + produce_size, consume_size, page_store, + context, wakeup_cb, client_data, ent); + } + + mutex_unlock(&qp_broker_list.mutex); + + if (swap) + *swap = (context_id == VMCI_HOST_CONTEXT_ID) && + !(create && is_local); + + return result; +} + +/* + * This function implements the kernel API for allocating a queue + * pair. + */ +static int qp_alloc_host_work(struct vmci_handle *handle, + struct vmci_queue **produce_q, + u64 produce_size, + struct vmci_queue **consume_q, + u64 consume_size, + u32 peer, + u32 flags, + u32 priv_flags, + vmci_event_release_cb wakeup_cb, + void *client_data) +{ + struct vmci_handle new_handle; + struct vmci_ctx *context; + struct qp_broker_entry *entry; + int result; + bool swap; + + if (vmci_handle_is_invalid(*handle)) { + new_handle = vmci_make_handle( + VMCI_HOST_CONTEXT_ID, VMCI_INVALID_ID); + } else + new_handle = *handle; + + context = vmci_ctx_get(VMCI_HOST_CONTEXT_ID); + entry = NULL; + result = + qp_broker_alloc(new_handle, peer, flags, priv_flags, + produce_size, consume_size, NULL, context, + wakeup_cb, client_data, &entry, &swap); + if (result == VMCI_SUCCESS) { + if (swap) { + /* + * If this is a local queue pair, the attacher + * will swap around produce and consume + * queues. + */ + + *produce_q = entry->consume_q; + *consume_q = entry->produce_q; + } else { + *produce_q = entry->produce_q; + *consume_q = entry->consume_q; + } + + *handle = vmci_resource_handle(&entry->resource); + } else { + *handle = VMCI_INVALID_HANDLE; + pr_devel("queue pair broker failed to alloc (result=%d)\n", + result); + } + vmci_ctx_put(context); + return result; +} + +/* + * Allocates a VMCI queue_pair. Only checks validity of input + * arguments. The real work is done in the host or guest + * specific function. + */ +int vmci_qp_alloc(struct vmci_handle *handle, + struct vmci_queue **produce_q, + u64 produce_size, + struct vmci_queue **consume_q, + u64 consume_size, + u32 peer, + u32 flags, + u32 priv_flags, + bool guest_endpoint, + vmci_event_release_cb wakeup_cb, + void *client_data) +{ + if (!handle || !produce_q || !consume_q || + (!produce_size && !consume_size) || (flags & ~VMCI_QP_ALL_FLAGS)) + return VMCI_ERROR_INVALID_ARGS; + + if (guest_endpoint) { + return qp_alloc_guest_work(handle, produce_q, + produce_size, consume_q, + consume_size, peer, + flags, priv_flags); + } else { + return qp_alloc_host_work(handle, produce_q, + produce_size, consume_q, + consume_size, peer, flags, + priv_flags, wakeup_cb, client_data); + } +} + +/* + * This function implements the host kernel API for detaching from + * a queue pair. + */ +static int qp_detatch_host_work(struct vmci_handle handle) +{ + int result; + struct vmci_ctx *context; + + context = vmci_ctx_get(VMCI_HOST_CONTEXT_ID); + + result = vmci_qp_broker_detach(handle, context); + + vmci_ctx_put(context); + return result; +} + +/* + * Detaches from a VMCI queue_pair. Only checks validity of input argument. + * Real work is done in the host or guest specific function. + */ +static int qp_detatch(struct vmci_handle handle, bool guest_endpoint) +{ + if (vmci_handle_is_invalid(handle)) + return VMCI_ERROR_INVALID_ARGS; + + if (guest_endpoint) + return qp_detatch_guest_work(handle); + else + return qp_detatch_host_work(handle); +} + +/* + * Returns the entry from the head of the list. Assumes that the list is + * locked. + */ +static struct qp_entry *qp_list_get_head(struct qp_list *qp_list) +{ + if (!list_empty(&qp_list->head)) { + struct qp_entry *entry = + list_first_entry(&qp_list->head, struct qp_entry, + list_item); + return entry; + } + + return NULL; +} + +void vmci_qp_broker_exit(void) +{ + struct qp_entry *entry; + struct qp_broker_entry *be; + + mutex_lock(&qp_broker_list.mutex); + + while ((entry = qp_list_get_head(&qp_broker_list))) { + be = (struct qp_broker_entry *)entry; + + qp_list_remove_entry(&qp_broker_list, entry); + kfree(be); + } + + mutex_unlock(&qp_broker_list.mutex); +} + +/* + * Requests that a queue pair be allocated with the VMCI queue + * pair broker. Allocates a queue pair entry if one does not + * exist. Attaches to one if it exists, and retrieves the page + * files backing that queue_pair. Assumes that the queue pair + * broker lock is held. + */ +int vmci_qp_broker_alloc(struct vmci_handle handle, + u32 peer, + u32 flags, + u32 priv_flags, + u64 produce_size, + u64 consume_size, + struct vmci_qp_page_store *page_store, + struct vmci_ctx *context) +{ + return qp_broker_alloc(handle, peer, flags, priv_flags, + produce_size, consume_size, + page_store, context, NULL, NULL, NULL, NULL); +} + +/* + * VMX'en with versions lower than VMCI_VERSION_NOVMVM use a separate + * step to add the UVAs of the VMX mapping of the queue pair. This function + * provides backwards compatibility with such VMX'en, and takes care of + * registering the page store for a queue pair previously allocated by the + * VMX during create or attach. This function will move the queue pair state + * to either from VMCIQBP_CREATED_NO_MEM to VMCIQBP_CREATED_MEM or + * VMCIQBP_ATTACHED_NO_MEM to VMCIQBP_ATTACHED_MEM. If moving to the + * attached state with memory, the queue pair is ready to be used by the + * host peer, and an attached event will be generated. + * + * Assumes that the queue pair broker lock is held. + * + * This function is only used by the hosted platform, since there is no + * issue with backwards compatibility for vmkernel. + */ +int vmci_qp_broker_set_page_store(struct vmci_handle handle, + u64 produce_uva, + u64 consume_uva, + struct vmci_ctx *context) +{ + struct qp_broker_entry *entry; + int result; + const u32 context_id = vmci_ctx_get_id(context); + + if (vmci_handle_is_invalid(handle) || !context || + context_id == VMCI_INVALID_ID) + return VMCI_ERROR_INVALID_ARGS; + + /* + * We only support guest to host queue pairs, so the VMX must + * supply UVAs for the mapped page files. + */ + + if (produce_uva == 0 || consume_uva == 0) + return VMCI_ERROR_INVALID_ARGS; + + mutex_lock(&qp_broker_list.mutex); + + if (!vmci_ctx_qp_exists(context, handle)) { + pr_warn("Context (ID=0x%x) not attached to queue pair (handle=0x%x:0x%x)\n", + context_id, handle.context, handle.resource); + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + entry = qp_broker_handle_to_entry(handle); + if (!entry) { + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + /* + * If I'm the owner then I can set the page store. + * + * Or, if a host created the queue_pair and I'm the attached peer + * then I can set the page store. + */ + if (entry->create_id != context_id && + (entry->create_id != VMCI_HOST_CONTEXT_ID || + entry->attach_id != context_id)) { + result = VMCI_ERROR_QUEUEPAIR_NOTOWNER; + goto out; + } + + if (entry->state != VMCIQPB_CREATED_NO_MEM && + entry->state != VMCIQPB_ATTACHED_NO_MEM) { + result = VMCI_ERROR_UNAVAILABLE; + goto out; + } + + result = qp_host_get_user_memory(produce_uva, consume_uva, + entry->produce_q, entry->consume_q); + if (result < VMCI_SUCCESS) + goto out; + + result = qp_host_map_queues(entry->produce_q, entry->consume_q); + if (result < VMCI_SUCCESS) { + qp_host_unregister_user_memory(entry->produce_q, + entry->consume_q); + goto out; + } + + if (entry->state == VMCIQPB_CREATED_NO_MEM) + entry->state = VMCIQPB_CREATED_MEM; + else + entry->state = VMCIQPB_ATTACHED_MEM; + + entry->vmci_page_files = true; + + if (entry->state == VMCIQPB_ATTACHED_MEM) { + result = + qp_notify_peer(true, handle, context_id, entry->create_id); + if (result < VMCI_SUCCESS) { + pr_warn("Failed to notify peer (ID=0x%x) of attach to queue pair (handle=0x%x:0x%x)\n", + entry->create_id, entry->qp.handle.context, + entry->qp.handle.resource); + } + } + + result = VMCI_SUCCESS; + out: + mutex_unlock(&qp_broker_list.mutex); + return result; +} + +/* + * Resets saved queue headers for the given QP broker + * entry. Should be used when guest memory becomes available + * again, or the guest detaches. + */ +static void qp_reset_saved_headers(struct qp_broker_entry *entry) +{ + entry->produce_q->saved_header = NULL; + entry->consume_q->saved_header = NULL; +} + +/* + * The main entry point for detaching from a queue pair registered with the + * queue pair broker. If more than one endpoint is attached to the queue + * pair, the first endpoint will mainly decrement a reference count and + * generate a notification to its peer. The last endpoint will clean up + * the queue pair state registered with the broker. + * + * When a guest endpoint detaches, it will unmap and unregister the guest + * memory backing the queue pair. If the host is still attached, it will + * no longer be able to access the queue pair content. + * + * If the queue pair is already in a state where there is no memory + * registered for the queue pair (any *_NO_MEM state), it will transition to + * the VMCIQPB_SHUTDOWN_NO_MEM state. This will also happen, if a guest + * endpoint is the first of two endpoints to detach. If the host endpoint is + * the first out of two to detach, the queue pair will move to the + * VMCIQPB_SHUTDOWN_MEM state. + */ +int vmci_qp_broker_detach(struct vmci_handle handle, struct vmci_ctx *context) +{ + struct qp_broker_entry *entry; + const u32 context_id = vmci_ctx_get_id(context); + u32 peer_id; + bool is_local = false; + int result; + + if (vmci_handle_is_invalid(handle) || !context || + context_id == VMCI_INVALID_ID) { + return VMCI_ERROR_INVALID_ARGS; + } + + mutex_lock(&qp_broker_list.mutex); + + if (!vmci_ctx_qp_exists(context, handle)) { + pr_devel("Context (ID=0x%x) not attached to queue pair (handle=0x%x:0x%x)\n", + context_id, handle.context, handle.resource); + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + entry = qp_broker_handle_to_entry(handle); + if (!entry) { + pr_devel("Context (ID=0x%x) reports being attached to queue pair(handle=0x%x:0x%x) that isn't present in broker\n", + context_id, handle.context, handle.resource); + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + if (context_id != entry->create_id && context_id != entry->attach_id) { + result = VMCI_ERROR_QUEUEPAIR_NOTATTACHED; + goto out; + } + + if (context_id == entry->create_id) { + peer_id = entry->attach_id; + entry->create_id = VMCI_INVALID_ID; + } else { + peer_id = entry->create_id; + entry->attach_id = VMCI_INVALID_ID; + } + entry->qp.ref_count--; + + is_local = entry->qp.flags & VMCI_QPFLAG_LOCAL; + + if (context_id != VMCI_HOST_CONTEXT_ID) { + bool headers_mapped; + + /* + * Pre NOVMVM vmx'en may detach from a queue pair + * before setting the page store, and in that case + * there is no user memory to detach from. Also, more + * recent VMX'en may detach from a queue pair in the + * quiesced state. + */ + + qp_acquire_queue_mutex(entry->produce_q); + headers_mapped = entry->produce_q->q_header || + entry->consume_q->q_header; + if (QPBROKERSTATE_HAS_MEM(entry)) { + result = + qp_host_unmap_queues(INVALID_VMCI_GUEST_MEM_ID, + entry->produce_q, + entry->consume_q); + if (result < VMCI_SUCCESS) + pr_warn("Failed to unmap queue headers for queue pair (handle=0x%x:0x%x,result=%d)\n", + handle.context, handle.resource, + result); + + if (entry->vmci_page_files) + qp_host_unregister_user_memory(entry->produce_q, + entry-> + consume_q); + else + qp_host_unregister_user_memory(entry->produce_q, + entry-> + consume_q); + + } + + if (!headers_mapped) + qp_reset_saved_headers(entry); + + qp_release_queue_mutex(entry->produce_q); + + if (!headers_mapped && entry->wakeup_cb) + entry->wakeup_cb(entry->client_data); + + } else { + if (entry->wakeup_cb) { + entry->wakeup_cb = NULL; + entry->client_data = NULL; + } + } + + if (entry->qp.ref_count == 0) { + qp_list_remove_entry(&qp_broker_list, &entry->qp); + + if (is_local) + kfree(entry->local_mem); + + qp_cleanup_queue_mutex(entry->produce_q, entry->consume_q); + qp_host_free_queue(entry->produce_q, entry->qp.produce_size); + qp_host_free_queue(entry->consume_q, entry->qp.consume_size); + /* Unlink from resource hash table and free callback */ + vmci_resource_remove(&entry->resource); + + kfree(entry); + + vmci_ctx_qp_destroy(context, handle); + } else { + qp_notify_peer(false, handle, context_id, peer_id); + if (context_id == VMCI_HOST_CONTEXT_ID && + QPBROKERSTATE_HAS_MEM(entry)) { + entry->state = VMCIQPB_SHUTDOWN_MEM; + } else { + entry->state = VMCIQPB_SHUTDOWN_NO_MEM; + } + + if (!is_local) + vmci_ctx_qp_destroy(context, handle); + + } + result = VMCI_SUCCESS; + out: + mutex_unlock(&qp_broker_list.mutex); + return result; +} + +/* + * Establishes the necessary mappings for a queue pair given a + * reference to the queue pair guest memory. This is usually + * called when a guest is unquiesced and the VMX is allowed to + * map guest memory once again. + */ +int vmci_qp_broker_map(struct vmci_handle handle, + struct vmci_ctx *context, + u64 guest_mem) +{ + struct qp_broker_entry *entry; + const u32 context_id = vmci_ctx_get_id(context); + bool is_local = false; + int result; + + if (vmci_handle_is_invalid(handle) || !context || + context_id == VMCI_INVALID_ID) + return VMCI_ERROR_INVALID_ARGS; + + mutex_lock(&qp_broker_list.mutex); + + if (!vmci_ctx_qp_exists(context, handle)) { + pr_devel("Context (ID=0x%x) not attached to queue pair (handle=0x%x:0x%x)\n", + context_id, handle.context, handle.resource); + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + entry = qp_broker_handle_to_entry(handle); + if (!entry) { + pr_devel("Context (ID=0x%x) reports being attached to queue pair (handle=0x%x:0x%x) that isn't present in broker\n", + context_id, handle.context, handle.resource); + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + if (context_id != entry->create_id && context_id != entry->attach_id) { + result = VMCI_ERROR_QUEUEPAIR_NOTATTACHED; + goto out; + } + + is_local = entry->qp.flags & VMCI_QPFLAG_LOCAL; + result = VMCI_SUCCESS; + + if (context_id != VMCI_HOST_CONTEXT_ID) { + struct vmci_qp_page_store page_store; + + page_store.pages = guest_mem; + page_store.len = QPE_NUM_PAGES(entry->qp); + + qp_acquire_queue_mutex(entry->produce_q); + qp_reset_saved_headers(entry); + result = + qp_host_register_user_memory(&page_store, + entry->produce_q, + entry->consume_q); + qp_release_queue_mutex(entry->produce_q); + if (result == VMCI_SUCCESS) { + /* Move state from *_NO_MEM to *_MEM */ + + entry->state++; + + if (entry->wakeup_cb) + entry->wakeup_cb(entry->client_data); + } + } + + out: + mutex_unlock(&qp_broker_list.mutex); + return result; +} + +/* + * Saves a snapshot of the queue headers for the given QP broker + * entry. Should be used when guest memory is unmapped. + * Results: + * VMCI_SUCCESS on success, appropriate error code if guest memory + * can't be accessed.. + */ +static int qp_save_headers(struct qp_broker_entry *entry) +{ + int result; + + if (entry->produce_q->saved_header != NULL && + entry->consume_q->saved_header != NULL) { + /* + * If the headers have already been saved, we don't need to do + * it again, and we don't want to map in the headers + * unnecessarily. + */ + + return VMCI_SUCCESS; + } + + if (NULL == entry->produce_q->q_header || + NULL == entry->consume_q->q_header) { + result = qp_host_map_queues(entry->produce_q, entry->consume_q); + if (result < VMCI_SUCCESS) + return result; + } + + memcpy(&entry->saved_produce_q, entry->produce_q->q_header, + sizeof(entry->saved_produce_q)); + entry->produce_q->saved_header = &entry->saved_produce_q; + memcpy(&entry->saved_consume_q, entry->consume_q->q_header, + sizeof(entry->saved_consume_q)); + entry->consume_q->saved_header = &entry->saved_consume_q; + + return VMCI_SUCCESS; +} + +/* + * Removes all references to the guest memory of a given queue pair, and + * will move the queue pair from state *_MEM to *_NO_MEM. It is usually + * called when a VM is being quiesced where access to guest memory should + * avoided. + */ +int vmci_qp_broker_unmap(struct vmci_handle handle, + struct vmci_ctx *context, + u32 gid) +{ + struct qp_broker_entry *entry; + const u32 context_id = vmci_ctx_get_id(context); + bool is_local = false; + int result; + + if (vmci_handle_is_invalid(handle) || !context || + context_id == VMCI_INVALID_ID) + return VMCI_ERROR_INVALID_ARGS; + + mutex_lock(&qp_broker_list.mutex); + + if (!vmci_ctx_qp_exists(context, handle)) { + pr_devel("Context (ID=0x%x) not attached to queue pair (handle=0x%x:0x%x)\n", + context_id, handle.context, handle.resource); + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + entry = qp_broker_handle_to_entry(handle); + if (!entry) { + pr_devel("Context (ID=0x%x) reports being attached to queue pair (handle=0x%x:0x%x) that isn't present in broker\n", + context_id, handle.context, handle.resource); + result = VMCI_ERROR_NOT_FOUND; + goto out; + } + + if (context_id != entry->create_id && context_id != entry->attach_id) { + result = VMCI_ERROR_QUEUEPAIR_NOTATTACHED; + goto out; + } + + is_local = entry->qp.flags & VMCI_QPFLAG_LOCAL; + + if (context_id != VMCI_HOST_CONTEXT_ID) { + qp_acquire_queue_mutex(entry->produce_q); + result = qp_save_headers(entry); + if (result < VMCI_SUCCESS) + pr_warn("Failed to save queue headers for queue pair (handle=0x%x:0x%x,result=%d)\n", + handle.context, handle.resource, result); + + qp_host_unmap_queues(gid, entry->produce_q, entry->consume_q); + + /* + * On hosted, when we unmap queue pairs, the VMX will also + * unmap the guest memory, so we invalidate the previously + * registered memory. If the queue pair is mapped again at a + * later point in time, we will need to reregister the user + * memory with a possibly new user VA. + */ + qp_host_unregister_user_memory(entry->produce_q, + entry->consume_q); + + /* + * Move state from *_MEM to *_NO_MEM. + */ + entry->state--; + + qp_release_queue_mutex(entry->produce_q); + } + + result = VMCI_SUCCESS; + + out: + mutex_unlock(&qp_broker_list.mutex); + return result; +} + +/* + * Destroys all guest queue pair endpoints. If active guest queue + * pairs still exist, hypercalls to attempt detach from these + * queue pairs will be made. Any failure to detach is silently + * ignored. + */ +void vmci_qp_guest_endpoints_exit(void) +{ + struct qp_entry *entry; + struct qp_guest_endpoint *ep; + + mutex_lock(&qp_guest_endpoints.mutex); + + while ((entry = qp_list_get_head(&qp_guest_endpoints))) { + ep = (struct qp_guest_endpoint *)entry; + + /* Don't make a hypercall for local queue_pairs. */ + if (!(entry->flags & VMCI_QPFLAG_LOCAL)) + qp_detatch_hypercall(entry->handle); + + /* We cannot fail the exit, so let's reset ref_count. */ + entry->ref_count = 0; + qp_list_remove_entry(&qp_guest_endpoints, entry); + + qp_guest_endpoint_destroy(ep); + } + + mutex_unlock(&qp_guest_endpoints.mutex); +} + +/* + * Helper routine that will lock the queue pair before subsequent + * operations. + * Note: Non-blocking on the host side is currently only implemented in ESX. + * Since non-blocking isn't yet implemented on the host personality we + * have no reason to acquire a spin lock. So to avoid the use of an + * unnecessary lock only acquire the mutex if we can block. + */ +static void qp_lock(const struct vmci_qp *qpair) +{ + qp_acquire_queue_mutex(qpair->produce_q); +} + +/* + * Helper routine that unlocks the queue pair after calling + * qp_lock. + */ +static void qp_unlock(const struct vmci_qp *qpair) +{ + qp_release_queue_mutex(qpair->produce_q); +} + +/* + * The queue headers may not be mapped at all times. If a queue is + * currently not mapped, it will be attempted to do so. + */ +static int qp_map_queue_headers(struct vmci_queue *produce_q, + struct vmci_queue *consume_q) +{ + int result; + + if (NULL == produce_q->q_header || NULL == consume_q->q_header) { + result = qp_host_map_queues(produce_q, consume_q); + if (result < VMCI_SUCCESS) + return (produce_q->saved_header && + consume_q->saved_header) ? + VMCI_ERROR_QUEUEPAIR_NOT_READY : + VMCI_ERROR_QUEUEPAIR_NOTATTACHED; + } + + return VMCI_SUCCESS; +} + +/* + * Helper routine that will retrieve the produce and consume + * headers of a given queue pair. If the guest memory of the + * queue pair is currently not available, the saved queue headers + * will be returned, if these are available. + */ +static int qp_get_queue_headers(const struct vmci_qp *qpair, + struct vmci_queue_header **produce_q_header, + struct vmci_queue_header **consume_q_header) +{ + int result; + + result = qp_map_queue_headers(qpair->produce_q, qpair->consume_q); + if (result == VMCI_SUCCESS) { + *produce_q_header = qpair->produce_q->q_header; + *consume_q_header = qpair->consume_q->q_header; + } else if (qpair->produce_q->saved_header && + qpair->consume_q->saved_header) { + *produce_q_header = qpair->produce_q->saved_header; + *consume_q_header = qpair->consume_q->saved_header; + result = VMCI_SUCCESS; + } + + return result; +} + +/* + * Callback from VMCI queue pair broker indicating that a queue + * pair that was previously not ready, now either is ready or + * gone forever. + */ +static int qp_wakeup_cb(void *client_data) +{ + struct vmci_qp *qpair = (struct vmci_qp *)client_data; + + qp_lock(qpair); + while (qpair->blocked > 0) { + qpair->blocked--; + qpair->generation++; + wake_up(&qpair->event); + } + qp_unlock(qpair); + + return VMCI_SUCCESS; +} + +/* + * Makes the calling thread wait for the queue pair to become + * ready for host side access. Returns true when thread is + * woken up after queue pair state change, false otherwise. + */ +static bool qp_wait_for_ready_queue(struct vmci_qp *qpair) +{ + unsigned int generation; + + qpair->blocked++; + generation = qpair->generation; + qp_unlock(qpair); + wait_event(qpair->event, generation != qpair->generation); + qp_lock(qpair); + + return true; +} + +/* + * Enqueues a given buffer to the produce queue using the provided + * function. As many bytes as possible (space available in the queue) + * are enqueued. Assumes the queue->mutex has been acquired. Returns + * VMCI_ERROR_QUEUEPAIR_NOSPACE if no space was available to enqueue + * data, VMCI_ERROR_INVALID_SIZE, if any queue pointer is outside the + * queue (as defined by the queue size), VMCI_ERROR_INVALID_ARGS, if + * an error occured when accessing the buffer, + * VMCI_ERROR_QUEUEPAIR_NOTATTACHED, if the queue pair pages aren't + * available. Otherwise, the number of bytes written to the queue is + * returned. Updates the tail pointer of the produce queue. + */ +static ssize_t qp_enqueue_locked(struct vmci_queue *produce_q, + struct vmci_queue *consume_q, + const u64 produce_q_size, + const void *buf, + size_t buf_size, + vmci_memcpy_to_queue_func memcpy_to_queue) +{ + s64 free_space; + u64 tail; + size_t written; + ssize_t result; + + result = qp_map_queue_headers(produce_q, consume_q); + if (unlikely(result != VMCI_SUCCESS)) + return result; + + free_space = vmci_q_header_free_space(produce_q->q_header, + consume_q->q_header, + produce_q_size); + if (free_space == 0) + return VMCI_ERROR_QUEUEPAIR_NOSPACE; + + if (free_space < VMCI_SUCCESS) + return (ssize_t) free_space; + + written = (size_t) (free_space > buf_size ? buf_size : free_space); + tail = vmci_q_header_producer_tail(produce_q->q_header); + if (likely(tail + written < produce_q_size)) { + result = memcpy_to_queue(produce_q, tail, buf, 0, written); + } else { + /* Tail pointer wraps around. */ + + const size_t tmp = (size_t) (produce_q_size - tail); + + result = memcpy_to_queue(produce_q, tail, buf, 0, tmp); + if (result >= VMCI_SUCCESS) + result = memcpy_to_queue(produce_q, 0, buf, tmp, + written - tmp); + } + + if (result < VMCI_SUCCESS) + return result; + + vmci_q_header_add_producer_tail(produce_q->q_header, written, + produce_q_size); + return written; +} + +/* + * Dequeues data (if available) from the given consume queue. Writes data + * to the user provided buffer using the provided function. + * Assumes the queue->mutex has been acquired. + * Results: + * VMCI_ERROR_QUEUEPAIR_NODATA if no data was available to dequeue. + * VMCI_ERROR_INVALID_SIZE, if any queue pointer is outside the queue + * (as defined by the queue size). + * VMCI_ERROR_INVALID_ARGS, if an error occured when accessing the buffer. + * Otherwise the number of bytes dequeued is returned. + * Side effects: + * Updates the head pointer of the consume queue. + */ +static ssize_t qp_dequeue_locked(struct vmci_queue *produce_q, + struct vmci_queue *consume_q, + const u64 consume_q_size, + void *buf, + size_t buf_size, + vmci_memcpy_from_queue_func memcpy_from_queue, + bool update_consumer) +{ + s64 buf_ready; + u64 head; + size_t read; + ssize_t result; + + result = qp_map_queue_headers(produce_q, consume_q); + if (unlikely(result != VMCI_SUCCESS)) + return result; + + buf_ready = vmci_q_header_buf_ready(consume_q->q_header, + produce_q->q_header, + consume_q_size); + if (buf_ready == 0) + return VMCI_ERROR_QUEUEPAIR_NODATA; + + if (buf_ready < VMCI_SUCCESS) + return (ssize_t) buf_ready; + + read = (size_t) (buf_ready > buf_size ? buf_size : buf_ready); + head = vmci_q_header_consumer_head(produce_q->q_header); + if (likely(head + read < consume_q_size)) { + result = memcpy_from_queue(buf, 0, consume_q, head, read); + } else { + /* Head pointer wraps around. */ + + const size_t tmp = (size_t) (consume_q_size - head); + + result = memcpy_from_queue(buf, 0, consume_q, head, tmp); + if (result >= VMCI_SUCCESS) + result = memcpy_from_queue(buf, tmp, consume_q, 0, + read - tmp); + + } + + if (result < VMCI_SUCCESS) + return result; + + if (update_consumer) + vmci_q_header_add_consumer_head(produce_q->q_header, + read, consume_q_size); + + return read; +} + +/* + * vmci_qpair_alloc() - Allocates a queue pair. + * @qpair: Pointer for the new vmci_qp struct. + * @handle: Handle to track the resource. + * @produce_qsize: Desired size of the producer queue. + * @consume_qsize: Desired size of the consumer queue. + * @peer: ContextID of the peer. + * @flags: VMCI flags. + * @priv_flags: VMCI priviledge flags. + * + * This is the client interface for allocating the memory for a + * vmci_qp structure and then attaching to the underlying + * queue. If an error occurs allocating the memory for the + * vmci_qp structure no attempt is made to attach. If an + * error occurs attaching, then the structure is freed. + */ +int vmci_qpair_alloc(struct vmci_qp **qpair, + struct vmci_handle *handle, + u64 produce_qsize, + u64 consume_qsize, + u32 peer, + u32 flags, + u32 priv_flags) +{ + struct vmci_qp *my_qpair; + int retval; + struct vmci_handle src = VMCI_INVALID_HANDLE; + struct vmci_handle dst = vmci_make_handle(peer, VMCI_INVALID_ID); + enum vmci_route route; + vmci_event_release_cb wakeup_cb; + void *client_data; + + /* + * Restrict the size of a queuepair. The device already + * enforces a limit on the total amount of memory that can be + * allocated to queuepairs for a guest. However, we try to + * allocate this memory before we make the queuepair + * allocation hypercall. On Linux, we allocate each page + * separately, which means rather than fail, the guest will + * thrash while it tries to allocate, and will become + * increasingly unresponsive to the point where it appears to + * be hung. So we place a limit on the size of an individual + * queuepair here, and leave the device to enforce the + * restriction on total queuepair memory. (Note that this + * doesn't prevent all cases; a user with only this much + * physical memory could still get into trouble.) The error + * used by the device is NO_RESOURCES, so use that here too. + */ + + if (produce_qsize + consume_qsize < max(produce_qsize, consume_qsize) || + produce_qsize + consume_qsize > VMCI_MAX_GUEST_QP_MEMORY) + return VMCI_ERROR_NO_RESOURCES; + + retval = vmci_route(&src, &dst, false, &route); + if (retval < VMCI_SUCCESS) + route = vmci_guest_code_active() ? + VMCI_ROUTE_AS_GUEST : VMCI_ROUTE_AS_HOST; + + if (flags & (VMCI_QPFLAG_NONBLOCK | VMCI_QPFLAG_PINNED)) { + pr_devel("NONBLOCK OR PINNED set"); + return VMCI_ERROR_INVALID_ARGS; + } + + my_qpair = kzalloc(sizeof(*my_qpair), GFP_KERNEL); + if (!my_qpair) + return VMCI_ERROR_NO_MEM; + + my_qpair->produce_q_size = produce_qsize; + my_qpair->consume_q_size = consume_qsize; + my_qpair->peer = peer; + my_qpair->flags = flags; + my_qpair->priv_flags = priv_flags; + + wakeup_cb = NULL; + client_data = NULL; + + if (VMCI_ROUTE_AS_HOST == route) { + my_qpair->guest_endpoint = false; + if (!(flags & VMCI_QPFLAG_LOCAL)) { + my_qpair->blocked = 0; + my_qpair->generation = 0; + init_waitqueue_head(&my_qpair->event); + wakeup_cb = qp_wakeup_cb; + client_data = (void *)my_qpair; + } + } else { + my_qpair->guest_endpoint = true; + } + + retval = vmci_qp_alloc(handle, + &my_qpair->produce_q, + my_qpair->produce_q_size, + &my_qpair->consume_q, + my_qpair->consume_q_size, + my_qpair->peer, + my_qpair->flags, + my_qpair->priv_flags, + my_qpair->guest_endpoint, + wakeup_cb, client_data); + + if (retval < VMCI_SUCCESS) { + kfree(my_qpair); + return retval; + } + + *qpair = my_qpair; + my_qpair->handle = *handle; + + return retval; +} +EXPORT_SYMBOL_GPL(vmci_qpair_alloc); + +/* + * vmci_qpair_detach() - Detatches the client from a queue pair. + * @qpair: Reference of a pointer to the qpair struct. + * + * This is the client interface for detaching from a VMCIQPair. + * Note that this routine will free the memory allocated for the + * vmci_qp structure too. + */ +int vmci_qpair_detach(struct vmci_qp **qpair) +{ + int result; + struct vmci_qp *old_qpair; + + if (!qpair || !(*qpair)) + return VMCI_ERROR_INVALID_ARGS; + + old_qpair = *qpair; + result = qp_detatch(old_qpair->handle, old_qpair->guest_endpoint); + + /* + * The guest can fail to detach for a number of reasons, and + * if it does so, it will cleanup the entry (if there is one). + * The host can fail too, but it won't cleanup the entry + * immediately, it will do that later when the context is + * freed. Either way, we need to release the qpair struct + * here; there isn't much the caller can do, and we don't want + * to leak. + */ + + memset(old_qpair, 0, sizeof(*old_qpair)); + old_qpair->handle = VMCI_INVALID_HANDLE; + old_qpair->peer = VMCI_INVALID_ID; + kfree(old_qpair); + *qpair = NULL; + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_detach); + +/* + * vmci_qpair_get_produce_indexes() - Retrieves the indexes of the producer. + * @qpair: Pointer to the queue pair struct. + * @producer_tail: Reference used for storing producer tail index. + * @consumer_head: Reference used for storing the consumer head index. + * + * This is the client interface for getting the current indexes of the + * QPair from the point of the view of the caller as the producer. + */ +int vmci_qpair_get_produce_indexes(const struct vmci_qp *qpair, + u64 *producer_tail, + u64 *consumer_head) +{ + struct vmci_queue_header *produce_q_header; + struct vmci_queue_header *consume_q_header; + int result; + + if (!qpair) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + result = + qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header); + if (result == VMCI_SUCCESS) + vmci_q_header_get_pointers(produce_q_header, consume_q_header, + producer_tail, consumer_head); + qp_unlock(qpair); + + if (result == VMCI_SUCCESS && + ((producer_tail && *producer_tail >= qpair->produce_q_size) || + (consumer_head && *consumer_head >= qpair->produce_q_size))) + return VMCI_ERROR_INVALID_SIZE; + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_get_produce_indexes); + +/* + * vmci_qpair_get_consume_indexes() - Retrieves the indexes of the comsumer. + * @qpair: Pointer to the queue pair struct. + * @consumer_tail: Reference used for storing consumer tail index. + * @producer_head: Reference used for storing the producer head index. + * + * This is the client interface for getting the current indexes of the + * QPair from the point of the view of the caller as the consumer. + */ +int vmci_qpair_get_consume_indexes(const struct vmci_qp *qpair, + u64 *consumer_tail, + u64 *producer_head) +{ + struct vmci_queue_header *produce_q_header; + struct vmci_queue_header *consume_q_header; + int result; + + if (!qpair) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + result = + qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header); + if (result == VMCI_SUCCESS) + vmci_q_header_get_pointers(consume_q_header, produce_q_header, + consumer_tail, producer_head); + qp_unlock(qpair); + + if (result == VMCI_SUCCESS && + ((consumer_tail && *consumer_tail >= qpair->consume_q_size) || + (producer_head && *producer_head >= qpair->consume_q_size))) + return VMCI_ERROR_INVALID_SIZE; + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_get_consume_indexes); + +/* + * vmci_qpair_produce_free_space() - Retrieves free space in producer queue. + * @qpair: Pointer to the queue pair struct. + * + * This is the client interface for getting the amount of free + * space in the QPair from the point of the view of the caller as + * the producer which is the common case. Returns < 0 if err, else + * available bytes into which data can be enqueued if > 0. + */ +s64 vmci_qpair_produce_free_space(const struct vmci_qp *qpair) +{ + struct vmci_queue_header *produce_q_header; + struct vmci_queue_header *consume_q_header; + s64 result; + + if (!qpair) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + result = + qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header); + if (result == VMCI_SUCCESS) + result = vmci_q_header_free_space(produce_q_header, + consume_q_header, + qpair->produce_q_size); + else + result = 0; + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_produce_free_space); + +/* + * vmci_qpair_consume_free_space() - Retrieves free space in consumer queue. + * @qpair: Pointer to the queue pair struct. + * + * This is the client interface for getting the amount of free + * space in the QPair from the point of the view of the caller as + * the consumer which is not the common case. Returns < 0 if err, else + * available bytes into which data can be enqueued if > 0. + */ +s64 vmci_qpair_consume_free_space(const struct vmci_qp *qpair) +{ + struct vmci_queue_header *produce_q_header; + struct vmci_queue_header *consume_q_header; + s64 result; + + if (!qpair) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + result = + qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header); + if (result == VMCI_SUCCESS) + result = vmci_q_header_free_space(consume_q_header, + produce_q_header, + qpair->consume_q_size); + else + result = 0; + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_consume_free_space); + +/* + * vmci_qpair_produce_buf_ready() - Gets bytes ready to read from + * producer queue. + * @qpair: Pointer to the queue pair struct. + * + * This is the client interface for getting the amount of + * enqueued data in the QPair from the point of the view of the + * caller as the producer which is not the common case. Returns < 0 if err, + * else available bytes that may be read. + */ +s64 vmci_qpair_produce_buf_ready(const struct vmci_qp *qpair) +{ + struct vmci_queue_header *produce_q_header; + struct vmci_queue_header *consume_q_header; + s64 result; + + if (!qpair) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + result = + qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header); + if (result == VMCI_SUCCESS) + result = vmci_q_header_buf_ready(produce_q_header, + consume_q_header, + qpair->produce_q_size); + else + result = 0; + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_produce_buf_ready); + +/* + * vmci_qpair_consume_buf_ready() - Gets bytes ready to read from + * consumer queue. + * @qpair: Pointer to the queue pair struct. + * + * This is the client interface for getting the amount of + * enqueued data in the QPair from the point of the view of the + * caller as the consumer which is the normal case. Returns < 0 if err, + * else available bytes that may be read. + */ +s64 vmci_qpair_consume_buf_ready(const struct vmci_qp *qpair) +{ + struct vmci_queue_header *produce_q_header; + struct vmci_queue_header *consume_q_header; + s64 result; + + if (!qpair) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + result = + qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header); + if (result == VMCI_SUCCESS) + result = vmci_q_header_buf_ready(consume_q_header, + produce_q_header, + qpair->consume_q_size); + else + result = 0; + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_consume_buf_ready); + +/* + * vmci_qpair_enqueue() - Throw data on the queue. + * @qpair: Pointer to the queue pair struct. + * @buf: Pointer to buffer containing data + * @buf_size: Length of buffer. + * @buf_type: Buffer type (Unused). + * + * This is the client interface for enqueueing data into the queue. + * Returns number of bytes enqueued or < 0 on error. + */ +ssize_t vmci_qpair_enqueue(struct vmci_qp *qpair, + const void *buf, + size_t buf_size, + int buf_type) +{ + ssize_t result; + + if (!qpair || !buf) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + + do { + result = qp_enqueue_locked(qpair->produce_q, + qpair->consume_q, + qpair->produce_q_size, + buf, buf_size, + qp_memcpy_to_queue); + + if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY && + !qp_wait_for_ready_queue(qpair)) + result = VMCI_ERROR_WOULD_BLOCK; + + } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY); + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_enqueue); + +/* + * vmci_qpair_dequeue() - Get data from the queue. + * @qpair: Pointer to the queue pair struct. + * @buf: Pointer to buffer for the data + * @buf_size: Length of buffer. + * @buf_type: Buffer type (Unused). + * + * This is the client interface for dequeueing data from the queue. + * Returns number of bytes dequeued or < 0 on error. + */ +ssize_t vmci_qpair_dequeue(struct vmci_qp *qpair, + void *buf, + size_t buf_size, + int buf_type) +{ + ssize_t result; + + if (!qpair || !buf) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + + do { + result = qp_dequeue_locked(qpair->produce_q, + qpair->consume_q, + qpair->consume_q_size, + buf, buf_size, + qp_memcpy_from_queue, true); + + if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY && + !qp_wait_for_ready_queue(qpair)) + result = VMCI_ERROR_WOULD_BLOCK; + + } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY); + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_dequeue); + +/* + * vmci_qpair_peek() - Peek at the data in the queue. + * @qpair: Pointer to the queue pair struct. + * @buf: Pointer to buffer for the data + * @buf_size: Length of buffer. + * @buf_type: Buffer type (Unused on Linux). + * + * This is the client interface for peeking into a queue. (I.e., + * copy data from the queue without updating the head pointer.) + * Returns number of bytes dequeued or < 0 on error. + */ +ssize_t vmci_qpair_peek(struct vmci_qp *qpair, + void *buf, + size_t buf_size, + int buf_type) +{ + ssize_t result; + + if (!qpair || !buf) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + + do { + result = qp_dequeue_locked(qpair->produce_q, + qpair->consume_q, + qpair->consume_q_size, + buf, buf_size, + qp_memcpy_from_queue, false); + + if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY && + !qp_wait_for_ready_queue(qpair)) + result = VMCI_ERROR_WOULD_BLOCK; + + } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY); + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_peek); + +/* + * vmci_qpair_enquev() - Throw data on the queue using iov. + * @qpair: Pointer to the queue pair struct. + * @iov: Pointer to buffer containing data + * @iov_size: Length of buffer. + * @buf_type: Buffer type (Unused). + * + * This is the client interface for enqueueing data into the queue. + * This function uses IO vectors to handle the work. Returns number + * of bytes enqueued or < 0 on error. + */ +ssize_t vmci_qpair_enquev(struct vmci_qp *qpair, + void *iov, + size_t iov_size, + int buf_type) +{ + ssize_t result; + + if (!qpair || !iov) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + + do { + result = qp_enqueue_locked(qpair->produce_q, + qpair->consume_q, + qpair->produce_q_size, + iov, iov_size, + qp_memcpy_to_queue_iov); + + if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY && + !qp_wait_for_ready_queue(qpair)) + result = VMCI_ERROR_WOULD_BLOCK; + + } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY); + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_enquev); + +/* + * vmci_qpair_dequev() - Get data from the queue using iov. + * @qpair: Pointer to the queue pair struct. + * @iov: Pointer to buffer for the data + * @iov_size: Length of buffer. + * @buf_type: Buffer type (Unused). + * + * This is the client interface for dequeueing data from the queue. + * This function uses IO vectors to handle the work. Returns number + * of bytes dequeued or < 0 on error. + */ +ssize_t vmci_qpair_dequev(struct vmci_qp *qpair, + void *iov, + size_t iov_size, + int buf_type) +{ + ssize_t result; + + if (!qpair || !iov) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + + do { + result = qp_dequeue_locked(qpair->produce_q, + qpair->consume_q, + qpair->consume_q_size, + iov, iov_size, + qp_memcpy_from_queue_iov, + true); + + if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY && + !qp_wait_for_ready_queue(qpair)) + result = VMCI_ERROR_WOULD_BLOCK; + + } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY); + + qp_unlock(qpair); + + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_dequev); + +/* + * vmci_qpair_peekv() - Peek at the data in the queue using iov. + * @qpair: Pointer to the queue pair struct. + * @iov: Pointer to buffer for the data + * @iov_size: Length of buffer. + * @buf_type: Buffer type (Unused on Linux). + * + * This is the client interface for peeking into a queue. (I.e., + * copy data from the queue without updating the head pointer.) + * This function uses IO vectors to handle the work. Returns number + * of bytes peeked or < 0 on error. + */ +ssize_t vmci_qpair_peekv(struct vmci_qp *qpair, + void *iov, + size_t iov_size, + int buf_type) +{ + ssize_t result; + + if (!qpair || !iov) + return VMCI_ERROR_INVALID_ARGS; + + qp_lock(qpair); + + do { + result = qp_dequeue_locked(qpair->produce_q, + qpair->consume_q, + qpair->consume_q_size, + iov, iov_size, + qp_memcpy_from_queue_iov, + false); + + if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY && + !qp_wait_for_ready_queue(qpair)) + result = VMCI_ERROR_WOULD_BLOCK; + + } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY); + + qp_unlock(qpair); + return result; +} +EXPORT_SYMBOL_GPL(vmci_qpair_peekv); diff --git a/drivers/misc/vmw_vmci/vmci_queue_pair.h b/drivers/misc/vmw_vmci/vmci_queue_pair.h new file mode 100644 index 00000000000..ed177f04ef2 --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_queue_pair.h @@ -0,0 +1,173 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef _VMCI_QUEUE_PAIR_H_ +#define _VMCI_QUEUE_PAIR_H_ + +#include <linux/vmw_vmci_defs.h> +#include <linux/types.h> + +#include "vmci_context.h" + +/* Callback needed for correctly waiting on events. */ +typedef int (*vmci_event_release_cb) (void *client_data); + +/* Guest device port I/O. */ +struct ppn_set { + u64 num_produce_pages; + u64 num_consume_pages; + u32 *produce_ppns; + u32 *consume_ppns; + bool initialized; +}; + +/* VMCIqueue_pairAllocInfo */ +struct vmci_qp_alloc_info { + struct vmci_handle handle; + u32 peer; + u32 flags; + u64 produce_size; + u64 consume_size; + u64 ppn_va; /* Start VA of queue pair PPNs. */ + u64 num_ppns; + s32 result; + u32 version; +}; + +/* VMCIqueue_pairSetVAInfo */ +struct vmci_qp_set_va_info { + struct vmci_handle handle; + u64 va; /* Start VA of queue pair PPNs. */ + u64 num_ppns; + u32 version; + s32 result; +}; + +/* + * For backwards compatibility, here is a version of the + * VMCIqueue_pairPageFileInfo before host support end-points was added. + * Note that the current version of that structure requires VMX to + * pass down the VA of the mapped file. Before host support was added + * there was nothing of the sort. So, when the driver sees the ioctl + * with a parameter that is the sizeof + * VMCIqueue_pairPageFileInfo_NoHostQP then it can infer that the version + * of VMX running can't attach to host end points because it doesn't + * provide the VA of the mapped files. + * + * The Linux driver doesn't get an indication of the size of the + * structure passed down from user space. So, to fix a long standing + * but unfiled bug, the _pad field has been renamed to version. + * Existing versions of VMX always initialize the PageFileInfo + * structure so that _pad, er, version is set to 0. + * + * A version value of 1 indicates that the size of the structure has + * been increased to include two UVA's: produce_uva and consume_uva. + * These UVA's are of the mmap()'d queue contents backing files. + * + * In addition, if when VMX is sending down the + * VMCIqueue_pairPageFileInfo structure it gets an error then it will + * try again with the _NoHostQP version of the file to see if an older + * VMCI kernel module is running. + */ + +/* VMCIqueue_pairPageFileInfo */ +struct vmci_qp_page_file_info { + struct vmci_handle handle; + u64 produce_page_file; /* User VA. */ + u64 consume_page_file; /* User VA. */ + u64 produce_page_file_size; /* Size of the file name array. */ + u64 consume_page_file_size; /* Size of the file name array. */ + s32 result; + u32 version; /* Was _pad. */ + u64 produce_va; /* User VA of the mapped file. */ + u64 consume_va; /* User VA of the mapped file. */ +}; + +/* vmci queuepair detach info */ +struct vmci_qp_dtch_info { + struct vmci_handle handle; + s32 result; + u32 _pad; +}; + +/* + * struct vmci_qp_page_store describes how the memory of a given queue pair + * is backed. When the queue pair is between the host and a guest, the + * page store consists of references to the guest pages. On vmkernel, + * this is a list of PPNs, and on hosted, it is a user VA where the + * queue pair is mapped into the VMX address space. + */ +struct vmci_qp_page_store { + /* Reference to pages backing the queue pair. */ + u64 pages; + /* Length of pageList/virtual addres range (in pages). */ + u32 len; +}; + +/* + * This data type contains the information about a queue. + * There are two queues (hence, queue pairs) per transaction model between a + * pair of end points, A & B. One queue is used by end point A to transmit + * commands and responses to B. The other queue is used by B to transmit + * commands and responses. + * + * struct vmci_queue_kern_if is a per-OS defined Queue structure. It contains + * either a direct pointer to the linear address of the buffer contents or a + * pointer to structures which help the OS locate those data pages. See + * vmciKernelIf.c for each platform for its definition. + */ +struct vmci_queue { + struct vmci_queue_header *q_header; + struct vmci_queue_header *saved_header; + struct vmci_queue_kern_if *kernel_if; +}; + +/* + * Utility function that checks whether the fields of the page + * store contain valid values. + * Result: + * true if the page store is wellformed. false otherwise. + */ +static inline bool +VMCI_QP_PAGESTORE_IS_WELLFORMED(struct vmci_qp_page_store *page_store) +{ + return page_store->len >= 2; +} + +void vmci_qp_broker_exit(void); +int vmci_qp_broker_alloc(struct vmci_handle handle, u32 peer, + u32 flags, u32 priv_flags, + u64 produce_size, u64 consume_size, + struct vmci_qp_page_store *page_store, + struct vmci_ctx *context); +int vmci_qp_broker_set_page_store(struct vmci_handle handle, + u64 produce_uva, u64 consume_uva, + struct vmci_ctx *context); +int vmci_qp_broker_detach(struct vmci_handle handle, struct vmci_ctx *context); + +void vmci_qp_guest_endpoints_exit(void); + +int vmci_qp_alloc(struct vmci_handle *handle, + struct vmci_queue **produce_q, u64 produce_size, + struct vmci_queue **consume_q, u64 consume_size, + u32 peer, u32 flags, u32 priv_flags, + bool guest_endpoint, vmci_event_release_cb wakeup_cb, + void *client_data); +int vmci_qp_broker_map(struct vmci_handle handle, + struct vmci_ctx *context, u64 guest_mem); +int vmci_qp_broker_unmap(struct vmci_handle handle, + struct vmci_ctx *context, u32 gid); + +#endif /* _VMCI_QUEUE_PAIR_H_ */ diff --git a/drivers/misc/vmw_vmci/vmci_resource.c b/drivers/misc/vmw_vmci/vmci_resource.c new file mode 100644 index 00000000000..9a53a30de44 --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_resource.c @@ -0,0 +1,227 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/hash.h> +#include <linux/types.h> +#include <linux/rculist.h> + +#include "vmci_resource.h" +#include "vmci_driver.h" + + +#define VMCI_RESOURCE_HASH_BITS 7 +#define VMCI_RESOURCE_HASH_BUCKETS (1 << VMCI_RESOURCE_HASH_BITS) + +struct vmci_hash_table { + spinlock_t lock; + struct hlist_head entries[VMCI_RESOURCE_HASH_BUCKETS]; +}; + +static struct vmci_hash_table vmci_resource_table = { + .lock = __SPIN_LOCK_UNLOCKED(vmci_resource_table.lock), +}; + +static unsigned int vmci_resource_hash(struct vmci_handle handle) +{ + return hash_32(handle.resource, VMCI_RESOURCE_HASH_BITS); +} + +/* + * Gets a resource (if one exists) matching given handle from the hash table. + */ +static struct vmci_resource *vmci_resource_lookup(struct vmci_handle handle, + enum vmci_resource_type type) +{ + struct vmci_resource *r, *resource = NULL; + unsigned int idx = vmci_resource_hash(handle); + + rcu_read_lock(); + hlist_for_each_entry_rcu(r, + &vmci_resource_table.entries[idx], node) { + u32 cid = r->handle.context; + u32 rid = r->handle.resource; + + if (r->type == type && + rid == handle.resource && + (cid == handle.context || cid == VMCI_INVALID_ID)) { + resource = r; + break; + } + } + rcu_read_unlock(); + + return resource; +} + +/* + * Find an unused resource ID and return it. The first + * VMCI_RESERVED_RESOURCE_ID_MAX are reserved so we start from + * its value + 1. + * Returns VMCI resource id on success, VMCI_INVALID_ID on failure. + */ +static u32 vmci_resource_find_id(u32 context_id, + enum vmci_resource_type resource_type) +{ + static u32 resource_id = VMCI_RESERVED_RESOURCE_ID_MAX + 1; + u32 old_rid = resource_id; + u32 current_rid; + + /* + * Generate a unique resource ID. Keep on trying until we wrap around + * in the RID space. + */ + do { + struct vmci_handle handle; + + current_rid = resource_id; + resource_id++; + if (unlikely(resource_id == VMCI_INVALID_ID)) { + /* Skip the reserved rids. */ + resource_id = VMCI_RESERVED_RESOURCE_ID_MAX + 1; + } + + handle = vmci_make_handle(context_id, current_rid); + if (!vmci_resource_lookup(handle, resource_type)) + return current_rid; + } while (resource_id != old_rid); + + return VMCI_INVALID_ID; +} + + +int vmci_resource_add(struct vmci_resource *resource, + enum vmci_resource_type resource_type, + struct vmci_handle handle) + +{ + unsigned int idx; + int result; + + spin_lock(&vmci_resource_table.lock); + + if (handle.resource == VMCI_INVALID_ID) { + handle.resource = vmci_resource_find_id(handle.context, + resource_type); + if (handle.resource == VMCI_INVALID_ID) { + result = VMCI_ERROR_NO_HANDLE; + goto out; + } + } else if (vmci_resource_lookup(handle, resource_type)) { + result = VMCI_ERROR_ALREADY_EXISTS; + goto out; + } + + resource->handle = handle; + resource->type = resource_type; + INIT_HLIST_NODE(&resource->node); + kref_init(&resource->kref); + init_completion(&resource->done); + + idx = vmci_resource_hash(resource->handle); + hlist_add_head_rcu(&resource->node, &vmci_resource_table.entries[idx]); + + result = VMCI_SUCCESS; + +out: + spin_unlock(&vmci_resource_table.lock); + return result; +} + +void vmci_resource_remove(struct vmci_resource *resource) +{ + struct vmci_handle handle = resource->handle; + unsigned int idx = vmci_resource_hash(handle); + struct vmci_resource *r; + + /* Remove resource from hash table. */ + spin_lock(&vmci_resource_table.lock); + + hlist_for_each_entry(r, &vmci_resource_table.entries[idx], node) { + if (vmci_handle_is_equal(r->handle, resource->handle)) { + hlist_del_init_rcu(&r->node); + break; + } + } + + spin_unlock(&vmci_resource_table.lock); + synchronize_rcu(); + + vmci_resource_put(resource); + wait_for_completion(&resource->done); +} + +struct vmci_resource * +vmci_resource_by_handle(struct vmci_handle resource_handle, + enum vmci_resource_type resource_type) +{ + struct vmci_resource *r, *resource = NULL; + + rcu_read_lock(); + + r = vmci_resource_lookup(resource_handle, resource_type); + if (r && + (resource_type == r->type || + resource_type == VMCI_RESOURCE_TYPE_ANY)) { + resource = vmci_resource_get(r); + } + + rcu_read_unlock(); + + return resource; +} + +/* + * Get a reference to given resource. + */ +struct vmci_resource *vmci_resource_get(struct vmci_resource *resource) +{ + kref_get(&resource->kref); + + return resource; +} + +static void vmci_release_resource(struct kref *kref) +{ + struct vmci_resource *resource = + container_of(kref, struct vmci_resource, kref); + + /* Verify the resource has been unlinked from hash table */ + WARN_ON(!hlist_unhashed(&resource->node)); + + /* Signal that container of this resource can now be destroyed */ + complete(&resource->done); +} + +/* + * Resource's release function will get called if last reference. + * If it is the last reference, then we are sure that nobody else + * can increment the count again (it's gone from the resource hash + * table), so there's no need for locking here. + */ +int vmci_resource_put(struct vmci_resource *resource) +{ + /* + * We propagate the information back to caller in case it wants to know + * whether entry was freed. + */ + return kref_put(&resource->kref, vmci_release_resource) ? + VMCI_SUCCESS_ENTRY_DEAD : VMCI_SUCCESS; +} + +struct vmci_handle vmci_resource_handle(struct vmci_resource *resource) +{ + return resource->handle; +} diff --git a/drivers/misc/vmw_vmci/vmci_resource.h b/drivers/misc/vmw_vmci/vmci_resource.h new file mode 100644 index 00000000000..9190cd298be --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_resource.h @@ -0,0 +1,59 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef _VMCI_RESOURCE_H_ +#define _VMCI_RESOURCE_H_ + +#include <linux/vmw_vmci_defs.h> +#include <linux/types.h> + +#include "vmci_context.h" + + +enum vmci_resource_type { + VMCI_RESOURCE_TYPE_ANY, + VMCI_RESOURCE_TYPE_API, + VMCI_RESOURCE_TYPE_GROUP, + VMCI_RESOURCE_TYPE_DATAGRAM, + VMCI_RESOURCE_TYPE_DOORBELL, + VMCI_RESOURCE_TYPE_QPAIR_GUEST, + VMCI_RESOURCE_TYPE_QPAIR_HOST +}; + +struct vmci_resource { + struct vmci_handle handle; + enum vmci_resource_type type; + struct hlist_node node; + struct kref kref; + struct completion done; +}; + + +int vmci_resource_add(struct vmci_resource *resource, + enum vmci_resource_type resource_type, + struct vmci_handle handle); + +void vmci_resource_remove(struct vmci_resource *resource); + +struct vmci_resource * +vmci_resource_by_handle(struct vmci_handle resource_handle, + enum vmci_resource_type resource_type); + +struct vmci_resource *vmci_resource_get(struct vmci_resource *resource); +int vmci_resource_put(struct vmci_resource *resource); + +struct vmci_handle vmci_resource_handle(struct vmci_resource *resource); + +#endif /* _VMCI_RESOURCE_H_ */ diff --git a/drivers/misc/vmw_vmci/vmci_route.c b/drivers/misc/vmw_vmci/vmci_route.c new file mode 100644 index 00000000000..91090658b92 --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_route.c @@ -0,0 +1,226 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> + +#include "vmci_context.h" +#include "vmci_driver.h" +#include "vmci_route.h" + +/* + * Make a routing decision for the given source and destination handles. + * This will try to determine the route using the handles and the available + * devices. Will set the source context if it is invalid. + */ +int vmci_route(struct vmci_handle *src, + const struct vmci_handle *dst, + bool from_guest, + enum vmci_route *route) +{ + bool has_host_device = vmci_host_code_active(); + bool has_guest_device = vmci_guest_code_active(); + + *route = VMCI_ROUTE_NONE; + + /* + * "from_guest" is only ever set to true by + * IOCTL_VMCI_DATAGRAM_SEND (or by the vmkernel equivalent), + * which comes from the VMX, so we know it is coming from a + * guest. + * + * To avoid inconsistencies, test these once. We will test + * them again when we do the actual send to ensure that we do + * not touch a non-existent device. + */ + + /* Must have a valid destination context. */ + if (VMCI_INVALID_ID == dst->context) + return VMCI_ERROR_INVALID_ARGS; + + /* Anywhere to hypervisor. */ + if (VMCI_HYPERVISOR_CONTEXT_ID == dst->context) { + + /* + * If this message already came from a guest then we + * cannot send it to the hypervisor. It must come + * from a local client. + */ + if (from_guest) + return VMCI_ERROR_DST_UNREACHABLE; + + /* + * We must be acting as a guest in order to send to + * the hypervisor. + */ + if (!has_guest_device) + return VMCI_ERROR_DEVICE_NOT_FOUND; + + /* And we cannot send if the source is the host context. */ + if (VMCI_HOST_CONTEXT_ID == src->context) + return VMCI_ERROR_INVALID_ARGS; + + /* + * If the client passed the ANON source handle then + * respect it (both context and resource are invalid). + * However, if they passed only an invalid context, + * then they probably mean ANY, in which case we + * should set the real context here before passing it + * down. + */ + if (VMCI_INVALID_ID == src->context && + VMCI_INVALID_ID != src->resource) + src->context = vmci_get_context_id(); + + /* Send from local client down to the hypervisor. */ + *route = VMCI_ROUTE_AS_GUEST; + return VMCI_SUCCESS; + } + + /* Anywhere to local client on host. */ + if (VMCI_HOST_CONTEXT_ID == dst->context) { + /* + * If it is not from a guest but we are acting as a + * guest, then we need to send it down to the host. + * Note that if we are also acting as a host then this + * will prevent us from sending from local client to + * local client, but we accept that restriction as a + * way to remove any ambiguity from the host context. + */ + if (src->context == VMCI_HYPERVISOR_CONTEXT_ID) { + /* + * If the hypervisor is the source, this is + * host local communication. The hypervisor + * may send vmci event datagrams to the host + * itself, but it will never send datagrams to + * an "outer host" through the guest device. + */ + + if (has_host_device) { + *route = VMCI_ROUTE_AS_HOST; + return VMCI_SUCCESS; + } else { + return VMCI_ERROR_DEVICE_NOT_FOUND; + } + } + + if (!from_guest && has_guest_device) { + /* If no source context then use the current. */ + if (VMCI_INVALID_ID == src->context) + src->context = vmci_get_context_id(); + + /* Send it from local client down to the host. */ + *route = VMCI_ROUTE_AS_GUEST; + return VMCI_SUCCESS; + } + + /* + * Otherwise we already received it from a guest and + * it is destined for a local client on this host, or + * it is from another local client on this host. We + * must be acting as a host to service it. + */ + if (!has_host_device) + return VMCI_ERROR_DEVICE_NOT_FOUND; + + if (VMCI_INVALID_ID == src->context) { + /* + * If it came from a guest then it must have a + * valid context. Otherwise we can use the + * host context. + */ + if (from_guest) + return VMCI_ERROR_INVALID_ARGS; + + src->context = VMCI_HOST_CONTEXT_ID; + } + + /* Route to local client. */ + *route = VMCI_ROUTE_AS_HOST; + return VMCI_SUCCESS; + } + + /* + * If we are acting as a host then this might be destined for + * a guest. + */ + if (has_host_device) { + /* It will have a context if it is meant for a guest. */ + if (vmci_ctx_exists(dst->context)) { + if (VMCI_INVALID_ID == src->context) { + /* + * If it came from a guest then it + * must have a valid context. + * Otherwise we can use the host + * context. + */ + + if (from_guest) + return VMCI_ERROR_INVALID_ARGS; + + src->context = VMCI_HOST_CONTEXT_ID; + } else if (VMCI_CONTEXT_IS_VM(src->context) && + src->context != dst->context) { + /* + * VM to VM communication is not + * allowed. Since we catch all + * communication destined for the host + * above, this must be destined for a + * VM since there is a valid context. + */ + + return VMCI_ERROR_DST_UNREACHABLE; + } + + /* Pass it up to the guest. */ + *route = VMCI_ROUTE_AS_HOST; + return VMCI_SUCCESS; + } else if (!has_guest_device) { + /* + * The host is attempting to reach a CID + * without an active context, and we can't + * send it down, since we have no guest + * device. + */ + + return VMCI_ERROR_DST_UNREACHABLE; + } + } + + /* + * We must be a guest trying to send to another guest, which means + * we need to send it down to the host. We do not filter out VM to + * VM communication here, since we want to be able to use the guest + * driver on older versions that do support VM to VM communication. + */ + if (!has_guest_device) { + /* + * Ending up here means we have neither guest nor host + * device. + */ + return VMCI_ERROR_DEVICE_NOT_FOUND; + } + + /* If no source context then use the current context. */ + if (VMCI_INVALID_ID == src->context) + src->context = vmci_get_context_id(); + + /* + * Send it from local client down to the host, which will + * route it to the other guest for us. + */ + *route = VMCI_ROUTE_AS_GUEST; + return VMCI_SUCCESS; +} diff --git a/drivers/misc/vmw_vmci/vmci_route.h b/drivers/misc/vmw_vmci/vmci_route.h new file mode 100644 index 00000000000..3b30e82419c --- /dev/null +++ b/drivers/misc/vmw_vmci/vmci_route.h @@ -0,0 +1,30 @@ +/* + * VMware VMCI Driver + * + * Copyright (C) 2012 VMware, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2 and no later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY + * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#ifndef _VMCI_ROUTE_H_ +#define _VMCI_ROUTE_H_ + +#include <linux/vmw_vmci_defs.h> + +enum vmci_route { + VMCI_ROUTE_NONE, + VMCI_ROUTE_AS_HOST, + VMCI_ROUTE_AS_GUEST, +}; + +int vmci_route(struct vmci_handle *src, const struct vmci_handle *dst, + bool from_guest, enum vmci_route *route); + +#endif /* _VMCI_ROUTE_H_ */ |