diff options
Diffstat (limited to 'arch/sparc')
365 files changed, 15487 insertions, 8663 deletions
diff --git a/arch/sparc/Kbuild b/arch/sparc/Kbuild index 5cd01161fd0..675afa285dd 100644 --- a/arch/sparc/Kbuild +++ b/arch/sparc/Kbuild @@ -6,3 +6,4 @@ obj-y += kernel/ obj-y += mm/ obj-y += math-emu/ obj-y += net/ +obj-y += crypto/ diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index e74ff137762..407c87d9879 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -12,39 +12,46 @@ config 64BIT config SPARC bool default y + select ARCH_MIGHT_HAVE_PC_PARPORT if SPARC64 && PCI + select ARCH_MIGHT_HAVE_PC_SERIO select OF select OF_PROMTREE select HAVE_IDE select HAVE_OPROFILE select HAVE_ARCH_KGDB if !SMP || SPARC64 select HAVE_ARCH_TRACEHOOK + select SYSCTL_EXCEPTION_TRACE select ARCH_WANT_OPTIONAL_GPIOLIB + select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE select RTC_CLASS select RTC_DRV_M48T59 - select HAVE_IRQ_WORK select HAVE_DMA_ATTRS select HAVE_DMA_API_DEBUG - select HAVE_ARCH_JUMP_LABEL - select HAVE_GENERIC_HARDIRQS + select HAVE_ARCH_JUMP_LABEL if SPARC64 select GENERIC_IRQ_SHOW - select USE_GENERIC_SMP_HELPERS if SMP + select ARCH_WANT_IPC_PARSE_VERSION select GENERIC_PCI_IOMAP select HAVE_NMI_WATCHDOG if SPARC64 select HAVE_BPF_JIT + select HAVE_DEBUG_BUGVERBOSE select GENERIC_SMP_IDLE_THREAD select GENERIC_CMOS_UPDATE select GENERIC_CLOCKEVENTS select GENERIC_STRNCPY_FROM_USER select GENERIC_STRNLEN_USER + select MODULES_USE_ELF_RELA + select ODD_RT_SIGACTION + select OLD_SIGSUSPEND config SPARC32 def_bool !64BIT select GENERIC_ATOMIC64 select CLZ_TAB + select HAVE_UID16 + select OLD_SIGACTION config SPARC64 def_bool 64BIT - select ARCH_SUPPORTS_MSI select HAVE_FUNCTION_TRACER select HAVE_FUNCTION_GRAPH_TRACER select HAVE_FUNCTION_GRAPH_FP_TEST @@ -54,10 +61,12 @@ config SPARC64 select HAVE_RCU_TABLE_FREE if SMP select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP - select HAVE_SYSCALL_WRAPPERS + select HAVE_ARCH_TRANSPARENT_HUGEPAGE select HAVE_DYNAMIC_FTRACE select HAVE_FTRACE_MCOUNT_RECORD select HAVE_SYSCALL_TRACEPOINTS + select HAVE_CONTEXT_TRACKING + select HAVE_DEBUG_KMEMLEAK select RTC_DRV_CMOS select RTC_DRV_BQ4802 select RTC_DRV_SUN4V @@ -68,18 +77,14 @@ config SPARC64 select ARCH_HAVE_NMI_SAFE_CMPXCHG select HAVE_C_RECORDMCOUNT select NO_BOOTMEM + select HAVE_ARCH_AUDITSYSCALL + select ARCH_SUPPORTS_ATOMIC_RMW config ARCH_DEFCONFIG string default "arch/sparc/configs/sparc32_defconfig" if SPARC32 default "arch/sparc/configs/sparc64_defconfig" if SPARC64 -# CONFIG_BITS can be used at source level to get 32/64 bits -config BITS - int - default 32 if SPARC32 - default 64 if SPARC64 - config IOMMU_HELPER bool default y if SPARC64 @@ -96,6 +101,9 @@ config HAVE_LATENCYTOP_SUPPORT bool default y if SPARC64 +config ARCH_HIBERNATION_POSSIBLE + def_bool y if SPARC64 + config AUDIT_ARCH bool default y @@ -131,14 +139,6 @@ config GENERIC_ISA_DMA bool default y if SPARC32 -config GENERIC_GPIO - bool - help - Generic GPIO API support - -config ARCH_NO_VIRT_TO_BUS - def_bool y - config ARCH_SUPPORTS_DEBUG_PAGEALLOC def_bool y if SPARC64 @@ -155,10 +155,10 @@ config SMP a system with only one CPU, say N. If you have a system with more than one CPU, say Y. - If you say N here, the kernel will run on single and multiprocessor + If you say N here, the kernel will run on uni- and multiprocessor machines, but will use only one CPU of a multiprocessor machine. If you say Y here, the kernel will run on many, but not all, - singleprocessor machines. On a singleprocessor machine, the kernel + uniprocessor machines. On a uniprocessor machine, the kernel will run faster if you say N here. People using multiprocessor machines who say Y here should also say @@ -190,7 +190,7 @@ config RWSEM_XCHGADD_ALGORITHM config GENERIC_HWEIGHT bool - default y if !ULTRA_HAS_POPULATION_COUNT + default y config GENERIC_CALIBRATE_DELAY bool @@ -225,25 +225,6 @@ config EARLYFB help Say Y here to enable a faster early framebuffer boot console. -choice - prompt "Kernel page size" if SPARC64 - default SPARC64_PAGE_SIZE_8KB - -config SPARC64_PAGE_SIZE_8KB - bool "8KB" - help - This lets you select the page size of the kernel. - - 8KB and 64KB work quite well, since SPARC ELF sections - provide for up to 64KB alignment. - - If you don't know what to do, choose 8KB. - -config SPARC64_PAGE_SIZE_64KB - bool "64KB" - -endchoice - config SECCOMP bool "Enable seccomp to safely compute untrusted bytecode" depends on SPARC64 && PROC_FS @@ -264,7 +245,6 @@ config SECCOMP config HOTPLUG_CPU bool "Support for hot-pluggable CPUs" depends on SPARC64 && SMP - select HOTPLUG help Say Y here to experiment with turning CPUs off and on. CPUs can be controlled through /sys/devices/system/cpu/cpu#. @@ -272,29 +252,6 @@ config HOTPLUG_CPU if SPARC64 source "drivers/cpufreq/Kconfig" - -config US3_FREQ - tristate "UltraSPARC-III CPU Frequency driver" - depends on CPU_FREQ - select CPU_FREQ_TABLE - help - This adds the CPUFreq driver for UltraSPARC-III processors. - - For details, take a look at <file:Documentation/cpu-freq>. - - If in doubt, say N. - -config US2E_FREQ - tristate "UltraSPARC-IIe CPU Frequency driver" - depends on CPU_FREQ - select CPU_FREQ_TABLE - help - This adds the CPUFreq driver for UltraSPARC-IIe processors. - - For details, take a look at <file:Documentation/cpu-freq>. - - If in doubt, say N. - endif config US3_MC @@ -315,23 +272,6 @@ config GENERIC_LOCKBREAK default y depends on SPARC64 && SMP && PREEMPT -choice - prompt "SPARC64 Huge TLB Page Size" - depends on SPARC64 && HUGETLB_PAGE - default HUGETLB_PAGE_SIZE_4MB - -config HUGETLB_PAGE_SIZE_4MB - bool "4MB" - -config HUGETLB_PAGE_SIZE_512K - bool "512K" - -config HUGETLB_PAGE_SIZE_64K - depends on !SPARC64_PAGE_SIZE_64KB - bool "64K" - -endchoice - config NUMA bool "NUMA support" depends on SPARC64 && SMP @@ -362,6 +302,10 @@ config ARCH_SPARSEMEM_DEFAULT source "mm/Kconfig" +if SPARC64 +source "kernel/power/Kconfig" +endif + config SCHED_SMT bool "SMT (Hyperthreading) scheduler support" depends on SPARC64 && SMP @@ -442,6 +386,8 @@ config SERIAL_CONSOLE config SPARC_LEON bool "Sparc Leon processor family" depends on SPARC32 + select USB_EHCI_BIG_ENDIAN_MMIO + select USB_EHCI_BIG_ENDIAN_DESC ---help--- If you say Y here if you are running on a SPARC-LEON processor. The LEON processor is a synthesizable VHDL model of the @@ -529,7 +475,18 @@ config LEON_PCI depends on PCI && SPARC_LEON default y -config GRPCI2 +config SPARC_GRPCI1 + bool "GRPCI Host Bridge Support" + depends on LEON_PCI + default y + help + Say Y here to include the GRPCI Host Bridge Driver. The GRPCI + PCI host controller is typically found in GRLIB SPARC32/LEON + systems. The driver has one property (all_pci_errors) controlled + from the bootloader that makes the GRPCI to generate interrupts + on detected PCI Parity and System errors. + +config SPARC_GRPCI2 bool "GRPCI2 Host Bridge Support" depends on LEON_PCI default y @@ -553,12 +510,17 @@ config SUN_OPENPROMFS Only choose N if you know in advance that you will not need to modify OpenPROM settings on the running system. -# Makefile helper +# Makefile helpers config SPARC64_PCI bool default y depends on SPARC64 && PCI +config SPARC64_PCI_MSI + bool + default y + depends on SPARC64_PCI && PCI_MSI + endmenu menu "Executable file formats" @@ -570,7 +532,9 @@ config COMPAT depends on SPARC64 default y select COMPAT_BINFMT_ELF + select HAVE_UID16 select ARCH_WANT_OLD_COMPAT_IPC + select COMPAT_OLD_SIGACTION config SYSVIPC_COMPAT bool diff --git a/arch/sparc/Makefile b/arch/sparc/Makefile index 541b8b075c7..9ff423678cb 100644 --- a/arch/sparc/Makefile +++ b/arch/sparc/Makefile @@ -57,6 +57,7 @@ core-y += arch/sparc/ libs-y += arch/sparc/prom/ libs-y += arch/sparc/lib/ +drivers-$(CONFIG_PM) += arch/sparc/power/ drivers-$(CONFIG_OPROFILE) += arch/sparc/oprofile/ boot := arch/sparc/boot diff --git a/arch/sparc/boot/piggyback.c b/arch/sparc/boot/piggyback.c index c0a798fcf03..bb7c95161d7 100644 --- a/arch/sparc/boot/piggyback.c +++ b/arch/sparc/boot/piggyback.c @@ -81,18 +81,18 @@ static void usage(void) static int start_line(const char *line) { - if (strcmp(line + 8, " T _start\n") == 0) + if (strcmp(line + 10, " _start\n") == 0) return 1; - else if (strcmp(line + 16, " T _start\n") == 0) + else if (strcmp(line + 18, " _start\n") == 0) return 1; return 0; } static int end_line(const char *line) { - if (strcmp(line + 8, " A _end\n") == 0) + if (strcmp(line + 10, " _end\n") == 0) return 1; - else if (strcmp (line + 16, " A _end\n") == 0) + else if (strcmp (line + 18, " _end\n") == 0) return 1; return 0; } @@ -100,8 +100,8 @@ static int end_line(const char *line) /* * Find address for start and end in System.map. * The file looks like this: - * f0004000 T _start - * f0379f79 A _end + * f0004000 ... _start + * f0379f79 ... _end * 1234567890123456 * ^coloumn 1 * There is support for 64 bit addresses too. diff --git a/arch/sparc/crypto/Makefile b/arch/sparc/crypto/Makefile new file mode 100644 index 00000000000..5d469d81761 --- /dev/null +++ b/arch/sparc/crypto/Makefile @@ -0,0 +1,25 @@ +# +# Arch-specific CryptoAPI modules. +# + +obj-$(CONFIG_CRYPTO_SHA1_SPARC64) += sha1-sparc64.o +obj-$(CONFIG_CRYPTO_SHA256_SPARC64) += sha256-sparc64.o +obj-$(CONFIG_CRYPTO_SHA512_SPARC64) += sha512-sparc64.o +obj-$(CONFIG_CRYPTO_MD5_SPARC64) += md5-sparc64.o + +obj-$(CONFIG_CRYPTO_AES_SPARC64) += aes-sparc64.o +obj-$(CONFIG_CRYPTO_DES_SPARC64) += des-sparc64.o +obj-$(CONFIG_CRYPTO_DES_SPARC64) += camellia-sparc64.o + +obj-$(CONFIG_CRYPTO_CRC32C_SPARC64) += crc32c-sparc64.o + +sha1-sparc64-y := sha1_asm.o sha1_glue.o +sha256-sparc64-y := sha256_asm.o sha256_glue.o +sha512-sparc64-y := sha512_asm.o sha512_glue.o +md5-sparc64-y := md5_asm.o md5_glue.o + +aes-sparc64-y := aes_asm.o aes_glue.o +des-sparc64-y := des_asm.o des_glue.o +camellia-sparc64-y := camellia_asm.o camellia_glue.o + +crc32c-sparc64-y := crc32c_asm.o crc32c_glue.o diff --git a/arch/sparc/crypto/aes_asm.S b/arch/sparc/crypto/aes_asm.S new file mode 100644 index 00000000000..1cda8aa7cb8 --- /dev/null +++ b/arch/sparc/crypto/aes_asm.S @@ -0,0 +1,1543 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> + +#include "opcodes.h" + +#define ENCRYPT_TWO_ROUNDS(KEY_BASE, I0, I1, T0, T1) \ + AES_EROUND01(KEY_BASE + 0, I0, I1, T0) \ + AES_EROUND23(KEY_BASE + 2, I0, I1, T1) \ + AES_EROUND01(KEY_BASE + 4, T0, T1, I0) \ + AES_EROUND23(KEY_BASE + 6, T0, T1, I1) + +#define ENCRYPT_TWO_ROUNDS_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + AES_EROUND01(KEY_BASE + 0, I0, I1, T0) \ + AES_EROUND23(KEY_BASE + 2, I0, I1, T1) \ + AES_EROUND01(KEY_BASE + 0, I2, I3, T2) \ + AES_EROUND23(KEY_BASE + 2, I2, I3, T3) \ + AES_EROUND01(KEY_BASE + 4, T0, T1, I0) \ + AES_EROUND23(KEY_BASE + 6, T0, T1, I1) \ + AES_EROUND01(KEY_BASE + 4, T2, T3, I2) \ + AES_EROUND23(KEY_BASE + 6, T2, T3, I3) + +#define ENCRYPT_TWO_ROUNDS_LAST(KEY_BASE, I0, I1, T0, T1) \ + AES_EROUND01(KEY_BASE + 0, I0, I1, T0) \ + AES_EROUND23(KEY_BASE + 2, I0, I1, T1) \ + AES_EROUND01_L(KEY_BASE + 4, T0, T1, I0) \ + AES_EROUND23_L(KEY_BASE + 6, T0, T1, I1) + +#define ENCRYPT_TWO_ROUNDS_LAST_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + AES_EROUND01(KEY_BASE + 0, I0, I1, T0) \ + AES_EROUND23(KEY_BASE + 2, I0, I1, T1) \ + AES_EROUND01(KEY_BASE + 0, I2, I3, T2) \ + AES_EROUND23(KEY_BASE + 2, I2, I3, T3) \ + AES_EROUND01_L(KEY_BASE + 4, T0, T1, I0) \ + AES_EROUND23_L(KEY_BASE + 6, T0, T1, I1) \ + AES_EROUND01_L(KEY_BASE + 4, T2, T3, I2) \ + AES_EROUND23_L(KEY_BASE + 6, T2, T3, I3) + + /* 10 rounds */ +#define ENCRYPT_128(KEY_BASE, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 0, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 8, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 16, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 24, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS_LAST(KEY_BASE + 32, I0, I1, T0, T1) + +#define ENCRYPT_128_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 0, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 8, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 16, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 24, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_LAST_2(KEY_BASE + 32, I0, I1, I2, I3, T0, T1, T2, T3) + + /* 12 rounds */ +#define ENCRYPT_192(KEY_BASE, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 0, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 8, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 16, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 24, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 32, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS_LAST(KEY_BASE + 40, I0, I1, T0, T1) + +#define ENCRYPT_192_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 0, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 8, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 16, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 24, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE + 32, I0, I1, I2, I3, T0, T1, T2, T3) \ + ENCRYPT_TWO_ROUNDS_LAST_2(KEY_BASE + 40, I0, I1, I2, I3, T0, T1, T2, T3) + + /* 14 rounds */ +#define ENCRYPT_256(KEY_BASE, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 0, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 8, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 16, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 24, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 32, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS(KEY_BASE + 40, I0, I1, T0, T1) \ + ENCRYPT_TWO_ROUNDS_LAST(KEY_BASE + 48, I0, I1, T0, T1) + +#define ENCRYPT_256_TWO_ROUNDS_2(KEY_BASE, I0, I1, I2, I3, TMP_BASE) \ + ENCRYPT_TWO_ROUNDS_2(KEY_BASE, I0, I1, I2, I3, \ + TMP_BASE + 0, TMP_BASE + 2, TMP_BASE + 4, TMP_BASE + 6) + +#define ENCRYPT_256_2(KEY_BASE, I0, I1, I2, I3) \ + ENCRYPT_256_TWO_ROUNDS_2(KEY_BASE + 0, I0, I1, I2, I3, KEY_BASE + 48) \ + ldd [%o0 + 0xd0], %f56; \ + ldd [%o0 + 0xd8], %f58; \ + ENCRYPT_256_TWO_ROUNDS_2(KEY_BASE + 8, I0, I1, I2, I3, KEY_BASE + 0) \ + ldd [%o0 + 0xe0], %f60; \ + ldd [%o0 + 0xe8], %f62; \ + ENCRYPT_256_TWO_ROUNDS_2(KEY_BASE + 16, I0, I1, I2, I3, KEY_BASE + 0) \ + ENCRYPT_256_TWO_ROUNDS_2(KEY_BASE + 24, I0, I1, I2, I3, KEY_BASE + 0) \ + ENCRYPT_256_TWO_ROUNDS_2(KEY_BASE + 32, I0, I1, I2, I3, KEY_BASE + 0) \ + ENCRYPT_256_TWO_ROUNDS_2(KEY_BASE + 40, I0, I1, I2, I3, KEY_BASE + 0) \ + AES_EROUND01(KEY_BASE + 48, I0, I1, KEY_BASE + 0) \ + AES_EROUND23(KEY_BASE + 50, I0, I1, KEY_BASE + 2) \ + AES_EROUND01(KEY_BASE + 48, I2, I3, KEY_BASE + 4) \ + AES_EROUND23(KEY_BASE + 50, I2, I3, KEY_BASE + 6) \ + AES_EROUND01_L(KEY_BASE + 52, KEY_BASE + 0, KEY_BASE + 2, I0) \ + AES_EROUND23_L(KEY_BASE + 54, KEY_BASE + 0, KEY_BASE + 2, I1) \ + ldd [%o0 + 0x10], %f8; \ + ldd [%o0 + 0x18], %f10; \ + AES_EROUND01_L(KEY_BASE + 52, KEY_BASE + 4, KEY_BASE + 6, I2) \ + AES_EROUND23_L(KEY_BASE + 54, KEY_BASE + 4, KEY_BASE + 6, I3) \ + ldd [%o0 + 0x20], %f12; \ + ldd [%o0 + 0x28], %f14; + +#define DECRYPT_TWO_ROUNDS(KEY_BASE, I0, I1, T0, T1) \ + AES_DROUND23(KEY_BASE + 0, I0, I1, T1) \ + AES_DROUND01(KEY_BASE + 2, I0, I1, T0) \ + AES_DROUND23(KEY_BASE + 4, T0, T1, I1) \ + AES_DROUND01(KEY_BASE + 6, T0, T1, I0) + +#define DECRYPT_TWO_ROUNDS_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + AES_DROUND23(KEY_BASE + 0, I0, I1, T1) \ + AES_DROUND01(KEY_BASE + 2, I0, I1, T0) \ + AES_DROUND23(KEY_BASE + 0, I2, I3, T3) \ + AES_DROUND01(KEY_BASE + 2, I2, I3, T2) \ + AES_DROUND23(KEY_BASE + 4, T0, T1, I1) \ + AES_DROUND01(KEY_BASE + 6, T0, T1, I0) \ + AES_DROUND23(KEY_BASE + 4, T2, T3, I3) \ + AES_DROUND01(KEY_BASE + 6, T2, T3, I2) + +#define DECRYPT_TWO_ROUNDS_LAST(KEY_BASE, I0, I1, T0, T1) \ + AES_DROUND23(KEY_BASE + 0, I0, I1, T1) \ + AES_DROUND01(KEY_BASE + 2, I0, I1, T0) \ + AES_DROUND23_L(KEY_BASE + 4, T0, T1, I1) \ + AES_DROUND01_L(KEY_BASE + 6, T0, T1, I0) + +#define DECRYPT_TWO_ROUNDS_LAST_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + AES_DROUND23(KEY_BASE + 0, I0, I1, T1) \ + AES_DROUND01(KEY_BASE + 2, I0, I1, T0) \ + AES_DROUND23(KEY_BASE + 0, I2, I3, T3) \ + AES_DROUND01(KEY_BASE + 2, I2, I3, T2) \ + AES_DROUND23_L(KEY_BASE + 4, T0, T1, I1) \ + AES_DROUND01_L(KEY_BASE + 6, T0, T1, I0) \ + AES_DROUND23_L(KEY_BASE + 4, T2, T3, I3) \ + AES_DROUND01_L(KEY_BASE + 6, T2, T3, I2) + + /* 10 rounds */ +#define DECRYPT_128(KEY_BASE, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 0, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 8, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 16, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 24, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS_LAST(KEY_BASE + 32, I0, I1, T0, T1) + +#define DECRYPT_128_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 0, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 8, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 16, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 24, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_LAST_2(KEY_BASE + 32, I0, I1, I2, I3, T0, T1, T2, T3) + + /* 12 rounds */ +#define DECRYPT_192(KEY_BASE, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 0, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 8, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 16, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 24, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 32, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS_LAST(KEY_BASE + 40, I0, I1, T0, T1) + +#define DECRYPT_192_2(KEY_BASE, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 0, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 8, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 16, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 24, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE + 32, I0, I1, I2, I3, T0, T1, T2, T3) \ + DECRYPT_TWO_ROUNDS_LAST_2(KEY_BASE + 40, I0, I1, I2, I3, T0, T1, T2, T3) + + /* 14 rounds */ +#define DECRYPT_256(KEY_BASE, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 0, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 8, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 16, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 24, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 32, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS(KEY_BASE + 40, I0, I1, T0, T1) \ + DECRYPT_TWO_ROUNDS_LAST(KEY_BASE + 48, I0, I1, T0, T1) + +#define DECRYPT_256_TWO_ROUNDS_2(KEY_BASE, I0, I1, I2, I3, TMP_BASE) \ + DECRYPT_TWO_ROUNDS_2(KEY_BASE, I0, I1, I2, I3, \ + TMP_BASE + 0, TMP_BASE + 2, TMP_BASE + 4, TMP_BASE + 6) + +#define DECRYPT_256_2(KEY_BASE, I0, I1, I2, I3) \ + DECRYPT_256_TWO_ROUNDS_2(KEY_BASE + 0, I0, I1, I2, I3, KEY_BASE + 48) \ + ldd [%o0 + 0x18], %f56; \ + ldd [%o0 + 0x10], %f58; \ + DECRYPT_256_TWO_ROUNDS_2(KEY_BASE + 8, I0, I1, I2, I3, KEY_BASE + 0) \ + ldd [%o0 + 0x08], %f60; \ + ldd [%o0 + 0x00], %f62; \ + DECRYPT_256_TWO_ROUNDS_2(KEY_BASE + 16, I0, I1, I2, I3, KEY_BASE + 0) \ + DECRYPT_256_TWO_ROUNDS_2(KEY_BASE + 24, I0, I1, I2, I3, KEY_BASE + 0) \ + DECRYPT_256_TWO_ROUNDS_2(KEY_BASE + 32, I0, I1, I2, I3, KEY_BASE + 0) \ + DECRYPT_256_TWO_ROUNDS_2(KEY_BASE + 40, I0, I1, I2, I3, KEY_BASE + 0) \ + AES_DROUND23(KEY_BASE + 48, I0, I1, KEY_BASE + 2) \ + AES_DROUND01(KEY_BASE + 50, I0, I1, KEY_BASE + 0) \ + AES_DROUND23(KEY_BASE + 48, I2, I3, KEY_BASE + 6) \ + AES_DROUND01(KEY_BASE + 50, I2, I3, KEY_BASE + 4) \ + AES_DROUND23_L(KEY_BASE + 52, KEY_BASE + 0, KEY_BASE + 2, I1) \ + AES_DROUND01_L(KEY_BASE + 54, KEY_BASE + 0, KEY_BASE + 2, I0) \ + ldd [%o0 + 0xd8], %f8; \ + ldd [%o0 + 0xd0], %f10; \ + AES_DROUND23_L(KEY_BASE + 52, KEY_BASE + 4, KEY_BASE + 6, I3) \ + AES_DROUND01_L(KEY_BASE + 54, KEY_BASE + 4, KEY_BASE + 6, I2) \ + ldd [%o0 + 0xc8], %f12; \ + ldd [%o0 + 0xc0], %f14; + + .align 32 +ENTRY(aes_sparc64_key_expand) + /* %o0=input_key, %o1=output_key, %o2=key_len */ + VISEntry + ld [%o0 + 0x00], %f0 + ld [%o0 + 0x04], %f1 + ld [%o0 + 0x08], %f2 + ld [%o0 + 0x0c], %f3 + + std %f0, [%o1 + 0x00] + std %f2, [%o1 + 0x08] + add %o1, 0x10, %o1 + + cmp %o2, 24 + bl 2f + nop + + be 1f + nop + + /* 256-bit key expansion */ + ld [%o0 + 0x10], %f4 + ld [%o0 + 0x14], %f5 + ld [%o0 + 0x18], %f6 + ld [%o0 + 0x1c], %f7 + + std %f4, [%o1 + 0x00] + std %f6, [%o1 + 0x08] + add %o1, 0x10, %o1 + + AES_KEXPAND1(0, 6, 0x0, 8) + AES_KEXPAND2(2, 8, 10) + AES_KEXPAND0(4, 10, 12) + AES_KEXPAND2(6, 12, 14) + AES_KEXPAND1(8, 14, 0x1, 16) + AES_KEXPAND2(10, 16, 18) + AES_KEXPAND0(12, 18, 20) + AES_KEXPAND2(14, 20, 22) + AES_KEXPAND1(16, 22, 0x2, 24) + AES_KEXPAND2(18, 24, 26) + AES_KEXPAND0(20, 26, 28) + AES_KEXPAND2(22, 28, 30) + AES_KEXPAND1(24, 30, 0x3, 32) + AES_KEXPAND2(26, 32, 34) + AES_KEXPAND0(28, 34, 36) + AES_KEXPAND2(30, 36, 38) + AES_KEXPAND1(32, 38, 0x4, 40) + AES_KEXPAND2(34, 40, 42) + AES_KEXPAND0(36, 42, 44) + AES_KEXPAND2(38, 44, 46) + AES_KEXPAND1(40, 46, 0x5, 48) + AES_KEXPAND2(42, 48, 50) + AES_KEXPAND0(44, 50, 52) + AES_KEXPAND2(46, 52, 54) + AES_KEXPAND1(48, 54, 0x6, 56) + AES_KEXPAND2(50, 56, 58) + + std %f8, [%o1 + 0x00] + std %f10, [%o1 + 0x08] + std %f12, [%o1 + 0x10] + std %f14, [%o1 + 0x18] + std %f16, [%o1 + 0x20] + std %f18, [%o1 + 0x28] + std %f20, [%o1 + 0x30] + std %f22, [%o1 + 0x38] + std %f24, [%o1 + 0x40] + std %f26, [%o1 + 0x48] + std %f28, [%o1 + 0x50] + std %f30, [%o1 + 0x58] + std %f32, [%o1 + 0x60] + std %f34, [%o1 + 0x68] + std %f36, [%o1 + 0x70] + std %f38, [%o1 + 0x78] + std %f40, [%o1 + 0x80] + std %f42, [%o1 + 0x88] + std %f44, [%o1 + 0x90] + std %f46, [%o1 + 0x98] + std %f48, [%o1 + 0xa0] + std %f50, [%o1 + 0xa8] + std %f52, [%o1 + 0xb0] + std %f54, [%o1 + 0xb8] + std %f56, [%o1 + 0xc0] + ba,pt %xcc, 80f + std %f58, [%o1 + 0xc8] + +1: + /* 192-bit key expansion */ + ld [%o0 + 0x10], %f4 + ld [%o0 + 0x14], %f5 + + std %f4, [%o1 + 0x00] + add %o1, 0x08, %o1 + + AES_KEXPAND1(0, 4, 0x0, 6) + AES_KEXPAND2(2, 6, 8) + AES_KEXPAND2(4, 8, 10) + AES_KEXPAND1(6, 10, 0x1, 12) + AES_KEXPAND2(8, 12, 14) + AES_KEXPAND2(10, 14, 16) + AES_KEXPAND1(12, 16, 0x2, 18) + AES_KEXPAND2(14, 18, 20) + AES_KEXPAND2(16, 20, 22) + AES_KEXPAND1(18, 22, 0x3, 24) + AES_KEXPAND2(20, 24, 26) + AES_KEXPAND2(22, 26, 28) + AES_KEXPAND1(24, 28, 0x4, 30) + AES_KEXPAND2(26, 30, 32) + AES_KEXPAND2(28, 32, 34) + AES_KEXPAND1(30, 34, 0x5, 36) + AES_KEXPAND2(32, 36, 38) + AES_KEXPAND2(34, 38, 40) + AES_KEXPAND1(36, 40, 0x6, 42) + AES_KEXPAND2(38, 42, 44) + AES_KEXPAND2(40, 44, 46) + AES_KEXPAND1(42, 46, 0x7, 48) + AES_KEXPAND2(44, 48, 50) + + std %f6, [%o1 + 0x00] + std %f8, [%o1 + 0x08] + std %f10, [%o1 + 0x10] + std %f12, [%o1 + 0x18] + std %f14, [%o1 + 0x20] + std %f16, [%o1 + 0x28] + std %f18, [%o1 + 0x30] + std %f20, [%o1 + 0x38] + std %f22, [%o1 + 0x40] + std %f24, [%o1 + 0x48] + std %f26, [%o1 + 0x50] + std %f28, [%o1 + 0x58] + std %f30, [%o1 + 0x60] + std %f32, [%o1 + 0x68] + std %f34, [%o1 + 0x70] + std %f36, [%o1 + 0x78] + std %f38, [%o1 + 0x80] + std %f40, [%o1 + 0x88] + std %f42, [%o1 + 0x90] + std %f44, [%o1 + 0x98] + std %f46, [%o1 + 0xa0] + std %f48, [%o1 + 0xa8] + ba,pt %xcc, 80f + std %f50, [%o1 + 0xb0] + +2: + /* 128-bit key expansion */ + AES_KEXPAND1(0, 2, 0x0, 4) + AES_KEXPAND2(2, 4, 6) + AES_KEXPAND1(4, 6, 0x1, 8) + AES_KEXPAND2(6, 8, 10) + AES_KEXPAND1(8, 10, 0x2, 12) + AES_KEXPAND2(10, 12, 14) + AES_KEXPAND1(12, 14, 0x3, 16) + AES_KEXPAND2(14, 16, 18) + AES_KEXPAND1(16, 18, 0x4, 20) + AES_KEXPAND2(18, 20, 22) + AES_KEXPAND1(20, 22, 0x5, 24) + AES_KEXPAND2(22, 24, 26) + AES_KEXPAND1(24, 26, 0x6, 28) + AES_KEXPAND2(26, 28, 30) + AES_KEXPAND1(28, 30, 0x7, 32) + AES_KEXPAND2(30, 32, 34) + AES_KEXPAND1(32, 34, 0x8, 36) + AES_KEXPAND2(34, 36, 38) + AES_KEXPAND1(36, 38, 0x9, 40) + AES_KEXPAND2(38, 40, 42) + + std %f4, [%o1 + 0x00] + std %f6, [%o1 + 0x08] + std %f8, [%o1 + 0x10] + std %f10, [%o1 + 0x18] + std %f12, [%o1 + 0x20] + std %f14, [%o1 + 0x28] + std %f16, [%o1 + 0x30] + std %f18, [%o1 + 0x38] + std %f20, [%o1 + 0x40] + std %f22, [%o1 + 0x48] + std %f24, [%o1 + 0x50] + std %f26, [%o1 + 0x58] + std %f28, [%o1 + 0x60] + std %f30, [%o1 + 0x68] + std %f32, [%o1 + 0x70] + std %f34, [%o1 + 0x78] + std %f36, [%o1 + 0x80] + std %f38, [%o1 + 0x88] + std %f40, [%o1 + 0x90] + std %f42, [%o1 + 0x98] +80: + retl + VISExit +ENDPROC(aes_sparc64_key_expand) + + .align 32 +ENTRY(aes_sparc64_encrypt_128) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ld [%o1 + 0x00], %f4 + ld [%o1 + 0x04], %f5 + ld [%o1 + 0x08], %f6 + ld [%o1 + 0x0c], %f7 + ldd [%o0 + 0x00], %f8 + ldd [%o0 + 0x08], %f10 + ldd [%o0 + 0x10], %f12 + ldd [%o0 + 0x18], %f14 + ldd [%o0 + 0x20], %f16 + ldd [%o0 + 0x28], %f18 + ldd [%o0 + 0x30], %f20 + ldd [%o0 + 0x38], %f22 + ldd [%o0 + 0x40], %f24 + ldd [%o0 + 0x48], %f26 + ldd [%o0 + 0x50], %f28 + ldd [%o0 + 0x58], %f30 + ldd [%o0 + 0x60], %f32 + ldd [%o0 + 0x68], %f34 + ldd [%o0 + 0x70], %f36 + ldd [%o0 + 0x78], %f38 + ldd [%o0 + 0x80], %f40 + ldd [%o0 + 0x88], %f42 + ldd [%o0 + 0x90], %f44 + ldd [%o0 + 0x98], %f46 + ldd [%o0 + 0xa0], %f48 + ldd [%o0 + 0xa8], %f50 + fxor %f8, %f4, %f4 + fxor %f10, %f6, %f6 + ENCRYPT_128(12, 4, 6, 0, 2) + st %f4, [%o2 + 0x00] + st %f5, [%o2 + 0x04] + st %f6, [%o2 + 0x08] + st %f7, [%o2 + 0x0c] + retl + VISExit +ENDPROC(aes_sparc64_encrypt_128) + + .align 32 +ENTRY(aes_sparc64_encrypt_192) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ld [%o1 + 0x00], %f4 + ld [%o1 + 0x04], %f5 + ld [%o1 + 0x08], %f6 + ld [%o1 + 0x0c], %f7 + + ldd [%o0 + 0x00], %f8 + ldd [%o0 + 0x08], %f10 + + fxor %f8, %f4, %f4 + fxor %f10, %f6, %f6 + + ldd [%o0 + 0x10], %f8 + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + add %o0, 0x20, %o0 + + ENCRYPT_TWO_ROUNDS(8, 4, 6, 0, 2) + + ldd [%o0 + 0x10], %f12 + ldd [%o0 + 0x18], %f14 + ldd [%o0 + 0x20], %f16 + ldd [%o0 + 0x28], %f18 + ldd [%o0 + 0x30], %f20 + ldd [%o0 + 0x38], %f22 + ldd [%o0 + 0x40], %f24 + ldd [%o0 + 0x48], %f26 + ldd [%o0 + 0x50], %f28 + ldd [%o0 + 0x58], %f30 + ldd [%o0 + 0x60], %f32 + ldd [%o0 + 0x68], %f34 + ldd [%o0 + 0x70], %f36 + ldd [%o0 + 0x78], %f38 + ldd [%o0 + 0x80], %f40 + ldd [%o0 + 0x88], %f42 + ldd [%o0 + 0x90], %f44 + ldd [%o0 + 0x98], %f46 + ldd [%o0 + 0xa0], %f48 + ldd [%o0 + 0xa8], %f50 + + + ENCRYPT_128(12, 4, 6, 0, 2) + + st %f4, [%o2 + 0x00] + st %f5, [%o2 + 0x04] + st %f6, [%o2 + 0x08] + st %f7, [%o2 + 0x0c] + + retl + VISExit +ENDPROC(aes_sparc64_encrypt_192) + + .align 32 +ENTRY(aes_sparc64_encrypt_256) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ld [%o1 + 0x00], %f4 + ld [%o1 + 0x04], %f5 + ld [%o1 + 0x08], %f6 + ld [%o1 + 0x0c], %f7 + + ldd [%o0 + 0x00], %f8 + ldd [%o0 + 0x08], %f10 + + fxor %f8, %f4, %f4 + fxor %f10, %f6, %f6 + + ldd [%o0 + 0x10], %f8 + + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + add %o0, 0x20, %o0 + + ENCRYPT_TWO_ROUNDS(8, 4, 6, 0, 2) + + ldd [%o0 + 0x10], %f8 + + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + add %o0, 0x20, %o0 + + ENCRYPT_TWO_ROUNDS(8, 4, 6, 0, 2) + + ldd [%o0 + 0x10], %f12 + ldd [%o0 + 0x18], %f14 + ldd [%o0 + 0x20], %f16 + ldd [%o0 + 0x28], %f18 + ldd [%o0 + 0x30], %f20 + ldd [%o0 + 0x38], %f22 + ldd [%o0 + 0x40], %f24 + ldd [%o0 + 0x48], %f26 + ldd [%o0 + 0x50], %f28 + ldd [%o0 + 0x58], %f30 + ldd [%o0 + 0x60], %f32 + ldd [%o0 + 0x68], %f34 + ldd [%o0 + 0x70], %f36 + ldd [%o0 + 0x78], %f38 + ldd [%o0 + 0x80], %f40 + ldd [%o0 + 0x88], %f42 + ldd [%o0 + 0x90], %f44 + ldd [%o0 + 0x98], %f46 + ldd [%o0 + 0xa0], %f48 + ldd [%o0 + 0xa8], %f50 + + ENCRYPT_128(12, 4, 6, 0, 2) + + st %f4, [%o2 + 0x00] + st %f5, [%o2 + 0x04] + st %f6, [%o2 + 0x08] + st %f7, [%o2 + 0x0c] + + retl + VISExit +ENDPROC(aes_sparc64_encrypt_256) + + .align 32 +ENTRY(aes_sparc64_decrypt_128) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ld [%o1 + 0x00], %f4 + ld [%o1 + 0x04], %f5 + ld [%o1 + 0x08], %f6 + ld [%o1 + 0x0c], %f7 + ldd [%o0 + 0xa0], %f8 + ldd [%o0 + 0xa8], %f10 + ldd [%o0 + 0x98], %f12 + ldd [%o0 + 0x90], %f14 + ldd [%o0 + 0x88], %f16 + ldd [%o0 + 0x80], %f18 + ldd [%o0 + 0x78], %f20 + ldd [%o0 + 0x70], %f22 + ldd [%o0 + 0x68], %f24 + ldd [%o0 + 0x60], %f26 + ldd [%o0 + 0x58], %f28 + ldd [%o0 + 0x50], %f30 + ldd [%o0 + 0x48], %f32 + ldd [%o0 + 0x40], %f34 + ldd [%o0 + 0x38], %f36 + ldd [%o0 + 0x30], %f38 + ldd [%o0 + 0x28], %f40 + ldd [%o0 + 0x20], %f42 + ldd [%o0 + 0x18], %f44 + ldd [%o0 + 0x10], %f46 + ldd [%o0 + 0x08], %f48 + ldd [%o0 + 0x00], %f50 + fxor %f8, %f4, %f4 + fxor %f10, %f6, %f6 + DECRYPT_128(12, 4, 6, 0, 2) + st %f4, [%o2 + 0x00] + st %f5, [%o2 + 0x04] + st %f6, [%o2 + 0x08] + st %f7, [%o2 + 0x0c] + retl + VISExit +ENDPROC(aes_sparc64_decrypt_128) + + .align 32 +ENTRY(aes_sparc64_decrypt_192) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ld [%o1 + 0x00], %f4 + ld [%o1 + 0x04], %f5 + ld [%o1 + 0x08], %f6 + ld [%o1 + 0x0c], %f7 + ldd [%o0 + 0xc0], %f8 + ldd [%o0 + 0xc8], %f10 + ldd [%o0 + 0xb8], %f12 + ldd [%o0 + 0xb0], %f14 + ldd [%o0 + 0xa8], %f16 + ldd [%o0 + 0xa0], %f18 + fxor %f8, %f4, %f4 + fxor %f10, %f6, %f6 + ldd [%o0 + 0x98], %f20 + ldd [%o0 + 0x90], %f22 + ldd [%o0 + 0x88], %f24 + ldd [%o0 + 0x80], %f26 + DECRYPT_TWO_ROUNDS(12, 4, 6, 0, 2) + ldd [%o0 + 0x78], %f28 + ldd [%o0 + 0x70], %f30 + ldd [%o0 + 0x68], %f32 + ldd [%o0 + 0x60], %f34 + ldd [%o0 + 0x58], %f36 + ldd [%o0 + 0x50], %f38 + ldd [%o0 + 0x48], %f40 + ldd [%o0 + 0x40], %f42 + ldd [%o0 + 0x38], %f44 + ldd [%o0 + 0x30], %f46 + ldd [%o0 + 0x28], %f48 + ldd [%o0 + 0x20], %f50 + ldd [%o0 + 0x18], %f52 + ldd [%o0 + 0x10], %f54 + ldd [%o0 + 0x08], %f56 + ldd [%o0 + 0x00], %f58 + DECRYPT_128(20, 4, 6, 0, 2) + st %f4, [%o2 + 0x00] + st %f5, [%o2 + 0x04] + st %f6, [%o2 + 0x08] + st %f7, [%o2 + 0x0c] + retl + VISExit +ENDPROC(aes_sparc64_decrypt_192) + + .align 32 +ENTRY(aes_sparc64_decrypt_256) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ld [%o1 + 0x00], %f4 + ld [%o1 + 0x04], %f5 + ld [%o1 + 0x08], %f6 + ld [%o1 + 0x0c], %f7 + ldd [%o0 + 0xe0], %f8 + ldd [%o0 + 0xe8], %f10 + ldd [%o0 + 0xd8], %f12 + ldd [%o0 + 0xd0], %f14 + ldd [%o0 + 0xc8], %f16 + fxor %f8, %f4, %f4 + ldd [%o0 + 0xc0], %f18 + fxor %f10, %f6, %f6 + ldd [%o0 + 0xb8], %f20 + AES_DROUND23(12, 4, 6, 2) + ldd [%o0 + 0xb0], %f22 + AES_DROUND01(14, 4, 6, 0) + ldd [%o0 + 0xa8], %f24 + AES_DROUND23(16, 0, 2, 6) + ldd [%o0 + 0xa0], %f26 + AES_DROUND01(18, 0, 2, 4) + ldd [%o0 + 0x98], %f12 + AES_DROUND23(20, 4, 6, 2) + ldd [%o0 + 0x90], %f14 + AES_DROUND01(22, 4, 6, 0) + ldd [%o0 + 0x88], %f16 + AES_DROUND23(24, 0, 2, 6) + ldd [%o0 + 0x80], %f18 + AES_DROUND01(26, 0, 2, 4) + ldd [%o0 + 0x78], %f20 + AES_DROUND23(12, 4, 6, 2) + ldd [%o0 + 0x70], %f22 + AES_DROUND01(14, 4, 6, 0) + ldd [%o0 + 0x68], %f24 + AES_DROUND23(16, 0, 2, 6) + ldd [%o0 + 0x60], %f26 + AES_DROUND01(18, 0, 2, 4) + ldd [%o0 + 0x58], %f28 + AES_DROUND23(20, 4, 6, 2) + ldd [%o0 + 0x50], %f30 + AES_DROUND01(22, 4, 6, 0) + ldd [%o0 + 0x48], %f32 + AES_DROUND23(24, 0, 2, 6) + ldd [%o0 + 0x40], %f34 + AES_DROUND01(26, 0, 2, 4) + ldd [%o0 + 0x38], %f36 + AES_DROUND23(28, 4, 6, 2) + ldd [%o0 + 0x30], %f38 + AES_DROUND01(30, 4, 6, 0) + ldd [%o0 + 0x28], %f40 + AES_DROUND23(32, 0, 2, 6) + ldd [%o0 + 0x20], %f42 + AES_DROUND01(34, 0, 2, 4) + ldd [%o0 + 0x18], %f44 + AES_DROUND23(36, 4, 6, 2) + ldd [%o0 + 0x10], %f46 + AES_DROUND01(38, 4, 6, 0) + ldd [%o0 + 0x08], %f48 + AES_DROUND23(40, 0, 2, 6) + ldd [%o0 + 0x00], %f50 + AES_DROUND01(42, 0, 2, 4) + AES_DROUND23(44, 4, 6, 2) + AES_DROUND01(46, 4, 6, 0) + AES_DROUND23_L(48, 0, 2, 6) + AES_DROUND01_L(50, 0, 2, 4) + st %f4, [%o2 + 0x00] + st %f5, [%o2 + 0x04] + st %f6, [%o2 + 0x08] + st %f7, [%o2 + 0x0c] + retl + VISExit +ENDPROC(aes_sparc64_decrypt_256) + + .align 32 +ENTRY(aes_sparc64_load_encrypt_keys_128) + /* %o0=key */ + VISEntry + ldd [%o0 + 0x10], %f8 + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + ldd [%o0 + 0x30], %f16 + ldd [%o0 + 0x38], %f18 + ldd [%o0 + 0x40], %f20 + ldd [%o0 + 0x48], %f22 + ldd [%o0 + 0x50], %f24 + ldd [%o0 + 0x58], %f26 + ldd [%o0 + 0x60], %f28 + ldd [%o0 + 0x68], %f30 + ldd [%o0 + 0x70], %f32 + ldd [%o0 + 0x78], %f34 + ldd [%o0 + 0x80], %f36 + ldd [%o0 + 0x88], %f38 + ldd [%o0 + 0x90], %f40 + ldd [%o0 + 0x98], %f42 + ldd [%o0 + 0xa0], %f44 + retl + ldd [%o0 + 0xa8], %f46 +ENDPROC(aes_sparc64_load_encrypt_keys_128) + + .align 32 +ENTRY(aes_sparc64_load_encrypt_keys_192) + /* %o0=key */ + VISEntry + ldd [%o0 + 0x10], %f8 + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + ldd [%o0 + 0x30], %f16 + ldd [%o0 + 0x38], %f18 + ldd [%o0 + 0x40], %f20 + ldd [%o0 + 0x48], %f22 + ldd [%o0 + 0x50], %f24 + ldd [%o0 + 0x58], %f26 + ldd [%o0 + 0x60], %f28 + ldd [%o0 + 0x68], %f30 + ldd [%o0 + 0x70], %f32 + ldd [%o0 + 0x78], %f34 + ldd [%o0 + 0x80], %f36 + ldd [%o0 + 0x88], %f38 + ldd [%o0 + 0x90], %f40 + ldd [%o0 + 0x98], %f42 + ldd [%o0 + 0xa0], %f44 + ldd [%o0 + 0xa8], %f46 + ldd [%o0 + 0xb0], %f48 + ldd [%o0 + 0xb8], %f50 + ldd [%o0 + 0xc0], %f52 + retl + ldd [%o0 + 0xc8], %f54 +ENDPROC(aes_sparc64_load_encrypt_keys_192) + + .align 32 +ENTRY(aes_sparc64_load_encrypt_keys_256) + /* %o0=key */ + VISEntry + ldd [%o0 + 0x10], %f8 + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + ldd [%o0 + 0x30], %f16 + ldd [%o0 + 0x38], %f18 + ldd [%o0 + 0x40], %f20 + ldd [%o0 + 0x48], %f22 + ldd [%o0 + 0x50], %f24 + ldd [%o0 + 0x58], %f26 + ldd [%o0 + 0x60], %f28 + ldd [%o0 + 0x68], %f30 + ldd [%o0 + 0x70], %f32 + ldd [%o0 + 0x78], %f34 + ldd [%o0 + 0x80], %f36 + ldd [%o0 + 0x88], %f38 + ldd [%o0 + 0x90], %f40 + ldd [%o0 + 0x98], %f42 + ldd [%o0 + 0xa0], %f44 + ldd [%o0 + 0xa8], %f46 + ldd [%o0 + 0xb0], %f48 + ldd [%o0 + 0xb8], %f50 + ldd [%o0 + 0xc0], %f52 + ldd [%o0 + 0xc8], %f54 + ldd [%o0 + 0xd0], %f56 + ldd [%o0 + 0xd8], %f58 + ldd [%o0 + 0xe0], %f60 + retl + ldd [%o0 + 0xe8], %f62 +ENDPROC(aes_sparc64_load_encrypt_keys_256) + + .align 32 +ENTRY(aes_sparc64_load_decrypt_keys_128) + /* %o0=key */ + VISEntry + ldd [%o0 + 0x98], %f8 + ldd [%o0 + 0x90], %f10 + ldd [%o0 + 0x88], %f12 + ldd [%o0 + 0x80], %f14 + ldd [%o0 + 0x78], %f16 + ldd [%o0 + 0x70], %f18 + ldd [%o0 + 0x68], %f20 + ldd [%o0 + 0x60], %f22 + ldd [%o0 + 0x58], %f24 + ldd [%o0 + 0x50], %f26 + ldd [%o0 + 0x48], %f28 + ldd [%o0 + 0x40], %f30 + ldd [%o0 + 0x38], %f32 + ldd [%o0 + 0x30], %f34 + ldd [%o0 + 0x28], %f36 + ldd [%o0 + 0x20], %f38 + ldd [%o0 + 0x18], %f40 + ldd [%o0 + 0x10], %f42 + ldd [%o0 + 0x08], %f44 + retl + ldd [%o0 + 0x00], %f46 +ENDPROC(aes_sparc64_load_decrypt_keys_128) + + .align 32 +ENTRY(aes_sparc64_load_decrypt_keys_192) + /* %o0=key */ + VISEntry + ldd [%o0 + 0xb8], %f8 + ldd [%o0 + 0xb0], %f10 + ldd [%o0 + 0xa8], %f12 + ldd [%o0 + 0xa0], %f14 + ldd [%o0 + 0x98], %f16 + ldd [%o0 + 0x90], %f18 + ldd [%o0 + 0x88], %f20 + ldd [%o0 + 0x80], %f22 + ldd [%o0 + 0x78], %f24 + ldd [%o0 + 0x70], %f26 + ldd [%o0 + 0x68], %f28 + ldd [%o0 + 0x60], %f30 + ldd [%o0 + 0x58], %f32 + ldd [%o0 + 0x50], %f34 + ldd [%o0 + 0x48], %f36 + ldd [%o0 + 0x40], %f38 + ldd [%o0 + 0x38], %f40 + ldd [%o0 + 0x30], %f42 + ldd [%o0 + 0x28], %f44 + ldd [%o0 + 0x20], %f46 + ldd [%o0 + 0x18], %f48 + ldd [%o0 + 0x10], %f50 + ldd [%o0 + 0x08], %f52 + retl + ldd [%o0 + 0x00], %f54 +ENDPROC(aes_sparc64_load_decrypt_keys_192) + + .align 32 +ENTRY(aes_sparc64_load_decrypt_keys_256) + /* %o0=key */ + VISEntry + ldd [%o0 + 0xd8], %f8 + ldd [%o0 + 0xd0], %f10 + ldd [%o0 + 0xc8], %f12 + ldd [%o0 + 0xc0], %f14 + ldd [%o0 + 0xb8], %f16 + ldd [%o0 + 0xb0], %f18 + ldd [%o0 + 0xa8], %f20 + ldd [%o0 + 0xa0], %f22 + ldd [%o0 + 0x98], %f24 + ldd [%o0 + 0x90], %f26 + ldd [%o0 + 0x88], %f28 + ldd [%o0 + 0x80], %f30 + ldd [%o0 + 0x78], %f32 + ldd [%o0 + 0x70], %f34 + ldd [%o0 + 0x68], %f36 + ldd [%o0 + 0x60], %f38 + ldd [%o0 + 0x58], %f40 + ldd [%o0 + 0x50], %f42 + ldd [%o0 + 0x48], %f44 + ldd [%o0 + 0x40], %f46 + ldd [%o0 + 0x38], %f48 + ldd [%o0 + 0x30], %f50 + ldd [%o0 + 0x28], %f52 + ldd [%o0 + 0x20], %f54 + ldd [%o0 + 0x18], %f56 + ldd [%o0 + 0x10], %f58 + ldd [%o0 + 0x08], %f60 + retl + ldd [%o0 + 0x00], %f62 +ENDPROC(aes_sparc64_load_decrypt_keys_256) + + .align 32 +ENTRY(aes_sparc64_ecb_encrypt_128) + /* %o0=key, %o1=input, %o2=output, %o3=len */ + ldx [%o0 + 0x00], %g1 + subcc %o3, 0x10, %o3 + be 10f + ldx [%o0 + 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + ldx [%o1 + 0x10], %o4 + ldx [%o1 + 0x18], %o5 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + xor %g1, %o4, %g3 + xor %g2, %o5, %g7 + MOVXTOD_G3_F60 + MOVXTOD_G7_F62 + ENCRYPT_128_2(8, 4, 6, 60, 62, 0, 2, 56, 58) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + std %f60, [%o2 + 0x10] + std %f62, [%o2 + 0x18] + sub %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + ENCRYPT_128(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: retl + nop +ENDPROC(aes_sparc64_ecb_encrypt_128) + + .align 32 +ENTRY(aes_sparc64_ecb_encrypt_192) + /* %o0=key, %o1=input, %o2=output, %o3=len */ + ldx [%o0 + 0x00], %g1 + subcc %o3, 0x10, %o3 + be 10f + ldx [%o0 + 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + ldx [%o1 + 0x10], %o4 + ldx [%o1 + 0x18], %o5 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + xor %g1, %o4, %g3 + xor %g2, %o5, %g7 + MOVXTOD_G3_F60 + MOVXTOD_G7_F62 + ENCRYPT_192_2(8, 4, 6, 60, 62, 0, 2, 56, 58) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + std %f60, [%o2 + 0x10] + std %f62, [%o2 + 0x18] + sub %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + ENCRYPT_192(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: retl + nop +ENDPROC(aes_sparc64_ecb_encrypt_192) + + .align 32 +ENTRY(aes_sparc64_ecb_encrypt_256) + /* %o0=key, %o1=input, %o2=output, %o3=len */ + ldx [%o0 + 0x00], %g1 + subcc %o3, 0x10, %o3 + be 10f + ldx [%o0 + 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + ldx [%o1 + 0x10], %o4 + ldx [%o1 + 0x18], %o5 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + xor %g1, %o4, %g3 + xor %g2, %o5, %g7 + MOVXTOD_G3_F0 + MOVXTOD_G7_F2 + ENCRYPT_256_2(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + std %f0, [%o2 + 0x10] + std %f2, [%o2 + 0x18] + sub %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: ldd [%o0 + 0xd0], %f56 + ldd [%o0 + 0xd8], %f58 + ldd [%o0 + 0xe0], %f60 + ldd [%o0 + 0xe8], %f62 + ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + ENCRYPT_256(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: retl + nop +ENDPROC(aes_sparc64_ecb_encrypt_256) + + .align 32 +ENTRY(aes_sparc64_ecb_decrypt_128) + /* %o0=&key[key_len], %o1=input, %o2=output, %o3=len */ + ldx [%o0 - 0x10], %g1 + subcc %o3, 0x10, %o3 + be 10f + ldx [%o0 - 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + ldx [%o1 + 0x10], %o4 + ldx [%o1 + 0x18], %o5 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + xor %g1, %o4, %g3 + xor %g2, %o5, %g7 + MOVXTOD_G3_F60 + MOVXTOD_G7_F62 + DECRYPT_128_2(8, 4, 6, 60, 62, 0, 2, 56, 58) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + std %f60, [%o2 + 0x10] + std %f62, [%o2 + 0x18] + sub %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz,pt %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + DECRYPT_128(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: retl + nop +ENDPROC(aes_sparc64_ecb_decrypt_128) + + .align 32 +ENTRY(aes_sparc64_ecb_decrypt_192) + /* %o0=&key[key_len], %o1=input, %o2=output, %o3=len */ + ldx [%o0 - 0x10], %g1 + subcc %o3, 0x10, %o3 + be 10f + ldx [%o0 - 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + ldx [%o1 + 0x10], %o4 + ldx [%o1 + 0x18], %o5 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + xor %g1, %o4, %g3 + xor %g2, %o5, %g7 + MOVXTOD_G3_F60 + MOVXTOD_G7_F62 + DECRYPT_192_2(8, 4, 6, 60, 62, 0, 2, 56, 58) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + std %f60, [%o2 + 0x10] + std %f62, [%o2 + 0x18] + sub %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz,pt %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + DECRYPT_192(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: retl + nop +ENDPROC(aes_sparc64_ecb_decrypt_192) + + .align 32 +ENTRY(aes_sparc64_ecb_decrypt_256) + /* %o0=&key[key_len], %o1=input, %o2=output, %o3=len */ + ldx [%o0 - 0x10], %g1 + subcc %o3, 0x10, %o3 + ldx [%o0 - 0x08], %g2 + be 10f + sub %o0, 0xf0, %o0 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + ldx [%o1 + 0x10], %o4 + ldx [%o1 + 0x18], %o5 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + xor %g1, %o4, %g3 + xor %g2, %o5, %g7 + MOVXTOD_G3_F0 + MOVXTOD_G7_F2 + DECRYPT_256_2(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + std %f0, [%o2 + 0x10] + std %f2, [%o2 + 0x18] + sub %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz,pt %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: ldd [%o0 + 0x18], %f56 + ldd [%o0 + 0x10], %f58 + ldd [%o0 + 0x08], %f60 + ldd [%o0 + 0x00], %f62 + ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + DECRYPT_256(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: retl + nop +ENDPROC(aes_sparc64_ecb_decrypt_256) + + .align 32 +ENTRY(aes_sparc64_cbc_encrypt_128) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldd [%o4 + 0x00], %f4 + ldd [%o4 + 0x08], %f6 + ldx [%o0 + 0x00], %g1 + ldx [%o0 + 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + add %o1, 0x10, %o1 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F0 + MOVXTOD_G7_F2 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + ENCRYPT_128(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + subcc %o3, 0x10, %o3 + bne,pt %xcc, 1b + add %o2, 0x10, %o2 + std %f4, [%o4 + 0x00] + std %f6, [%o4 + 0x08] + retl + nop +ENDPROC(aes_sparc64_cbc_encrypt_128) + + .align 32 +ENTRY(aes_sparc64_cbc_encrypt_192) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldd [%o4 + 0x00], %f4 + ldd [%o4 + 0x08], %f6 + ldx [%o0 + 0x00], %g1 + ldx [%o0 + 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + add %o1, 0x10, %o1 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F0 + MOVXTOD_G7_F2 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + ENCRYPT_192(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + subcc %o3, 0x10, %o3 + bne,pt %xcc, 1b + add %o2, 0x10, %o2 + std %f4, [%o4 + 0x00] + std %f6, [%o4 + 0x08] + retl + nop +ENDPROC(aes_sparc64_cbc_encrypt_192) + + .align 32 +ENTRY(aes_sparc64_cbc_encrypt_256) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldd [%o4 + 0x00], %f4 + ldd [%o4 + 0x08], %f6 + ldx [%o0 + 0x00], %g1 + ldx [%o0 + 0x08], %g2 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + add %o1, 0x10, %o1 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F0 + MOVXTOD_G7_F2 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + ENCRYPT_256(8, 4, 6, 0, 2) + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + subcc %o3, 0x10, %o3 + bne,pt %xcc, 1b + add %o2, 0x10, %o2 + std %f4, [%o4 + 0x00] + std %f6, [%o4 + 0x08] + retl + nop +ENDPROC(aes_sparc64_cbc_encrypt_256) + + .align 32 +ENTRY(aes_sparc64_cbc_decrypt_128) + /* %o0=&key[key_len], %o1=input, %o2=output, %o3=len, %o4=iv */ + ldx [%o0 - 0x10], %g1 + ldx [%o0 - 0x08], %g2 + ldx [%o4 + 0x00], %o0 + ldx [%o4 + 0x08], %o5 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + add %o1, 0x10, %o1 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + DECRYPT_128(8, 4, 6, 0, 2) + MOVXTOD_O0_F0 + MOVXTOD_O5_F2 + xor %g1, %g3, %o0 + xor %g2, %g7, %o5 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + subcc %o3, 0x10, %o3 + bne,pt %xcc, 1b + add %o2, 0x10, %o2 + stx %o0, [%o4 + 0x00] + stx %o5, [%o4 + 0x08] + retl + nop +ENDPROC(aes_sparc64_cbc_decrypt_128) + + .align 32 +ENTRY(aes_sparc64_cbc_decrypt_192) + /* %o0=&key[key_len], %o1=input, %o2=output, %o3=len, %o4=iv */ + ldx [%o0 - 0x10], %g1 + ldx [%o0 - 0x08], %g2 + ldx [%o4 + 0x00], %o0 + ldx [%o4 + 0x08], %o5 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + add %o1, 0x10, %o1 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + DECRYPT_192(8, 4, 6, 0, 2) + MOVXTOD_O0_F0 + MOVXTOD_O5_F2 + xor %g1, %g3, %o0 + xor %g2, %g7, %o5 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + subcc %o3, 0x10, %o3 + bne,pt %xcc, 1b + add %o2, 0x10, %o2 + stx %o0, [%o4 + 0x00] + stx %o5, [%o4 + 0x08] + retl + nop +ENDPROC(aes_sparc64_cbc_decrypt_192) + + .align 32 +ENTRY(aes_sparc64_cbc_decrypt_256) + /* %o0=&key[key_len], %o1=input, %o2=output, %o3=len, %o4=iv */ + ldx [%o0 - 0x10], %g1 + ldx [%o0 - 0x08], %g2 + ldx [%o4 + 0x00], %o0 + ldx [%o4 + 0x08], %o5 +1: ldx [%o1 + 0x00], %g3 + ldx [%o1 + 0x08], %g7 + add %o1, 0x10, %o1 + xor %g1, %g3, %g3 + xor %g2, %g7, %g7 + MOVXTOD_G3_F4 + MOVXTOD_G7_F6 + DECRYPT_256(8, 4, 6, 0, 2) + MOVXTOD_O0_F0 + MOVXTOD_O5_F2 + xor %g1, %g3, %o0 + xor %g2, %g7, %o5 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] + subcc %o3, 0x10, %o3 + bne,pt %xcc, 1b + add %o2, 0x10, %o2 + stx %o0, [%o4 + 0x00] + stx %o5, [%o4 + 0x08] + retl + nop +ENDPROC(aes_sparc64_cbc_decrypt_256) + + .align 32 +ENTRY(aes_sparc64_ctr_crypt_128) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldx [%o4 + 0x00], %g3 + ldx [%o4 + 0x08], %g7 + subcc %o3, 0x10, %o3 + ldx [%o0 + 0x00], %g1 + be 10f + ldx [%o0 + 0x08], %g2 +1: xor %g1, %g3, %o5 + MOVXTOD_O5_F0 + xor %g2, %g7, %o5 + MOVXTOD_O5_F2 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + xor %g1, %g3, %o5 + MOVXTOD_O5_F4 + xor %g2, %g7, %o5 + MOVXTOD_O5_F6 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + ENCRYPT_128_2(8, 0, 2, 4, 6, 56, 58, 60, 62) + ldd [%o1 + 0x00], %f56 + ldd [%o1 + 0x08], %f58 + ldd [%o1 + 0x10], %f60 + ldd [%o1 + 0x18], %f62 + fxor %f56, %f0, %f56 + fxor %f58, %f2, %f58 + fxor %f60, %f4, %f60 + fxor %f62, %f6, %f62 + std %f56, [%o2 + 0x00] + std %f58, [%o2 + 0x08] + std %f60, [%o2 + 0x10] + std %f62, [%o2 + 0x18] + subcc %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: xor %g1, %g3, %o5 + MOVXTOD_O5_F0 + xor %g2, %g7, %o5 + MOVXTOD_O5_F2 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + ENCRYPT_128(8, 0, 2, 4, 6) + ldd [%o1 + 0x00], %f4 + ldd [%o1 + 0x08], %f6 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: stx %g3, [%o4 + 0x00] + retl + stx %g7, [%o4 + 0x08] +ENDPROC(aes_sparc64_ctr_crypt_128) + + .align 32 +ENTRY(aes_sparc64_ctr_crypt_192) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldx [%o4 + 0x00], %g3 + ldx [%o4 + 0x08], %g7 + subcc %o3, 0x10, %o3 + ldx [%o0 + 0x00], %g1 + be 10f + ldx [%o0 + 0x08], %g2 +1: xor %g1, %g3, %o5 + MOVXTOD_O5_F0 + xor %g2, %g7, %o5 + MOVXTOD_O5_F2 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + xor %g1, %g3, %o5 + MOVXTOD_O5_F4 + xor %g2, %g7, %o5 + MOVXTOD_O5_F6 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + ENCRYPT_192_2(8, 0, 2, 4, 6, 56, 58, 60, 62) + ldd [%o1 + 0x00], %f56 + ldd [%o1 + 0x08], %f58 + ldd [%o1 + 0x10], %f60 + ldd [%o1 + 0x18], %f62 + fxor %f56, %f0, %f56 + fxor %f58, %f2, %f58 + fxor %f60, %f4, %f60 + fxor %f62, %f6, %f62 + std %f56, [%o2 + 0x00] + std %f58, [%o2 + 0x08] + std %f60, [%o2 + 0x10] + std %f62, [%o2 + 0x18] + subcc %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: xor %g1, %g3, %o5 + MOVXTOD_O5_F0 + xor %g2, %g7, %o5 + MOVXTOD_O5_F2 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + ENCRYPT_192(8, 0, 2, 4, 6) + ldd [%o1 + 0x00], %f4 + ldd [%o1 + 0x08], %f6 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: stx %g3, [%o4 + 0x00] + retl + stx %g7, [%o4 + 0x08] +ENDPROC(aes_sparc64_ctr_crypt_192) + + .align 32 +ENTRY(aes_sparc64_ctr_crypt_256) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldx [%o4 + 0x00], %g3 + ldx [%o4 + 0x08], %g7 + subcc %o3, 0x10, %o3 + ldx [%o0 + 0x00], %g1 + be 10f + ldx [%o0 + 0x08], %g2 +1: xor %g1, %g3, %o5 + MOVXTOD_O5_F0 + xor %g2, %g7, %o5 + MOVXTOD_O5_F2 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + xor %g1, %g3, %o5 + MOVXTOD_O5_F4 + xor %g2, %g7, %o5 + MOVXTOD_O5_F6 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + ENCRYPT_256_2(8, 0, 2, 4, 6) + ldd [%o1 + 0x00], %f56 + ldd [%o1 + 0x08], %f58 + ldd [%o1 + 0x10], %f60 + ldd [%o1 + 0x18], %f62 + fxor %f56, %f0, %f56 + fxor %f58, %f2, %f58 + fxor %f60, %f4, %f60 + fxor %f62, %f6, %f62 + std %f56, [%o2 + 0x00] + std %f58, [%o2 + 0x08] + std %f60, [%o2 + 0x10] + std %f62, [%o2 + 0x18] + subcc %o3, 0x20, %o3 + add %o1, 0x20, %o1 + brgz %o3, 1b + add %o2, 0x20, %o2 + brlz,pt %o3, 11f + nop +10: ldd [%o0 + 0xd0], %f56 + ldd [%o0 + 0xd8], %f58 + ldd [%o0 + 0xe0], %f60 + ldd [%o0 + 0xe8], %f62 + xor %g1, %g3, %o5 + MOVXTOD_O5_F0 + xor %g2, %g7, %o5 + MOVXTOD_O5_F2 + add %g7, 1, %g7 + add %g3, 1, %o5 + movrz %g7, %o5, %g3 + ENCRYPT_256(8, 0, 2, 4, 6) + ldd [%o1 + 0x00], %f4 + ldd [%o1 + 0x08], %f6 + fxor %f4, %f0, %f4 + fxor %f6, %f2, %f6 + std %f4, [%o2 + 0x00] + std %f6, [%o2 + 0x08] +11: stx %g3, [%o4 + 0x00] + retl + stx %g7, [%o4 + 0x08] +ENDPROC(aes_sparc64_ctr_crypt_256) diff --git a/arch/sparc/crypto/aes_glue.c b/arch/sparc/crypto/aes_glue.c new file mode 100644 index 00000000000..df922f52d76 --- /dev/null +++ b/arch/sparc/crypto/aes_glue.c @@ -0,0 +1,504 @@ +/* Glue code for AES encryption optimized for sparc64 crypto opcodes. + * + * This is based largely upon arch/x86/crypto/aesni-intel_glue.c + * + * Copyright (C) 2008, Intel Corp. + * Author: Huang Ying <ying.huang@intel.com> + * + * Added RFC4106 AES-GCM support for 128-bit keys under the AEAD + * interface for 64-bit kernels. + * Authors: Adrian Hoban <adrian.hoban@intel.com> + * Gabriele Paoloni <gabriele.paoloni@intel.com> + * Tadeusz Struk (tadeusz.struk@intel.com) + * Aidan O'Mahony (aidan.o.mahony@intel.com) + * Copyright (c) 2010, Intel Corporation. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/crypto.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/types.h> +#include <crypto/algapi.h> +#include <crypto/aes.h> + +#include <asm/fpumacro.h> +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +struct aes_ops { + void (*encrypt)(const u64 *key, const u32 *input, u32 *output); + void (*decrypt)(const u64 *key, const u32 *input, u32 *output); + void (*load_encrypt_keys)(const u64 *key); + void (*load_decrypt_keys)(const u64 *key); + void (*ecb_encrypt)(const u64 *key, const u64 *input, u64 *output, + unsigned int len); + void (*ecb_decrypt)(const u64 *key, const u64 *input, u64 *output, + unsigned int len); + void (*cbc_encrypt)(const u64 *key, const u64 *input, u64 *output, + unsigned int len, u64 *iv); + void (*cbc_decrypt)(const u64 *key, const u64 *input, u64 *output, + unsigned int len, u64 *iv); + void (*ctr_crypt)(const u64 *key, const u64 *input, u64 *output, + unsigned int len, u64 *iv); +}; + +struct crypto_sparc64_aes_ctx { + struct aes_ops *ops; + u64 key[AES_MAX_KEYLENGTH / sizeof(u64)]; + u32 key_length; + u32 expanded_key_length; +}; + +extern void aes_sparc64_encrypt_128(const u64 *key, const u32 *input, + u32 *output); +extern void aes_sparc64_encrypt_192(const u64 *key, const u32 *input, + u32 *output); +extern void aes_sparc64_encrypt_256(const u64 *key, const u32 *input, + u32 *output); + +extern void aes_sparc64_decrypt_128(const u64 *key, const u32 *input, + u32 *output); +extern void aes_sparc64_decrypt_192(const u64 *key, const u32 *input, + u32 *output); +extern void aes_sparc64_decrypt_256(const u64 *key, const u32 *input, + u32 *output); + +extern void aes_sparc64_load_encrypt_keys_128(const u64 *key); +extern void aes_sparc64_load_encrypt_keys_192(const u64 *key); +extern void aes_sparc64_load_encrypt_keys_256(const u64 *key); + +extern void aes_sparc64_load_decrypt_keys_128(const u64 *key); +extern void aes_sparc64_load_decrypt_keys_192(const u64 *key); +extern void aes_sparc64_load_decrypt_keys_256(const u64 *key); + +extern void aes_sparc64_ecb_encrypt_128(const u64 *key, const u64 *input, + u64 *output, unsigned int len); +extern void aes_sparc64_ecb_encrypt_192(const u64 *key, const u64 *input, + u64 *output, unsigned int len); +extern void aes_sparc64_ecb_encrypt_256(const u64 *key, const u64 *input, + u64 *output, unsigned int len); + +extern void aes_sparc64_ecb_decrypt_128(const u64 *key, const u64 *input, + u64 *output, unsigned int len); +extern void aes_sparc64_ecb_decrypt_192(const u64 *key, const u64 *input, + u64 *output, unsigned int len); +extern void aes_sparc64_ecb_decrypt_256(const u64 *key, const u64 *input, + u64 *output, unsigned int len); + +extern void aes_sparc64_cbc_encrypt_128(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +extern void aes_sparc64_cbc_encrypt_192(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +extern void aes_sparc64_cbc_encrypt_256(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +extern void aes_sparc64_cbc_decrypt_128(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +extern void aes_sparc64_cbc_decrypt_192(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +extern void aes_sparc64_cbc_decrypt_256(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +extern void aes_sparc64_ctr_crypt_128(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); +extern void aes_sparc64_ctr_crypt_192(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); +extern void aes_sparc64_ctr_crypt_256(const u64 *key, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +static struct aes_ops aes128_ops = { + .encrypt = aes_sparc64_encrypt_128, + .decrypt = aes_sparc64_decrypt_128, + .load_encrypt_keys = aes_sparc64_load_encrypt_keys_128, + .load_decrypt_keys = aes_sparc64_load_decrypt_keys_128, + .ecb_encrypt = aes_sparc64_ecb_encrypt_128, + .ecb_decrypt = aes_sparc64_ecb_decrypt_128, + .cbc_encrypt = aes_sparc64_cbc_encrypt_128, + .cbc_decrypt = aes_sparc64_cbc_decrypt_128, + .ctr_crypt = aes_sparc64_ctr_crypt_128, +}; + +static struct aes_ops aes192_ops = { + .encrypt = aes_sparc64_encrypt_192, + .decrypt = aes_sparc64_decrypt_192, + .load_encrypt_keys = aes_sparc64_load_encrypt_keys_192, + .load_decrypt_keys = aes_sparc64_load_decrypt_keys_192, + .ecb_encrypt = aes_sparc64_ecb_encrypt_192, + .ecb_decrypt = aes_sparc64_ecb_decrypt_192, + .cbc_encrypt = aes_sparc64_cbc_encrypt_192, + .cbc_decrypt = aes_sparc64_cbc_decrypt_192, + .ctr_crypt = aes_sparc64_ctr_crypt_192, +}; + +static struct aes_ops aes256_ops = { + .encrypt = aes_sparc64_encrypt_256, + .decrypt = aes_sparc64_decrypt_256, + .load_encrypt_keys = aes_sparc64_load_encrypt_keys_256, + .load_decrypt_keys = aes_sparc64_load_decrypt_keys_256, + .ecb_encrypt = aes_sparc64_ecb_encrypt_256, + .ecb_decrypt = aes_sparc64_ecb_decrypt_256, + .cbc_encrypt = aes_sparc64_cbc_encrypt_256, + .cbc_decrypt = aes_sparc64_cbc_decrypt_256, + .ctr_crypt = aes_sparc64_ctr_crypt_256, +}; + +extern void aes_sparc64_key_expand(const u32 *in_key, u64 *output_key, + unsigned int key_len); + +static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + + switch (key_len) { + case AES_KEYSIZE_128: + ctx->expanded_key_length = 0xb0; + ctx->ops = &aes128_ops; + break; + + case AES_KEYSIZE_192: + ctx->expanded_key_length = 0xd0; + ctx->ops = &aes192_ops; + break; + + case AES_KEYSIZE_256: + ctx->expanded_key_length = 0xf0; + ctx->ops = &aes256_ops; + break; + + default: + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + + aes_sparc64_key_expand((const u32 *)in_key, &ctx->key[0], key_len); + ctx->key_length = key_len; + + return 0; +} + +static void aes_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_tfm_ctx(tfm); + + ctx->ops->encrypt(&ctx->key[0], (const u32 *) src, (u32 *) dst); +} + +static void aes_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_tfm_ctx(tfm); + + ctx->ops->decrypt(&ctx->key[0], (const u32 *) src, (u32 *) dst); +} + +#define AES_BLOCK_MASK (~(AES_BLOCK_SIZE-1)) + +static int ecb_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + ctx->ops->load_encrypt_keys(&ctx->key[0]); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & AES_BLOCK_MASK; + + if (likely(block_len)) { + ctx->ops->ecb_encrypt(&ctx->key[0], + (const u64 *)walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, + block_len); + } + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static int ecb_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + u64 *key_end; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + ctx->ops->load_decrypt_keys(&ctx->key[0]); + key_end = &ctx->key[ctx->expanded_key_length / sizeof(u64)]; + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & AES_BLOCK_MASK; + + if (likely(block_len)) { + ctx->ops->ecb_decrypt(key_end, + (const u64 *) walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, block_len); + } + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + + return err; +} + +static int cbc_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + ctx->ops->load_encrypt_keys(&ctx->key[0]); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & AES_BLOCK_MASK; + + if (likely(block_len)) { + ctx->ops->cbc_encrypt(&ctx->key[0], + (const u64 *)walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, + block_len, (u64 *) walk.iv); + } + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static int cbc_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + u64 *key_end; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + ctx->ops->load_decrypt_keys(&ctx->key[0]); + key_end = &ctx->key[ctx->expanded_key_length / sizeof(u64)]; + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & AES_BLOCK_MASK; + + if (likely(block_len)) { + ctx->ops->cbc_decrypt(key_end, + (const u64 *) walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, + block_len, (u64 *) walk.iv); + } + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + + return err; +} + +static void ctr_crypt_final(struct crypto_sparc64_aes_ctx *ctx, + struct blkcipher_walk *walk) +{ + u8 *ctrblk = walk->iv; + u64 keystream[AES_BLOCK_SIZE / sizeof(u64)]; + u8 *src = walk->src.virt.addr; + u8 *dst = walk->dst.virt.addr; + unsigned int nbytes = walk->nbytes; + + ctx->ops->ecb_encrypt(&ctx->key[0], (const u64 *)ctrblk, + keystream, AES_BLOCK_SIZE); + crypto_xor((u8 *) keystream, src, nbytes); + memcpy(dst, keystream, nbytes); + crypto_inc(ctrblk, AES_BLOCK_SIZE); +} + +static int ctr_crypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_sparc64_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, AES_BLOCK_SIZE); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + ctx->ops->load_encrypt_keys(&ctx->key[0]); + while ((nbytes = walk.nbytes) >= AES_BLOCK_SIZE) { + unsigned int block_len = nbytes & AES_BLOCK_MASK; + + if (likely(block_len)) { + ctx->ops->ctr_crypt(&ctx->key[0], + (const u64 *)walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, + block_len, (u64 *) walk.iv); + } + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + if (walk.nbytes) { + ctr_crypt_final(ctx, &walk); + err = blkcipher_walk_done(desc, &walk, 0); + } + fprs_write(0); + return err; +} + +static struct crypto_alg algs[] = { { + .cra_name = "aes", + .cra_driver_name = "aes-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto_sparc64_aes_ctx), + .cra_alignmask = 3, + .cra_module = THIS_MODULE, + .cra_u = { + .cipher = { + .cia_min_keysize = AES_MIN_KEY_SIZE, + .cia_max_keysize = AES_MAX_KEY_SIZE, + .cia_setkey = aes_set_key, + .cia_encrypt = aes_encrypt, + .cia_decrypt = aes_decrypt + } + } +}, { + .cra_name = "ecb(aes)", + .cra_driver_name = "ecb-aes-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto_sparc64_aes_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = aes_set_key, + .encrypt = ecb_encrypt, + .decrypt = ecb_decrypt, + }, + }, +}, { + .cra_name = "cbc(aes)", + .cra_driver_name = "cbc-aes-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto_sparc64_aes_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = aes_set_key, + .encrypt = cbc_encrypt, + .decrypt = cbc_decrypt, + }, + }, +}, { + .cra_name = "ctr(aes)", + .cra_driver_name = "ctr-aes-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct crypto_sparc64_aes_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = aes_set_key, + .encrypt = ctr_crypt, + .decrypt = ctr_crypt, + }, + }, +} }; + +static bool __init sparc64_has_aes_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_AES)) + return false; + + return true; +} + +static int __init aes_sparc64_mod_init(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(algs); i++) + INIT_LIST_HEAD(&algs[i].cra_list); + + if (sparc64_has_aes_opcode()) { + pr_info("Using sparc64 aes opcodes optimized AES implementation\n"); + return crypto_register_algs(algs, ARRAY_SIZE(algs)); + } + pr_info("sparc64 aes opcodes not available.\n"); + return -ENODEV; +} + +static void __exit aes_sparc64_mod_fini(void) +{ + crypto_unregister_algs(algs, ARRAY_SIZE(algs)); +} + +module_init(aes_sparc64_mod_init); +module_exit(aes_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("AES Secure Hash Algorithm, sparc64 aes opcode accelerated"); + +MODULE_ALIAS("aes"); + +#include "crop_devid.c" diff --git a/arch/sparc/crypto/camellia_asm.S b/arch/sparc/crypto/camellia_asm.S new file mode 100644 index 00000000000..cc39553a4e4 --- /dev/null +++ b/arch/sparc/crypto/camellia_asm.S @@ -0,0 +1,563 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> + +#include "opcodes.h" + +#define CAMELLIA_6ROUNDS(KEY_BASE, I0, I1) \ + CAMELLIA_F(KEY_BASE + 0, I1, I0, I1) \ + CAMELLIA_F(KEY_BASE + 2, I0, I1, I0) \ + CAMELLIA_F(KEY_BASE + 4, I1, I0, I1) \ + CAMELLIA_F(KEY_BASE + 6, I0, I1, I0) \ + CAMELLIA_F(KEY_BASE + 8, I1, I0, I1) \ + CAMELLIA_F(KEY_BASE + 10, I0, I1, I0) + +#define CAMELLIA_6ROUNDS_FL_FLI(KEY_BASE, I0, I1) \ + CAMELLIA_6ROUNDS(KEY_BASE, I0, I1) \ + CAMELLIA_FL(KEY_BASE + 12, I0, I0) \ + CAMELLIA_FLI(KEY_BASE + 14, I1, I1) + + .data + + .align 8 +SIGMA: .xword 0xA09E667F3BCC908B + .xword 0xB67AE8584CAA73B2 + .xword 0xC6EF372FE94F82BE + .xword 0x54FF53A5F1D36F1C + .xword 0x10E527FADE682D1D + .xword 0xB05688C2B3E6C1FD + + .text + + .align 32 +ENTRY(camellia_sparc64_key_expand) + /* %o0=in_key, %o1=encrypt_key, %o2=key_len, %o3=decrypt_key */ + VISEntry + ld [%o0 + 0x00], %f0 ! i0, k[0] + ld [%o0 + 0x04], %f1 ! i1, k[1] + ld [%o0 + 0x08], %f2 ! i2, k[2] + ld [%o0 + 0x0c], %f3 ! i3, k[3] + std %f0, [%o1 + 0x00] ! k[0, 1] + fsrc2 %f0, %f28 + std %f2, [%o1 + 0x08] ! k[2, 3] + cmp %o2, 16 + be 10f + fsrc2 %f2, %f30 + + ld [%o0 + 0x10], %f0 + ld [%o0 + 0x14], %f1 + std %f0, [%o1 + 0x20] ! k[8, 9] + cmp %o2, 24 + fone %f10 + be,a 1f + fxor %f10, %f0, %f2 + ld [%o0 + 0x18], %f2 + ld [%o0 + 0x1c], %f3 +1: + std %f2, [%o1 + 0x28] ! k[10, 11] + fxor %f28, %f0, %f0 + fxor %f30, %f2, %f2 + +10: + sethi %hi(SIGMA), %g3 + or %g3, %lo(SIGMA), %g3 + ldd [%g3 + 0x00], %f16 + ldd [%g3 + 0x08], %f18 + ldd [%g3 + 0x10], %f20 + ldd [%g3 + 0x18], %f22 + ldd [%g3 + 0x20], %f24 + ldd [%g3 + 0x28], %f26 + CAMELLIA_F(16, 2, 0, 2) + CAMELLIA_F(18, 0, 2, 0) + fxor %f28, %f0, %f0 + fxor %f30, %f2, %f2 + CAMELLIA_F(20, 2, 0, 2) + CAMELLIA_F(22, 0, 2, 0) + +#define ROTL128(S01, S23, TMP1, TMP2, N) \ + srlx S01, (64 - N), TMP1; \ + sllx S01, N, S01; \ + srlx S23, (64 - N), TMP2; \ + sllx S23, N, S23; \ + or S01, TMP2, S01; \ + or S23, TMP1, S23 + + cmp %o2, 16 + bne 1f + nop + /* 128-bit key */ + std %f0, [%o1 + 0x10] ! k[ 4, 5] + std %f2, [%o1 + 0x18] ! k[ 6, 7] + MOVDTOX_F0_O4 + MOVDTOX_F2_O5 + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x30] ! k[12, 13] + stx %o5, [%o1 + 0x38] ! k[14, 15] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x40] ! k[16, 17] + stx %o5, [%o1 + 0x48] ! k[18, 19] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x60] ! k[24, 25] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x70] ! k[28, 29] + stx %o5, [%o1 + 0x78] ! k[30, 31] + ROTL128(%o4, %o5, %g2, %g3, 34) + stx %o4, [%o1 + 0xa0] ! k[40, 41] + stx %o5, [%o1 + 0xa8] ! k[42, 43] + ROTL128(%o4, %o5, %g2, %g3, 17) + stx %o4, [%o1 + 0xc0] ! k[48, 49] + stx %o5, [%o1 + 0xc8] ! k[50, 51] + + ldx [%o1 + 0x00], %o4 ! k[ 0, 1] + ldx [%o1 + 0x08], %o5 ! k[ 2, 3] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x20] ! k[ 8, 9] + stx %o5, [%o1 + 0x28] ! k[10, 11] + ROTL128(%o4, %o5, %g2, %g3, 30) + stx %o4, [%o1 + 0x50] ! k[20, 21] + stx %o5, [%o1 + 0x58] ! k[22, 23] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o5, [%o1 + 0x68] ! k[26, 27] + ROTL128(%o4, %o5, %g2, %g3, 17) + stx %o4, [%o1 + 0x80] ! k[32, 33] + stx %o5, [%o1 + 0x88] ! k[34, 35] + ROTL128(%o4, %o5, %g2, %g3, 17) + stx %o4, [%o1 + 0x90] ! k[36, 37] + stx %o5, [%o1 + 0x98] ! k[38, 39] + ROTL128(%o4, %o5, %g2, %g3, 17) + stx %o4, [%o1 + 0xb0] ! k[44, 45] + stx %o5, [%o1 + 0xb8] ! k[46, 47] + + ba,pt %xcc, 2f + mov (3 * 16 * 4), %o0 + +1: + /* 192-bit or 256-bit key */ + std %f0, [%o1 + 0x30] ! k[12, 13] + std %f2, [%o1 + 0x38] ! k[14, 15] + ldd [%o1 + 0x20], %f4 ! k[ 8, 9] + ldd [%o1 + 0x28], %f6 ! k[10, 11] + fxor %f0, %f4, %f0 + fxor %f2, %f6, %f2 + CAMELLIA_F(24, 2, 0, 2) + CAMELLIA_F(26, 0, 2, 0) + std %f0, [%o1 + 0x10] ! k[ 4, 5] + std %f2, [%o1 + 0x18] ! k[ 6, 7] + MOVDTOX_F0_O4 + MOVDTOX_F2_O5 + ROTL128(%o4, %o5, %g2, %g3, 30) + stx %o4, [%o1 + 0x50] ! k[20, 21] + stx %o5, [%o1 + 0x58] ! k[22, 23] + ROTL128(%o4, %o5, %g2, %g3, 30) + stx %o4, [%o1 + 0xa0] ! k[40, 41] + stx %o5, [%o1 + 0xa8] ! k[42, 43] + ROTL128(%o4, %o5, %g2, %g3, 51) + stx %o4, [%o1 + 0x100] ! k[64, 65] + stx %o5, [%o1 + 0x108] ! k[66, 67] + ldx [%o1 + 0x20], %o4 ! k[ 8, 9] + ldx [%o1 + 0x28], %o5 ! k[10, 11] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x20] ! k[ 8, 9] + stx %o5, [%o1 + 0x28] ! k[10, 11] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x40] ! k[16, 17] + stx %o5, [%o1 + 0x48] ! k[18, 19] + ROTL128(%o4, %o5, %g2, %g3, 30) + stx %o4, [%o1 + 0x90] ! k[36, 37] + stx %o5, [%o1 + 0x98] ! k[38, 39] + ROTL128(%o4, %o5, %g2, %g3, 34) + stx %o4, [%o1 + 0xd0] ! k[52, 53] + stx %o5, [%o1 + 0xd8] ! k[54, 55] + ldx [%o1 + 0x30], %o4 ! k[12, 13] + ldx [%o1 + 0x38], %o5 ! k[14, 15] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x30] ! k[12, 13] + stx %o5, [%o1 + 0x38] ! k[14, 15] + ROTL128(%o4, %o5, %g2, %g3, 30) + stx %o4, [%o1 + 0x70] ! k[28, 29] + stx %o5, [%o1 + 0x78] ! k[30, 31] + srlx %o4, 32, %g2 + srlx %o5, 32, %g3 + stw %o4, [%o1 + 0xc0] ! k[48] + stw %g3, [%o1 + 0xc4] ! k[49] + stw %o5, [%o1 + 0xc8] ! k[50] + stw %g2, [%o1 + 0xcc] ! k[51] + ROTL128(%o4, %o5, %g2, %g3, 49) + stx %o4, [%o1 + 0xe0] ! k[56, 57] + stx %o5, [%o1 + 0xe8] ! k[58, 59] + ldx [%o1 + 0x00], %o4 ! k[ 0, 1] + ldx [%o1 + 0x08], %o5 ! k[ 2, 3] + ROTL128(%o4, %o5, %g2, %g3, 45) + stx %o4, [%o1 + 0x60] ! k[24, 25] + stx %o5, [%o1 + 0x68] ! k[26, 27] + ROTL128(%o4, %o5, %g2, %g3, 15) + stx %o4, [%o1 + 0x80] ! k[32, 33] + stx %o5, [%o1 + 0x88] ! k[34, 35] + ROTL128(%o4, %o5, %g2, %g3, 17) + stx %o4, [%o1 + 0xb0] ! k[44, 45] + stx %o5, [%o1 + 0xb8] ! k[46, 47] + ROTL128(%o4, %o5, %g2, %g3, 34) + stx %o4, [%o1 + 0xf0] ! k[60, 61] + stx %o5, [%o1 + 0xf8] ! k[62, 63] + mov (4 * 16 * 4), %o0 +2: + add %o1, %o0, %o1 + ldd [%o1 + 0x00], %f0 + ldd [%o1 + 0x08], %f2 + std %f0, [%o3 + 0x00] + std %f2, [%o3 + 0x08] + add %o3, 0x10, %o3 +1: + sub %o1, (16 * 4), %o1 + ldd [%o1 + 0x38], %f0 + ldd [%o1 + 0x30], %f2 + ldd [%o1 + 0x28], %f4 + ldd [%o1 + 0x20], %f6 + ldd [%o1 + 0x18], %f8 + ldd [%o1 + 0x10], %f10 + std %f0, [%o3 + 0x00] + std %f2, [%o3 + 0x08] + std %f4, [%o3 + 0x10] + std %f6, [%o3 + 0x18] + std %f8, [%o3 + 0x20] + std %f10, [%o3 + 0x28] + + ldd [%o1 + 0x08], %f0 + ldd [%o1 + 0x00], %f2 + std %f0, [%o3 + 0x30] + std %f2, [%o3 + 0x38] + subcc %o0, (16 * 4), %o0 + bne,pt %icc, 1b + add %o3, (16 * 4), %o3 + + std %f2, [%o3 - 0x10] + std %f0, [%o3 - 0x08] + + retl + VISExit +ENDPROC(camellia_sparc64_key_expand) + + .align 32 +ENTRY(camellia_sparc64_crypt) + /* %o0=key, %o1=input, %o2=output, %o3=key_len */ + VISEntry + + ld [%o1 + 0x00], %f0 + ld [%o1 + 0x04], %f1 + ld [%o1 + 0x08], %f2 + ld [%o1 + 0x0c], %f3 + + ldd [%o0 + 0x00], %f4 + ldd [%o0 + 0x08], %f6 + + cmp %o3, 16 + fxor %f4, %f0, %f0 + be 1f + fxor %f6, %f2, %f2 + + ldd [%o0 + 0x10], %f8 + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + ldd [%o0 + 0x30], %f16 + ldd [%o0 + 0x38], %f18 + ldd [%o0 + 0x40], %f20 + ldd [%o0 + 0x48], %f22 + add %o0, 0x40, %o0 + + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + +1: + ldd [%o0 + 0x10], %f8 + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + ldd [%o0 + 0x30], %f16 + ldd [%o0 + 0x38], %f18 + ldd [%o0 + 0x40], %f20 + ldd [%o0 + 0x48], %f22 + ldd [%o0 + 0x50], %f24 + ldd [%o0 + 0x58], %f26 + ldd [%o0 + 0x60], %f28 + ldd [%o0 + 0x68], %f30 + ldd [%o0 + 0x70], %f32 + ldd [%o0 + 0x78], %f34 + ldd [%o0 + 0x80], %f36 + ldd [%o0 + 0x88], %f38 + ldd [%o0 + 0x90], %f40 + ldd [%o0 + 0x98], %f42 + ldd [%o0 + 0xa0], %f44 + ldd [%o0 + 0xa8], %f46 + ldd [%o0 + 0xb0], %f48 + ldd [%o0 + 0xb8], %f50 + ldd [%o0 + 0xc0], %f52 + ldd [%o0 + 0xc8], %f54 + + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + CAMELLIA_6ROUNDS_FL_FLI(24, 0, 2) + CAMELLIA_6ROUNDS(40, 0, 2) + fxor %f52, %f2, %f2 + fxor %f54, %f0, %f0 + + st %f2, [%o2 + 0x00] + st %f3, [%o2 + 0x04] + st %f0, [%o2 + 0x08] + st %f1, [%o2 + 0x0c] + + retl + VISExit +ENDPROC(camellia_sparc64_crypt) + + .align 32 +ENTRY(camellia_sparc64_load_keys) + /* %o0=key, %o1=key_len */ + VISEntry + ldd [%o0 + 0x00], %f4 + ldd [%o0 + 0x08], %f6 + ldd [%o0 + 0x10], %f8 + ldd [%o0 + 0x18], %f10 + ldd [%o0 + 0x20], %f12 + ldd [%o0 + 0x28], %f14 + ldd [%o0 + 0x30], %f16 + ldd [%o0 + 0x38], %f18 + ldd [%o0 + 0x40], %f20 + ldd [%o0 + 0x48], %f22 + ldd [%o0 + 0x50], %f24 + ldd [%o0 + 0x58], %f26 + ldd [%o0 + 0x60], %f28 + ldd [%o0 + 0x68], %f30 + ldd [%o0 + 0x70], %f32 + ldd [%o0 + 0x78], %f34 + ldd [%o0 + 0x80], %f36 + ldd [%o0 + 0x88], %f38 + ldd [%o0 + 0x90], %f40 + ldd [%o0 + 0x98], %f42 + ldd [%o0 + 0xa0], %f44 + ldd [%o0 + 0xa8], %f46 + ldd [%o0 + 0xb0], %f48 + ldd [%o0 + 0xb8], %f50 + ldd [%o0 + 0xc0], %f52 + retl + ldd [%o0 + 0xc8], %f54 +ENDPROC(camellia_sparc64_load_keys) + + .align 32 +ENTRY(camellia_sparc64_ecb_crypt_3_grand_rounds) + /* %o0=input, %o1=output, %o2=len, %o3=key */ +1: ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + add %o0, 0x10, %o0 + fxor %f4, %f0, %f0 + fxor %f6, %f2, %f2 + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + CAMELLIA_6ROUNDS_FL_FLI(24, 0, 2) + CAMELLIA_6ROUNDS(40, 0, 2) + fxor %f52, %f2, %f2 + fxor %f54, %f0, %f0 + std %f2, [%o1 + 0x00] + std %f0, [%o1 + 0x08] + subcc %o2, 0x10, %o2 + bne,pt %icc, 1b + add %o1, 0x10, %o1 + retl + nop +ENDPROC(camellia_sparc64_ecb_crypt_3_grand_rounds) + + .align 32 +ENTRY(camellia_sparc64_ecb_crypt_4_grand_rounds) + /* %o0=input, %o1=output, %o2=len, %o3=key */ +1: ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + add %o0, 0x10, %o0 + fxor %f4, %f0, %f0 + fxor %f6, %f2, %f2 + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + ldd [%o3 + 0xd0], %f8 + ldd [%o3 + 0xd8], %f10 + ldd [%o3 + 0xe0], %f12 + ldd [%o3 + 0xe8], %f14 + ldd [%o3 + 0xf0], %f16 + ldd [%o3 + 0xf8], %f18 + ldd [%o3 + 0x100], %f20 + ldd [%o3 + 0x108], %f22 + CAMELLIA_6ROUNDS_FL_FLI(24, 0, 2) + CAMELLIA_6ROUNDS_FL_FLI(40, 0, 2) + CAMELLIA_F(8, 2, 0, 2) + CAMELLIA_F(10, 0, 2, 0) + ldd [%o3 + 0x10], %f8 + ldd [%o3 + 0x18], %f10 + CAMELLIA_F(12, 2, 0, 2) + CAMELLIA_F(14, 0, 2, 0) + ldd [%o3 + 0x20], %f12 + ldd [%o3 + 0x28], %f14 + CAMELLIA_F(16, 2, 0, 2) + CAMELLIA_F(18, 0, 2, 0) + ldd [%o3 + 0x30], %f16 + ldd [%o3 + 0x38], %f18 + fxor %f20, %f2, %f2 + fxor %f22, %f0, %f0 + ldd [%o3 + 0x40], %f20 + ldd [%o3 + 0x48], %f22 + std %f2, [%o1 + 0x00] + std %f0, [%o1 + 0x08] + subcc %o2, 0x10, %o2 + bne,pt %icc, 1b + add %o1, 0x10, %o1 + retl + nop +ENDPROC(camellia_sparc64_ecb_crypt_4_grand_rounds) + + .align 32 +ENTRY(camellia_sparc64_cbc_encrypt_3_grand_rounds) + /* %o0=input, %o1=output, %o2=len, %o3=key, %o4=IV */ + ldd [%o4 + 0x00], %f60 + ldd [%o4 + 0x08], %f62 +1: ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + add %o0, 0x10, %o0 + fxor %f60, %f0, %f0 + fxor %f62, %f2, %f2 + fxor %f4, %f0, %f0 + fxor %f6, %f2, %f2 + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + CAMELLIA_6ROUNDS_FL_FLI(24, 0, 2) + CAMELLIA_6ROUNDS(40, 0, 2) + fxor %f52, %f2, %f60 + fxor %f54, %f0, %f62 + std %f60, [%o1 + 0x00] + std %f62, [%o1 + 0x08] + subcc %o2, 0x10, %o2 + bne,pt %icc, 1b + add %o1, 0x10, %o1 + std %f60, [%o4 + 0x00] + retl + std %f62, [%o4 + 0x08] +ENDPROC(camellia_sparc64_cbc_encrypt_3_grand_rounds) + + .align 32 +ENTRY(camellia_sparc64_cbc_encrypt_4_grand_rounds) + /* %o0=input, %o1=output, %o2=len, %o3=key, %o4=IV */ + ldd [%o4 + 0x00], %f60 + ldd [%o4 + 0x08], %f62 +1: ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + add %o0, 0x10, %o0 + fxor %f60, %f0, %f0 + fxor %f62, %f2, %f2 + fxor %f4, %f0, %f0 + fxor %f6, %f2, %f2 + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + ldd [%o3 + 0xd0], %f8 + ldd [%o3 + 0xd8], %f10 + ldd [%o3 + 0xe0], %f12 + ldd [%o3 + 0xe8], %f14 + ldd [%o3 + 0xf0], %f16 + ldd [%o3 + 0xf8], %f18 + ldd [%o3 + 0x100], %f20 + ldd [%o3 + 0x108], %f22 + CAMELLIA_6ROUNDS_FL_FLI(24, 0, 2) + CAMELLIA_6ROUNDS_FL_FLI(40, 0, 2) + CAMELLIA_F(8, 2, 0, 2) + CAMELLIA_F(10, 0, 2, 0) + ldd [%o3 + 0x10], %f8 + ldd [%o3 + 0x18], %f10 + CAMELLIA_F(12, 2, 0, 2) + CAMELLIA_F(14, 0, 2, 0) + ldd [%o3 + 0x20], %f12 + ldd [%o3 + 0x28], %f14 + CAMELLIA_F(16, 2, 0, 2) + CAMELLIA_F(18, 0, 2, 0) + ldd [%o3 + 0x30], %f16 + ldd [%o3 + 0x38], %f18 + fxor %f20, %f2, %f60 + fxor %f22, %f0, %f62 + ldd [%o3 + 0x40], %f20 + ldd [%o3 + 0x48], %f22 + std %f60, [%o1 + 0x00] + std %f62, [%o1 + 0x08] + subcc %o2, 0x10, %o2 + bne,pt %icc, 1b + add %o1, 0x10, %o1 + std %f60, [%o4 + 0x00] + retl + std %f62, [%o4 + 0x08] +ENDPROC(camellia_sparc64_cbc_encrypt_4_grand_rounds) + + .align 32 +ENTRY(camellia_sparc64_cbc_decrypt_3_grand_rounds) + /* %o0=input, %o1=output, %o2=len, %o3=key, %o4=IV */ + ldd [%o4 + 0x00], %f60 + ldd [%o4 + 0x08], %f62 +1: ldd [%o0 + 0x00], %f56 + ldd [%o0 + 0x08], %f58 + add %o0, 0x10, %o0 + fxor %f4, %f56, %f0 + fxor %f6, %f58, %f2 + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + CAMELLIA_6ROUNDS_FL_FLI(24, 0, 2) + CAMELLIA_6ROUNDS(40, 0, 2) + fxor %f52, %f2, %f2 + fxor %f54, %f0, %f0 + fxor %f60, %f2, %f2 + fxor %f62, %f0, %f0 + fsrc2 %f56, %f60 + fsrc2 %f58, %f62 + std %f2, [%o1 + 0x00] + std %f0, [%o1 + 0x08] + subcc %o2, 0x10, %o2 + bne,pt %icc, 1b + add %o1, 0x10, %o1 + std %f60, [%o4 + 0x00] + retl + std %f62, [%o4 + 0x08] +ENDPROC(camellia_sparc64_cbc_decrypt_3_grand_rounds) + + .align 32 +ENTRY(camellia_sparc64_cbc_decrypt_4_grand_rounds) + /* %o0=input, %o1=output, %o2=len, %o3=key, %o4=IV */ + ldd [%o4 + 0x00], %f60 + ldd [%o4 + 0x08], %f62 +1: ldd [%o0 + 0x00], %f56 + ldd [%o0 + 0x08], %f58 + add %o0, 0x10, %o0 + fxor %f4, %f56, %f0 + fxor %f6, %f58, %f2 + CAMELLIA_6ROUNDS_FL_FLI( 8, 0, 2) + ldd [%o3 + 0xd0], %f8 + ldd [%o3 + 0xd8], %f10 + ldd [%o3 + 0xe0], %f12 + ldd [%o3 + 0xe8], %f14 + ldd [%o3 + 0xf0], %f16 + ldd [%o3 + 0xf8], %f18 + ldd [%o3 + 0x100], %f20 + ldd [%o3 + 0x108], %f22 + CAMELLIA_6ROUNDS_FL_FLI(24, 0, 2) + CAMELLIA_6ROUNDS_FL_FLI(40, 0, 2) + CAMELLIA_F(8, 2, 0, 2) + CAMELLIA_F(10, 0, 2, 0) + ldd [%o3 + 0x10], %f8 + ldd [%o3 + 0x18], %f10 + CAMELLIA_F(12, 2, 0, 2) + CAMELLIA_F(14, 0, 2, 0) + ldd [%o3 + 0x20], %f12 + ldd [%o3 + 0x28], %f14 + CAMELLIA_F(16, 2, 0, 2) + CAMELLIA_F(18, 0, 2, 0) + ldd [%o3 + 0x30], %f16 + ldd [%o3 + 0x38], %f18 + fxor %f20, %f2, %f2 + fxor %f22, %f0, %f0 + ldd [%o3 + 0x40], %f20 + ldd [%o3 + 0x48], %f22 + fxor %f60, %f2, %f2 + fxor %f62, %f0, %f0 + fsrc2 %f56, %f60 + fsrc2 %f58, %f62 + std %f2, [%o1 + 0x00] + std %f0, [%o1 + 0x08] + subcc %o2, 0x10, %o2 + bne,pt %icc, 1b + add %o1, 0x10, %o1 + std %f60, [%o4 + 0x00] + retl + std %f62, [%o4 + 0x08] +ENDPROC(camellia_sparc64_cbc_decrypt_4_grand_rounds) diff --git a/arch/sparc/crypto/camellia_glue.c b/arch/sparc/crypto/camellia_glue.c new file mode 100644 index 00000000000..888f6260b4e --- /dev/null +++ b/arch/sparc/crypto/camellia_glue.c @@ -0,0 +1,327 @@ +/* Glue code for CAMELLIA encryption optimized for sparc64 crypto opcodes. + * + * Copyright (C) 2012 David S. Miller <davem@davemloft.net> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/crypto.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/types.h> +#include <crypto/algapi.h> + +#include <asm/fpumacro.h> +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +#define CAMELLIA_MIN_KEY_SIZE 16 +#define CAMELLIA_MAX_KEY_SIZE 32 +#define CAMELLIA_BLOCK_SIZE 16 +#define CAMELLIA_TABLE_BYTE_LEN 272 + +struct camellia_sparc64_ctx { + u64 encrypt_key[CAMELLIA_TABLE_BYTE_LEN / sizeof(u64)]; + u64 decrypt_key[CAMELLIA_TABLE_BYTE_LEN / sizeof(u64)]; + int key_len; +}; + +extern void camellia_sparc64_key_expand(const u32 *in_key, u64 *encrypt_key, + unsigned int key_len, u64 *decrypt_key); + +static int camellia_set_key(struct crypto_tfm *tfm, const u8 *_in_key, + unsigned int key_len) +{ + struct camellia_sparc64_ctx *ctx = crypto_tfm_ctx(tfm); + const u32 *in_key = (const u32 *) _in_key; + u32 *flags = &tfm->crt_flags; + + if (key_len != 16 && key_len != 24 && key_len != 32) { + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + + ctx->key_len = key_len; + + camellia_sparc64_key_expand(in_key, &ctx->encrypt_key[0], + key_len, &ctx->decrypt_key[0]); + return 0; +} + +extern void camellia_sparc64_crypt(const u64 *key, const u32 *input, + u32 *output, unsigned int key_len); + +static void camellia_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct camellia_sparc64_ctx *ctx = crypto_tfm_ctx(tfm); + + camellia_sparc64_crypt(&ctx->encrypt_key[0], + (const u32 *) src, + (u32 *) dst, ctx->key_len); +} + +static void camellia_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct camellia_sparc64_ctx *ctx = crypto_tfm_ctx(tfm); + + camellia_sparc64_crypt(&ctx->decrypt_key[0], + (const u32 *) src, + (u32 *) dst, ctx->key_len); +} + +extern void camellia_sparc64_load_keys(const u64 *key, unsigned int key_len); + +typedef void ecb_crypt_op(const u64 *input, u64 *output, unsigned int len, + const u64 *key); + +extern ecb_crypt_op camellia_sparc64_ecb_crypt_3_grand_rounds; +extern ecb_crypt_op camellia_sparc64_ecb_crypt_4_grand_rounds; + +#define CAMELLIA_BLOCK_MASK (~(CAMELLIA_BLOCK_SIZE - 1)) + +static int __ecb_crypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes, bool encrypt) +{ + struct camellia_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + ecb_crypt_op *op; + const u64 *key; + int err; + + op = camellia_sparc64_ecb_crypt_3_grand_rounds; + if (ctx->key_len != 16) + op = camellia_sparc64_ecb_crypt_4_grand_rounds; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + if (encrypt) + key = &ctx->encrypt_key[0]; + else + key = &ctx->decrypt_key[0]; + camellia_sparc64_load_keys(key, ctx->key_len); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & CAMELLIA_BLOCK_MASK; + + if (likely(block_len)) { + const u64 *src64; + u64 *dst64; + + src64 = (const u64 *)walk.src.virt.addr; + dst64 = (u64 *) walk.dst.virt.addr; + op(src64, dst64, block_len, key); + } + nbytes &= CAMELLIA_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static int ecb_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + return __ecb_crypt(desc, dst, src, nbytes, true); +} + +static int ecb_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + return __ecb_crypt(desc, dst, src, nbytes, false); +} + +typedef void cbc_crypt_op(const u64 *input, u64 *output, unsigned int len, + const u64 *key, u64 *iv); + +extern cbc_crypt_op camellia_sparc64_cbc_encrypt_3_grand_rounds; +extern cbc_crypt_op camellia_sparc64_cbc_encrypt_4_grand_rounds; +extern cbc_crypt_op camellia_sparc64_cbc_decrypt_3_grand_rounds; +extern cbc_crypt_op camellia_sparc64_cbc_decrypt_4_grand_rounds; + +static int cbc_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct camellia_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + cbc_crypt_op *op; + const u64 *key; + int err; + + op = camellia_sparc64_cbc_encrypt_3_grand_rounds; + if (ctx->key_len != 16) + op = camellia_sparc64_cbc_encrypt_4_grand_rounds; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + key = &ctx->encrypt_key[0]; + camellia_sparc64_load_keys(key, ctx->key_len); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & CAMELLIA_BLOCK_MASK; + + if (likely(block_len)) { + const u64 *src64; + u64 *dst64; + + src64 = (const u64 *)walk.src.virt.addr; + dst64 = (u64 *) walk.dst.virt.addr; + op(src64, dst64, block_len, key, + (u64 *) walk.iv); + } + nbytes &= CAMELLIA_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static int cbc_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct camellia_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + cbc_crypt_op *op; + const u64 *key; + int err; + + op = camellia_sparc64_cbc_decrypt_3_grand_rounds; + if (ctx->key_len != 16) + op = camellia_sparc64_cbc_decrypt_4_grand_rounds; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + key = &ctx->decrypt_key[0]; + camellia_sparc64_load_keys(key, ctx->key_len); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & CAMELLIA_BLOCK_MASK; + + if (likely(block_len)) { + const u64 *src64; + u64 *dst64; + + src64 = (const u64 *)walk.src.virt.addr; + dst64 = (u64 *) walk.dst.virt.addr; + op(src64, dst64, block_len, key, + (u64 *) walk.iv); + } + nbytes &= CAMELLIA_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static struct crypto_alg algs[] = { { + .cra_name = "camellia", + .cra_driver_name = "camellia-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER, + .cra_blocksize = CAMELLIA_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct camellia_sparc64_ctx), + .cra_alignmask = 3, + .cra_module = THIS_MODULE, + .cra_u = { + .cipher = { + .cia_min_keysize = CAMELLIA_MIN_KEY_SIZE, + .cia_max_keysize = CAMELLIA_MAX_KEY_SIZE, + .cia_setkey = camellia_set_key, + .cia_encrypt = camellia_encrypt, + .cia_decrypt = camellia_decrypt + } + } +}, { + .cra_name = "ecb(camellia)", + .cra_driver_name = "ecb-camellia-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = CAMELLIA_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct camellia_sparc64_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = CAMELLIA_MIN_KEY_SIZE, + .max_keysize = CAMELLIA_MAX_KEY_SIZE, + .setkey = camellia_set_key, + .encrypt = ecb_encrypt, + .decrypt = ecb_decrypt, + }, + }, +}, { + .cra_name = "cbc(camellia)", + .cra_driver_name = "cbc-camellia-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = CAMELLIA_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct camellia_sparc64_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = CAMELLIA_MIN_KEY_SIZE, + .max_keysize = CAMELLIA_MAX_KEY_SIZE, + .setkey = camellia_set_key, + .encrypt = cbc_encrypt, + .decrypt = cbc_decrypt, + }, + }, +} +}; + +static bool __init sparc64_has_camellia_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_CAMELLIA)) + return false; + + return true; +} + +static int __init camellia_sparc64_mod_init(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(algs); i++) + INIT_LIST_HEAD(&algs[i].cra_list); + + if (sparc64_has_camellia_opcode()) { + pr_info("Using sparc64 camellia opcodes optimized CAMELLIA implementation\n"); + return crypto_register_algs(algs, ARRAY_SIZE(algs)); + } + pr_info("sparc64 camellia opcodes not available.\n"); + return -ENODEV; +} + +static void __exit camellia_sparc64_mod_fini(void) +{ + crypto_unregister_algs(algs, ARRAY_SIZE(algs)); +} + +module_init(camellia_sparc64_mod_init); +module_exit(camellia_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Camellia Cipher Algorithm, sparc64 camellia opcode accelerated"); + +MODULE_ALIAS("aes"); + +#include "crop_devid.c" diff --git a/arch/sparc/crypto/crc32c_asm.S b/arch/sparc/crypto/crc32c_asm.S new file mode 100644 index 00000000000..2b1976e765b --- /dev/null +++ b/arch/sparc/crypto/crc32c_asm.S @@ -0,0 +1,20 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> +#include <asm/asi.h> + +#include "opcodes.h" + +ENTRY(crc32c_sparc64) + /* %o0=crc32p, %o1=data_ptr, %o2=len */ + VISEntryHalf + lda [%o0] ASI_PL, %f1 +1: ldd [%o1], %f2 + CRC32C(0,2,0) + subcc %o2, 8, %o2 + bne,pt %icc, 1b + add %o1, 0x8, %o1 + sta %f1, [%o0] ASI_PL + VISExitHalf +2: retl + nop +ENDPROC(crc32c_sparc64) diff --git a/arch/sparc/crypto/crc32c_glue.c b/arch/sparc/crypto/crc32c_glue.c new file mode 100644 index 00000000000..5162fad912c --- /dev/null +++ b/arch/sparc/crypto/crc32c_glue.c @@ -0,0 +1,181 @@ +/* Glue code for CRC32C optimized for sparc64 crypto opcodes. + * + * This is based largely upon arch/x86/crypto/crc32c-intel.c + * + * Copyright (C) 2008 Intel Corporation + * Authors: Austin Zhang <austin_zhang@linux.intel.com> + * Kent Liu <kent.liu@intel.com> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/init.h> +#include <linux/module.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/crc32.h> + +#include <crypto/internal/hash.h> + +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +/* + * Setting the seed allows arbitrary accumulators and flexible XOR policy + * If your algorithm starts with ~0, then XOR with ~0 before you set + * the seed. + */ +static int crc32c_sparc64_setkey(struct crypto_shash *hash, const u8 *key, + unsigned int keylen) +{ + u32 *mctx = crypto_shash_ctx(hash); + + if (keylen != sizeof(u32)) { + crypto_shash_set_flags(hash, CRYPTO_TFM_RES_BAD_KEY_LEN); + return -EINVAL; + } + *(__le32 *)mctx = le32_to_cpup((__le32 *)key); + return 0; +} + +static int crc32c_sparc64_init(struct shash_desc *desc) +{ + u32 *mctx = crypto_shash_ctx(desc->tfm); + u32 *crcp = shash_desc_ctx(desc); + + *crcp = *mctx; + + return 0; +} + +extern void crc32c_sparc64(u32 *crcp, const u64 *data, unsigned int len); + +static void crc32c_compute(u32 *crcp, const u64 *data, unsigned int len) +{ + unsigned int asm_len; + + asm_len = len & ~7U; + if (asm_len) { + crc32c_sparc64(crcp, data, asm_len); + data += asm_len / 8; + len -= asm_len; + } + if (len) + *crcp = __crc32c_le(*crcp, (const unsigned char *) data, len); +} + +static int crc32c_sparc64_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + u32 *crcp = shash_desc_ctx(desc); + + crc32c_compute(crcp, (const u64 *) data, len); + + return 0; +} + +static int __crc32c_sparc64_finup(u32 *crcp, const u8 *data, unsigned int len, + u8 *out) +{ + u32 tmp = *crcp; + + crc32c_compute(&tmp, (const u64 *) data, len); + + *(__le32 *) out = ~cpu_to_le32(tmp); + return 0; +} + +static int crc32c_sparc64_finup(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return __crc32c_sparc64_finup(shash_desc_ctx(desc), data, len, out); +} + +static int crc32c_sparc64_final(struct shash_desc *desc, u8 *out) +{ + u32 *crcp = shash_desc_ctx(desc); + + *(__le32 *) out = ~cpu_to_le32p(crcp); + return 0; +} + +static int crc32c_sparc64_digest(struct shash_desc *desc, const u8 *data, + unsigned int len, u8 *out) +{ + return __crc32c_sparc64_finup(crypto_shash_ctx(desc->tfm), data, len, + out); +} + +static int crc32c_sparc64_cra_init(struct crypto_tfm *tfm) +{ + u32 *key = crypto_tfm_ctx(tfm); + + *key = ~0; + + return 0; +} + +#define CHKSUM_BLOCK_SIZE 1 +#define CHKSUM_DIGEST_SIZE 4 + +static struct shash_alg alg = { + .setkey = crc32c_sparc64_setkey, + .init = crc32c_sparc64_init, + .update = crc32c_sparc64_update, + .final = crc32c_sparc64_final, + .finup = crc32c_sparc64_finup, + .digest = crc32c_sparc64_digest, + .descsize = sizeof(u32), + .digestsize = CHKSUM_DIGEST_SIZE, + .base = { + .cra_name = "crc32c", + .cra_driver_name = "crc32c-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_blocksize = CHKSUM_BLOCK_SIZE, + .cra_ctxsize = sizeof(u32), + .cra_alignmask = 7, + .cra_module = THIS_MODULE, + .cra_init = crc32c_sparc64_cra_init, + } +}; + +static bool __init sparc64_has_crc32c_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_CRC32C)) + return false; + + return true; +} + +static int __init crc32c_sparc64_mod_init(void) +{ + if (sparc64_has_crc32c_opcode()) { + pr_info("Using sparc64 crc32c opcode optimized CRC32C implementation\n"); + return crypto_register_shash(&alg); + } + pr_info("sparc64 crc32c opcode not available.\n"); + return -ENODEV; +} + +static void __exit crc32c_sparc64_mod_fini(void) +{ + crypto_unregister_shash(&alg); +} + +module_init(crc32c_sparc64_mod_init); +module_exit(crc32c_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("CRC32c (Castagnoli), sparc64 crc32c opcode accelerated"); + +MODULE_ALIAS("crc32c"); + +#include "crop_devid.c" diff --git a/arch/sparc/crypto/crop_devid.c b/arch/sparc/crypto/crop_devid.c new file mode 100644 index 00000000000..5f5724a0ae2 --- /dev/null +++ b/arch/sparc/crypto/crop_devid.c @@ -0,0 +1,14 @@ +#include <linux/module.h> +#include <linux/of_device.h> + +/* This is a dummy device table linked into all of the crypto + * opcode drivers. It serves to trigger the module autoloading + * mechanisms in userspace which scan the OF device tree and + * load any modules which have device table entries that + * match OF device nodes. + */ +static const struct of_device_id crypto_opcode_match[] = { + { .name = "cpu", .compatible = "sun4v", }, + {}, +}; +MODULE_DEVICE_TABLE(of, crypto_opcode_match); diff --git a/arch/sparc/crypto/des_asm.S b/arch/sparc/crypto/des_asm.S new file mode 100644 index 00000000000..b5c8fc269b5 --- /dev/null +++ b/arch/sparc/crypto/des_asm.S @@ -0,0 +1,419 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> + +#include "opcodes.h" + + .align 32 +ENTRY(des_sparc64_key_expand) + /* %o0=input_key, %o1=output_key */ + VISEntryHalf + ld [%o0 + 0x00], %f0 + ld [%o0 + 0x04], %f1 + DES_KEXPAND(0, 0, 0) + DES_KEXPAND(0, 1, 2) + DES_KEXPAND(2, 3, 6) + DES_KEXPAND(2, 2, 4) + DES_KEXPAND(6, 3, 10) + DES_KEXPAND(6, 2, 8) + DES_KEXPAND(10, 3, 14) + DES_KEXPAND(10, 2, 12) + DES_KEXPAND(14, 1, 16) + DES_KEXPAND(16, 3, 20) + DES_KEXPAND(16, 2, 18) + DES_KEXPAND(20, 3, 24) + DES_KEXPAND(20, 2, 22) + DES_KEXPAND(24, 3, 28) + DES_KEXPAND(24, 2, 26) + DES_KEXPAND(28, 1, 30) + std %f0, [%o1 + 0x00] + std %f2, [%o1 + 0x08] + std %f4, [%o1 + 0x10] + std %f6, [%o1 + 0x18] + std %f8, [%o1 + 0x20] + std %f10, [%o1 + 0x28] + std %f12, [%o1 + 0x30] + std %f14, [%o1 + 0x38] + std %f16, [%o1 + 0x40] + std %f18, [%o1 + 0x48] + std %f20, [%o1 + 0x50] + std %f22, [%o1 + 0x58] + std %f24, [%o1 + 0x60] + std %f26, [%o1 + 0x68] + std %f28, [%o1 + 0x70] + std %f30, [%o1 + 0x78] + retl + VISExitHalf +ENDPROC(des_sparc64_key_expand) + + .align 32 +ENTRY(des_sparc64_crypt) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ldd [%o1 + 0x00], %f32 + ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + ldd [%o0 + 0x10], %f4 + ldd [%o0 + 0x18], %f6 + ldd [%o0 + 0x20], %f8 + ldd [%o0 + 0x28], %f10 + ldd [%o0 + 0x30], %f12 + ldd [%o0 + 0x38], %f14 + ldd [%o0 + 0x40], %f16 + ldd [%o0 + 0x48], %f18 + ldd [%o0 + 0x50], %f20 + ldd [%o0 + 0x58], %f22 + ldd [%o0 + 0x60], %f24 + ldd [%o0 + 0x68], %f26 + ldd [%o0 + 0x70], %f28 + ldd [%o0 + 0x78], %f30 + DES_IP(32, 32) + DES_ROUND(0, 2, 32, 32) + DES_ROUND(4, 6, 32, 32) + DES_ROUND(8, 10, 32, 32) + DES_ROUND(12, 14, 32, 32) + DES_ROUND(16, 18, 32, 32) + DES_ROUND(20, 22, 32, 32) + DES_ROUND(24, 26, 32, 32) + DES_ROUND(28, 30, 32, 32) + DES_IIP(32, 32) + std %f32, [%o2 + 0x00] + retl + VISExit +ENDPROC(des_sparc64_crypt) + + .align 32 +ENTRY(des_sparc64_load_keys) + /* %o0=key */ + VISEntry + ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + ldd [%o0 + 0x10], %f4 + ldd [%o0 + 0x18], %f6 + ldd [%o0 + 0x20], %f8 + ldd [%o0 + 0x28], %f10 + ldd [%o0 + 0x30], %f12 + ldd [%o0 + 0x38], %f14 + ldd [%o0 + 0x40], %f16 + ldd [%o0 + 0x48], %f18 + ldd [%o0 + 0x50], %f20 + ldd [%o0 + 0x58], %f22 + ldd [%o0 + 0x60], %f24 + ldd [%o0 + 0x68], %f26 + ldd [%o0 + 0x70], %f28 + retl + ldd [%o0 + 0x78], %f30 +ENDPROC(des_sparc64_load_keys) + + .align 32 +ENTRY(des_sparc64_ecb_crypt) + /* %o0=input, %o1=output, %o2=len */ +1: ldd [%o0 + 0x00], %f32 + add %o0, 0x08, %o0 + DES_IP(32, 32) + DES_ROUND(0, 2, 32, 32) + DES_ROUND(4, 6, 32, 32) + DES_ROUND(8, 10, 32, 32) + DES_ROUND(12, 14, 32, 32) + DES_ROUND(16, 18, 32, 32) + DES_ROUND(20, 22, 32, 32) + DES_ROUND(24, 26, 32, 32) + DES_ROUND(28, 30, 32, 32) + DES_IIP(32, 32) + std %f32, [%o1 + 0x00] + subcc %o2, 0x08, %o2 + bne,pt %icc, 1b + add %o1, 0x08, %o1 + retl + nop +ENDPROC(des_sparc64_ecb_crypt) + + .align 32 +ENTRY(des_sparc64_cbc_encrypt) + /* %o0=input, %o1=output, %o2=len, %o3=IV */ + ldd [%o3 + 0x00], %f32 +1: ldd [%o0 + 0x00], %f34 + fxor %f32, %f34, %f32 + DES_IP(32, 32) + DES_ROUND(0, 2, 32, 32) + DES_ROUND(4, 6, 32, 32) + DES_ROUND(8, 10, 32, 32) + DES_ROUND(12, 14, 32, 32) + DES_ROUND(16, 18, 32, 32) + DES_ROUND(20, 22, 32, 32) + DES_ROUND(24, 26, 32, 32) + DES_ROUND(28, 30, 32, 32) + DES_IIP(32, 32) + std %f32, [%o1 + 0x00] + add %o0, 0x08, %o0 + subcc %o2, 0x08, %o2 + bne,pt %icc, 1b + add %o1, 0x08, %o1 + retl + std %f32, [%o3 + 0x00] +ENDPROC(des_sparc64_cbc_encrypt) + + .align 32 +ENTRY(des_sparc64_cbc_decrypt) + /* %o0=input, %o1=output, %o2=len, %o3=IV */ + ldd [%o3 + 0x00], %f34 +1: ldd [%o0 + 0x00], %f36 + DES_IP(36, 32) + DES_ROUND(0, 2, 32, 32) + DES_ROUND(4, 6, 32, 32) + DES_ROUND(8, 10, 32, 32) + DES_ROUND(12, 14, 32, 32) + DES_ROUND(16, 18, 32, 32) + DES_ROUND(20, 22, 32, 32) + DES_ROUND(24, 26, 32, 32) + DES_ROUND(28, 30, 32, 32) + DES_IIP(32, 32) + fxor %f32, %f34, %f32 + fsrc2 %f36, %f34 + std %f32, [%o1 + 0x00] + add %o0, 0x08, %o0 + subcc %o2, 0x08, %o2 + bne,pt %icc, 1b + add %o1, 0x08, %o1 + retl + std %f36, [%o3 + 0x00] +ENDPROC(des_sparc64_cbc_decrypt) + + .align 32 +ENTRY(des3_ede_sparc64_crypt) + /* %o0=key, %o1=input, %o2=output */ + VISEntry + ldd [%o1 + 0x00], %f32 + ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + ldd [%o0 + 0x10], %f4 + ldd [%o0 + 0x18], %f6 + ldd [%o0 + 0x20], %f8 + ldd [%o0 + 0x28], %f10 + ldd [%o0 + 0x30], %f12 + ldd [%o0 + 0x38], %f14 + ldd [%o0 + 0x40], %f16 + ldd [%o0 + 0x48], %f18 + ldd [%o0 + 0x50], %f20 + ldd [%o0 + 0x58], %f22 + ldd [%o0 + 0x60], %f24 + ldd [%o0 + 0x68], %f26 + ldd [%o0 + 0x70], %f28 + ldd [%o0 + 0x78], %f30 + DES_IP(32, 32) + DES_ROUND(0, 2, 32, 32) + ldd [%o0 + 0x80], %f0 + ldd [%o0 + 0x88], %f2 + DES_ROUND(4, 6, 32, 32) + ldd [%o0 + 0x90], %f4 + ldd [%o0 + 0x98], %f6 + DES_ROUND(8, 10, 32, 32) + ldd [%o0 + 0xa0], %f8 + ldd [%o0 + 0xa8], %f10 + DES_ROUND(12, 14, 32, 32) + ldd [%o0 + 0xb0], %f12 + ldd [%o0 + 0xb8], %f14 + DES_ROUND(16, 18, 32, 32) + ldd [%o0 + 0xc0], %f16 + ldd [%o0 + 0xc8], %f18 + DES_ROUND(20, 22, 32, 32) + ldd [%o0 + 0xd0], %f20 + ldd [%o0 + 0xd8], %f22 + DES_ROUND(24, 26, 32, 32) + ldd [%o0 + 0xe0], %f24 + ldd [%o0 + 0xe8], %f26 + DES_ROUND(28, 30, 32, 32) + ldd [%o0 + 0xf0], %f28 + ldd [%o0 + 0xf8], %f30 + DES_IIP(32, 32) + DES_IP(32, 32) + DES_ROUND(0, 2, 32, 32) + ldd [%o0 + 0x100], %f0 + ldd [%o0 + 0x108], %f2 + DES_ROUND(4, 6, 32, 32) + ldd [%o0 + 0x110], %f4 + ldd [%o0 + 0x118], %f6 + DES_ROUND(8, 10, 32, 32) + ldd [%o0 + 0x120], %f8 + ldd [%o0 + 0x128], %f10 + DES_ROUND(12, 14, 32, 32) + ldd [%o0 + 0x130], %f12 + ldd [%o0 + 0x138], %f14 + DES_ROUND(16, 18, 32, 32) + ldd [%o0 + 0x140], %f16 + ldd [%o0 + 0x148], %f18 + DES_ROUND(20, 22, 32, 32) + ldd [%o0 + 0x150], %f20 + ldd [%o0 + 0x158], %f22 + DES_ROUND(24, 26, 32, 32) + ldd [%o0 + 0x160], %f24 + ldd [%o0 + 0x168], %f26 + DES_ROUND(28, 30, 32, 32) + ldd [%o0 + 0x170], %f28 + ldd [%o0 + 0x178], %f30 + DES_IIP(32, 32) + DES_IP(32, 32) + DES_ROUND(0, 2, 32, 32) + DES_ROUND(4, 6, 32, 32) + DES_ROUND(8, 10, 32, 32) + DES_ROUND(12, 14, 32, 32) + DES_ROUND(16, 18, 32, 32) + DES_ROUND(20, 22, 32, 32) + DES_ROUND(24, 26, 32, 32) + DES_ROUND(28, 30, 32, 32) + DES_IIP(32, 32) + + std %f32, [%o2 + 0x00] + retl + VISExit +ENDPROC(des3_ede_sparc64_crypt) + + .align 32 +ENTRY(des3_ede_sparc64_load_keys) + /* %o0=key */ + VISEntry + ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + ldd [%o0 + 0x10], %f4 + ldd [%o0 + 0x18], %f6 + ldd [%o0 + 0x20], %f8 + ldd [%o0 + 0x28], %f10 + ldd [%o0 + 0x30], %f12 + ldd [%o0 + 0x38], %f14 + ldd [%o0 + 0x40], %f16 + ldd [%o0 + 0x48], %f18 + ldd [%o0 + 0x50], %f20 + ldd [%o0 + 0x58], %f22 + ldd [%o0 + 0x60], %f24 + ldd [%o0 + 0x68], %f26 + ldd [%o0 + 0x70], %f28 + ldd [%o0 + 0x78], %f30 + ldd [%o0 + 0x80], %f32 + ldd [%o0 + 0x88], %f34 + ldd [%o0 + 0x90], %f36 + ldd [%o0 + 0x98], %f38 + ldd [%o0 + 0xa0], %f40 + ldd [%o0 + 0xa8], %f42 + ldd [%o0 + 0xb0], %f44 + ldd [%o0 + 0xb8], %f46 + ldd [%o0 + 0xc0], %f48 + ldd [%o0 + 0xc8], %f50 + ldd [%o0 + 0xd0], %f52 + ldd [%o0 + 0xd8], %f54 + ldd [%o0 + 0xe0], %f56 + retl + ldd [%o0 + 0xe8], %f58 +ENDPROC(des3_ede_sparc64_load_keys) + +#define DES3_LOOP_BODY(X) \ + DES_IP(X, X) \ + DES_ROUND(0, 2, X, X) \ + DES_ROUND(4, 6, X, X) \ + DES_ROUND(8, 10, X, X) \ + DES_ROUND(12, 14, X, X) \ + DES_ROUND(16, 18, X, X) \ + ldd [%o0 + 0xf0], %f16; \ + ldd [%o0 + 0xf8], %f18; \ + DES_ROUND(20, 22, X, X) \ + ldd [%o0 + 0x100], %f20; \ + ldd [%o0 + 0x108], %f22; \ + DES_ROUND(24, 26, X, X) \ + ldd [%o0 + 0x110], %f24; \ + ldd [%o0 + 0x118], %f26; \ + DES_ROUND(28, 30, X, X) \ + ldd [%o0 + 0x120], %f28; \ + ldd [%o0 + 0x128], %f30; \ + DES_IIP(X, X) \ + DES_IP(X, X) \ + DES_ROUND(32, 34, X, X) \ + ldd [%o0 + 0x130], %f0; \ + ldd [%o0 + 0x138], %f2; \ + DES_ROUND(36, 38, X, X) \ + ldd [%o0 + 0x140], %f4; \ + ldd [%o0 + 0x148], %f6; \ + DES_ROUND(40, 42, X, X) \ + ldd [%o0 + 0x150], %f8; \ + ldd [%o0 + 0x158], %f10; \ + DES_ROUND(44, 46, X, X) \ + ldd [%o0 + 0x160], %f12; \ + ldd [%o0 + 0x168], %f14; \ + DES_ROUND(48, 50, X, X) \ + DES_ROUND(52, 54, X, X) \ + DES_ROUND(56, 58, X, X) \ + DES_ROUND(16, 18, X, X) \ + ldd [%o0 + 0x170], %f16; \ + ldd [%o0 + 0x178], %f18; \ + DES_IIP(X, X) \ + DES_IP(X, X) \ + DES_ROUND(20, 22, X, X) \ + ldd [%o0 + 0x50], %f20; \ + ldd [%o0 + 0x58], %f22; \ + DES_ROUND(24, 26, X, X) \ + ldd [%o0 + 0x60], %f24; \ + ldd [%o0 + 0x68], %f26; \ + DES_ROUND(28, 30, X, X) \ + ldd [%o0 + 0x70], %f28; \ + ldd [%o0 + 0x78], %f30; \ + DES_ROUND(0, 2, X, X) \ + ldd [%o0 + 0x00], %f0; \ + ldd [%o0 + 0x08], %f2; \ + DES_ROUND(4, 6, X, X) \ + ldd [%o0 + 0x10], %f4; \ + ldd [%o0 + 0x18], %f6; \ + DES_ROUND(8, 10, X, X) \ + ldd [%o0 + 0x20], %f8; \ + ldd [%o0 + 0x28], %f10; \ + DES_ROUND(12, 14, X, X) \ + ldd [%o0 + 0x30], %f12; \ + ldd [%o0 + 0x38], %f14; \ + DES_ROUND(16, 18, X, X) \ + ldd [%o0 + 0x40], %f16; \ + ldd [%o0 + 0x48], %f18; \ + DES_IIP(X, X) + + .align 32 +ENTRY(des3_ede_sparc64_ecb_crypt) + /* %o0=key, %o1=input, %o2=output, %o3=len */ +1: ldd [%o1 + 0x00], %f60 + DES3_LOOP_BODY(60) + std %f60, [%o2 + 0x00] + add %o1, 0x08, %o1 + subcc %o3, 0x08, %o3 + bne,pt %icc, 1b + add %o2, 0x08, %o2 + retl + nop +ENDPROC(des3_ede_sparc64_ecb_crypt) + + .align 32 +ENTRY(des3_ede_sparc64_cbc_encrypt) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldd [%o4 + 0x00], %f60 +1: ldd [%o1 + 0x00], %f62 + fxor %f60, %f62, %f60 + DES3_LOOP_BODY(60) + std %f60, [%o2 + 0x00] + add %o1, 0x08, %o1 + subcc %o3, 0x08, %o3 + bne,pt %icc, 1b + add %o2, 0x08, %o2 + retl + std %f60, [%o4 + 0x00] +ENDPROC(des3_ede_sparc64_cbc_encrypt) + + .align 32 +ENTRY(des3_ede_sparc64_cbc_decrypt) + /* %o0=key, %o1=input, %o2=output, %o3=len, %o4=IV */ + ldd [%o4 + 0x00], %f62 +1: ldx [%o1 + 0x00], %g1 + MOVXTOD_G1_F60 + DES3_LOOP_BODY(60) + fxor %f62, %f60, %f60 + MOVXTOD_G1_F62 + std %f60, [%o2 + 0x00] + add %o1, 0x08, %o1 + subcc %o3, 0x08, %o3 + bne,pt %icc, 1b + add %o2, 0x08, %o2 + retl + stx %g1, [%o4 + 0x00] +ENDPROC(des3_ede_sparc64_cbc_decrypt) diff --git a/arch/sparc/crypto/des_glue.c b/arch/sparc/crypto/des_glue.c new file mode 100644 index 00000000000..3065bc61f9d --- /dev/null +++ b/arch/sparc/crypto/des_glue.c @@ -0,0 +1,537 @@ +/* Glue code for DES encryption optimized for sparc64 crypto opcodes. + * + * Copyright (C) 2012 David S. Miller <davem@davemloft.net> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/crypto.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/types.h> +#include <crypto/algapi.h> +#include <crypto/des.h> + +#include <asm/fpumacro.h> +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +struct des_sparc64_ctx { + u64 encrypt_expkey[DES_EXPKEY_WORDS / 2]; + u64 decrypt_expkey[DES_EXPKEY_WORDS / 2]; +}; + +struct des3_ede_sparc64_ctx { + u64 encrypt_expkey[DES3_EDE_EXPKEY_WORDS / 2]; + u64 decrypt_expkey[DES3_EDE_EXPKEY_WORDS / 2]; +}; + +static void encrypt_to_decrypt(u64 *d, const u64 *e) +{ + const u64 *s = e + (DES_EXPKEY_WORDS / 2) - 1; + int i; + + for (i = 0; i < DES_EXPKEY_WORDS / 2; i++) + *d++ = *s--; +} + +extern void des_sparc64_key_expand(const u32 *input_key, u64 *key); + +static int des_set_key(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct des_sparc64_ctx *dctx = crypto_tfm_ctx(tfm); + u32 *flags = &tfm->crt_flags; + u32 tmp[DES_EXPKEY_WORDS]; + int ret; + + /* Even though we have special instructions for key expansion, + * we call des_ekey() so that we don't have to write our own + * weak key detection code. + */ + ret = des_ekey(tmp, key); + if (unlikely(ret == 0) && (*flags & CRYPTO_TFM_REQ_WEAK_KEY)) { + *flags |= CRYPTO_TFM_RES_WEAK_KEY; + return -EINVAL; + } + + des_sparc64_key_expand((const u32 *) key, &dctx->encrypt_expkey[0]); + encrypt_to_decrypt(&dctx->decrypt_expkey[0], &dctx->encrypt_expkey[0]); + + return 0; +} + +extern void des_sparc64_crypt(const u64 *key, const u64 *input, + u64 *output); + +static void des_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct des_sparc64_ctx *ctx = crypto_tfm_ctx(tfm); + const u64 *K = ctx->encrypt_expkey; + + des_sparc64_crypt(K, (const u64 *) src, (u64 *) dst); +} + +static void des_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct des_sparc64_ctx *ctx = crypto_tfm_ctx(tfm); + const u64 *K = ctx->decrypt_expkey; + + des_sparc64_crypt(K, (const u64 *) src, (u64 *) dst); +} + +extern void des_sparc64_load_keys(const u64 *key); + +extern void des_sparc64_ecb_crypt(const u64 *input, u64 *output, + unsigned int len); + +#define DES_BLOCK_MASK (~(DES_BLOCK_SIZE - 1)) + +static int __ecb_crypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes, bool encrypt) +{ + struct des_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + if (encrypt) + des_sparc64_load_keys(&ctx->encrypt_expkey[0]); + else + des_sparc64_load_keys(&ctx->decrypt_expkey[0]); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & DES_BLOCK_MASK; + + if (likely(block_len)) { + des_sparc64_ecb_crypt((const u64 *)walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, + block_len); + } + nbytes &= DES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static int ecb_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + return __ecb_crypt(desc, dst, src, nbytes, true); +} + +static int ecb_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + return __ecb_crypt(desc, dst, src, nbytes, false); +} + +extern void des_sparc64_cbc_encrypt(const u64 *input, u64 *output, + unsigned int len, u64 *iv); + +static int cbc_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct des_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + des_sparc64_load_keys(&ctx->encrypt_expkey[0]); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & DES_BLOCK_MASK; + + if (likely(block_len)) { + des_sparc64_cbc_encrypt((const u64 *)walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, + block_len, (u64 *) walk.iv); + } + nbytes &= DES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +extern void des_sparc64_cbc_decrypt(const u64 *input, u64 *output, + unsigned int len, u64 *iv); + +static int cbc_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct des_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + des_sparc64_load_keys(&ctx->decrypt_expkey[0]); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & DES_BLOCK_MASK; + + if (likely(block_len)) { + des_sparc64_cbc_decrypt((const u64 *)walk.src.virt.addr, + (u64 *) walk.dst.virt.addr, + block_len, (u64 *) walk.iv); + } + nbytes &= DES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static int des3_ede_set_key(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen) +{ + struct des3_ede_sparc64_ctx *dctx = crypto_tfm_ctx(tfm); + const u32 *K = (const u32 *)key; + u32 *flags = &tfm->crt_flags; + u64 k1[DES_EXPKEY_WORDS / 2]; + u64 k2[DES_EXPKEY_WORDS / 2]; + u64 k3[DES_EXPKEY_WORDS / 2]; + + if (unlikely(!((K[0] ^ K[2]) | (K[1] ^ K[3])) || + !((K[2] ^ K[4]) | (K[3] ^ K[5]))) && + (*flags & CRYPTO_TFM_REQ_WEAK_KEY)) { + *flags |= CRYPTO_TFM_RES_WEAK_KEY; + return -EINVAL; + } + + des_sparc64_key_expand((const u32 *)key, k1); + key += DES_KEY_SIZE; + des_sparc64_key_expand((const u32 *)key, k2); + key += DES_KEY_SIZE; + des_sparc64_key_expand((const u32 *)key, k3); + + memcpy(&dctx->encrypt_expkey[0], &k1[0], sizeof(k1)); + encrypt_to_decrypt(&dctx->encrypt_expkey[DES_EXPKEY_WORDS / 2], &k2[0]); + memcpy(&dctx->encrypt_expkey[(DES_EXPKEY_WORDS / 2) * 2], + &k3[0], sizeof(k3)); + + encrypt_to_decrypt(&dctx->decrypt_expkey[0], &k3[0]); + memcpy(&dctx->decrypt_expkey[DES_EXPKEY_WORDS / 2], + &k2[0], sizeof(k2)); + encrypt_to_decrypt(&dctx->decrypt_expkey[(DES_EXPKEY_WORDS / 2) * 2], + &k1[0]); + + return 0; +} + +extern void des3_ede_sparc64_crypt(const u64 *key, const u64 *input, + u64 *output); + +static void des3_ede_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct des3_ede_sparc64_ctx *ctx = crypto_tfm_ctx(tfm); + const u64 *K = ctx->encrypt_expkey; + + des3_ede_sparc64_crypt(K, (const u64 *) src, (u64 *) dst); +} + +static void des3_ede_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct des3_ede_sparc64_ctx *ctx = crypto_tfm_ctx(tfm); + const u64 *K = ctx->decrypt_expkey; + + des3_ede_sparc64_crypt(K, (const u64 *) src, (u64 *) dst); +} + +extern void des3_ede_sparc64_load_keys(const u64 *key); + +extern void des3_ede_sparc64_ecb_crypt(const u64 *expkey, const u64 *input, + u64 *output, unsigned int len); + +static int __ecb3_crypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes, bool encrypt) +{ + struct des3_ede_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + const u64 *K; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + if (encrypt) + K = &ctx->encrypt_expkey[0]; + else + K = &ctx->decrypt_expkey[0]; + des3_ede_sparc64_load_keys(K); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & DES_BLOCK_MASK; + + if (likely(block_len)) { + const u64 *src64 = (const u64 *)walk.src.virt.addr; + des3_ede_sparc64_ecb_crypt(K, src64, + (u64 *) walk.dst.virt.addr, + block_len); + } + nbytes &= DES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static int ecb3_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + return __ecb3_crypt(desc, dst, src, nbytes, true); +} + +static int ecb3_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + return __ecb3_crypt(desc, dst, src, nbytes, false); +} + +extern void des3_ede_sparc64_cbc_encrypt(const u64 *expkey, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +static int cbc3_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct des3_ede_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + const u64 *K; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + K = &ctx->encrypt_expkey[0]; + des3_ede_sparc64_load_keys(K); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & DES_BLOCK_MASK; + + if (likely(block_len)) { + const u64 *src64 = (const u64 *)walk.src.virt.addr; + des3_ede_sparc64_cbc_encrypt(K, src64, + (u64 *) walk.dst.virt.addr, + block_len, + (u64 *) walk.iv); + } + nbytes &= DES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +extern void des3_ede_sparc64_cbc_decrypt(const u64 *expkey, const u64 *input, + u64 *output, unsigned int len, + u64 *iv); + +static int cbc3_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct des3_ede_sparc64_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + const u64 *K; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; + + K = &ctx->decrypt_expkey[0]; + des3_ede_sparc64_load_keys(K); + while ((nbytes = walk.nbytes)) { + unsigned int block_len = nbytes & DES_BLOCK_MASK; + + if (likely(block_len)) { + const u64 *src64 = (const u64 *)walk.src.virt.addr; + des3_ede_sparc64_cbc_decrypt(K, src64, + (u64 *) walk.dst.virt.addr, + block_len, + (u64 *) walk.iv); + } + nbytes &= DES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + fprs_write(0); + return err; +} + +static struct crypto_alg algs[] = { { + .cra_name = "des", + .cra_driver_name = "des-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct des_sparc64_ctx), + .cra_alignmask = 7, + .cra_module = THIS_MODULE, + .cra_u = { + .cipher = { + .cia_min_keysize = DES_KEY_SIZE, + .cia_max_keysize = DES_KEY_SIZE, + .cia_setkey = des_set_key, + .cia_encrypt = des_encrypt, + .cia_decrypt = des_decrypt + } + } +}, { + .cra_name = "ecb(des)", + .cra_driver_name = "ecb-des-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct des_sparc64_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES_KEY_SIZE, + .max_keysize = DES_KEY_SIZE, + .setkey = des_set_key, + .encrypt = ecb_encrypt, + .decrypt = ecb_decrypt, + }, + }, +}, { + .cra_name = "cbc(des)", + .cra_driver_name = "cbc-des-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct des_sparc64_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES_KEY_SIZE, + .max_keysize = DES_KEY_SIZE, + .setkey = des_set_key, + .encrypt = cbc_encrypt, + .decrypt = cbc_decrypt, + }, + }, +}, { + .cra_name = "des3_ede", + .cra_driver_name = "des3_ede-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER, + .cra_blocksize = DES3_EDE_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct des3_ede_sparc64_ctx), + .cra_alignmask = 7, + .cra_module = THIS_MODULE, + .cra_u = { + .cipher = { + .cia_min_keysize = DES3_EDE_KEY_SIZE, + .cia_max_keysize = DES3_EDE_KEY_SIZE, + .cia_setkey = des3_ede_set_key, + .cia_encrypt = des3_ede_encrypt, + .cia_decrypt = des3_ede_decrypt + } + } +}, { + .cra_name = "ecb(des3_ede)", + .cra_driver_name = "ecb-des3_ede-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES3_EDE_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct des3_ede_sparc64_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES3_EDE_KEY_SIZE, + .max_keysize = DES3_EDE_KEY_SIZE, + .setkey = des3_ede_set_key, + .encrypt = ecb3_encrypt, + .decrypt = ecb3_decrypt, + }, + }, +}, { + .cra_name = "cbc(des3_ede)", + .cra_driver_name = "cbc-des3_ede-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = DES3_EDE_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct des3_ede_sparc64_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_u = { + .blkcipher = { + .min_keysize = DES3_EDE_KEY_SIZE, + .max_keysize = DES3_EDE_KEY_SIZE, + .setkey = des3_ede_set_key, + .encrypt = cbc3_encrypt, + .decrypt = cbc3_decrypt, + }, + }, +} }; + +static bool __init sparc64_has_des_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_DES)) + return false; + + return true; +} + +static int __init des_sparc64_mod_init(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(algs); i++) + INIT_LIST_HEAD(&algs[i].cra_list); + + if (sparc64_has_des_opcode()) { + pr_info("Using sparc64 des opcodes optimized DES implementation\n"); + return crypto_register_algs(algs, ARRAY_SIZE(algs)); + } + pr_info("sparc64 des opcodes not available.\n"); + return -ENODEV; +} + +static void __exit des_sparc64_mod_fini(void) +{ + crypto_unregister_algs(algs, ARRAY_SIZE(algs)); +} + +module_init(des_sparc64_mod_init); +module_exit(des_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("DES & Triple DES EDE Cipher Algorithms, sparc64 des opcode accelerated"); + +MODULE_ALIAS("des"); + +#include "crop_devid.c" diff --git a/arch/sparc/crypto/md5_asm.S b/arch/sparc/crypto/md5_asm.S new file mode 100644 index 00000000000..3150404e602 --- /dev/null +++ b/arch/sparc/crypto/md5_asm.S @@ -0,0 +1,70 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> + +#include "opcodes.h" + +ENTRY(md5_sparc64_transform) + /* %o0 = digest, %o1 = data, %o2 = rounds */ + VISEntryHalf + ld [%o0 + 0x00], %f0 + ld [%o0 + 0x04], %f1 + andcc %o1, 0x7, %g0 + ld [%o0 + 0x08], %f2 + bne,pn %xcc, 10f + ld [%o0 + 0x0c], %f3 + +1: + ldd [%o1 + 0x00], %f8 + ldd [%o1 + 0x08], %f10 + ldd [%o1 + 0x10], %f12 + ldd [%o1 + 0x18], %f14 + ldd [%o1 + 0x20], %f16 + ldd [%o1 + 0x28], %f18 + ldd [%o1 + 0x30], %f20 + ldd [%o1 + 0x38], %f22 + + MD5 + + subcc %o2, 1, %o2 + bne,pt %xcc, 1b + add %o1, 0x40, %o1 + +5: + st %f0, [%o0 + 0x00] + st %f1, [%o0 + 0x04] + st %f2, [%o0 + 0x08] + st %f3, [%o0 + 0x0c] + retl + VISExitHalf +10: + alignaddr %o1, %g0, %o1 + + ldd [%o1 + 0x00], %f10 +1: + ldd [%o1 + 0x08], %f12 + ldd [%o1 + 0x10], %f14 + ldd [%o1 + 0x18], %f16 + ldd [%o1 + 0x20], %f18 + ldd [%o1 + 0x28], %f20 + ldd [%o1 + 0x30], %f22 + ldd [%o1 + 0x38], %f24 + ldd [%o1 + 0x40], %f26 + + faligndata %f10, %f12, %f8 + faligndata %f12, %f14, %f10 + faligndata %f14, %f16, %f12 + faligndata %f16, %f18, %f14 + faligndata %f18, %f20, %f16 + faligndata %f20, %f22, %f18 + faligndata %f22, %f24, %f20 + faligndata %f24, %f26, %f22 + + MD5 + + subcc %o2, 1, %o2 + fsrc2 %f26, %f10 + bne,pt %xcc, 1b + add %o1, 0x40, %o1 + + ba,a,pt %xcc, 5b +ENDPROC(md5_sparc64_transform) diff --git a/arch/sparc/crypto/md5_glue.c b/arch/sparc/crypto/md5_glue.c new file mode 100644 index 00000000000..09a9ea1dfb6 --- /dev/null +++ b/arch/sparc/crypto/md5_glue.c @@ -0,0 +1,190 @@ +/* Glue code for MD5 hashing optimized for sparc64 crypto opcodes. + * + * This is based largely upon arch/x86/crypto/sha1_ssse3_glue.c + * and crypto/md5.c which are: + * + * Copyright (c) Alan Smithee. + * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk> + * Copyright (c) Jean-Francois Dive <jef@linuxbe.org> + * Copyright (c) Mathias Krause <minipli@googlemail.com> + * Copyright (c) Cryptoapi developers. + * Copyright (c) 2002 James Morris <jmorris@intercode.com.au> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <crypto/md5.h> + +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +asmlinkage void md5_sparc64_transform(u32 *digest, const char *data, + unsigned int rounds); + +static int md5_sparc64_init(struct shash_desc *desc) +{ + struct md5_state *mctx = shash_desc_ctx(desc); + + mctx->hash[0] = cpu_to_le32(0x67452301); + mctx->hash[1] = cpu_to_le32(0xefcdab89); + mctx->hash[2] = cpu_to_le32(0x98badcfe); + mctx->hash[3] = cpu_to_le32(0x10325476); + mctx->byte_count = 0; + + return 0; +} + +static void __md5_sparc64_update(struct md5_state *sctx, const u8 *data, + unsigned int len, unsigned int partial) +{ + unsigned int done = 0; + + sctx->byte_count += len; + if (partial) { + done = MD5_HMAC_BLOCK_SIZE - partial; + memcpy((u8 *)sctx->block + partial, data, done); + md5_sparc64_transform(sctx->hash, (u8 *)sctx->block, 1); + } + if (len - done >= MD5_HMAC_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / MD5_HMAC_BLOCK_SIZE; + + md5_sparc64_transform(sctx->hash, data + done, rounds); + done += rounds * MD5_HMAC_BLOCK_SIZE; + } + + memcpy(sctx->block, data + done, len - done); +} + +static int md5_sparc64_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct md5_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->byte_count % MD5_HMAC_BLOCK_SIZE; + + /* Handle the fast case right here */ + if (partial + len < MD5_HMAC_BLOCK_SIZE) { + sctx->byte_count += len; + memcpy((u8 *)sctx->block + partial, data, len); + } else + __md5_sparc64_update(sctx, data, len, partial); + + return 0; +} + +/* Add padding and return the message digest. */ +static int md5_sparc64_final(struct shash_desc *desc, u8 *out) +{ + struct md5_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + u32 *dst = (u32 *)out; + __le64 bits; + static const u8 padding[MD5_HMAC_BLOCK_SIZE] = { 0x80, }; + + bits = cpu_to_le64(sctx->byte_count << 3); + + /* Pad out to 56 mod 64 and append length */ + index = sctx->byte_count % MD5_HMAC_BLOCK_SIZE; + padlen = (index < 56) ? (56 - index) : ((MD5_HMAC_BLOCK_SIZE+56) - index); + + /* We need to fill a whole block for __md5_sparc64_update() */ + if (padlen <= 56) { + sctx->byte_count += padlen; + memcpy((u8 *)sctx->block + index, padding, padlen); + } else { + __md5_sparc64_update(sctx, padding, padlen, index); + } + __md5_sparc64_update(sctx, (const u8 *)&bits, sizeof(bits), 56); + + /* Store state in digest */ + for (i = 0; i < MD5_HASH_WORDS; i++) + dst[i] = sctx->hash[i]; + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int md5_sparc64_export(struct shash_desc *desc, void *out) +{ + struct md5_state *sctx = shash_desc_ctx(desc); + + memcpy(out, sctx, sizeof(*sctx)); + + return 0; +} + +static int md5_sparc64_import(struct shash_desc *desc, const void *in) +{ + struct md5_state *sctx = shash_desc_ctx(desc); + + memcpy(sctx, in, sizeof(*sctx)); + + return 0; +} + +static struct shash_alg alg = { + .digestsize = MD5_DIGEST_SIZE, + .init = md5_sparc64_init, + .update = md5_sparc64_update, + .final = md5_sparc64_final, + .export = md5_sparc64_export, + .import = md5_sparc64_import, + .descsize = sizeof(struct md5_state), + .statesize = sizeof(struct md5_state), + .base = { + .cra_name = "md5", + .cra_driver_name= "md5-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = MD5_HMAC_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static bool __init sparc64_has_md5_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_MD5)) + return false; + + return true; +} + +static int __init md5_sparc64_mod_init(void) +{ + if (sparc64_has_md5_opcode()) { + pr_info("Using sparc64 md5 opcode optimized MD5 implementation\n"); + return crypto_register_shash(&alg); + } + pr_info("sparc64 md5 opcode not available.\n"); + return -ENODEV; +} + +static void __exit md5_sparc64_mod_fini(void) +{ + crypto_unregister_shash(&alg); +} + +module_init(md5_sparc64_mod_init); +module_exit(md5_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("MD5 Secure Hash Algorithm, sparc64 md5 opcode accelerated"); + +MODULE_ALIAS("md5"); + +#include "crop_devid.c" diff --git a/arch/sparc/crypto/opcodes.h b/arch/sparc/crypto/opcodes.h new file mode 100644 index 00000000000..19cbaea6976 --- /dev/null +++ b/arch/sparc/crypto/opcodes.h @@ -0,0 +1,99 @@ +#ifndef _OPCODES_H +#define _OPCODES_H + +#define SPARC_CR_OPCODE_PRIORITY 300 + +#define F3F(x,y,z) (((x)<<30)|((y)<<19)|((z)<<5)) + +#define FPD_ENCODE(x) (((x) >> 5) | ((x) & ~(0x20))) + +#define RS1(x) (FPD_ENCODE(x) << 14) +#define RS2(x) (FPD_ENCODE(x) << 0) +#define RS3(x) (FPD_ENCODE(x) << 9) +#define RD(x) (FPD_ENCODE(x) << 25) +#define IMM5_0(x) ((x) << 0) +#define IMM5_9(x) ((x) << 9) + +#define CRC32C(a,b,c) \ + .word (F3F(2,0x36,0x147)|RS1(a)|RS2(b)|RD(c)); + +#define MD5 \ + .word 0x81b02800; +#define SHA1 \ + .word 0x81b02820; +#define SHA256 \ + .word 0x81b02840; +#define SHA512 \ + .word 0x81b02860; + +#define AES_EROUND01(a,b,c,d) \ + .word (F3F(2, 0x19, 0)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_EROUND23(a,b,c,d) \ + .word (F3F(2, 0x19, 1)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_DROUND01(a,b,c,d) \ + .word (F3F(2, 0x19, 2)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_DROUND23(a,b,c,d) \ + .word (F3F(2, 0x19, 3)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_EROUND01_L(a,b,c,d) \ + .word (F3F(2, 0x19, 4)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_EROUND23_L(a,b,c,d) \ + .word (F3F(2, 0x19, 5)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_DROUND01_L(a,b,c,d) \ + .word (F3F(2, 0x19, 6)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_DROUND23_L(a,b,c,d) \ + .word (F3F(2, 0x19, 7)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define AES_KEXPAND1(a,b,c,d) \ + .word (F3F(2, 0x19, 8)|RS1(a)|RS2(b)|IMM5_9(c)|RD(d)); +#define AES_KEXPAND0(a,b,c) \ + .word (F3F(2, 0x36, 0x130)|RS1(a)|RS2(b)|RD(c)); +#define AES_KEXPAND2(a,b,c) \ + .word (F3F(2, 0x36, 0x131)|RS1(a)|RS2(b)|RD(c)); + +#define DES_IP(a,b) \ + .word (F3F(2, 0x36, 0x134)|RS1(a)|RD(b)); +#define DES_IIP(a,b) \ + .word (F3F(2, 0x36, 0x135)|RS1(a)|RD(b)); +#define DES_KEXPAND(a,b,c) \ + .word (F3F(2, 0x36, 0x136)|RS1(a)|IMM5_0(b)|RD(c)); +#define DES_ROUND(a,b,c,d) \ + .word (F3F(2, 0x19, 0x009)|RS1(a)|RS2(b)|RS3(c)|RD(d)); + +#define CAMELLIA_F(a,b,c,d) \ + .word (F3F(2, 0x19, 0x00c)|RS1(a)|RS2(b)|RS3(c)|RD(d)); +#define CAMELLIA_FL(a,b,c) \ + .word (F3F(2, 0x36, 0x13c)|RS1(a)|RS2(b)|RD(c)); +#define CAMELLIA_FLI(a,b,c) \ + .word (F3F(2, 0x36, 0x13d)|RS1(a)|RS2(b)|RD(c)); + +#define MOVDTOX_F0_O4 \ + .word 0x99b02200 +#define MOVDTOX_F2_O5 \ + .word 0x9bb02202 +#define MOVXTOD_G1_F60 \ + .word 0xbbb02301 +#define MOVXTOD_G1_F62 \ + .word 0xbfb02301 +#define MOVXTOD_G3_F4 \ + .word 0x89b02303; +#define MOVXTOD_G7_F6 \ + .word 0x8db02307; +#define MOVXTOD_G3_F0 \ + .word 0x81b02303; +#define MOVXTOD_G7_F2 \ + .word 0x85b02307; +#define MOVXTOD_O0_F0 \ + .word 0x81b02308; +#define MOVXTOD_O5_F0 \ + .word 0x81b0230d; +#define MOVXTOD_O5_F2 \ + .word 0x85b0230d; +#define MOVXTOD_O5_F4 \ + .word 0x89b0230d; +#define MOVXTOD_O5_F6 \ + .word 0x8db0230d; +#define MOVXTOD_G3_F60 \ + .word 0xbbb02303; +#define MOVXTOD_G7_F62 \ + .word 0xbfb02307; + +#endif /* _OPCODES_H */ diff --git a/arch/sparc/crypto/sha1_asm.S b/arch/sparc/crypto/sha1_asm.S new file mode 100644 index 00000000000..219d10c5ae0 --- /dev/null +++ b/arch/sparc/crypto/sha1_asm.S @@ -0,0 +1,72 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> + +#include "opcodes.h" + +ENTRY(sha1_sparc64_transform) + /* %o0 = digest, %o1 = data, %o2 = rounds */ + VISEntryHalf + ld [%o0 + 0x00], %f0 + ld [%o0 + 0x04], %f1 + ld [%o0 + 0x08], %f2 + andcc %o1, 0x7, %g0 + ld [%o0 + 0x0c], %f3 + bne,pn %xcc, 10f + ld [%o0 + 0x10], %f4 + +1: + ldd [%o1 + 0x00], %f8 + ldd [%o1 + 0x08], %f10 + ldd [%o1 + 0x10], %f12 + ldd [%o1 + 0x18], %f14 + ldd [%o1 + 0x20], %f16 + ldd [%o1 + 0x28], %f18 + ldd [%o1 + 0x30], %f20 + ldd [%o1 + 0x38], %f22 + + SHA1 + + subcc %o2, 1, %o2 + bne,pt %xcc, 1b + add %o1, 0x40, %o1 + +5: + st %f0, [%o0 + 0x00] + st %f1, [%o0 + 0x04] + st %f2, [%o0 + 0x08] + st %f3, [%o0 + 0x0c] + st %f4, [%o0 + 0x10] + retl + VISExitHalf +10: + alignaddr %o1, %g0, %o1 + + ldd [%o1 + 0x00], %f10 +1: + ldd [%o1 + 0x08], %f12 + ldd [%o1 + 0x10], %f14 + ldd [%o1 + 0x18], %f16 + ldd [%o1 + 0x20], %f18 + ldd [%o1 + 0x28], %f20 + ldd [%o1 + 0x30], %f22 + ldd [%o1 + 0x38], %f24 + ldd [%o1 + 0x40], %f26 + + faligndata %f10, %f12, %f8 + faligndata %f12, %f14, %f10 + faligndata %f14, %f16, %f12 + faligndata %f16, %f18, %f14 + faligndata %f18, %f20, %f16 + faligndata %f20, %f22, %f18 + faligndata %f22, %f24, %f20 + faligndata %f24, %f26, %f22 + + SHA1 + + subcc %o2, 1, %o2 + fsrc2 %f26, %f10 + bne,pt %xcc, 1b + add %o1, 0x40, %o1 + + ba,a,pt %xcc, 5b +ENDPROC(sha1_sparc64_transform) diff --git a/arch/sparc/crypto/sha1_glue.c b/arch/sparc/crypto/sha1_glue.c new file mode 100644 index 00000000000..6cd5f29e1e0 --- /dev/null +++ b/arch/sparc/crypto/sha1_glue.c @@ -0,0 +1,185 @@ +/* Glue code for SHA1 hashing optimized for sparc64 crypto opcodes. + * + * This is based largely upon arch/x86/crypto/sha1_ssse3_glue.c + * + * Copyright (c) Alan Smithee. + * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk> + * Copyright (c) Jean-Francois Dive <jef@linuxbe.org> + * Copyright (c) Mathias Krause <minipli@googlemail.com> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <crypto/sha.h> + +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +asmlinkage void sha1_sparc64_transform(u32 *digest, const char *data, + unsigned int rounds); + +static int sha1_sparc64_init(struct shash_desc *desc) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + + *sctx = (struct sha1_state){ + .state = { SHA1_H0, SHA1_H1, SHA1_H2, SHA1_H3, SHA1_H4 }, + }; + + return 0; +} + +static void __sha1_sparc64_update(struct sha1_state *sctx, const u8 *data, + unsigned int len, unsigned int partial) +{ + unsigned int done = 0; + + sctx->count += len; + if (partial) { + done = SHA1_BLOCK_SIZE - partial; + memcpy(sctx->buffer + partial, data, done); + sha1_sparc64_transform(sctx->state, sctx->buffer, 1); + } + if (len - done >= SHA1_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / SHA1_BLOCK_SIZE; + + sha1_sparc64_transform(sctx->state, data + done, rounds); + done += rounds * SHA1_BLOCK_SIZE; + } + + memcpy(sctx->buffer, data + done, len - done); +} + +static int sha1_sparc64_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->count % SHA1_BLOCK_SIZE; + + /* Handle the fast case right here */ + if (partial + len < SHA1_BLOCK_SIZE) { + sctx->count += len; + memcpy(sctx->buffer + partial, data, len); + } else + __sha1_sparc64_update(sctx, data, len, partial); + + return 0; +} + +/* Add padding and return the message digest. */ +static int sha1_sparc64_final(struct shash_desc *desc, u8 *out) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + __be32 *dst = (__be32 *)out; + __be64 bits; + static const u8 padding[SHA1_BLOCK_SIZE] = { 0x80, }; + + bits = cpu_to_be64(sctx->count << 3); + + /* Pad out to 56 mod 64 and append length */ + index = sctx->count % SHA1_BLOCK_SIZE; + padlen = (index < 56) ? (56 - index) : ((SHA1_BLOCK_SIZE+56) - index); + + /* We need to fill a whole block for __sha1_sparc64_update() */ + if (padlen <= 56) { + sctx->count += padlen; + memcpy(sctx->buffer + index, padding, padlen); + } else { + __sha1_sparc64_update(sctx, padding, padlen, index); + } + __sha1_sparc64_update(sctx, (const u8 *)&bits, sizeof(bits), 56); + + /* Store state in digest */ + for (i = 0; i < 5; i++) + dst[i] = cpu_to_be32(sctx->state[i]); + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int sha1_sparc64_export(struct shash_desc *desc, void *out) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + + memcpy(out, sctx, sizeof(*sctx)); + + return 0; +} + +static int sha1_sparc64_import(struct shash_desc *desc, const void *in) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + + memcpy(sctx, in, sizeof(*sctx)); + + return 0; +} + +static struct shash_alg alg = { + .digestsize = SHA1_DIGEST_SIZE, + .init = sha1_sparc64_init, + .update = sha1_sparc64_update, + .final = sha1_sparc64_final, + .export = sha1_sparc64_export, + .import = sha1_sparc64_import, + .descsize = sizeof(struct sha1_state), + .statesize = sizeof(struct sha1_state), + .base = { + .cra_name = "sha1", + .cra_driver_name= "sha1-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static bool __init sparc64_has_sha1_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_SHA1)) + return false; + + return true; +} + +static int __init sha1_sparc64_mod_init(void) +{ + if (sparc64_has_sha1_opcode()) { + pr_info("Using sparc64 sha1 opcode optimized SHA-1 implementation\n"); + return crypto_register_shash(&alg); + } + pr_info("sparc64 sha1 opcode not available.\n"); + return -ENODEV; +} + +static void __exit sha1_sparc64_mod_fini(void) +{ + crypto_unregister_shash(&alg); +} + +module_init(sha1_sparc64_mod_init); +module_exit(sha1_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA1 Secure Hash Algorithm, sparc64 sha1 opcode accelerated"); + +MODULE_ALIAS("sha1"); + +#include "crop_devid.c" diff --git a/arch/sparc/crypto/sha256_asm.S b/arch/sparc/crypto/sha256_asm.S new file mode 100644 index 00000000000..b5f3d5826eb --- /dev/null +++ b/arch/sparc/crypto/sha256_asm.S @@ -0,0 +1,78 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> + +#include "opcodes.h" + +ENTRY(sha256_sparc64_transform) + /* %o0 = digest, %o1 = data, %o2 = rounds */ + VISEntryHalf + ld [%o0 + 0x00], %f0 + ld [%o0 + 0x04], %f1 + ld [%o0 + 0x08], %f2 + ld [%o0 + 0x0c], %f3 + ld [%o0 + 0x10], %f4 + ld [%o0 + 0x14], %f5 + andcc %o1, 0x7, %g0 + ld [%o0 + 0x18], %f6 + bne,pn %xcc, 10f + ld [%o0 + 0x1c], %f7 + +1: + ldd [%o1 + 0x00], %f8 + ldd [%o1 + 0x08], %f10 + ldd [%o1 + 0x10], %f12 + ldd [%o1 + 0x18], %f14 + ldd [%o1 + 0x20], %f16 + ldd [%o1 + 0x28], %f18 + ldd [%o1 + 0x30], %f20 + ldd [%o1 + 0x38], %f22 + + SHA256 + + subcc %o2, 1, %o2 + bne,pt %xcc, 1b + add %o1, 0x40, %o1 + +5: + st %f0, [%o0 + 0x00] + st %f1, [%o0 + 0x04] + st %f2, [%o0 + 0x08] + st %f3, [%o0 + 0x0c] + st %f4, [%o0 + 0x10] + st %f5, [%o0 + 0x14] + st %f6, [%o0 + 0x18] + st %f7, [%o0 + 0x1c] + retl + VISExitHalf +10: + alignaddr %o1, %g0, %o1 + + ldd [%o1 + 0x00], %f10 +1: + ldd [%o1 + 0x08], %f12 + ldd [%o1 + 0x10], %f14 + ldd [%o1 + 0x18], %f16 + ldd [%o1 + 0x20], %f18 + ldd [%o1 + 0x28], %f20 + ldd [%o1 + 0x30], %f22 + ldd [%o1 + 0x38], %f24 + ldd [%o1 + 0x40], %f26 + + faligndata %f10, %f12, %f8 + faligndata %f12, %f14, %f10 + faligndata %f14, %f16, %f12 + faligndata %f16, %f18, %f14 + faligndata %f18, %f20, %f16 + faligndata %f20, %f22, %f18 + faligndata %f22, %f24, %f20 + faligndata %f24, %f26, %f22 + + SHA256 + + subcc %o2, 1, %o2 + fsrc2 %f26, %f10 + bne,pt %xcc, 1b + add %o1, 0x40, %o1 + + ba,a,pt %xcc, 5b +ENDPROC(sha256_sparc64_transform) diff --git a/arch/sparc/crypto/sha256_glue.c b/arch/sparc/crypto/sha256_glue.c new file mode 100644 index 00000000000..04f555ab268 --- /dev/null +++ b/arch/sparc/crypto/sha256_glue.c @@ -0,0 +1,243 @@ +/* Glue code for SHA256 hashing optimized for sparc64 crypto opcodes. + * + * This is based largely upon crypto/sha256_generic.c + * + * Copyright (c) Jean-Luc Cooke <jlcooke@certainkey.com> + * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk> + * Copyright (c) 2002 James Morris <jmorris@intercode.com.au> + * SHA224 Support Copyright 2007 Intel Corporation <jonathan.lynch@intel.com> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <crypto/sha.h> + +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +asmlinkage void sha256_sparc64_transform(u32 *digest, const char *data, + unsigned int rounds); + +static int sha224_sparc64_init(struct shash_desc *desc) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + sctx->state[0] = SHA224_H0; + sctx->state[1] = SHA224_H1; + sctx->state[2] = SHA224_H2; + sctx->state[3] = SHA224_H3; + sctx->state[4] = SHA224_H4; + sctx->state[5] = SHA224_H5; + sctx->state[6] = SHA224_H6; + sctx->state[7] = SHA224_H7; + sctx->count = 0; + + return 0; +} + +static int sha256_sparc64_init(struct shash_desc *desc) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + sctx->state[0] = SHA256_H0; + sctx->state[1] = SHA256_H1; + sctx->state[2] = SHA256_H2; + sctx->state[3] = SHA256_H3; + sctx->state[4] = SHA256_H4; + sctx->state[5] = SHA256_H5; + sctx->state[6] = SHA256_H6; + sctx->state[7] = SHA256_H7; + sctx->count = 0; + + return 0; +} + +static void __sha256_sparc64_update(struct sha256_state *sctx, const u8 *data, + unsigned int len, unsigned int partial) +{ + unsigned int done = 0; + + sctx->count += len; + if (partial) { + done = SHA256_BLOCK_SIZE - partial; + memcpy(sctx->buf + partial, data, done); + sha256_sparc64_transform(sctx->state, sctx->buf, 1); + } + if (len - done >= SHA256_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / SHA256_BLOCK_SIZE; + + sha256_sparc64_transform(sctx->state, data + done, rounds); + done += rounds * SHA256_BLOCK_SIZE; + } + + memcpy(sctx->buf, data + done, len - done); +} + +static int sha256_sparc64_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->count % SHA256_BLOCK_SIZE; + + /* Handle the fast case right here */ + if (partial + len < SHA256_BLOCK_SIZE) { + sctx->count += len; + memcpy(sctx->buf + partial, data, len); + } else + __sha256_sparc64_update(sctx, data, len, partial); + + return 0; +} + +static int sha256_sparc64_final(struct shash_desc *desc, u8 *out) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + __be32 *dst = (__be32 *)out; + __be64 bits; + static const u8 padding[SHA256_BLOCK_SIZE] = { 0x80, }; + + bits = cpu_to_be64(sctx->count << 3); + + /* Pad out to 56 mod 64 and append length */ + index = sctx->count % SHA256_BLOCK_SIZE; + padlen = (index < 56) ? (56 - index) : ((SHA256_BLOCK_SIZE+56) - index); + + /* We need to fill a whole block for __sha256_sparc64_update() */ + if (padlen <= 56) { + sctx->count += padlen; + memcpy(sctx->buf + index, padding, padlen); + } else { + __sha256_sparc64_update(sctx, padding, padlen, index); + } + __sha256_sparc64_update(sctx, (const u8 *)&bits, sizeof(bits), 56); + + /* Store state in digest */ + for (i = 0; i < 8; i++) + dst[i] = cpu_to_be32(sctx->state[i]); + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int sha224_sparc64_final(struct shash_desc *desc, u8 *hash) +{ + u8 D[SHA256_DIGEST_SIZE]; + + sha256_sparc64_final(desc, D); + + memcpy(hash, D, SHA224_DIGEST_SIZE); + memset(D, 0, SHA256_DIGEST_SIZE); + + return 0; +} + +static int sha256_sparc64_export(struct shash_desc *desc, void *out) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + + memcpy(out, sctx, sizeof(*sctx)); + return 0; +} + +static int sha256_sparc64_import(struct shash_desc *desc, const void *in) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + + memcpy(sctx, in, sizeof(*sctx)); + return 0; +} + +static struct shash_alg sha256 = { + .digestsize = SHA256_DIGEST_SIZE, + .init = sha256_sparc64_init, + .update = sha256_sparc64_update, + .final = sha256_sparc64_final, + .export = sha256_sparc64_export, + .import = sha256_sparc64_import, + .descsize = sizeof(struct sha256_state), + .statesize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha256", + .cra_driver_name= "sha256-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA256_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static struct shash_alg sha224 = { + .digestsize = SHA224_DIGEST_SIZE, + .init = sha224_sparc64_init, + .update = sha256_sparc64_update, + .final = sha224_sparc64_final, + .descsize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha224", + .cra_driver_name= "sha224-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA224_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static bool __init sparc64_has_sha256_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_SHA256)) + return false; + + return true; +} + +static int __init sha256_sparc64_mod_init(void) +{ + if (sparc64_has_sha256_opcode()) { + int ret = crypto_register_shash(&sha224); + if (ret < 0) + return ret; + + ret = crypto_register_shash(&sha256); + if (ret < 0) { + crypto_unregister_shash(&sha224); + return ret; + } + + pr_info("Using sparc64 sha256 opcode optimized SHA-256/SHA-224 implementation\n"); + return 0; + } + pr_info("sparc64 sha256 opcode not available.\n"); + return -ENODEV; +} + +static void __exit sha256_sparc64_mod_fini(void) +{ + crypto_unregister_shash(&sha224); + crypto_unregister_shash(&sha256); +} + +module_init(sha256_sparc64_mod_init); +module_exit(sha256_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA-224 and SHA-256 Secure Hash Algorithm, sparc64 sha256 opcode accelerated"); + +MODULE_ALIAS("sha224"); +MODULE_ALIAS("sha256"); + +#include "crop_devid.c" diff --git a/arch/sparc/crypto/sha512_asm.S b/arch/sparc/crypto/sha512_asm.S new file mode 100644 index 00000000000..54bfba713c0 --- /dev/null +++ b/arch/sparc/crypto/sha512_asm.S @@ -0,0 +1,102 @@ +#include <linux/linkage.h> +#include <asm/visasm.h> + +#include "opcodes.h" + +ENTRY(sha512_sparc64_transform) + /* %o0 = digest, %o1 = data, %o2 = rounds */ + VISEntry + ldd [%o0 + 0x00], %f0 + ldd [%o0 + 0x08], %f2 + ldd [%o0 + 0x10], %f4 + ldd [%o0 + 0x18], %f6 + ldd [%o0 + 0x20], %f8 + ldd [%o0 + 0x28], %f10 + andcc %o1, 0x7, %g0 + ldd [%o0 + 0x30], %f12 + bne,pn %xcc, 10f + ldd [%o0 + 0x38], %f14 + +1: + ldd [%o1 + 0x00], %f16 + ldd [%o1 + 0x08], %f18 + ldd [%o1 + 0x10], %f20 + ldd [%o1 + 0x18], %f22 + ldd [%o1 + 0x20], %f24 + ldd [%o1 + 0x28], %f26 + ldd [%o1 + 0x30], %f28 + ldd [%o1 + 0x38], %f30 + ldd [%o1 + 0x40], %f32 + ldd [%o1 + 0x48], %f34 + ldd [%o1 + 0x50], %f36 + ldd [%o1 + 0x58], %f38 + ldd [%o1 + 0x60], %f40 + ldd [%o1 + 0x68], %f42 + ldd [%o1 + 0x70], %f44 + ldd [%o1 + 0x78], %f46 + + SHA512 + + subcc %o2, 1, %o2 + bne,pt %xcc, 1b + add %o1, 0x80, %o1 + +5: + std %f0, [%o0 + 0x00] + std %f2, [%o0 + 0x08] + std %f4, [%o0 + 0x10] + std %f6, [%o0 + 0x18] + std %f8, [%o0 + 0x20] + std %f10, [%o0 + 0x28] + std %f12, [%o0 + 0x30] + std %f14, [%o0 + 0x38] + retl + VISExit +10: + alignaddr %o1, %g0, %o1 + + ldd [%o1 + 0x00], %f18 +1: + ldd [%o1 + 0x08], %f20 + ldd [%o1 + 0x10], %f22 + ldd [%o1 + 0x18], %f24 + ldd [%o1 + 0x20], %f26 + ldd [%o1 + 0x28], %f28 + ldd [%o1 + 0x30], %f30 + ldd [%o1 + 0x38], %f32 + ldd [%o1 + 0x40], %f34 + ldd [%o1 + 0x48], %f36 + ldd [%o1 + 0x50], %f38 + ldd [%o1 + 0x58], %f40 + ldd [%o1 + 0x60], %f42 + ldd [%o1 + 0x68], %f44 + ldd [%o1 + 0x70], %f46 + ldd [%o1 + 0x78], %f48 + ldd [%o1 + 0x80], %f50 + + faligndata %f18, %f20, %f16 + faligndata %f20, %f22, %f18 + faligndata %f22, %f24, %f20 + faligndata %f24, %f26, %f22 + faligndata %f26, %f28, %f24 + faligndata %f28, %f30, %f26 + faligndata %f30, %f32, %f28 + faligndata %f32, %f34, %f30 + faligndata %f34, %f36, %f32 + faligndata %f36, %f38, %f34 + faligndata %f38, %f40, %f36 + faligndata %f40, %f42, %f38 + faligndata %f42, %f44, %f40 + faligndata %f44, %f46, %f42 + faligndata %f46, %f48, %f44 + faligndata %f48, %f50, %f46 + + SHA512 + + subcc %o2, 1, %o2 + fsrc2 %f50, %f18 + bne,pt %xcc, 1b + add %o1, 0x80, %o1 + + ba,a,pt %xcc, 5b +ENDPROC(sha512_sparc64_transform) diff --git a/arch/sparc/crypto/sha512_glue.c b/arch/sparc/crypto/sha512_glue.c new file mode 100644 index 00000000000..f04d1994d19 --- /dev/null +++ b/arch/sparc/crypto/sha512_glue.c @@ -0,0 +1,228 @@ +/* Glue code for SHA512 hashing optimized for sparc64 crypto opcodes. + * + * This is based largely upon crypto/sha512_generic.c + * + * Copyright (c) Jean-Luc Cooke <jlcooke@certainkey.com> + * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk> + * Copyright (c) 2003 Kyle McMartin <kyle@debian.org> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <crypto/sha.h> + +#include <asm/pstate.h> +#include <asm/elf.h> + +#include "opcodes.h" + +asmlinkage void sha512_sparc64_transform(u64 *digest, const char *data, + unsigned int rounds); + +static int sha512_sparc64_init(struct shash_desc *desc) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + sctx->state[0] = SHA512_H0; + sctx->state[1] = SHA512_H1; + sctx->state[2] = SHA512_H2; + sctx->state[3] = SHA512_H3; + sctx->state[4] = SHA512_H4; + sctx->state[5] = SHA512_H5; + sctx->state[6] = SHA512_H6; + sctx->state[7] = SHA512_H7; + sctx->count[0] = sctx->count[1] = 0; + + return 0; +} + +static int sha384_sparc64_init(struct shash_desc *desc) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + sctx->state[0] = SHA384_H0; + sctx->state[1] = SHA384_H1; + sctx->state[2] = SHA384_H2; + sctx->state[3] = SHA384_H3; + sctx->state[4] = SHA384_H4; + sctx->state[5] = SHA384_H5; + sctx->state[6] = SHA384_H6; + sctx->state[7] = SHA384_H7; + sctx->count[0] = sctx->count[1] = 0; + + return 0; +} + +static void __sha512_sparc64_update(struct sha512_state *sctx, const u8 *data, + unsigned int len, unsigned int partial) +{ + unsigned int done = 0; + + if ((sctx->count[0] += len) < len) + sctx->count[1]++; + if (partial) { + done = SHA512_BLOCK_SIZE - partial; + memcpy(sctx->buf + partial, data, done); + sha512_sparc64_transform(sctx->state, sctx->buf, 1); + } + if (len - done >= SHA512_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / SHA512_BLOCK_SIZE; + + sha512_sparc64_transform(sctx->state, data + done, rounds); + done += rounds * SHA512_BLOCK_SIZE; + } + + memcpy(sctx->buf, data + done, len - done); +} + +static int sha512_sparc64_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->count[0] % SHA512_BLOCK_SIZE; + + /* Handle the fast case right here */ + if (partial + len < SHA512_BLOCK_SIZE) { + if ((sctx->count[0] += len) < len) + sctx->count[1]++; + memcpy(sctx->buf + partial, data, len); + } else + __sha512_sparc64_update(sctx, data, len, partial); + + return 0; +} + +static int sha512_sparc64_final(struct shash_desc *desc, u8 *out) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + __be64 *dst = (__be64 *)out; + __be64 bits[2]; + static const u8 padding[SHA512_BLOCK_SIZE] = { 0x80, }; + + /* Save number of bits */ + bits[1] = cpu_to_be64(sctx->count[0] << 3); + bits[0] = cpu_to_be64(sctx->count[1] << 3 | sctx->count[0] >> 61); + + /* Pad out to 112 mod 128 and append length */ + index = sctx->count[0] % SHA512_BLOCK_SIZE; + padlen = (index < 112) ? (112 - index) : ((SHA512_BLOCK_SIZE+112) - index); + + /* We need to fill a whole block for __sha512_sparc64_update() */ + if (padlen <= 112) { + if ((sctx->count[0] += padlen) < padlen) + sctx->count[1]++; + memcpy(sctx->buf + index, padding, padlen); + } else { + __sha512_sparc64_update(sctx, padding, padlen, index); + } + __sha512_sparc64_update(sctx, (const u8 *)&bits, sizeof(bits), 112); + + /* Store state in digest */ + for (i = 0; i < 8; i++) + dst[i] = cpu_to_be64(sctx->state[i]); + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int sha384_sparc64_final(struct shash_desc *desc, u8 *hash) +{ + u8 D[64]; + + sha512_sparc64_final(desc, D); + + memcpy(hash, D, 48); + memset(D, 0, 64); + + return 0; +} + +static struct shash_alg sha512 = { + .digestsize = SHA512_DIGEST_SIZE, + .init = sha512_sparc64_init, + .update = sha512_sparc64_update, + .final = sha512_sparc64_final, + .descsize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha512", + .cra_driver_name= "sha512-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA512_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static struct shash_alg sha384 = { + .digestsize = SHA384_DIGEST_SIZE, + .init = sha384_sparc64_init, + .update = sha512_sparc64_update, + .final = sha384_sparc64_final, + .descsize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha384", + .cra_driver_name= "sha384-sparc64", + .cra_priority = SPARC_CR_OPCODE_PRIORITY, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA384_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static bool __init sparc64_has_sha512_opcode(void) +{ + unsigned long cfr; + + if (!(sparc64_elf_hwcap & HWCAP_SPARC_CRYPTO)) + return false; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + if (!(cfr & CFR_SHA512)) + return false; + + return true; +} + +static int __init sha512_sparc64_mod_init(void) +{ + if (sparc64_has_sha512_opcode()) { + int ret = crypto_register_shash(&sha384); + if (ret < 0) + return ret; + + ret = crypto_register_shash(&sha512); + if (ret < 0) { + crypto_unregister_shash(&sha384); + return ret; + } + + pr_info("Using sparc64 sha512 opcode optimized SHA-512/SHA-384 implementation\n"); + return 0; + } + pr_info("sparc64 sha512 opcode not available.\n"); + return -ENODEV; +} + +static void __exit sha512_sparc64_mod_fini(void) +{ + crypto_unregister_shash(&sha384); + crypto_unregister_shash(&sha512); +} + +module_init(sha512_sparc64_mod_init); +module_exit(sha512_sparc64_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA-384 and SHA-512 Secure Hash Algorithm, sparc64 sha512 opcode accelerated"); + +MODULE_ALIAS("sha384"); +MODULE_ALIAS("sha512"); + +#include "crop_devid.c" diff --git a/arch/sparc/include/asm/Kbuild b/arch/sparc/include/asm/Kbuild index 67f83e0a0d6..a4582181800 100644 --- a/arch/sparc/include/asm/Kbuild +++ b/arch/sparc/include/asm/Kbuild @@ -1,24 +1,21 @@ # User exported sparc header files -include include/asm-generic/Kbuild.asm -header-y += apc.h -header-y += asi.h -header-y += display7seg.h -header-y += envctrl.h -header-y += fbio.h -header-y += jsflash.h -header-y += openpromio.h -header-y += perfctr.h -header-y += psrcompat.h -header-y += psr.h -header-y += pstate.h -header-y += traps.h -header-y += uctx.h -header-y += utrap.h -header-y += watchdog.h +generic-y += clkdev.h +generic-y += cputime.h generic-y += div64.h -generic-y += local64.h +generic-y += emergency-restart.h +generic-y += exec.h +generic-y += hash.h generic-y += irq_regs.h +generic-y += linkage.h generic-y += local.h +generic-y += local64.h +generic-y += mcs_spinlock.h +generic-y += module.h +generic-y += mutex.h +generic-y += preempt.h +generic-y += serial.h +generic-y += trace_clock.h +generic-y += types.h generic-y += word-at-a-time.h diff --git a/arch/sparc/include/asm/atomic_32.h b/arch/sparc/include/asm/atomic_32.h index 905832aa9e9..7aed2be45b4 100644 --- a/arch/sparc/include/asm/atomic_32.h +++ b/arch/sparc/include/asm/atomic_32.h @@ -14,16 +14,17 @@ #include <linux/types.h> #include <asm/cmpxchg.h> +#include <asm/barrier.h> #include <asm-generic/atomic64.h> #define ATOMIC_INIT(i) { (i) } -extern int __atomic_add_return(int, atomic_t *); -extern int atomic_cmpxchg(atomic_t *, int, int); +int __atomic_add_return(int, atomic_t *); +int atomic_cmpxchg(atomic_t *, int, int); #define atomic_xchg(v, new) (xchg(&((v)->counter), new)) -extern int __atomic_add_unless(atomic_t *, int, int); -extern void atomic_set(atomic_t *, int); +int __atomic_add_unless(atomic_t *, int, int); +void atomic_set(atomic_t *, int); #define atomic_read(v) (*(volatile int *)&(v)->counter) @@ -52,10 +53,4 @@ extern void atomic_set(atomic_t *, int); #define atomic_dec_and_test(v) (atomic_dec_return(v) == 0) #define atomic_sub_and_test(i, v) (atomic_sub_return(i, v) == 0) -/* Atomic operations are already serializing */ -#define smp_mb__before_atomic_dec() barrier() -#define smp_mb__after_atomic_dec() barrier() -#define smp_mb__before_atomic_inc() barrier() -#define smp_mb__after_atomic_inc() barrier() - #endif /* !(__ARCH_SPARC_ATOMIC__) */ diff --git a/arch/sparc/include/asm/atomic_64.h b/arch/sparc/include/asm/atomic_64.h index ce35a1cf1a2..bb894c8bec5 100644 --- a/arch/sparc/include/asm/atomic_64.h +++ b/arch/sparc/include/asm/atomic_64.h @@ -1,7 +1,7 @@ /* atomic.h: Thankfully the V9 is at least reasonable for this * stuff. * - * Copyright (C) 1996, 1997, 2000 David S. Miller (davem@redhat.com) + * Copyright (C) 1996, 1997, 2000, 2012 David S. Miller (davem@redhat.com) */ #ifndef __ARCH_SPARC64_ATOMIC__ @@ -9,6 +9,7 @@ #include <linux/types.h> #include <asm/cmpxchg.h> +#include <asm/barrier.h> #define ATOMIC_INIT(i) { (i) } #define ATOMIC64_INIT(i) { (i) } @@ -19,15 +20,15 @@ #define atomic_set(v, i) (((v)->counter) = i) #define atomic64_set(v, i) (((v)->counter) = i) -extern void atomic_add(int, atomic_t *); -extern void atomic64_add(long, atomic64_t *); -extern void atomic_sub(int, atomic_t *); -extern void atomic64_sub(long, atomic64_t *); +void atomic_add(int, atomic_t *); +void atomic64_add(long, atomic64_t *); +void atomic_sub(int, atomic_t *); +void atomic64_sub(long, atomic64_t *); -extern int atomic_add_ret(int, atomic_t *); -extern long atomic64_add_ret(long, atomic64_t *); -extern int atomic_sub_ret(int, atomic_t *); -extern long atomic64_sub_ret(long, atomic64_t *); +int atomic_add_ret(int, atomic_t *); +long atomic64_add_ret(long, atomic64_t *); +int atomic_sub_ret(int, atomic_t *); +long atomic64_sub_ret(long, atomic64_t *); #define atomic_dec_return(v) atomic_sub_ret(1, v) #define atomic64_dec_return(v) atomic64_sub_ret(1, v) @@ -106,10 +107,6 @@ static inline long atomic64_add_unless(atomic64_t *v, long a, long u) #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0) -/* Atomic operations are already serializing */ -#define smp_mb__before_atomic_dec() barrier() -#define smp_mb__after_atomic_dec() barrier() -#define smp_mb__before_atomic_inc() barrier() -#define smp_mb__after_atomic_inc() barrier() +long atomic64_dec_if_positive(atomic64_t *v); #endif /* !(__ARCH_SPARC64_ATOMIC__) */ diff --git a/arch/sparc/include/asm/auxio.h b/arch/sparc/include/asm/auxio.h index 13dc67f0301..3e09a07b77e 100644 --- a/arch/sparc/include/asm/auxio.h +++ b/arch/sparc/include/asm/auxio.h @@ -1,5 +1,12 @@ #ifndef ___ASM_SPARC_AUXIO_H #define ___ASM_SPARC_AUXIO_H + +#ifndef __ASSEMBLY__ + +extern void __iomem *auxio_register; + +#endif /* ifndef __ASSEMBLY__ */ + #if defined(__sparc__) && defined(__arch64__) #include <asm/auxio_64.h> #else diff --git a/arch/sparc/include/asm/auxio_32.h b/arch/sparc/include/asm/auxio_32.h index 3a319775ae3..5d685df427b 100644 --- a/arch/sparc/include/asm/auxio_32.h +++ b/arch/sparc/include/asm/auxio_32.h @@ -34,8 +34,8 @@ * NOTE: these routines are implementation dependent-- * understand the hardware you are querying! */ -extern void set_auxio(unsigned char bits_on, unsigned char bits_off); -extern unsigned char get_auxio(void); /* .../asm/floppy.h */ +void set_auxio(unsigned char bits_on, unsigned char bits_off); +unsigned char get_auxio(void); /* .../asm/floppy.h */ /* * The following routines are provided for driver-compatibility @@ -78,7 +78,7 @@ do { \ /* AUXIO2 (Power Off Control) */ -extern __volatile__ unsigned char * auxio_power_register; +extern volatile u8 __iomem *auxio_power_register; #define AUXIO_POWER_DETECT_FAILURE 32 #define AUXIO_POWER_CLEAR_FAILURE 2 diff --git a/arch/sparc/include/asm/auxio_64.h b/arch/sparc/include/asm/auxio_64.h index f61cd1e3e39..6079e59a7ad 100644 --- a/arch/sparc/include/asm/auxio_64.h +++ b/arch/sparc/include/asm/auxio_64.h @@ -75,8 +75,6 @@ #ifndef __ASSEMBLY__ -extern void __iomem *auxio_register; - #define AUXIO_LTE_ON 1 #define AUXIO_LTE_OFF 0 @@ -84,7 +82,7 @@ extern void __iomem *auxio_register; * * on - AUXIO_LTE_ON or AUXIO_LTE_OFF */ -extern void auxio_set_lte(int on); +void auxio_set_lte(int on); #define AUXIO_LED_ON 1 #define AUXIO_LED_OFF 0 @@ -93,7 +91,7 @@ extern void auxio_set_lte(int on); * * on - AUXIO_LED_ON or AUXIO_LED_OFF */ -extern void auxio_set_led(int on); +void auxio_set_led(int on); #endif /* ifndef __ASSEMBLY__ */ diff --git a/arch/sparc/include/asm/backoff.h b/arch/sparc/include/asm/backoff.h index db3af0d30fb..4e02086b839 100644 --- a/arch/sparc/include/asm/backoff.h +++ b/arch/sparc/include/asm/backoff.h @@ -1,6 +1,46 @@ #ifndef _SPARC64_BACKOFF_H #define _SPARC64_BACKOFF_H +/* The macros in this file implement an exponential backoff facility + * for atomic operations. + * + * When multiple threads compete on an atomic operation, it is + * possible for one thread to be continually denied a successful + * completion of the compare-and-swap instruction. Heavily + * threaded cpu implementations like Niagara can compound this + * problem even further. + * + * When an atomic operation fails and needs to be retried, we spin a + * certain number of times. At each subsequent failure of the same + * operation we double the spin count, realizing an exponential + * backoff. + * + * When we spin, we try to use an operation that will cause the + * current cpu strand to block, and therefore make the core fully + * available to any other other runnable strands. There are two + * options, based upon cpu capabilities. + * + * On all cpus prior to SPARC-T4 we do three dummy reads of the + * condition code register. Each read blocks the strand for something + * between 40 and 50 cpu cycles. + * + * For SPARC-T4 and later we have a special "pause" instruction + * available. This is implemented using writes to register %asr27. + * The cpu will block the number of cycles written into the register, + * unless a disrupting trap happens first. SPARC-T4 specifically + * implements pause with a granularity of 8 cycles. Each strand has + * an internal pause counter which decrements every 8 cycles. So the + * chip shifts the %asr27 value down by 3 bits, and writes the result + * into the pause counter. If a value smaller than 8 is written, the + * chip blocks for 1 cycle. + * + * To achieve the same amount of backoff as the three %ccr reads give + * on earlier chips, we shift the backoff value up by 7 bits. (Three + * %ccr reads block for about 128 cycles, 1 << 7 == 128) We write the + * whole amount we want to block into the pause register, rather than + * loop writing 128 each time. + */ + #define BACKOFF_LIMIT (4 * 1024) #ifdef CONFIG_SMP @@ -11,16 +51,25 @@ #define BACKOFF_LABEL(spin_label, continue_label) \ spin_label -#define BACKOFF_SPIN(reg, tmp, label) \ - mov reg, tmp; \ -88: brnz,pt tmp, 88b; \ - sub tmp, 1, tmp; \ - set BACKOFF_LIMIT, tmp; \ - cmp reg, tmp; \ - bg,pn %xcc, label; \ - nop; \ - ba,pt %xcc, label; \ - sllx reg, 1, reg; +#define BACKOFF_SPIN(reg, tmp, label) \ + mov reg, tmp; \ +88: rd %ccr, %g0; \ + rd %ccr, %g0; \ + rd %ccr, %g0; \ + .section .pause_3insn_patch,"ax";\ + .word 88b; \ + sllx tmp, 7, tmp; \ + wr tmp, 0, %asr27; \ + clr tmp; \ + .previous; \ + brnz,pt tmp, 88b; \ + sub tmp, 1, tmp; \ + set BACKOFF_LIMIT, tmp; \ + cmp reg, tmp; \ + bg,pn %xcc, label; \ + nop; \ + ba,pt %xcc, label; \ + sllx reg, 1, reg; #else diff --git a/arch/sparc/include/asm/barrier_32.h b/arch/sparc/include/asm/barrier_32.h index c1b76654ee7..ae69eda288f 100644 --- a/arch/sparc/include/asm/barrier_32.h +++ b/arch/sparc/include/asm/barrier_32.h @@ -1,15 +1,7 @@ #ifndef __SPARC_BARRIER_H #define __SPARC_BARRIER_H -/* XXX Change this if we ever use a PSO mode kernel. */ -#define mb() __asm__ __volatile__ ("" : : : "memory") -#define rmb() mb() -#define wmb() mb() -#define read_barrier_depends() do { } while(0) -#define set_mb(__var, __value) do { __var = __value; mb(); } while(0) -#define smp_mb() __asm__ __volatile__("":::"memory") -#define smp_rmb() __asm__ __volatile__("":::"memory") -#define smp_wmb() __asm__ __volatile__("":::"memory") -#define smp_read_barrier_depends() do { } while(0) +#include <asm/processor.h> /* for nop() */ +#include <asm-generic/barrier.h> #endif /* !(__SPARC_BARRIER_H) */ diff --git a/arch/sparc/include/asm/barrier_64.h b/arch/sparc/include/asm/barrier_64.h index 95d45986f90..305dcc3dc72 100644 --- a/arch/sparc/include/asm/barrier_64.h +++ b/arch/sparc/include/asm/barrier_64.h @@ -53,4 +53,22 @@ do { __asm__ __volatile__("ba,pt %%xcc, 1f\n\t" \ #define smp_read_barrier_depends() do { } while(0) +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ___p1; \ +}) + +#define smp_mb__before_atomic() barrier() +#define smp_mb__after_atomic() barrier() + #endif /* !(__SPARC64_BARRIER_H) */ diff --git a/arch/sparc/include/asm/bitext.h b/arch/sparc/include/asm/bitext.h index 297b2f2fcb4..9c988bf3adb 100644 --- a/arch/sparc/include/asm/bitext.h +++ b/arch/sparc/include/asm/bitext.h @@ -20,8 +20,8 @@ struct bit_map { int num_colors; }; -extern int bit_map_string_get(struct bit_map *t, int len, int align); -extern void bit_map_clear(struct bit_map *t, int offset, int len); -extern void bit_map_init(struct bit_map *t, unsigned long *map, int size); +int bit_map_string_get(struct bit_map *t, int len, int align); +void bit_map_clear(struct bit_map *t, int offset, int len); +void bit_map_init(struct bit_map *t, unsigned long *map, int size); #endif /* defined(_SPARC_BITEXT_H) */ diff --git a/arch/sparc/include/asm/bitops_32.h b/arch/sparc/include/asm/bitops_32.h index 25a676653d4..600ed1d9c8c 100644 --- a/arch/sparc/include/asm/bitops_32.h +++ b/arch/sparc/include/asm/bitops_32.h @@ -18,9 +18,9 @@ #error only <linux/bitops.h> can be included directly #endif -extern unsigned long ___set_bit(unsigned long *addr, unsigned long mask); -extern unsigned long ___clear_bit(unsigned long *addr, unsigned long mask); -extern unsigned long ___change_bit(unsigned long *addr, unsigned long mask); +unsigned long ___set_bit(unsigned long *addr, unsigned long mask); +unsigned long ___clear_bit(unsigned long *addr, unsigned long mask); +unsigned long ___change_bit(unsigned long *addr, unsigned long mask); /* * Set bit 'nr' in 32-bit quantity at address 'addr' where bit '0' @@ -90,9 +90,6 @@ static inline void change_bit(unsigned long nr, volatile unsigned long *addr) #include <asm-generic/bitops/non-atomic.h> -#define smp_mb__before_clear_bit() do { } while(0) -#define smp_mb__after_clear_bit() do { } while(0) - #include <asm-generic/bitops/ffz.h> #include <asm-generic/bitops/__ffs.h> #include <asm-generic/bitops/sched.h> diff --git a/arch/sparc/include/asm/bitops_64.h b/arch/sparc/include/asm/bitops_64.h index 29011cc0e4b..2d522402a93 100644 --- a/arch/sparc/include/asm/bitops_64.h +++ b/arch/sparc/include/asm/bitops_64.h @@ -13,27 +13,25 @@ #include <linux/compiler.h> #include <asm/byteorder.h> +#include <asm/barrier.h> -extern int test_and_set_bit(unsigned long nr, volatile unsigned long *addr); -extern int test_and_clear_bit(unsigned long nr, volatile unsigned long *addr); -extern int test_and_change_bit(unsigned long nr, volatile unsigned long *addr); -extern void set_bit(unsigned long nr, volatile unsigned long *addr); -extern void clear_bit(unsigned long nr, volatile unsigned long *addr); -extern void change_bit(unsigned long nr, volatile unsigned long *addr); +int test_and_set_bit(unsigned long nr, volatile unsigned long *addr); +int test_and_clear_bit(unsigned long nr, volatile unsigned long *addr); +int test_and_change_bit(unsigned long nr, volatile unsigned long *addr); +void set_bit(unsigned long nr, volatile unsigned long *addr); +void clear_bit(unsigned long nr, volatile unsigned long *addr); +void change_bit(unsigned long nr, volatile unsigned long *addr); #include <asm-generic/bitops/non-atomic.h> -#define smp_mb__before_clear_bit() barrier() -#define smp_mb__after_clear_bit() barrier() - #include <asm-generic/bitops/fls.h> #include <asm-generic/bitops/__fls.h> #include <asm-generic/bitops/fls64.h> #ifdef __KERNEL__ -extern int ffs(int x); -extern unsigned long __ffs(unsigned long); +int ffs(int x); +unsigned long __ffs(unsigned long); #include <asm-generic/bitops/ffz.h> #include <asm-generic/bitops/sched.h> @@ -43,10 +41,10 @@ extern unsigned long __ffs(unsigned long); * of bits set) of a N-bit word */ -extern unsigned long __arch_hweight64(__u64 w); -extern unsigned int __arch_hweight32(unsigned int w); -extern unsigned int __arch_hweight16(unsigned int w); -extern unsigned int __arch_hweight8(unsigned int w); +unsigned long __arch_hweight64(__u64 w); +unsigned int __arch_hweight32(unsigned int w); +unsigned int __arch_hweight16(unsigned int w); +unsigned int __arch_hweight8(unsigned int w); #include <asm-generic/bitops/const_hweight.h> #include <asm-generic/bitops/lock.h> diff --git a/arch/sparc/include/asm/btext.h b/arch/sparc/include/asm/btext.h index 9b2bc6b6ed0..75a32b109e1 100644 --- a/arch/sparc/include/asm/btext.h +++ b/arch/sparc/include/asm/btext.h @@ -1,6 +1,6 @@ #ifndef _SPARC_BTEXT_H #define _SPARC_BTEXT_H -extern int btext_find_display(void); +int btext_find_display(void); #endif /* _SPARC_BTEXT_H */ diff --git a/arch/sparc/include/asm/bug.h b/arch/sparc/include/asm/bug.h index 6bd9f43cb5a..eaa8f8d3812 100644 --- a/arch/sparc/include/asm/bug.h +++ b/arch/sparc/include/asm/bug.h @@ -5,7 +5,7 @@ #include <linux/compiler.h> #ifdef CONFIG_DEBUG_BUGVERBOSE -extern void do_BUG(const char *file, int line); +void do_BUG(const char *file, int line); #define BUG() do { \ do_BUG(__FILE__, __LINE__); \ __builtin_trap(); \ @@ -20,6 +20,6 @@ extern void do_BUG(const char *file, int line); #include <asm-generic/bug.h> struct pt_regs; -extern void die_if_kernel(char *str, struct pt_regs *regs) __attribute__ ((noreturn)); +void __noreturn die_if_kernel(char *str, struct pt_regs *regs); #endif diff --git a/arch/sparc/include/asm/cacheflush_32.h b/arch/sparc/include/asm/cacheflush_32.h index bb014c24f31..12164006181 100644 --- a/arch/sparc/include/asm/cacheflush_32.h +++ b/arch/sparc/include/asm/cacheflush_32.h @@ -36,7 +36,7 @@ #define flush_page_for_dma(addr) \ sparc32_cachetlb_ops->page_for_dma(addr) -extern void sparc_flush_page_to_ram(struct page *page); +void sparc_flush_page_to_ram(struct page *page); #define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 1 #define flush_dcache_page(page) sparc_flush_page_to_ram(page) @@ -51,8 +51,8 @@ extern void sparc_flush_page_to_ram(struct page *page); * way the windows are all clean for the next process and the stack * frames are up to date. */ -extern void flush_user_windows(void); -extern void kill_user_windows(void); -extern void flushw_all(void); +void flush_user_windows(void); +void kill_user_windows(void); +void flushw_all(void); #endif /* _SPARC_CACHEFLUSH_H */ diff --git a/arch/sparc/include/asm/cacheflush_64.h b/arch/sparc/include/asm/cacheflush_64.h index 301736d9e7a..38965379e35 100644 --- a/arch/sparc/include/asm/cacheflush_64.h +++ b/arch/sparc/include/asm/cacheflush_64.h @@ -10,7 +10,7 @@ /* Cache flush operations. */ #define flushw_all() __asm__ __volatile__("flushw") -extern void __flushw_user(void); +void __flushw_user(void); #define flushw_user() __flushw_user() #define flush_user_windows flushw_user @@ -30,29 +30,29 @@ extern void __flushw_user(void); * use block commit stores (which invalidate icache lines) during * module load, so we need this. */ -extern void flush_icache_range(unsigned long start, unsigned long end); -extern void __flush_icache_page(unsigned long); +void flush_icache_range(unsigned long start, unsigned long end); +void __flush_icache_page(unsigned long); -extern void __flush_dcache_page(void *addr, int flush_icache); -extern void flush_dcache_page_impl(struct page *page); +void __flush_dcache_page(void *addr, int flush_icache); +void flush_dcache_page_impl(struct page *page); #ifdef CONFIG_SMP -extern void smp_flush_dcache_page_impl(struct page *page, int cpu); -extern void flush_dcache_page_all(struct mm_struct *mm, struct page *page); +void smp_flush_dcache_page_impl(struct page *page, int cpu); +void flush_dcache_page_all(struct mm_struct *mm, struct page *page); #else #define smp_flush_dcache_page_impl(page,cpu) flush_dcache_page_impl(page) #define flush_dcache_page_all(mm,page) flush_dcache_page_impl(page) #endif -extern void __flush_dcache_range(unsigned long start, unsigned long end); +void __flush_dcache_range(unsigned long start, unsigned long end); #define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 1 -extern void flush_dcache_page(struct page *page); +void flush_dcache_page(struct page *page); #define flush_icache_page(vma, pg) do { } while(0) #define flush_icache_user_range(vma,pg,adr,len) do { } while (0) -extern void flush_ptrace_access(struct vm_area_struct *, struct page *, - unsigned long uaddr, void *kaddr, - unsigned long len, int write); +void flush_ptrace_access(struct vm_area_struct *, struct page *, + unsigned long uaddr, void *kaddr, + unsigned long len, int write); #define copy_to_user_page(vma, page, vaddr, dst, src, len) \ do { \ diff --git a/arch/sparc/include/asm/checksum_32.h b/arch/sparc/include/asm/checksum_32.h index bdbda1453aa..426b2389a1c 100644 --- a/arch/sparc/include/asm/checksum_32.h +++ b/arch/sparc/include/asm/checksum_32.h @@ -29,7 +29,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -extern __wsum csum_partial(const void *buff, int len, __wsum sum); +__wsum csum_partial(const void *buff, int len, __wsum sum); /* the same as csum_partial, but copies from fs:src while it * checksums @@ -38,7 +38,7 @@ extern __wsum csum_partial(const void *buff, int len, __wsum sum); * better 64-bit) boundary */ -extern unsigned int __csum_partial_copy_sparc_generic (const unsigned char *, unsigned char *); +unsigned int __csum_partial_copy_sparc_generic (const unsigned char *, unsigned char *); static inline __wsum csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum) @@ -238,4 +238,16 @@ static inline __sum16 ip_compute_csum(const void *buff, int len) return csum_fold(csum_partial(buff, len, 0)); } +#define HAVE_ARCH_CSUM_ADD +static inline __wsum csum_add(__wsum csum, __wsum addend) +{ + __asm__ __volatile__( + "addcc %0, %1, %0\n" + "addx %0, %%g0, %0" + : "=r" (csum) + : "r" (addend), "0" (csum)); + + return csum; +} + #endif /* !(__SPARC_CHECKSUM_H) */ diff --git a/arch/sparc/include/asm/checksum_64.h b/arch/sparc/include/asm/checksum_64.h index 019b9615e43..b8779a6a591 100644 --- a/arch/sparc/include/asm/checksum_64.h +++ b/arch/sparc/include/asm/checksum_64.h @@ -29,7 +29,7 @@ * * it's best to have buff aligned on a 32-bit boundary */ -extern __wsum csum_partial(const void * buff, int len, __wsum sum); +__wsum csum_partial(const void * buff, int len, __wsum sum); /* the same as csum_partial, but copies from user space while it * checksums @@ -37,12 +37,12 @@ extern __wsum csum_partial(const void * buff, int len, __wsum sum); * here even more important to align src and dst on a 32-bit (or even * better 64-bit) boundary */ -extern __wsum csum_partial_copy_nocheck(const void *src, void *dst, - int len, __wsum sum); +__wsum csum_partial_copy_nocheck(const void *src, void *dst, + int len, __wsum sum); -extern long __csum_partial_copy_from_user(const void __user *src, - void *dst, int len, - __wsum sum); +long __csum_partial_copy_from_user(const void __user *src, + void *dst, int len, + __wsum sum); static inline __wsum csum_partial_copy_from_user(const void __user *src, @@ -59,9 +59,9 @@ csum_partial_copy_from_user(const void __user *src, * Copy and checksum to user */ #define HAVE_CSUM_COPY_USER -extern long __csum_partial_copy_to_user(const void *src, - void __user *dst, int len, - __wsum sum); +long __csum_partial_copy_to_user(const void *src, + void __user *dst, int len, + __wsum sum); static inline __wsum csum_and_copy_to_user(const void *src, @@ -77,7 +77,7 @@ csum_and_copy_to_user(const void *src, /* ihl is always 5 or greater, almost always is 5, and iph is word aligned * the majority of the time. */ -extern __sum16 ip_fast_csum(const void *iph, unsigned int ihl); +__sum16 ip_fast_csum(const void *iph, unsigned int ihl); /* Fold a partial checksum without adding pseudo headers. */ static inline __sum16 csum_fold(__wsum sum) @@ -96,9 +96,9 @@ static inline __sum16 csum_fold(__wsum sum) } static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, - unsigned int len, - unsigned short proto, - __wsum sum) + unsigned int len, + unsigned short proto, + __wsum sum) { __asm__ __volatile__( " addcc %1, %0, %0\n" @@ -116,9 +116,9 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr, * returns a 16-bit checksum, already complemented */ static inline __sum16 csum_tcpudp_magic(__be32 saddr, __be32 daddr, - unsigned short len, - unsigned short proto, - __wsum sum) + unsigned short len, + unsigned short proto, + __wsum sum) { return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); } @@ -164,4 +164,16 @@ static inline __sum16 ip_compute_csum(const void *buff, int len) return csum_fold(csum_partial(buff, len, 0)); } +#define HAVE_ARCH_CSUM_ADD +static inline __wsum csum_add(__wsum csum, __wsum addend) +{ + __asm__ __volatile__( + "addcc %0, %1, %0\n" + "addx %0, %%g0, %0" + : "=r" (csum) + : "r" (addend), "0" (csum)); + + return csum; +} + #endif /* !(__SPARC64_CHECKSUM_H) */ diff --git a/arch/sparc/include/asm/cmpxchg_32.h b/arch/sparc/include/asm/cmpxchg_32.h index 1fae1a02e3c..32c29a133f9 100644 --- a/arch/sparc/include/asm/cmpxchg_32.h +++ b/arch/sparc/include/asm/cmpxchg_32.h @@ -20,7 +20,7 @@ static inline unsigned long xchg_u32(__volatile__ unsigned long *m, unsigned lon return val; } -extern void __xchg_called_with_bad_pointer(void); +void __xchg_called_with_bad_pointer(void); static inline unsigned long __xchg(unsigned long x, __volatile__ void * ptr, int size) { @@ -45,9 +45,9 @@ static inline unsigned long __xchg(unsigned long x, __volatile__ void * ptr, int #define __HAVE_ARCH_CMPXCHG 1 /* bug catcher for when unsupported size is used - won't link */ -extern void __cmpxchg_called_with_bad_pointer(void); +void __cmpxchg_called_with_bad_pointer(void); /* we only need to support cmpxchg of a u32 on sparc */ -extern unsigned long __cmpxchg_u32(volatile u32 *m, u32 old, u32 new_); +unsigned long __cmpxchg_u32(volatile u32 *m, u32 old, u32 new_); /* don't worry...optimizer will get rid of most of this */ static inline unsigned long diff --git a/arch/sparc/include/asm/cmpxchg_64.h b/arch/sparc/include/asm/cmpxchg_64.h index b30eb37294c..0e1ed6cfbf6 100644 --- a/arch/sparc/include/asm/cmpxchg_64.h +++ b/arch/sparc/include/asm/cmpxchg_64.h @@ -42,7 +42,7 @@ static inline unsigned long xchg64(__volatile__ unsigned long *m, unsigned long #define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr)))) -extern void __xchg_called_with_bad_pointer(void); +void __xchg_called_with_bad_pointer(void); static inline unsigned long __xchg(unsigned long x, __volatile__ void * ptr, int size) @@ -91,7 +91,7 @@ __cmpxchg_u64(volatile long *m, unsigned long old, unsigned long new) /* This function doesn't exist, so you'll get a linker error if something tries to do an invalid cmpxchg(). */ -extern void __cmpxchg_called_with_bad_pointer(void); +void __cmpxchg_called_with_bad_pointer(void); static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size) @@ -141,5 +141,6 @@ static inline unsigned long __cmpxchg_local(volatile void *ptr, BUILD_BUG_ON(sizeof(*(ptr)) != 8); \ cmpxchg_local((ptr), (o), (n)); \ }) +#define cmpxchg64(ptr, o, n) cmpxchg64_local((ptr), (o), (n)) #endif /* __ARCH_SPARC64_CMPXCHG__ */ diff --git a/arch/sparc/include/asm/compat.h b/arch/sparc/include/asm/compat.h index b8be20d42a0..830502fe62b 100644 --- a/arch/sparc/include/asm/compat.h +++ b/arch/sparc/include/asm/compat.h @@ -36,6 +36,7 @@ typedef s64 compat_s64; typedef u32 compat_uint_t; typedef u32 compat_ulong_t; typedef u64 compat_u64; +typedef u32 compat_uptr_t; struct compat_timespec { compat_time_t tv_sec; @@ -147,6 +148,65 @@ typedef u32 compat_old_sigset_t; typedef u32 compat_sigset_word; +typedef union compat_sigval { + compat_int_t sival_int; + compat_uptr_t sival_ptr; +} compat_sigval_t; + +#define SI_PAD_SIZE32 (128/sizeof(int) - 3) + +typedef struct compat_siginfo { + int si_signo; + int si_errno; + int si_code; + + union { + int _pad[SI_PAD_SIZE32]; + + /* kill() */ + struct { + compat_pid_t _pid; /* sender's pid */ + unsigned int _uid; /* sender's uid */ + } _kill; + + /* POSIX.1b timers */ + struct { + compat_timer_t _tid; /* timer id */ + int _overrun; /* overrun count */ + compat_sigval_t _sigval; /* same as below */ + int _sys_private; /* not to be passed to user */ + } _timer; + + /* POSIX.1b signals */ + struct { + compat_pid_t _pid; /* sender's pid */ + unsigned int _uid; /* sender's uid */ + compat_sigval_t _sigval; + } _rt; + + /* SIGCHLD */ + struct { + compat_pid_t _pid; /* which child */ + unsigned int _uid; /* sender's uid */ + int _status; /* exit code */ + compat_clock_t _utime; + compat_clock_t _stime; + } _sigchld; + + /* SIGILL, SIGFPE, SIGSEGV, SIGBUS, SIGEMT */ + struct { + u32 _addr; /* faulting insn/memory ref. */ + int _trapno; + } _sigfault; + + /* SIGPOLL */ + struct { + int _band; /* POLL_IN, POLL_OUT, POLL_MSG */ + int _fd; + } _sigpoll; + } _sifields; +} compat_siginfo_t; + #define COMPAT_OFF_T_MAX 0x7fffffff #define COMPAT_LOFF_T_MAX 0x7fffffffffffffffL @@ -156,7 +216,6 @@ typedef u32 compat_sigset_word; * as pointers because the syscall entry code will have * appropriately converted them already. */ -typedef u32 compat_uptr_t; static inline void __user *compat_ptr(compat_uptr_t uptr) { @@ -173,9 +232,10 @@ static inline void __user *arch_compat_alloc_user_space(long len) struct pt_regs *regs = current_thread_info()->kregs; unsigned long usp = regs->u_regs[UREG_I6]; - if (!(test_thread_flag(TIF_32BIT))) + if (test_thread_64bit_stack(usp)) usp += STACK_BIAS; - else + + if (test_thread_flag(TIF_32BIT)) usp &= 0xffffffffUL; usp -= len; diff --git a/arch/sparc/include/asm/compat_signal.h b/arch/sparc/include/asm/compat_signal.h index b759eab9b51..9ed1f128b4d 100644 --- a/arch/sparc/include/asm/compat_signal.h +++ b/arch/sparc/include/asm/compat_signal.h @@ -18,12 +18,6 @@ struct __old_sigaction32 { unsigned int sa_flags; unsigned sa_restorer; /* not used by Linux/SPARC yet */ }; - -typedef struct sigaltstack32 { - u32 ss_sp; - int ss_flags; - compat_size_t ss_size; -} stack_t32; #endif #endif /* !(_COMPAT_SIGNAL_H) */ diff --git a/arch/sparc/include/asm/cpudata.h b/arch/sparc/include/asm/cpudata.h index b5976de7cac..128b56b0867 100644 --- a/arch/sparc/include/asm/cpudata.h +++ b/arch/sparc/include/asm/cpudata.h @@ -1,5 +1,15 @@ #ifndef ___ASM_SPARC_CPUDATA_H #define ___ASM_SPARC_CPUDATA_H + +#ifndef __ASSEMBLY__ + +#include <linux/threads.h> +#include <linux/percpu.h> + +extern const struct seq_operations cpuinfo_op; + +#endif /* !(__ASSEMBLY__) */ + #if defined(__sparc__) && defined(__arch64__) #include <asm/cpudata_64.h> #else diff --git a/arch/sparc/include/asm/cpudata_64.h b/arch/sparc/include/asm/cpudata_64.h index 050ef35b9dc..0e594076912 100644 --- a/arch/sparc/include/asm/cpudata_64.h +++ b/arch/sparc/include/asm/cpudata_64.h @@ -8,9 +8,6 @@ #ifndef __ASSEMBLY__ -#include <linux/percpu.h> -#include <linux/threads.h> - typedef struct { /* Dcache line 1 */ unsigned int __softirq_pending; /* must be 1st, see rtrap.S */ @@ -35,8 +32,6 @@ DECLARE_PER_CPU(cpuinfo_sparc, __cpu_data); #define cpu_data(__cpu) per_cpu(__cpu_data, (__cpu)) #define local_cpu_data() __get_cpu_var(__cpu_data) -extern const struct seq_operations cpuinfo_op; - #endif /* !(__ASSEMBLY__) */ #include <asm/trap_block.h> diff --git a/arch/sparc/include/asm/cputime.h b/arch/sparc/include/asm/cputime.h deleted file mode 100644 index 1a642b81e01..00000000000 --- a/arch/sparc/include/asm/cputime.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __SPARC_CPUTIME_H -#define __SPARC_CPUTIME_H - -#include <asm-generic/cputime.h> - -#endif /* __SPARC_CPUTIME_H */ diff --git a/arch/sparc/include/asm/delay_32.h b/arch/sparc/include/asm/delay_32.h index bc9aba2bead..3fb8ca144b4 100644 --- a/arch/sparc/include/asm/delay_32.h +++ b/arch/sparc/include/asm/delay_32.h @@ -20,8 +20,8 @@ static inline void __delay(unsigned long loops) } /* This is too messy with inline asm on the Sparc. */ -extern void __udelay(unsigned long usecs, unsigned long lpj); -extern void __ndelay(unsigned long nsecs, unsigned long lpj); +void __udelay(unsigned long usecs, unsigned long lpj); +void __ndelay(unsigned long nsecs, unsigned long lpj); #ifdef CONFIG_SMP #define __udelay_val cpu_data(smp_processor_id()).udelay_val diff --git a/arch/sparc/include/asm/delay_64.h b/arch/sparc/include/asm/delay_64.h index a77aa622d76..0ba5424856d 100644 --- a/arch/sparc/include/asm/delay_64.h +++ b/arch/sparc/include/asm/delay_64.h @@ -8,8 +8,8 @@ #ifndef __ASSEMBLY__ -extern void __delay(unsigned long loops); -extern void udelay(unsigned long usecs); +void __delay(unsigned long loops); +void udelay(unsigned long usecs); #define mdelay(n) udelay((n) * 1000) #endif /* !__ASSEMBLY__ */ diff --git a/arch/sparc/include/asm/device.h b/arch/sparc/include/asm/device.h index daa6a8a5e9c..bb3f0b0c675 100644 --- a/arch/sparc/include/asm/device.h +++ b/arch/sparc/include/asm/device.h @@ -19,7 +19,7 @@ struct dev_archdata { int numa_node; }; -extern void of_propagate_archdata(struct platform_device *bus); +void of_propagate_archdata(struct platform_device *bus); struct pdev_archdata { struct resource resource[PROMREG_MAX]; diff --git a/arch/sparc/include/asm/dma-mapping.h b/arch/sparc/include/asm/dma-mapping.h index 8493fd3c7ba..1ee02710b2d 100644 --- a/arch/sparc/include/asm/dma-mapping.h +++ b/arch/sparc/include/asm/dma-mapping.h @@ -7,7 +7,7 @@ #define DMA_ERROR_CODE (~(dma_addr_t)0x0) -extern int dma_supported(struct device *dev, u64 mask); +int dma_supported(struct device *dev, u64 mask); #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) @@ -59,6 +59,7 @@ static inline void dma_free_attrs(struct device *dev, size_t size, static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) { + debug_dma_mapping_error(dev, dma_addr); return (dma_addr == DMA_ERROR_CODE); } diff --git a/arch/sparc/include/asm/ebus_dma.h b/arch/sparc/include/asm/ebus_dma.h index f07a5b541c9..fcfb4948147 100644 --- a/arch/sparc/include/asm/ebus_dma.h +++ b/arch/sparc/include/asm/ebus_dma.h @@ -22,14 +22,14 @@ struct ebus_dma_info { unsigned char name[64]; }; -extern int ebus_dma_register(struct ebus_dma_info *p); -extern int ebus_dma_irq_enable(struct ebus_dma_info *p, int on); -extern void ebus_dma_unregister(struct ebus_dma_info *p); -extern int ebus_dma_request(struct ebus_dma_info *p, dma_addr_t bus_addr, +int ebus_dma_register(struct ebus_dma_info *p); +int ebus_dma_irq_enable(struct ebus_dma_info *p, int on); +void ebus_dma_unregister(struct ebus_dma_info *p); +int ebus_dma_request(struct ebus_dma_info *p, dma_addr_t bus_addr, size_t len); -extern void ebus_dma_prepare(struct ebus_dma_info *p, int write); -extern unsigned int ebus_dma_residue(struct ebus_dma_info *p); -extern unsigned int ebus_dma_addr(struct ebus_dma_info *p); -extern void ebus_dma_enable(struct ebus_dma_info *p, int on); +void ebus_dma_prepare(struct ebus_dma_info *p, int write); +unsigned int ebus_dma_residue(struct ebus_dma_info *p); +unsigned int ebus_dma_addr(struct ebus_dma_info *p); +void ebus_dma_enable(struct ebus_dma_info *p, int on); #endif /* __ASM_SPARC_EBUS_DMA_H */ diff --git a/arch/sparc/include/asm/elf_32.h b/arch/sparc/include/asm/elf_32.h index 2d4d755cba9..a24e41fcdde 100644 --- a/arch/sparc/include/asm/elf_32.h +++ b/arch/sparc/include/asm/elf_32.h @@ -128,6 +128,4 @@ typedef struct { #define ELF_PLATFORM (NULL) -#define SET_PERSONALITY(ex) set_personality(PER_LINUX) - #endif /* !(__ASMSPARC_ELF_H) */ diff --git a/arch/sparc/include/asm/elf_64.h b/arch/sparc/include/asm/elf_64.h index 7df8b7f544d..370ca1e71ff 100644 --- a/arch/sparc/include/asm/elf_64.h +++ b/arch/sparc/include/asm/elf_64.h @@ -86,6 +86,15 @@ #define AV_SPARC_IMA 0x00400000 /* integer multiply-add */ #define AV_SPARC_ASI_CACHE_SPARING \ 0x00800000 /* cache sparing ASIs available */ +#define AV_SPARC_PAUSE 0x01000000 /* PAUSE available */ +#define AV_SPARC_CBCOND 0x02000000 /* CBCOND insns available */ + +/* Solaris decided to enumerate every single crypto instruction type + * in the AT_HWCAP bits. This is wasteful, since if crypto is present, + * you still need to look in the CFR register to see if the opcode is + * really available. So we simply advertise only "crypto" support. + */ +#define HWCAP_SPARC_CRYPTO 0x04000000 /* CRYPTO insns available */ #define CORE_DUMP_USE_REGSET diff --git a/arch/sparc/include/asm/emergency-restart.h b/arch/sparc/include/asm/emergency-restart.h deleted file mode 100644 index 108d8c48e42..00000000000 --- a/arch/sparc/include/asm/emergency-restart.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef _ASM_EMERGENCY_RESTART_H -#define _ASM_EMERGENCY_RESTART_H - -#include <asm-generic/emergency-restart.h> - -#endif /* _ASM_EMERGENCY_RESTART_H */ diff --git a/arch/sparc/include/asm/exec.h b/arch/sparc/include/asm/exec.h deleted file mode 100644 index 2e085881e0d..00000000000 --- a/arch/sparc/include/asm/exec.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __SPARC_EXEC_H -#define __SPARC_EXEC_H - -#define arch_align_stack(x) (x) - -#endif /* __SPARC_EXEC_H */ diff --git a/arch/sparc/include/asm/fbio.h b/arch/sparc/include/asm/fbio.h index 0a21da87f7d..1d9afe277e9 100644 --- a/arch/sparc/include/asm/fbio.h +++ b/arch/sparc/include/asm/fbio.h @@ -1,225 +1,10 @@ #ifndef __LINUX_FBIO_H #define __LINUX_FBIO_H -#include <linux/compiler.h> -#include <linux/types.h> +#include <uapi/asm/fbio.h> -/* Constants used for fbio SunOS compatibility */ -/* (C) 1996 Miguel de Icaza */ - -/* Frame buffer types */ -#define FBTYPE_NOTYPE -1 -#define FBTYPE_SUN1BW 0 /* mono */ -#define FBTYPE_SUN1COLOR 1 -#define FBTYPE_SUN2BW 2 -#define FBTYPE_SUN2COLOR 3 -#define FBTYPE_SUN2GP 4 -#define FBTYPE_SUN5COLOR 5 -#define FBTYPE_SUN3COLOR 6 -#define FBTYPE_MEMCOLOR 7 -#define FBTYPE_SUN4COLOR 8 - -#define FBTYPE_NOTSUN1 9 -#define FBTYPE_NOTSUN2 10 -#define FBTYPE_NOTSUN3 11 - -#define FBTYPE_SUNFAST_COLOR 12 /* cg6 */ -#define FBTYPE_SUNROP_COLOR 13 -#define FBTYPE_SUNFB_VIDEO 14 -#define FBTYPE_SUNGIFB 15 -#define FBTYPE_SUNGPLAS 16 -#define FBTYPE_SUNGP3 17 -#define FBTYPE_SUNGT 18 -#define FBTYPE_SUNLEO 19 /* zx Leo card */ -#define FBTYPE_MDICOLOR 20 /* cg14 */ -#define FBTYPE_TCXCOLOR 21 /* SUNW,tcx card */ - -#define FBTYPE_LASTPLUSONE 21 /* This is not last + 1 in fact... */ - -/* Does not seem to be listed in the Sun file either */ -#define FBTYPE_CREATOR 22 -#define FBTYPE_PCI_IGA1682 23 -#define FBTYPE_P9100COLOR 24 - -#define FBTYPE_PCI_GENERIC 1000 -#define FBTYPE_PCI_MACH64 1001 - -/* fbio ioctls */ -/* Returned by FBIOGTYPE */ -struct fbtype { - int fb_type; /* fb type, see above */ - int fb_height; /* pixels */ - int fb_width; /* pixels */ - int fb_depth; - int fb_cmsize; /* color map entries */ - int fb_size; /* fb size in bytes */ -}; -#define FBIOGTYPE _IOR('F', 0, struct fbtype) - -struct fbcmap { - int index; /* first element (0 origin) */ - int count; - unsigned char __user *red; - unsigned char __user *green; - unsigned char __user *blue; -}; - -#ifdef __KERNEL__ #define FBIOPUTCMAP_SPARC _IOW('F', 3, struct fbcmap) #define FBIOGETCMAP_SPARC _IOW('F', 4, struct fbcmap) -#else -#define FBIOPUTCMAP _IOW('F', 3, struct fbcmap) -#define FBIOGETCMAP _IOW('F', 4, struct fbcmap) -#endif - -/* # of device specific values */ -#define FB_ATTR_NDEVSPECIFIC 8 -/* # of possible emulations */ -#define FB_ATTR_NEMUTYPES 4 - -struct fbsattr { - int flags; - int emu_type; /* -1 if none */ - int dev_specific[FB_ATTR_NDEVSPECIFIC]; -}; - -struct fbgattr { - int real_type; /* real frame buffer type */ - int owner; /* unknown */ - struct fbtype fbtype; /* real frame buffer fbtype */ - struct fbsattr sattr; - int emu_types[FB_ATTR_NEMUTYPES]; /* supported emulations */ -}; -#define FBIOSATTR _IOW('F', 5, struct fbgattr) /* Unsupported: */ -#define FBIOGATTR _IOR('F', 6, struct fbgattr) /* supported */ - -#define FBIOSVIDEO _IOW('F', 7, int) -#define FBIOGVIDEO _IOR('F', 8, int) - -struct fbcursor { - short set; /* what to set, choose from the list above */ - short enable; /* cursor on/off */ - struct fbcurpos pos; /* cursor position */ - struct fbcurpos hot; /* cursor hot spot */ - struct fbcmap cmap; /* color map info */ - struct fbcurpos size; /* cursor bit map size */ - char __user *image; /* cursor image bits */ - char __user *mask; /* cursor mask bits */ -}; - -/* set/get cursor attributes/shape */ -#define FBIOSCURSOR _IOW('F', 24, struct fbcursor) -#define FBIOGCURSOR _IOWR('F', 25, struct fbcursor) - -/* set/get cursor position */ -#define FBIOSCURPOS _IOW('F', 26, struct fbcurpos) -#define FBIOGCURPOS _IOW('F', 27, struct fbcurpos) - -/* get max cursor size */ -#define FBIOGCURMAX _IOR('F', 28, struct fbcurpos) - -/* wid manipulation */ -struct fb_wid_alloc { -#define FB_WID_SHARED_8 0 -#define FB_WID_SHARED_24 1 -#define FB_WID_DBL_8 2 -#define FB_WID_DBL_24 3 - __u32 wa_type; - __s32 wa_index; /* Set on return */ - __u32 wa_count; -}; -struct fb_wid_item { - __u32 wi_type; - __s32 wi_index; - __u32 wi_attrs; - __u32 wi_values[32]; -}; -struct fb_wid_list { - __u32 wl_flags; - __u32 wl_count; - struct fb_wid_item *wl_list; -}; - -#define FBIO_WID_ALLOC _IOWR('F', 30, struct fb_wid_alloc) -#define FBIO_WID_FREE _IOW('F', 31, struct fb_wid_alloc) -#define FBIO_WID_PUT _IOW('F', 32, struct fb_wid_list) -#define FBIO_WID_GET _IOWR('F', 33, struct fb_wid_list) - -/* Creator ioctls */ -#define FFB_IOCTL ('F'<<8) -#define FFB_SYS_INFO (FFB_IOCTL|80) -#define FFB_CLUTREAD (FFB_IOCTL|81) -#define FFB_CLUTPOST (FFB_IOCTL|82) -#define FFB_SETDIAGMODE (FFB_IOCTL|83) -#define FFB_GETMONITORID (FFB_IOCTL|84) -#define FFB_GETVIDEOMODE (FFB_IOCTL|85) -#define FFB_SETVIDEOMODE (FFB_IOCTL|86) -#define FFB_SETSERVER (FFB_IOCTL|87) -#define FFB_SETOVCTL (FFB_IOCTL|88) -#define FFB_GETOVCTL (FFB_IOCTL|89) -#define FFB_GETSAXNUM (FFB_IOCTL|90) -#define FFB_FBDEBUG (FFB_IOCTL|91) - -/* Cg14 ioctls */ -#define MDI_IOCTL ('M'<<8) -#define MDI_RESET (MDI_IOCTL|1) -#define MDI_GET_CFGINFO (MDI_IOCTL|2) -#define MDI_SET_PIXELMODE (MDI_IOCTL|3) -# define MDI_32_PIX 32 -# define MDI_16_PIX 16 -# define MDI_8_PIX 8 - -struct mdi_cfginfo { - int mdi_ncluts; /* Number of implemented CLUTs in this MDI */ - int mdi_type; /* FBTYPE name */ - int mdi_height; /* height */ - int mdi_width; /* width */ - int mdi_size; /* available ram */ - int mdi_mode; /* 8bpp, 16bpp or 32bpp */ - int mdi_pixfreq; /* pixel clock (from PROM) */ -}; - -/* SparcLinux specific ioctl for the MDI, should be replaced for - * the SET_XLUT/SET_CLUTn ioctls instead - */ -#define MDI_CLEAR_XLUT (MDI_IOCTL|9) - -/* leo & ffb ioctls */ -struct fb_clut_alloc { - __u32 clutid; /* Set on return */ - __u32 flag; - __u32 index; -}; - -struct fb_clut { -#define FB_CLUT_WAIT 0x00000001 /* Not yet implemented */ - __u32 flag; - __u32 clutid; - __u32 offset; - __u32 count; - char * red; - char * green; - char * blue; -}; - -struct fb_clut32 { - __u32 flag; - __u32 clutid; - __u32 offset; - __u32 count; - __u32 red; - __u32 green; - __u32 blue; -}; - -#define LEO_CLUTALLOC _IOWR('L', 53, struct fb_clut_alloc) -#define LEO_CLUTFREE _IOW('L', 54, struct fb_clut_alloc) -#define LEO_CLUTREAD _IOW('L', 55, struct fb_clut) -#define LEO_CLUTPOST _IOW('L', 56, struct fb_clut) -#define LEO_SETGAMMA _IOW('L', 68, int) /* Not yet implemented */ -#define LEO_GETGAMMA _IOR('L', 69, int) /* Not yet implemented */ - -#ifdef __KERNEL__ /* Addresses on the fd of a cgsix that are mappable */ #define CG6_FBC 0x70000000 #define CG6_TEC 0x70001000 @@ -260,47 +45,6 @@ struct fb_clut32 { #define CG14_CLUT3 0x6000 /* Color Look Up Table */ #define CG14_AUTO 0xf000 -#endif /* KERNEL */ - -/* These are exported to userland for applications to use */ -/* Mappable offsets for the cg14: control registers */ -#define MDI_DIRECT_MAP 0x10000000 -#define MDI_CTLREG_MAP 0x20000000 -#define MDI_CURSOR_MAP 0x30000000 -#define MDI_SHDW_VRT_MAP 0x40000000 - -/* Mappable offsets for the cg14: frame buffer resolutions */ -/* 32 bits */ -#define MDI_CHUNKY_XBGR_MAP 0x50000000 -#define MDI_CHUNKY_BGR_MAP 0x60000000 - -/* 16 bits */ -#define MDI_PLANAR_X16_MAP 0x70000000 -#define MDI_PLANAR_C16_MAP 0x80000000 - -/* 8 bit is done as CG3 MMAP offset */ -/* 32 bits, planar */ -#define MDI_PLANAR_X32_MAP 0x90000000 -#define MDI_PLANAR_B32_MAP 0xa0000000 -#define MDI_PLANAR_G32_MAP 0xb0000000 -#define MDI_PLANAR_R32_MAP 0xc0000000 - -/* Mappable offsets on leo */ -#define LEO_SS0_MAP 0x00000000 -#define LEO_LC_SS0_USR_MAP 0x00800000 -#define LEO_LD_SS0_MAP 0x00801000 -#define LEO_LX_CURSOR_MAP 0x00802000 -#define LEO_SS1_MAP 0x00803000 -#define LEO_LC_SS1_USR_MAP 0x01003000 -#define LEO_LD_SS1_MAP 0x01004000 -#define LEO_UNK_MAP 0x01005000 -#define LEO_LX_KRN_MAP 0x01006000 -#define LEO_LC_SS0_KRN_MAP 0x01007000 -#define LEO_LC_SS1_KRN_MAP 0x01008000 -#define LEO_LD_GBL_MAP 0x01009000 -#define LEO_UNK2_MAP 0x0100a000 - -#ifdef __KERNEL__ struct fbcmap32 { int index; /* first element (0 origin) */ int count; @@ -325,6 +69,4 @@ struct fbcursor32 { #define FBIOSCURSOR32 _IOW('F', 24, struct fbcursor32) #define FBIOGCURSOR32 _IOW('F', 25, struct fbcursor32) -#endif - #endif /* __LINUX_FBIO_H */ diff --git a/arch/sparc/include/asm/fixmap.h b/arch/sparc/include/asm/fixmap.h deleted file mode 100644 index f18fc0755ad..00000000000 --- a/arch/sparc/include/asm/fixmap.h +++ /dev/null @@ -1,110 +0,0 @@ -/* - * fixmap.h: compile-time virtual memory allocation - * - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (C) 1998 Ingo Molnar - * - * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999 - */ - -#ifndef _ASM_FIXMAP_H -#define _ASM_FIXMAP_H - -#include <linux/kernel.h> -#include <asm/page.h> -#ifdef CONFIG_HIGHMEM -#include <linux/threads.h> -#include <asm/kmap_types.h> -#endif - -/* - * Here we define all the compile-time 'special' virtual - * addresses. The point is to have a constant address at - * compile time, but to set the physical address only - * in the boot process. We allocate these special addresses - * from the top of unused virtual memory (0xfd000000 - 1 page) backwards. - * Also this lets us do fail-safe vmalloc(), we - * can guarantee that these special addresses and - * vmalloc()-ed addresses never overlap. - * - * these 'compile-time allocated' memory buffers are - * fixed-size 4k pages. (or larger if used with an increment - * highger than 1) use fixmap_set(idx,phys) to associate - * physical memory with fixmap indices. - * - * TLB entries of such buffers will not be flushed across - * task switches. - */ - -/* - * on UP currently we will have no trace of the fixmap mechanism, - * no page table allocations, etc. This might change in the - * future, say framebuffers for the console driver(s) could be - * fix-mapped? - */ -enum fixed_addresses { - FIX_HOLE, -#ifdef CONFIG_HIGHMEM - FIX_KMAP_BEGIN, - FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1, -#endif - __end_of_fixed_addresses -}; - -extern void __set_fixmap (enum fixed_addresses idx, - unsigned long phys, pgprot_t flags); - -#define set_fixmap(idx, phys) \ - __set_fixmap(idx, phys, PAGE_KERNEL) -/* - * Some hardware wants to get fixmapped without caching. - */ -#define set_fixmap_nocache(idx, phys) \ - __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE) -/* - * used by vmalloc.c. - * - * Leave one empty page between IO pages at 0xfd000000 and - * the start of the fixmap. - */ -#define FIXADDR_TOP (0xfcfff000UL) -#define FIXADDR_SIZE ((__end_of_fixed_addresses) << PAGE_SHIFT) -#define FIXADDR_START (FIXADDR_TOP - FIXADDR_SIZE) - -#define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT)) -#define __virt_to_fix(x) ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT) - -extern void __this_fixmap_does_not_exist(void); - -/* - * 'index to address' translation. If anyone tries to use the idx - * directly without tranlation, we catch the bug with a NULL-deference - * kernel oops. Illegal ranges of incoming indices are caught too. - */ -static inline unsigned long fix_to_virt(const unsigned int idx) -{ - /* - * this branch gets completely eliminated after inlining, - * except when someone tries to use fixaddr indices in an - * illegal way. (such as mixing up address types or using - * out-of-range indices). - * - * If it doesn't get removed, the linker will complain - * loudly with a reasonably clear error message.. - */ - if (idx >= __end_of_fixed_addresses) - __this_fixmap_does_not_exist(); - - return __fix_to_virt(idx); -} - -static inline unsigned long virt_to_fix(const unsigned long vaddr) -{ - BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START); - return __virt_to_fix(vaddr); -} - -#endif diff --git a/arch/sparc/include/asm/floppy_32.h b/arch/sparc/include/asm/floppy_32.h index fb3f16954c6..071b83e52f1 100644 --- a/arch/sparc/include/asm/floppy_32.h +++ b/arch/sparc/include/asm/floppy_32.h @@ -9,11 +9,12 @@ #include <linux/of.h> #include <linux/of_device.h> -#include <asm/page.h> #include <asm/pgtable.h> #include <asm/idprom.h> #include <asm/oplib.h> #include <asm/auxio.h> +#include <asm/setup.h> +#include <asm/page.h> #include <asm/irq.h> /* We don't need no stinkin' I/O port allocation crap. */ @@ -49,7 +50,6 @@ struct sun_flpy_controller { /* You'll only ever find one controller on a SparcStation anyways. */ static struct sun_flpy_controller *sun_fdc = NULL; -extern volatile unsigned char *fdc_status; struct sun_floppy_ops { unsigned char (*fd_inb)(int port); @@ -212,13 +212,6 @@ static void sun_82077_fd_outb(unsigned char value, int port) * underruns. If non-zero, doing_pdma encodes the direction of * the transfer for debugging. 1=read 2=write */ -extern char *pdma_vaddr; -extern unsigned long pdma_size; -extern volatile int doing_pdma; - -/* This is software state */ -extern char *pdma_base; -extern unsigned long pdma_areasize; /* Common routines to all controller types on the Sparc. */ static inline void virtual_dma_init(void) @@ -263,8 +256,7 @@ static inline void sun_fd_enable_dma(void) pdma_areasize = pdma_size; } -extern int sparc_floppy_request_irq(unsigned int irq, - irq_handler_t irq_handler); +int sparc_floppy_request_irq(unsigned int irq, irq_handler_t irq_handler); static int sun_fd_request_irq(void) { diff --git a/arch/sparc/include/asm/floppy_64.h b/arch/sparc/include/asm/floppy_64.h index e204f902e6c..625756406a7 100644 --- a/arch/sparc/include/asm/floppy_64.h +++ b/arch/sparc/include/asm/floppy_64.h @@ -254,7 +254,7 @@ static int sun_fd_request_irq(void) once = 1; error = request_irq(FLOPPY_IRQ, sparc_floppy_irq, - IRQF_DISABLED, "floppy", NULL); + 0, "floppy", NULL); return ((error == 0) ? 0 : -1); } @@ -296,7 +296,7 @@ struct sun_pci_dma_op { static struct sun_pci_dma_op sun_pci_dma_current = { -1U, 0, 0, NULL}; static struct sun_pci_dma_op sun_pci_dma_pending = { -1U, 0, 0, NULL}; -extern irqreturn_t floppy_interrupt(int irq, void *dev_id); +irqreturn_t floppy_interrupt(int irq, void *dev_id); static unsigned char sun_pci_fd_inb(unsigned long port) { diff --git a/arch/sparc/include/asm/ftrace.h b/arch/sparc/include/asm/ftrace.h index b0f18e9893d..9ec94ad116f 100644 --- a/arch/sparc/include/asm/ftrace.h +++ b/arch/sparc/include/asm/ftrace.h @@ -6,7 +6,7 @@ #define MCOUNT_INSN_SIZE 4 /* sizeof mcount call */ #ifndef __ASSEMBLY__ -extern void _mcount(void); +void _mcount(void); #endif #endif @@ -22,4 +22,8 @@ struct dyn_arch_ftrace { }; #endif /* CONFIG_DYNAMIC_FTRACE */ +unsigned long prepare_ftrace_return(unsigned long parent, + unsigned long self_addr, + unsigned long frame_pointer); + #endif /* _ASM_SPARC64_FTRACE */ diff --git a/arch/sparc/include/asm/hardirq_32.h b/arch/sparc/include/asm/hardirq_32.h index 162007643cd..ee93923b7f8 100644 --- a/arch/sparc/include/asm/hardirq_32.h +++ b/arch/sparc/include/asm/hardirq_32.h @@ -7,7 +7,6 @@ #ifndef __SPARC_HARDIRQ_H #define __SPARC_HARDIRQ_H -#define HARDIRQ_BITS 8 #include <asm-generic/hardirq.h> #endif /* __SPARC_HARDIRQ_H */ diff --git a/arch/sparc/include/asm/hardirq_64.h b/arch/sparc/include/asm/hardirq_64.h index 7c29fd1a87a..f478ff1ddd0 100644 --- a/arch/sparc/include/asm/hardirq_64.h +++ b/arch/sparc/include/asm/hardirq_64.h @@ -14,6 +14,4 @@ void ack_bad_irq(unsigned int irq); -#define HARDIRQ_BITS 8 - #endif /* !(__SPARC64_HARDIRQ_H) */ diff --git a/arch/sparc/include/asm/head_32.h b/arch/sparc/include/asm/head_32.h index a76874838f6..5f1dbe315bc 100644 --- a/arch/sparc/include/asm/head_32.h +++ b/arch/sparc/include/asm/head_32.h @@ -55,15 +55,15 @@ /* The Get Condition Codes software trap for userland. */ #define GETCC_TRAP \ - b getcc_trap_handler; mov %psr, %l0; nop; nop; + b getcc_trap_handler; rd %psr, %l0; nop; nop; /* The Set Condition Codes software trap for userland. */ #define SETCC_TRAP \ - b setcc_trap_handler; mov %psr, %l0; nop; nop; + b setcc_trap_handler; rd %psr, %l0; nop; nop; /* The Get PSR software trap for userland. */ #define GETPSR_TRAP \ - mov %psr, %i0; jmp %l2; rett %l2 + 4; nop; + rd %psr, %i0; jmp %l2; rett %l2 + 4; nop; /* This is for hard interrupts from level 1-14, 15 is non-maskable (nmi) and * gets handled with another macro. diff --git a/arch/sparc/include/asm/hibernate.h b/arch/sparc/include/asm/hibernate.h new file mode 100644 index 00000000000..2ec34f84224 --- /dev/null +++ b/arch/sparc/include/asm/hibernate.h @@ -0,0 +1,23 @@ +/* + * hibernate.h: Hibernaton support specific for sparc64. + * + * Copyright (C) 2013 Kirill V Tkhai (tkhai@yandex.ru) + */ + +#ifndef ___SPARC_HIBERNATE_H +#define ___SPARC_HIBERNATE_H + +struct saved_context { + unsigned long fp; + unsigned long cwp; + unsigned long wstate; + + unsigned long tick; + unsigned long pstate; + + unsigned long g4; + unsigned long g5; + unsigned long g6; +}; + +#endif diff --git a/arch/sparc/include/asm/highmem.h b/arch/sparc/include/asm/highmem.h index 3b6e00dd96e..92ded294a4e 100644 --- a/arch/sparc/include/asm/highmem.h +++ b/arch/sparc/include/asm/highmem.h @@ -21,7 +21,6 @@ #ifdef __KERNEL__ #include <linux/interrupt.h> -#include <asm/fixmap.h> #include <asm/vaddrs.h> #include <asm/kmap_types.h> #include <asm/pgtable.h> @@ -29,11 +28,10 @@ /* declarations for highmem.c */ extern unsigned long highstart_pfn, highend_pfn; -extern pte_t *kmap_pte; extern pgprot_t kmap_prot; extern pte_t *pkmap_page_table; -extern void kmap_init(void) __init; +void kmap_init(void) __init; /* * Right now we initialize only a single pte table. It can be extended @@ -51,8 +49,8 @@ extern void kmap_init(void) __init; #define PKMAP_END (PKMAP_ADDR(LAST_PKMAP)) -extern void *kmap_high(struct page *page); -extern void kunmap_high(struct page *page); +void *kmap_high(struct page *page); +void kunmap_high(struct page *page); static inline void *kmap(struct page *page) { @@ -70,9 +68,8 @@ static inline void kunmap(struct page *page) kunmap_high(page); } -extern void *kmap_atomic(struct page *page); -extern void __kunmap_atomic(void *kvaddr); -extern struct page *kmap_atomic_to_page(void *vaddr); +void *kmap_atomic(struct page *page); +void __kunmap_atomic(void *kvaddr); #define flush_cache_kmaps() flush_cache_all() diff --git a/arch/sparc/include/asm/hugetlb.h b/arch/sparc/include/asm/hugetlb.h index 177061064ee..e4cab465b81 100644 --- a/arch/sparc/include/asm/hugetlb.h +++ b/arch/sparc/include/asm/hugetlb.h @@ -2,6 +2,7 @@ #define _ASM_SPARC64_HUGETLB_H #include <asm/page.h> +#include <asm-generic/hugetlb.h> void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, @@ -10,7 +11,9 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep); -void hugetlb_prefault_arch_hook(struct mm_struct *mm); +static inline void hugetlb_prefault_arch_hook(struct mm_struct *mm) +{ +} static inline int is_hugepage_only_range(struct mm_struct *mm, unsigned long addr, @@ -58,14 +61,20 @@ static inline pte_t huge_pte_wrprotect(pte_t pte) static inline void huge_ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - ptep_set_wrprotect(mm, addr, ptep); + pte_t old_pte = *ptep; + set_huge_pte_at(mm, addr, ptep, pte_wrprotect(old_pte)); } static inline int huge_ptep_set_access_flags(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep, pte_t pte, int dirty) { - return ptep_set_access_flags(vma, addr, ptep, pte, dirty); + int changed = !pte_same(*ptep, pte); + if (changed) { + set_huge_pte_at(vma->vm_mm, addr, ptep, pte); + flush_tlb_page(vma, addr); + } + return changed; } static inline pte_t huge_ptep_get(pte_t *ptep) @@ -82,4 +91,8 @@ static inline void arch_release_hugepage(struct page *page) { } +static inline void arch_clear_hugepage_flags(struct page *page) +{ +} + #endif /* _ASM_SPARC64_HUGETLB_H */ diff --git a/arch/sparc/include/asm/hvtramp.h b/arch/sparc/include/asm/hvtramp.h index b2b9b947b3a..04b56f862bb 100644 --- a/arch/sparc/include/asm/hvtramp.h +++ b/arch/sparc/include/asm/hvtramp.h @@ -19,7 +19,7 @@ struct hvtramp_descr { struct hvtramp_mapping maps[1]; }; -extern void hv_cpu_startup(unsigned long hvdescr_pa); +void hv_cpu_startup(unsigned long hvdescr_pa); #endif diff --git a/arch/sparc/include/asm/hypervisor.h b/arch/sparc/include/asm/hypervisor.h index 015a761eaa3..94b39caea3e 100644 --- a/arch/sparc/include/asm/hypervisor.h +++ b/arch/sparc/include/asm/hypervisor.h @@ -98,7 +98,7 @@ #define HV_FAST_MACH_EXIT 0x00 #ifndef __ASSEMBLY__ -extern void sun4v_mach_exit(unsigned long exit_code); +void sun4v_mach_exit(unsigned long exit_code); #endif /* Domain services. */ @@ -127,9 +127,9 @@ extern void sun4v_mach_exit(unsigned long exit_code); #define HV_FAST_MACH_DESC 0x01 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_mach_desc(unsigned long buffer_pa, - unsigned long buf_len, - unsigned long *real_buf_len); +unsigned long sun4v_mach_desc(unsigned long buffer_pa, + unsigned long buf_len, + unsigned long *real_buf_len); #endif /* mach_sir() @@ -148,7 +148,7 @@ extern unsigned long sun4v_mach_desc(unsigned long buffer_pa, #define HV_FAST_MACH_SIR 0x02 #ifndef __ASSEMBLY__ -extern void sun4v_mach_sir(void); +void sun4v_mach_sir(void); #endif /* mach_set_watchdog() @@ -204,8 +204,8 @@ extern void sun4v_mach_sir(void); #define HV_FAST_MACH_SET_WATCHDOG 0x05 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_mach_set_watchdog(unsigned long timeout, - unsigned long *orig_timeout); +unsigned long sun4v_mach_set_watchdog(unsigned long timeout, + unsigned long *orig_timeout); #endif /* CPU services. @@ -250,10 +250,10 @@ extern unsigned long sun4v_mach_set_watchdog(unsigned long timeout, #define HV_FAST_CPU_START 0x10 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_cpu_start(unsigned long cpuid, - unsigned long pc, - unsigned long rtba, - unsigned long arg0); +unsigned long sun4v_cpu_start(unsigned long cpuid, + unsigned long pc, + unsigned long rtba, + unsigned long arg0); #endif /* cpu_stop() @@ -278,7 +278,7 @@ extern unsigned long sun4v_cpu_start(unsigned long cpuid, #define HV_FAST_CPU_STOP 0x11 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_cpu_stop(unsigned long cpuid); +unsigned long sun4v_cpu_stop(unsigned long cpuid); #endif /* cpu_yield() @@ -295,7 +295,7 @@ extern unsigned long sun4v_cpu_stop(unsigned long cpuid); #define HV_FAST_CPU_YIELD 0x12 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_cpu_yield(void); +unsigned long sun4v_cpu_yield(void); #endif /* cpu_qconf() @@ -341,9 +341,9 @@ extern unsigned long sun4v_cpu_yield(void); #define HV_CPU_QUEUE_NONRES_ERROR 0x3f #ifndef __ASSEMBLY__ -extern unsigned long sun4v_cpu_qconf(unsigned long type, - unsigned long queue_paddr, - unsigned long num_queue_entries); +unsigned long sun4v_cpu_qconf(unsigned long type, + unsigned long queue_paddr, + unsigned long num_queue_entries); #endif /* cpu_qinfo() @@ -394,7 +394,9 @@ extern unsigned long sun4v_cpu_qconf(unsigned long type, #define HV_FAST_CPU_MONDO_SEND 0x42 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_cpu_mondo_send(unsigned long cpu_count, unsigned long cpu_list_pa, unsigned long mondo_block_pa); +unsigned long sun4v_cpu_mondo_send(unsigned long cpu_count, + unsigned long cpu_list_pa, + unsigned long mondo_block_pa); #endif /* cpu_myid() @@ -425,7 +427,7 @@ extern unsigned long sun4v_cpu_mondo_send(unsigned long cpu_count, unsigned long #define HV_CPU_STATE_ERROR 0x03 #ifndef __ASSEMBLY__ -extern long sun4v_cpu_state(unsigned long cpuid); +long sun4v_cpu_state(unsigned long cpuid); #endif /* cpu_set_rtba() @@ -625,8 +627,8 @@ struct hv_fault_status { #define HV_FAST_MMU_TSB_CTX0 0x20 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions, - unsigned long tsb_desc_ra); +unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions, + unsigned long tsb_desc_ra); #endif /* mmu_tsb_ctxnon0() @@ -710,7 +712,7 @@ extern unsigned long sun4v_mmu_tsb_ctx0(unsigned long num_descriptions, #define HV_FAST_MMU_DEMAP_ALL 0x24 #ifndef __ASSEMBLY__ -extern void sun4v_mmu_demap_all(void); +void sun4v_mmu_demap_all(void); #endif /* mmu_map_perm_addr() @@ -740,10 +742,10 @@ extern void sun4v_mmu_demap_all(void); #define HV_FAST_MMU_MAP_PERM_ADDR 0x25 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr, - unsigned long set_to_zero, - unsigned long tte, - unsigned long flags); +unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr, + unsigned long set_to_zero, + unsigned long tte, + unsigned long flags); #endif /* mmu_fault_area_conf() @@ -945,7 +947,7 @@ extern unsigned long sun4v_mmu_map_perm_addr(unsigned long vaddr, #define HV_FAST_TOD_GET 0x50 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_tod_get(unsigned long *time); +unsigned long sun4v_tod_get(unsigned long *time); #endif /* tod_set() @@ -962,7 +964,7 @@ extern unsigned long sun4v_tod_get(unsigned long *time); #define HV_FAST_TOD_SET 0x51 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_tod_set(unsigned long time); +unsigned long sun4v_tod_set(unsigned long time); #endif /* Console services */ @@ -1038,14 +1040,14 @@ extern unsigned long sun4v_tod_set(unsigned long time); #define HV_FAST_CONS_WRITE 0x63 #ifndef __ASSEMBLY__ -extern long sun4v_con_getchar(long *status); -extern long sun4v_con_putchar(long c); -extern long sun4v_con_read(unsigned long buffer, - unsigned long size, - unsigned long *bytes_read); -extern unsigned long sun4v_con_write(unsigned long buffer, - unsigned long size, - unsigned long *bytes_written); +long sun4v_con_getchar(long *status); +long sun4v_con_putchar(long c); +long sun4v_con_read(unsigned long buffer, + unsigned long size, + unsigned long *bytes_read); +unsigned long sun4v_con_write(unsigned long buffer, + unsigned long size, + unsigned long *bytes_written); #endif /* mach_set_soft_state() @@ -1080,8 +1082,8 @@ extern unsigned long sun4v_con_write(unsigned long buffer, #define HV_SOFT_STATE_TRANSITION 0x02 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_mach_set_soft_state(unsigned long soft_state, - unsigned long msg_string_ra); +unsigned long sun4v_mach_set_soft_state(unsigned long soft_state, + unsigned long msg_string_ra); #endif /* mach_get_soft_state() @@ -1159,20 +1161,20 @@ extern unsigned long sun4v_mach_set_soft_state(unsigned long soft_state, #define HV_FAST_SVC_CLRSTATUS 0x84 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_svc_send(unsigned long svc_id, - unsigned long buffer, - unsigned long buffer_size, - unsigned long *sent_bytes); -extern unsigned long sun4v_svc_recv(unsigned long svc_id, - unsigned long buffer, - unsigned long buffer_size, - unsigned long *recv_bytes); -extern unsigned long sun4v_svc_getstatus(unsigned long svc_id, - unsigned long *status_bits); -extern unsigned long sun4v_svc_setstatus(unsigned long svc_id, - unsigned long status_bits); -extern unsigned long sun4v_svc_clrstatus(unsigned long svc_id, - unsigned long status_bits); +unsigned long sun4v_svc_send(unsigned long svc_id, + unsigned long buffer, + unsigned long buffer_size, + unsigned long *sent_bytes); +unsigned long sun4v_svc_recv(unsigned long svc_id, + unsigned long buffer, + unsigned long buffer_size, + unsigned long *recv_bytes); +unsigned long sun4v_svc_getstatus(unsigned long svc_id, + unsigned long *status_bits); +unsigned long sun4v_svc_setstatus(unsigned long svc_id, + unsigned long status_bits); +unsigned long sun4v_svc_clrstatus(unsigned long svc_id, + unsigned long status_bits); #endif /* Trap trace services. @@ -1458,8 +1460,8 @@ struct hv_trap_trace_entry { #define HV_FAST_INTR_DEVINO2SYSINO 0xa0 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_devino_to_sysino(unsigned long devhandle, - unsigned long devino); +unsigned long sun4v_devino_to_sysino(unsigned long devhandle, + unsigned long devino); #endif /* intr_getenabled() @@ -1476,7 +1478,7 @@ extern unsigned long sun4v_devino_to_sysino(unsigned long devhandle, #define HV_FAST_INTR_GETENABLED 0xa1 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_intr_getenabled(unsigned long sysino); +unsigned long sun4v_intr_getenabled(unsigned long sysino); #endif /* intr_setenabled() @@ -1492,7 +1494,8 @@ extern unsigned long sun4v_intr_getenabled(unsigned long sysino); #define HV_FAST_INTR_SETENABLED 0xa2 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_intr_setenabled(unsigned long sysino, unsigned long intr_enabled); +unsigned long sun4v_intr_setenabled(unsigned long sysino, + unsigned long intr_enabled); #endif /* intr_getstate() @@ -1508,7 +1511,7 @@ extern unsigned long sun4v_intr_setenabled(unsigned long sysino, unsigned long i #define HV_FAST_INTR_GETSTATE 0xa3 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_intr_getstate(unsigned long sysino); +unsigned long sun4v_intr_getstate(unsigned long sysino); #endif /* intr_setstate() @@ -1528,7 +1531,7 @@ extern unsigned long sun4v_intr_getstate(unsigned long sysino); #define HV_FAST_INTR_SETSTATE 0xa4 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_intr_setstate(unsigned long sysino, unsigned long intr_state); +unsigned long sun4v_intr_setstate(unsigned long sysino, unsigned long intr_state); #endif /* intr_gettarget() @@ -1546,7 +1549,7 @@ extern unsigned long sun4v_intr_setstate(unsigned long sysino, unsigned long int #define HV_FAST_INTR_GETTARGET 0xa5 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_intr_gettarget(unsigned long sysino); +unsigned long sun4v_intr_gettarget(unsigned long sysino); #endif /* intr_settarget() @@ -1563,7 +1566,7 @@ extern unsigned long sun4v_intr_gettarget(unsigned long sysino); #define HV_FAST_INTR_SETTARGET 0xa6 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cpuid); +unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cpuid); #endif /* vintr_get_cookie() @@ -1647,30 +1650,30 @@ extern unsigned long sun4v_intr_settarget(unsigned long sysino, unsigned long cp #define HV_FAST_VINTR_SET_TARGET 0xae #ifndef __ASSEMBLY__ -extern unsigned long sun4v_vintr_get_cookie(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long *cookie); -extern unsigned long sun4v_vintr_set_cookie(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long cookie); -extern unsigned long sun4v_vintr_get_valid(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long *valid); -extern unsigned long sun4v_vintr_set_valid(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long valid); -extern unsigned long sun4v_vintr_get_state(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long *state); -extern unsigned long sun4v_vintr_set_state(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long state); -extern unsigned long sun4v_vintr_get_target(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long *cpuid); -extern unsigned long sun4v_vintr_set_target(unsigned long dev_handle, - unsigned long dev_ino, - unsigned long cpuid); +unsigned long sun4v_vintr_get_cookie(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *cookie); +unsigned long sun4v_vintr_set_cookie(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long cookie); +unsigned long sun4v_vintr_get_valid(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *valid); +unsigned long sun4v_vintr_set_valid(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long valid); +unsigned long sun4v_vintr_get_state(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *state); +unsigned long sun4v_vintr_set_state(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long state); +unsigned long sun4v_vintr_get_target(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long *cpuid); +unsigned long sun4v_vintr_set_target(unsigned long dev_handle, + unsigned long dev_ino, + unsigned long cpuid); #endif /* PCI IO services. @@ -2627,50 +2630,50 @@ struct ldc_mtable_entry { #define HV_FAST_LDC_REVOKE 0xef #ifndef __ASSEMBLY__ -extern unsigned long sun4v_ldc_tx_qconf(unsigned long channel, - unsigned long ra, - unsigned long num_entries); -extern unsigned long sun4v_ldc_tx_qinfo(unsigned long channel, - unsigned long *ra, - unsigned long *num_entries); -extern unsigned long sun4v_ldc_tx_get_state(unsigned long channel, - unsigned long *head_off, - unsigned long *tail_off, - unsigned long *chan_state); -extern unsigned long sun4v_ldc_tx_set_qtail(unsigned long channel, - unsigned long tail_off); -extern unsigned long sun4v_ldc_rx_qconf(unsigned long channel, - unsigned long ra, - unsigned long num_entries); -extern unsigned long sun4v_ldc_rx_qinfo(unsigned long channel, - unsigned long *ra, - unsigned long *num_entries); -extern unsigned long sun4v_ldc_rx_get_state(unsigned long channel, - unsigned long *head_off, - unsigned long *tail_off, - unsigned long *chan_state); -extern unsigned long sun4v_ldc_rx_set_qhead(unsigned long channel, - unsigned long head_off); -extern unsigned long sun4v_ldc_set_map_table(unsigned long channel, - unsigned long ra, - unsigned long num_entries); -extern unsigned long sun4v_ldc_get_map_table(unsigned long channel, - unsigned long *ra, - unsigned long *num_entries); -extern unsigned long sun4v_ldc_copy(unsigned long channel, - unsigned long dir_code, - unsigned long tgt_raddr, - unsigned long lcl_raddr, - unsigned long len, - unsigned long *actual_len); -extern unsigned long sun4v_ldc_mapin(unsigned long channel, - unsigned long cookie, - unsigned long *ra, - unsigned long *perm); -extern unsigned long sun4v_ldc_unmap(unsigned long ra); -extern unsigned long sun4v_ldc_revoke(unsigned long channel, - unsigned long cookie, - unsigned long mte_cookie); +unsigned long sun4v_ldc_tx_qconf(unsigned long channel, + unsigned long ra, + unsigned long num_entries); +unsigned long sun4v_ldc_tx_qinfo(unsigned long channel, + unsigned long *ra, + unsigned long *num_entries); +unsigned long sun4v_ldc_tx_get_state(unsigned long channel, + unsigned long *head_off, + unsigned long *tail_off, + unsigned long *chan_state); +unsigned long sun4v_ldc_tx_set_qtail(unsigned long channel, + unsigned long tail_off); +unsigned long sun4v_ldc_rx_qconf(unsigned long channel, + unsigned long ra, + unsigned long num_entries); +unsigned long sun4v_ldc_rx_qinfo(unsigned long channel, + unsigned long *ra, + unsigned long *num_entries); +unsigned long sun4v_ldc_rx_get_state(unsigned long channel, + unsigned long *head_off, + unsigned long *tail_off, + unsigned long *chan_state); +unsigned long sun4v_ldc_rx_set_qhead(unsigned long channel, + unsigned long head_off); +unsigned long sun4v_ldc_set_map_table(unsigned long channel, + unsigned long ra, + unsigned long num_entries); +unsigned long sun4v_ldc_get_map_table(unsigned long channel, + unsigned long *ra, + unsigned long *num_entries); +unsigned long sun4v_ldc_copy(unsigned long channel, + unsigned long dir_code, + unsigned long tgt_raddr, + unsigned long lcl_raddr, + unsigned long len, + unsigned long *actual_len); +unsigned long sun4v_ldc_mapin(unsigned long channel, + unsigned long cookie, + unsigned long *ra, + unsigned long *perm); +unsigned long sun4v_ldc_unmap(unsigned long ra); +unsigned long sun4v_ldc_revoke(unsigned long channel, + unsigned long cookie, + unsigned long mte_cookie); #endif /* Performance counter services. */ @@ -2727,14 +2730,14 @@ extern unsigned long sun4v_ldc_revoke(unsigned long channel, #define HV_FAST_N2_SET_PERFREG 0x105 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_niagara_getperf(unsigned long reg, - unsigned long *val); -extern unsigned long sun4v_niagara_setperf(unsigned long reg, - unsigned long val); -extern unsigned long sun4v_niagara2_getperf(unsigned long reg, - unsigned long *val); -extern unsigned long sun4v_niagara2_setperf(unsigned long reg, - unsigned long val); +unsigned long sun4v_niagara_getperf(unsigned long reg, + unsigned long *val); +unsigned long sun4v_niagara_setperf(unsigned long reg, + unsigned long val); +unsigned long sun4v_niagara2_getperf(unsigned long reg, + unsigned long *val); +unsigned long sun4v_niagara2_setperf(unsigned long reg, + unsigned long val); #endif /* MMU statistics services. @@ -2829,8 +2832,8 @@ struct hv_mmu_statistics { #define HV_FAST_MMUSTAT_INFO 0x103 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_mmustat_conf(unsigned long ra, unsigned long *orig_ra); -extern unsigned long sun4v_mmustat_info(unsigned long *ra); +unsigned long sun4v_mmustat_conf(unsigned long ra, unsigned long *orig_ra); +unsigned long sun4v_mmustat_info(unsigned long *ra); #endif /* NCS crypto services */ @@ -2919,9 +2922,9 @@ struct hv_ncs_qtail_update_arg { #define HV_FAST_NCS_REQUEST 0x110 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_ncs_request(unsigned long request, - unsigned long arg_ra, - unsigned long arg_size); +unsigned long sun4v_ncs_request(unsigned long request, + unsigned long arg_ra, + unsigned long arg_size); #endif #define HV_FAST_FIRE_GET_PERFREG 0x120 @@ -2930,8 +2933,18 @@ extern unsigned long sun4v_ncs_request(unsigned long request, #define HV_FAST_REBOOT_DATA_SET 0x172 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_reboot_data_set(unsigned long ra, - unsigned long len); +unsigned long sun4v_reboot_data_set(unsigned long ra, + unsigned long len); +#endif + +#define HV_FAST_VT_GET_PERFREG 0x184 +#define HV_FAST_VT_SET_PERFREG 0x185 + +#ifndef __ASSEMBLY__ +unsigned long sun4v_vt_get_perfreg(unsigned long reg_num, + unsigned long *reg_val); +unsigned long sun4v_vt_set_perfreg(unsigned long reg_num, + unsigned long reg_val); #endif /* Function numbers for HV_CORE_TRAP. */ @@ -2964,24 +2977,25 @@ extern unsigned long sun4v_reboot_data_set(unsigned long ra, #define HV_GRP_NIU 0x0204 #define HV_GRP_VF_CPU 0x0205 #define HV_GRP_KT_CPU 0x0209 +#define HV_GRP_VT_CPU 0x020c #define HV_GRP_DIAG 0x0300 #ifndef __ASSEMBLY__ -extern unsigned long sun4v_get_version(unsigned long group, - unsigned long *major, - unsigned long *minor); -extern unsigned long sun4v_set_version(unsigned long group, - unsigned long major, - unsigned long minor, - unsigned long *actual_minor); - -extern int sun4v_hvapi_register(unsigned long group, unsigned long major, - unsigned long *minor); -extern void sun4v_hvapi_unregister(unsigned long group); -extern int sun4v_hvapi_get(unsigned long group, - unsigned long *major, - unsigned long *minor); -extern void sun4v_hvapi_init(void); +unsigned long sun4v_get_version(unsigned long group, + unsigned long *major, + unsigned long *minor); +unsigned long sun4v_set_version(unsigned long group, + unsigned long major, + unsigned long minor, + unsigned long *actual_minor); + +int sun4v_hvapi_register(unsigned long group, unsigned long major, + unsigned long *minor); +void sun4v_hvapi_unregister(unsigned long group); +int sun4v_hvapi_get(unsigned long group, + unsigned long *major, + unsigned long *minor); +void sun4v_hvapi_init(void); #endif #endif /* !(_SPARC64_HYPERVISOR_H) */ diff --git a/arch/sparc/include/asm/idprom.h b/arch/sparc/include/asm/idprom.h index 6976aa2439c..3793f7f91c4 100644 --- a/arch/sparc/include/asm/idprom.h +++ b/arch/sparc/include/asm/idprom.h @@ -20,6 +20,6 @@ struct idprom { }; extern struct idprom *idprom; -extern void idprom_init(void); +void idprom_init(void); #endif /* !(_SPARC_IDPROM_H) */ diff --git a/arch/sparc/include/asm/io-unit.h b/arch/sparc/include/asm/io-unit.h index 01ab2f613e9..04a9701e720 100644 --- a/arch/sparc/include/asm/io-unit.h +++ b/arch/sparc/include/asm/io-unit.h @@ -43,7 +43,7 @@ struct iounit_struct { unsigned long bmap[(IOUNIT_DMA_SIZE >> (PAGE_SHIFT + 3)) / sizeof(unsigned long)]; spinlock_t lock; - iopte_t *page_table; + iopte_t __iomem *page_table; unsigned long rotor[3]; unsigned long limit[4]; }; diff --git a/arch/sparc/include/asm/io_32.h b/arch/sparc/include/asm/io_32.h index c1acbd891cb..9f532902627 100644 --- a/arch/sparc/include/asm/io_32.h +++ b/arch/sparc/include/asm/io_32.h @@ -2,191 +2,94 @@ #define __SPARC_IO_H #include <linux/kernel.h> -#include <linux/types.h> #include <linux/ioport.h> /* struct resource */ -#include <asm/page.h> /* IO address mapping routines need this */ -#include <asm-generic/pci_iomap.h> - -#define page_to_phys(page) (page_to_pfn(page) << PAGE_SHIFT) - -static inline u32 flip_dword (u32 l) -{ - return ((l&0xff)<<24) | (((l>>8)&0xff)<<16) | (((l>>16)&0xff)<<8)| ((l>>24)&0xff); -} - -static inline u16 flip_word (u16 w) -{ - return ((w&0xff) << 8) | ((w>>8)&0xff); -} - -#define mmiowb() - -/* - * Memory mapped I/O to PCI - */ - -static inline u8 __raw_readb(const volatile void __iomem *addr) -{ - return *(__force volatile u8 *)addr; -} - -static inline u16 __raw_readw(const volatile void __iomem *addr) -{ - return *(__force volatile u16 *)addr; -} - -static inline u32 __raw_readl(const volatile void __iomem *addr) -{ - return *(__force volatile u32 *)addr; -} +#define readb_relaxed(__addr) readb(__addr) +#define readw_relaxed(__addr) readw(__addr) +#define readl_relaxed(__addr) readl(__addr) -static inline void __raw_writeb(u8 b, volatile void __iomem *addr) -{ - *(__force volatile u8 *)addr = b; -} +#define IO_SPACE_LIMIT 0xffffffff -static inline void __raw_writew(u16 w, volatile void __iomem *addr) -{ - *(__force volatile u16 *)addr = w; -} +#define memset_io(d,c,sz) _memset_io(d,c,sz) +#define memcpy_fromio(d,s,sz) _memcpy_fromio(d,s,sz) +#define memcpy_toio(d,s,sz) _memcpy_toio(d,s,sz) -static inline void __raw_writel(u32 l, volatile void __iomem *addr) -{ - *(__force volatile u32 *)addr = l; -} +#include <asm-generic/io.h> -static inline u8 __readb(const volatile void __iomem *addr) +static inline void _memset_io(volatile void __iomem *dst, + int c, __kernel_size_t n) { - return *(__force volatile u8 *)addr; -} + volatile void __iomem *d = dst; -static inline u16 __readw(const volatile void __iomem *addr) -{ - return flip_word(*(__force volatile u16 *)addr); + while (n--) { + writeb(c, d); + d++; + } } -static inline u32 __readl(const volatile void __iomem *addr) +static inline void _memcpy_fromio(void *dst, const volatile void __iomem *src, + __kernel_size_t n) { - return flip_dword(*(__force volatile u32 *)addr); -} + char *d = dst; -static inline void __writeb(u8 b, volatile void __iomem *addr) -{ - *(__force volatile u8 *)addr = b; + while (n--) { + char tmp = readb(src); + *d++ = tmp; + src++; + } } -static inline void __writew(u16 w, volatile void __iomem *addr) +static inline void _memcpy_toio(volatile void __iomem *dst, const void *src, + __kernel_size_t n) { - *(__force volatile u16 *)addr = flip_word(w); -} + const char *s = src; + volatile void __iomem *d = dst; -static inline void __writel(u32 l, volatile void __iomem *addr) -{ - *(__force volatile u32 *)addr = flip_dword(l); + while (n--) { + char tmp = *s++; + writeb(tmp, d); + d++; + } } -#define readb(__addr) __readb(__addr) -#define readw(__addr) __readw(__addr) -#define readl(__addr) __readl(__addr) -#define readb_relaxed(__addr) readb(__addr) -#define readw_relaxed(__addr) readw(__addr) -#define readl_relaxed(__addr) readl(__addr) - -#define writeb(__b, __addr) __writeb((__b),(__addr)) -#define writew(__w, __addr) __writew((__w),(__addr)) -#define writel(__l, __addr) __writel((__l),(__addr)) - -/* - * I/O space operations - * - * Arrangement on a Sun is somewhat complicated. - * - * First of all, we want to use standard Linux drivers - * for keyboard, PC serial, etc. These drivers think - * they access I/O space and use inb/outb. - * On the other hand, EBus bridge accepts PCI *memory* - * cycles and converts them into ISA *I/O* cycles. - * Ergo, we want inb & outb to generate PCI memory cycles. - * - * If we want to issue PCI *I/O* cycles, we do this - * with a low 64K fixed window in PCIC. This window gets - * mapped somewhere into virtual kernel space and we - * can use inb/outb again. - */ -#define inb_local(__addr) __readb((void __iomem *)(unsigned long)(__addr)) -#define inb(__addr) __readb((void __iomem *)(unsigned long)(__addr)) -#define inw(__addr) __readw((void __iomem *)(unsigned long)(__addr)) -#define inl(__addr) __readl((void __iomem *)(unsigned long)(__addr)) - -#define outb_local(__b, __addr) __writeb(__b, (void __iomem *)(unsigned long)(__addr)) -#define outb(__b, __addr) __writeb(__b, (void __iomem *)(unsigned long)(__addr)) -#define outw(__w, __addr) __writew(__w, (void __iomem *)(unsigned long)(__addr)) -#define outl(__l, __addr) __writel(__l, (void __iomem *)(unsigned long)(__addr)) - -#define inb_p(__addr) inb(__addr) -#define outb_p(__b, __addr) outb(__b, __addr) -#define inw_p(__addr) inw(__addr) -#define outw_p(__w, __addr) outw(__w, __addr) -#define inl_p(__addr) inl(__addr) -#define outl_p(__l, __addr) outl(__l, __addr) - -void outsb(unsigned long addr, const void *src, unsigned long cnt); -void outsw(unsigned long addr, const void *src, unsigned long cnt); -void outsl(unsigned long addr, const void *src, unsigned long cnt); -void insb(unsigned long addr, void *dst, unsigned long count); -void insw(unsigned long addr, void *dst, unsigned long count); -void insl(unsigned long addr, void *dst, unsigned long count); - -#define IO_SPACE_LIMIT 0xffffffff - /* * SBus accessors. * * SBus has only one, memory mapped, I/O space. * We do not need to flip bytes for SBus of course. */ -static inline u8 _sbus_readb(const volatile void __iomem *addr) +static inline u8 sbus_readb(const volatile void __iomem *addr) { return *(__force volatile u8 *)addr; } -static inline u16 _sbus_readw(const volatile void __iomem *addr) +static inline u16 sbus_readw(const volatile void __iomem *addr) { return *(__force volatile u16 *)addr; } -static inline u32 _sbus_readl(const volatile void __iomem *addr) +static inline u32 sbus_readl(const volatile void __iomem *addr) { return *(__force volatile u32 *)addr; } -static inline void _sbus_writeb(u8 b, volatile void __iomem *addr) +static inline void sbus_writeb(u8 b, volatile void __iomem *addr) { *(__force volatile u8 *)addr = b; } -static inline void _sbus_writew(u16 w, volatile void __iomem *addr) +static inline void sbus_writew(u16 w, volatile void __iomem *addr) { *(__force volatile u16 *)addr = w; } -static inline void _sbus_writel(u32 l, volatile void __iomem *addr) +static inline void sbus_writel(u32 l, volatile void __iomem *addr) { *(__force volatile u32 *)addr = l; } -/* - * The only reason for #define's is to hide casts to unsigned long. - */ -#define sbus_readb(__addr) _sbus_readb(__addr) -#define sbus_readw(__addr) _sbus_readw(__addr) -#define sbus_readl(__addr) _sbus_readl(__addr) -#define sbus_writeb(__b, __addr) _sbus_writeb(__b, __addr) -#define sbus_writew(__w, __addr) _sbus_writew(__w, __addr) -#define sbus_writel(__l, __addr) _sbus_writel(__l, __addr) - -static inline void sbus_memset_io(volatile void __iomem *__dst, int c, __kernel_size_t n) +static inline void sbus_memset_io(volatile void __iomem *__dst, int c, + __kernel_size_t n) { while(n--) { sbus_writeb(c, __dst); @@ -194,22 +97,9 @@ static inline void sbus_memset_io(volatile void __iomem *__dst, int c, __kernel_ } } -static inline void -_memset_io(volatile void __iomem *dst, int c, __kernel_size_t n) -{ - volatile void __iomem *d = dst; - - while (n--) { - writeb(c, d); - d++; - } -} - -#define memset_io(d,c,sz) _memset_io(d,c,sz) - -static inline void -_sbus_memcpy_fromio(void *dst, const volatile void __iomem *src, - __kernel_size_t n) +static inline void sbus_memcpy_fromio(void *dst, + const volatile void __iomem *src, + __kernel_size_t n) { char *d = dst; @@ -220,25 +110,9 @@ _sbus_memcpy_fromio(void *dst, const volatile void __iomem *src, } } -#define sbus_memcpy_fromio(d, s, sz) _sbus_memcpy_fromio(d, s, sz) - -static inline void -_memcpy_fromio(void *dst, const volatile void __iomem *src, __kernel_size_t n) -{ - char *d = dst; - - while (n--) { - char tmp = readb(src); - *d++ = tmp; - src++; - } -} - -#define memcpy_fromio(d,s,sz) _memcpy_fromio(d,s,sz) - -static inline void -_sbus_memcpy_toio(volatile void __iomem *dst, const void *src, - __kernel_size_t n) +static inline void sbus_memcpy_toio(volatile void __iomem *dst, + const void *src, + __kernel_size_t n) { const char *s = src; volatile void __iomem *d = dst; @@ -250,81 +124,26 @@ _sbus_memcpy_toio(volatile void __iomem *dst, const void *src, } } -#define sbus_memcpy_toio(d, s, sz) _sbus_memcpy_toio(d, s, sz) - -static inline void -_memcpy_toio(volatile void __iomem *dst, const void *src, __kernel_size_t n) -{ - const char *s = src; - volatile void __iomem *d = dst; - - while (n--) { - char tmp = *s++; - writeb(tmp, d); - d++; - } -} - -#define memcpy_toio(d,s,sz) _memcpy_toio(d,s,sz) - #ifdef __KERNEL__ /* * Bus number may be embedded in the higher bits of the physical address. * This is why we have no bus number argument to ioremap(). */ -extern void __iomem *ioremap(unsigned long offset, unsigned long size); +void __iomem *ioremap(unsigned long offset, unsigned long size); #define ioremap_nocache(X,Y) ioremap((X),(Y)) #define ioremap_wc(X,Y) ioremap((X),(Y)) -extern void iounmap(volatile void __iomem *addr); - -#define ioread8(X) readb(X) -#define ioread16(X) readw(X) -#define ioread16be(X) __raw_readw(X) -#define ioread32(X) readl(X) -#define ioread32be(X) __raw_readl(X) -#define iowrite8(val,X) writeb(val,X) -#define iowrite16(val,X) writew(val,X) -#define iowrite16be(val,X) __raw_writew(val,X) -#define iowrite32(val,X) writel(val,X) -#define iowrite32be(val,X) __raw_writel(val,X) - -static inline void ioread8_rep(void __iomem *port, void *buf, unsigned long count) -{ - insb((unsigned long __force)port, buf, count); -} -static inline void ioread16_rep(void __iomem *port, void *buf, unsigned long count) -{ - insw((unsigned long __force)port, buf, count); -} - -static inline void ioread32_rep(void __iomem *port, void *buf, unsigned long count) -{ - insl((unsigned long __force)port, buf, count); -} - -static inline void iowrite8_rep(void __iomem *port, const void *buf, unsigned long count) -{ - outsb((unsigned long __force)port, buf, count); -} - -static inline void iowrite16_rep(void __iomem *port, const void *buf, unsigned long count) -{ - outsw((unsigned long __force)port, buf, count); -} - -static inline void iowrite32_rep(void __iomem *port, const void *buf, unsigned long count) -{ - outsl((unsigned long __force)port, buf, count); -} +void iounmap(volatile void __iomem *addr); /* Create a virtual mapping cookie for an IO port range */ -extern void __iomem *ioport_map(unsigned long port, unsigned int nr); -extern void ioport_unmap(void __iomem *); +void __iomem *ioport_map(unsigned long port, unsigned int nr); +void ioport_unmap(void __iomem *); /* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ struct pci_dev; -extern void pci_iounmap(struct pci_dev *dev, void __iomem *); +void pci_iounmap(struct pci_dev *dev, void __iomem *); + + /* * At the moment, we do not use CMOS_READ anywhere outside of rtc.c, @@ -343,21 +162,11 @@ static inline int sbus_can_burst64(void) return 0; /* actually, sparc_cpu_model==sun4d */ } struct device; -extern void sbus_set_sbus64(struct device *, int); +void sbus_set_sbus64(struct device *, int); #endif #define __ARCH_HAS_NO_PAGE_ZERO_MAPPED 1 -/* - * Convert a physical pointer to a virtual kernel pointer for /dev/mem - * access - */ -#define xlate_dev_mem_ptr(p) __va(p) - -/* - * Convert a virtual cached pointer to an uncached pointer - */ -#define xlate_dev_kmem_ptr(p) p #endif /* !(__SPARC_IO_H) */ diff --git a/arch/sparc/include/asm/io_64.h b/arch/sparc/include/asm/io_64.h index 09b0b88aeb2..05381c3a422 100644 --- a/arch/sparc/include/asm/io_64.h +++ b/arch/sparc/include/asm/io_64.h @@ -15,7 +15,6 @@ /* BIO layer definitions. */ extern unsigned long kern_base, kern_size; -#define page_to_phys(page) (page_to_pfn(page) << PAGE_SHIFT) static inline u8 _inb(unsigned long addr) { @@ -91,12 +90,12 @@ static inline void _outl(u32 l, unsigned long addr) #define inl_p(__addr) inl(__addr) #define outl_p(__l, __addr) outl(__l, __addr) -extern void outsb(unsigned long, const void *, unsigned long); -extern void outsw(unsigned long, const void *, unsigned long); -extern void outsl(unsigned long, const void *, unsigned long); -extern void insb(unsigned long, void *, unsigned long); -extern void insw(unsigned long, void *, unsigned long); -extern void insl(unsigned long, void *, unsigned long); +void outsb(unsigned long, const void *, unsigned long); +void outsw(unsigned long, const void *, unsigned long); +void outsl(unsigned long, const void *, unsigned long); +void insb(unsigned long, void *, unsigned long); +void insw(unsigned long, void *, unsigned long); +void insl(unsigned long, void *, unsigned long); static inline void ioread8_rep(void __iomem *port, void *buf, unsigned long count) { @@ -509,12 +508,12 @@ static inline void iounmap(volatile void __iomem *addr) #define iowrite32be(val,X) __raw_writel(val,X) /* Create a virtual mapping cookie for an IO port range */ -extern void __iomem *ioport_map(unsigned long port, unsigned int nr); -extern void ioport_unmap(void __iomem *); +void __iomem *ioport_map(unsigned long port, unsigned int nr); +void ioport_unmap(void __iomem *); /* Create a virtual mapping cookie for a PCI BAR (memory or IO) */ struct pci_dev; -extern void pci_iounmap(struct pci_dev *dev, void __iomem *); +void pci_iounmap(struct pci_dev *dev, void __iomem *); static inline int sbus_can_dma_64bit(void) { @@ -525,7 +524,7 @@ static inline int sbus_can_burst64(void) return 1; } struct device; -extern void sbus_set_sbus64(struct device *, int); +void sbus_set_sbus64(struct device *, int); /* * Convert a physical pointer to a virtual kernel pointer for /dev/mem diff --git a/arch/sparc/include/asm/ioctls.h b/arch/sparc/include/asm/ioctls.h index 28d0c8b02cc..77413b7e3a1 100644 --- a/arch/sparc/include/asm/ioctls.h +++ b/arch/sparc/include/asm/ioctls.h @@ -1,123 +1,8 @@ #ifndef _ASM_SPARC_IOCTLS_H #define _ASM_SPARC_IOCTLS_H -#include <asm/ioctl.h> +#include <uapi/asm/ioctls.h> -/* Big T */ -#define TCGETA _IOR('T', 1, struct termio) -#define TCSETA _IOW('T', 2, struct termio) -#define TCSETAW _IOW('T', 3, struct termio) -#define TCSETAF _IOW('T', 4, struct termio) -#define TCSBRK _IO('T', 5) -#define TCXONC _IO('T', 6) -#define TCFLSH _IO('T', 7) -#define TCGETS _IOR('T', 8, struct termios) -#define TCSETS _IOW('T', 9, struct termios) -#define TCSETSW _IOW('T', 10, struct termios) -#define TCSETSF _IOW('T', 11, struct termios) -#define TCGETS2 _IOR('T', 12, struct termios2) -#define TCSETS2 _IOW('T', 13, struct termios2) -#define TCSETSW2 _IOW('T', 14, struct termios2) -#define TCSETSF2 _IOW('T', 15, struct termios2) -#define TIOCGDEV _IOR('T',0x32, unsigned int) /* Get primary device node of /dev/console */ -#define TIOCVHANGUP _IO('T', 0x37) - -/* Note that all the ioctls that are not available in Linux have a - * double underscore on the front to: a) avoid some programs to - * think we support some ioctls under Linux (autoconfiguration stuff) - */ -/* Little t */ -#define TIOCGETD _IOR('t', 0, int) -#define TIOCSETD _IOW('t', 1, int) -#define __TIOCHPCL _IO('t', 2) /* SunOS Specific */ -#define __TIOCMODG _IOR('t', 3, int) /* SunOS Specific */ -#define __TIOCMODS _IOW('t', 4, int) /* SunOS Specific */ -#define __TIOCGETP _IOR('t', 8, struct sgttyb) /* SunOS Specific */ -#define __TIOCSETP _IOW('t', 9, struct sgttyb) /* SunOS Specific */ -#define __TIOCSETN _IOW('t', 10, struct sgttyb) /* SunOS Specific */ -#define TIOCEXCL _IO('t', 13) -#define TIOCNXCL _IO('t', 14) -#define __TIOCFLUSH _IOW('t', 16, int) /* SunOS Specific */ -#define __TIOCSETC _IOW('t', 17, struct tchars) /* SunOS Specific */ -#define __TIOCGETC _IOR('t', 18, struct tchars) /* SunOS Specific */ -#define __TIOCTCNTL _IOW('t', 32, int) /* SunOS Specific */ -#define __TIOCSIGNAL _IOW('t', 33, int) /* SunOS Specific */ -#define __TIOCSETX _IOW('t', 34, int) /* SunOS Specific */ -#define __TIOCGETX _IOR('t', 35, int) /* SunOS Specific */ -#define TIOCCONS _IO('t', 36) -#define TIOCGSOFTCAR _IOR('t', 100, int) -#define TIOCSSOFTCAR _IOW('t', 101, int) -#define __TIOCUCNTL _IOW('t', 102, int) /* SunOS Specific */ -#define TIOCSWINSZ _IOW('t', 103, struct winsize) -#define TIOCGWINSZ _IOR('t', 104, struct winsize) -#define __TIOCREMOTE _IOW('t', 105, int) /* SunOS Specific */ -#define TIOCMGET _IOR('t', 106, int) -#define TIOCMBIC _IOW('t', 107, int) -#define TIOCMBIS _IOW('t', 108, int) -#define TIOCMSET _IOW('t', 109, int) -#define TIOCSTART _IO('t', 110) -#define TIOCSTOP _IO('t', 111) -#define TIOCPKT _IOW('t', 112, int) -#define TIOCNOTTY _IO('t', 113) -#define TIOCSTI _IOW('t', 114, char) -#define TIOCOUTQ _IOR('t', 115, int) -#define __TIOCGLTC _IOR('t', 116, struct ltchars) /* SunOS Specific */ -#define __TIOCSLTC _IOW('t', 117, struct ltchars) /* SunOS Specific */ -/* 118 is the non-posix setpgrp tty ioctl */ -/* 119 is the non-posix getpgrp tty ioctl */ -#define __TIOCCDTR _IO('t', 120) /* SunOS Specific */ -#define __TIOCSDTR _IO('t', 121) /* SunOS Specific */ -#define TIOCCBRK _IO('t', 122) -#define TIOCSBRK _IO('t', 123) -#define __TIOCLGET _IOW('t', 124, int) /* SunOS Specific */ -#define __TIOCLSET _IOW('t', 125, int) /* SunOS Specific */ -#define __TIOCLBIC _IOW('t', 126, int) /* SunOS Specific */ -#define __TIOCLBIS _IOW('t', 127, int) /* SunOS Specific */ -#define __TIOCISPACE _IOR('t', 128, int) /* SunOS Specific */ -#define __TIOCISIZE _IOR('t', 129, int) /* SunOS Specific */ -#define TIOCSPGRP _IOW('t', 130, int) -#define TIOCGPGRP _IOR('t', 131, int) -#define TIOCSCTTY _IO('t', 132) -#define TIOCGSID _IOR('t', 133, int) -/* Get minor device of a pty master's FD -- Solaris equiv is ISPTM */ -#define TIOCGPTN _IOR('t', 134, unsigned int) /* Get Pty Number */ -#define TIOCSPTLCK _IOW('t', 135, int) /* Lock/unlock PTY */ -#define TIOCSIG _IOW('t', 136, int) /* Generate signal on Pty slave */ - -/* Little f */ -#define FIOCLEX _IO('f', 1) -#define FIONCLEX _IO('f', 2) -#define FIOASYNC _IOW('f', 125, int) -#define FIONBIO _IOW('f', 126, int) -#define FIONREAD _IOR('f', 127, int) -#define TIOCINQ FIONREAD -#define FIOQSIZE _IOR('f', 128, loff_t) - -/* SCARY Rutgers local SunOS kernel hackery, perhaps I will support it - * someday. This is completely bogus, I know... - */ -#define __TCGETSTAT _IO('T', 200) /* Rutgers specific */ -#define __TCSETSTAT _IO('T', 201) /* Rutgers specific */ - -/* Linux specific, no SunOS equivalent. */ -#define TIOCLINUX 0x541C -#define TIOCGSERIAL 0x541E -#define TIOCSSERIAL 0x541F -#define TCSBRKP 0x5425 -#define TIOCSERCONFIG 0x5453 -#define TIOCSERGWILD 0x5454 -#define TIOCSERSWILD 0x5455 -#define TIOCGLCKTRMIOS 0x5456 -#define TIOCSLCKTRMIOS 0x5457 -#define TIOCSERGSTRUCT 0x5458 /* For debugging only */ -#define TIOCSERGETLSR 0x5459 /* Get line status register */ -#define TIOCSERGETMULTI 0x545A /* Get multiport config */ -#define TIOCSERSETMULTI 0x545B /* Set multiport config */ -#define TIOCMIWAIT 0x545C /* Wait for change on serial input line(s) */ -#define TIOCGICOUNT 0x545D /* Read serial port inline interrupt counts */ - -/* Kernel definitions */ -#ifdef __KERNEL__ #define TIOCGETC __TIOCGETC #define TIOCGETP __TIOCGETP #define TIOCGLTC __TIOCGLTC @@ -125,16 +10,4 @@ #define TIOCSETP __TIOCSETP #define TIOCSETN __TIOCSETN #define TIOCSETC __TIOCSETC -#endif - -/* Used for packet mode */ -#define TIOCPKT_DATA 0 -#define TIOCPKT_FLUSHREAD 1 -#define TIOCPKT_FLUSHWRITE 2 -#define TIOCPKT_STOP 4 -#define TIOCPKT_START 8 -#define TIOCPKT_NOSTOP 16 -#define TIOCPKT_DOSTOP 32 -#define TIOCPKT_IOCTL 64 - #endif /* !(_ASM_SPARC_IOCTLS_H) */ diff --git a/arch/sparc/include/asm/iommu_32.h b/arch/sparc/include/asm/iommu_32.h index 70c589c05a1..f6c066b52fd 100644 --- a/arch/sparc/include/asm/iommu_32.h +++ b/arch/sparc/include/asm/iommu_32.h @@ -99,7 +99,7 @@ struct iommu_regs { #define IOPTE_WAZ 0x00000001 /* Write as zeros */ struct iommu_struct { - struct iommu_regs *regs; + struct iommu_regs __iomem *regs; iopte_t *page_table; /* For convenience */ unsigned long start; /* First managed virtual address */ @@ -108,14 +108,14 @@ struct iommu_struct { struct bit_map usemap; }; -static inline void iommu_invalidate(struct iommu_regs *regs) +static inline void iommu_invalidate(struct iommu_regs __iomem *regs) { - regs->tlbflush = 0; + sbus_writel(0, ®s->tlbflush); } -static inline void iommu_invalidate_page(struct iommu_regs *regs, unsigned long ba) +static inline void iommu_invalidate_page(struct iommu_regs __iomem *regs, unsigned long ba) { - regs->pageflush = (ba & PAGE_MASK); + sbus_writel(ba & PAGE_MASK, ®s->pageflush); } #endif /* !(_SPARC_IOMMU_H) */ diff --git a/arch/sparc/include/asm/iommu_64.h b/arch/sparc/include/asm/iommu_64.h index caf798b5619..2b9321ab064 100644 --- a/arch/sparc/include/asm/iommu_64.h +++ b/arch/sparc/include/asm/iommu_64.h @@ -58,8 +58,8 @@ struct strbuf { volatile unsigned long __flushflag_buf[(64+(64-1)) / sizeof(long)]; }; -extern int iommu_table_init(struct iommu *iommu, int tsbsize, - u32 dma_offset, u32 dma_addr_mask, - int numa_node); +int iommu_table_init(struct iommu *iommu, int tsbsize, + u32 dma_offset, u32 dma_addr_mask, + int numa_node); #endif /* !(_SPARC64_IOMMU_H) */ diff --git a/arch/sparc/include/asm/irq_32.h b/arch/sparc/include/asm/irq_32.h index 2ae3acaeb1b..eecd3d8442c 100644 --- a/arch/sparc/include/asm/irq_32.h +++ b/arch/sparc/include/asm/irq_32.h @@ -16,7 +16,8 @@ #define irq_canonicalize(irq) (irq) -extern void __init init_IRQ(void); +void __init init_IRQ(void); +void __init sun4d_init_sbi_irq(void); #define NO_IRQ 0xffffffff diff --git a/arch/sparc/include/asm/irq_64.h b/arch/sparc/include/asm/irq_64.h index abf6afe82ca..91d21938130 100644 --- a/arch/sparc/include/asm/irq_64.h +++ b/arch/sparc/include/asm/irq_64.h @@ -39,32 +39,32 @@ */ #define NR_IRQS 255 -extern void irq_install_pre_handler(int irq, - void (*func)(unsigned int, void *, void *), - void *arg1, void *arg2); +void irq_install_pre_handler(int irq, + void (*func)(unsigned int, void *, void *), + void *arg1, void *arg2); #define irq_canonicalize(irq) (irq) -extern unsigned int build_irq(int inofixup, unsigned long iclr, unsigned long imap); -extern unsigned int sun4v_build_irq(u32 devhandle, unsigned int devino); -extern unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino); -extern unsigned int sun4v_build_msi(u32 devhandle, unsigned int *irq_p, - unsigned int msi_devino_start, - unsigned int msi_devino_end); -extern void sun4v_destroy_msi(unsigned int irq); -extern unsigned int sun4u_build_msi(u32 portid, unsigned int *irq_p, - unsigned int msi_devino_start, - unsigned int msi_devino_end, - unsigned long imap_base, - unsigned long iclr_base); -extern void sun4u_destroy_msi(unsigned int irq); - -extern unsigned char irq_alloc(unsigned int dev_handle, - unsigned int dev_ino); +unsigned int build_irq(int inofixup, unsigned long iclr, unsigned long imap); +unsigned int sun4v_build_irq(u32 devhandle, unsigned int devino); +unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino); +unsigned int sun4v_build_msi(u32 devhandle, unsigned int *irq_p, + unsigned int msi_devino_start, + unsigned int msi_devino_end); +void sun4v_destroy_msi(unsigned int irq); +unsigned int sun4u_build_msi(u32 portid, unsigned int *irq_p, + unsigned int msi_devino_start, + unsigned int msi_devino_end, + unsigned long imap_base, + unsigned long iclr_base); +void sun4u_destroy_msi(unsigned int irq); + +unsigned char irq_alloc(unsigned int dev_handle, + unsigned int dev_ino); #ifdef CONFIG_PCI_MSI -extern void irq_free(unsigned int irq); +void irq_free(unsigned int irq); #endif -extern void __init init_IRQ(void); -extern void fixup_irqs(void); +void __init init_IRQ(void); +void fixup_irqs(void); static inline void set_softint(unsigned long bits) { @@ -89,7 +89,7 @@ static inline unsigned long get_softint(void) return retval; } -void arch_trigger_all_cpu_backtrace(void); +void arch_trigger_all_cpu_backtrace(bool); #define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace extern void *hardirq_stack[NR_CPUS]; diff --git a/arch/sparc/include/asm/irqflags_32.h b/arch/sparc/include/asm/irqflags_32.h index e414c06615c..71cc284f55c 100644 --- a/arch/sparc/include/asm/irqflags_32.h +++ b/arch/sparc/include/asm/irqflags_32.h @@ -15,9 +15,9 @@ #include <linux/types.h> #include <asm/psr.h> -extern void arch_local_irq_restore(unsigned long); -extern unsigned long arch_local_irq_save(void); -extern void arch_local_irq_enable(void); +void arch_local_irq_restore(unsigned long); +unsigned long arch_local_irq_save(void); +void arch_local_irq_enable(void); static inline notrace unsigned long arch_local_save_flags(void) { diff --git a/arch/sparc/include/asm/jump_label.h b/arch/sparc/include/asm/jump_label.h index 5080d16a832..ec2e2e2aba7 100644 --- a/arch/sparc/include/asm/jump_label.h +++ b/arch/sparc/include/asm/jump_label.h @@ -9,7 +9,7 @@ static __always_inline bool arch_static_branch(struct static_key *key) { - asm goto("1:\n\t" + asm_volatile_goto("1:\n\t" "nop\n\t" "nop\n\t" ".pushsection __jump_table, \"aw\"\n\t" diff --git a/arch/sparc/include/asm/kdebug_64.h b/arch/sparc/include/asm/kdebug_64.h index feb3578e12c..04465de8f3b 100644 --- a/arch/sparc/include/asm/kdebug_64.h +++ b/arch/sparc/include/asm/kdebug_64.h @@ -3,7 +3,7 @@ struct pt_regs; -extern void bad_trap(struct pt_regs *, long); +void bad_trap(struct pt_regs *, long); /* Grossly misnamed. */ enum die_val { diff --git a/arch/sparc/include/asm/kgdb.h b/arch/sparc/include/asm/kgdb.h index b6ef301d05b..47366af7a58 100644 --- a/arch/sparc/include/asm/kgdb.h +++ b/arch/sparc/include/asm/kgdb.h @@ -28,9 +28,12 @@ enum regnames { #define NUMREGBYTES ((GDB_CSR + 1) * 4) #else #define NUMREGBYTES ((GDB_Y + 1) * 8) + +struct pt_regs; +asmlinkage void kgdb_trap(unsigned long trap_level, struct pt_regs *regs); #endif -extern void arch_kgdb_breakpoint(void); +void arch_kgdb_breakpoint(void); #define BREAK_INSTR_SIZE 4 #define CACHE_FLUSH_IS_SAFE 1 diff --git a/arch/sparc/include/asm/kprobes.h b/arch/sparc/include/asm/kprobes.h index 5879d71afda..a145d798e11 100644 --- a/arch/sparc/include/asm/kprobes.h +++ b/arch/sparc/include/asm/kprobes.h @@ -43,7 +43,9 @@ struct kprobe_ctlblk { struct prev_kprobe prev_kprobe; }; -extern int kprobe_exceptions_notify(struct notifier_block *self, - unsigned long val, void *data); -extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr); +int kprobe_exceptions_notify(struct notifier_block *self, + unsigned long val, void *data); +int kprobe_fault_handler(struct pt_regs *regs, int trapnr); +asmlinkage void __kprobes kprobe_trap(unsigned long trap_level, + struct pt_regs *regs); #endif /* _SPARC64_KPROBES_H */ diff --git a/arch/sparc/include/asm/ldc.h b/arch/sparc/include/asm/ldc.h index bdb524a7b81..c8c67f621f4 100644 --- a/arch/sparc/include/asm/ldc.h +++ b/arch/sparc/include/asm/ldc.h @@ -4,9 +4,9 @@ #include <asm/hypervisor.h> extern int ldom_domaining_enabled; -extern void ldom_set_var(const char *var, const char *value); -extern void ldom_reboot(const char *boot_command); -extern void ldom_power_off(void); +void ldom_set_var(const char *var, const char *value); +void ldom_reboot(const char *boot_command); +void ldom_power_off(void); /* The event handler will be evoked when link state changes * or data becomes available on the receive side. @@ -51,30 +51,30 @@ struct ldc_channel_config { struct ldc_channel; /* Allocate state for a channel. */ -extern struct ldc_channel *ldc_alloc(unsigned long id, - const struct ldc_channel_config *cfgp, - void *event_arg); +struct ldc_channel *ldc_alloc(unsigned long id, + const struct ldc_channel_config *cfgp, + void *event_arg); /* Shut down and free state for a channel. */ -extern void ldc_free(struct ldc_channel *lp); +void ldc_free(struct ldc_channel *lp); /* Register TX and RX queues of the link with the hypervisor. */ -extern int ldc_bind(struct ldc_channel *lp, const char *name); +int ldc_bind(struct ldc_channel *lp, const char *name); /* For non-RAW protocols we need to complete a handshake before * communication can proceed. ldc_connect() does that, if the * handshake completes successfully, an LDC_EVENT_UP event will * be sent up to the driver. */ -extern int ldc_connect(struct ldc_channel *lp); -extern int ldc_disconnect(struct ldc_channel *lp); +int ldc_connect(struct ldc_channel *lp); +int ldc_disconnect(struct ldc_channel *lp); -extern int ldc_state(struct ldc_channel *lp); +int ldc_state(struct ldc_channel *lp); /* Read and write operations. Only valid when the link is up. */ -extern int ldc_write(struct ldc_channel *lp, const void *buf, - unsigned int size); -extern int ldc_read(struct ldc_channel *lp, void *buf, unsigned int size); +int ldc_write(struct ldc_channel *lp, const void *buf, + unsigned int size); +int ldc_read(struct ldc_channel *lp, void *buf, unsigned int size); #define LDC_MAP_SHADOW 0x01 #define LDC_MAP_DIRECT 0x02 @@ -92,22 +92,22 @@ struct ldc_trans_cookie { }; struct scatterlist; -extern int ldc_map_sg(struct ldc_channel *lp, - struct scatterlist *sg, int num_sg, - struct ldc_trans_cookie *cookies, int ncookies, - unsigned int map_perm); +int ldc_map_sg(struct ldc_channel *lp, + struct scatterlist *sg, int num_sg, + struct ldc_trans_cookie *cookies, int ncookies, + unsigned int map_perm); -extern int ldc_map_single(struct ldc_channel *lp, - void *buf, unsigned int len, - struct ldc_trans_cookie *cookies, int ncookies, - unsigned int map_perm); +int ldc_map_single(struct ldc_channel *lp, + void *buf, unsigned int len, + struct ldc_trans_cookie *cookies, int ncookies, + unsigned int map_perm); -extern void ldc_unmap(struct ldc_channel *lp, struct ldc_trans_cookie *cookies, - int ncookies); +void ldc_unmap(struct ldc_channel *lp, struct ldc_trans_cookie *cookies, + int ncookies); -extern int ldc_copy(struct ldc_channel *lp, int copy_dir, - void *buf, unsigned int len, unsigned long offset, - struct ldc_trans_cookie *cookies, int ncookies); +int ldc_copy(struct ldc_channel *lp, int copy_dir, + void *buf, unsigned int len, unsigned long offset, + struct ldc_trans_cookie *cookies, int ncookies); static inline int ldc_get_dring_entry(struct ldc_channel *lp, void *buf, unsigned int len, @@ -127,12 +127,12 @@ static inline int ldc_put_dring_entry(struct ldc_channel *lp, return ldc_copy(lp, LDC_COPY_OUT, buf, len, offset, cookies, ncookies); } -extern void *ldc_alloc_exp_dring(struct ldc_channel *lp, unsigned int len, - struct ldc_trans_cookie *cookies, - int *ncookies, unsigned int map_perm); +void *ldc_alloc_exp_dring(struct ldc_channel *lp, unsigned int len, + struct ldc_trans_cookie *cookies, + int *ncookies, unsigned int map_perm); -extern void ldc_free_exp_dring(struct ldc_channel *lp, void *buf, - unsigned int len, - struct ldc_trans_cookie *cookies, int ncookies); +void ldc_free_exp_dring(struct ldc_channel *lp, void *buf, + unsigned int len, + struct ldc_trans_cookie *cookies, int ncookies); #endif /* _SPARC64_LDC_H */ diff --git a/arch/sparc/include/asm/leon.h b/arch/sparc/include/asm/leon.h index 3375c629389..204771cd74a 100644 --- a/arch/sparc/include/asm/leon.h +++ b/arch/sparc/include/asm/leon.h @@ -82,9 +82,8 @@ static inline unsigned long leon_load_reg(unsigned long paddr) #define LEON_BYPASS_LOAD_PA(x) leon_load_reg((unsigned long)(x)) #define LEON_BYPASS_STORE_PA(x, v) leon_store_reg((unsigned long)(x), (unsigned long)(v)) -extern void leon_init(void); -extern void leon_switch_mm(void); -extern void leon_init_IRQ(void); +void leon_switch_mm(void); +void leon_init_IRQ(void); static inline unsigned long sparc_leon3_get_dcachecfg(void) { @@ -109,7 +108,7 @@ static inline int sparc_leon3_snooping_enabled(void) { u32 cctrl; __asm__ __volatile__("lda [%%g0] 2, %0\n\t" : "=r"(cctrl)); - return (cctrl >> 23) & 1; + return ((cctrl >> 23) & 1) && ((cctrl >> 17) & 1); }; static inline void sparc_leon3_disable_cache(void) @@ -136,7 +135,7 @@ static inline int sparc_leon3_cpuid(void) #ifdef CONFIG_SMP # define LEON3_IRQ_IPI_DEFAULT 13 -# define LEON3_IRQ_TICKER (leon3_ticker_irq) +# define LEON3_IRQ_TICKER (leon3_gptimer_irq) # define LEON3_IRQ_CROSS_CALL 15 #endif @@ -197,14 +196,14 @@ static inline int sparc_leon3_cpuid(void) #ifndef __ASSEMBLY__ struct vm_area_struct; -extern unsigned long leon_swprobe(unsigned long vaddr, unsigned long *paddr); -extern void leon_flush_icache_all(void); -extern void leon_flush_dcache_all(void); -extern void leon_flush_cache_all(void); -extern void leon_flush_tlb_all(void); +unsigned long leon_swprobe(unsigned long vaddr, unsigned long *paddr); +void leon_flush_icache_all(void); +void leon_flush_dcache_all(void); +void leon_flush_cache_all(void); +void leon_flush_tlb_all(void); extern int leon_flush_during_switch; -extern int leon_flush_needed(void); -extern void leon_flush_pcache_all(struct vm_area_struct *vma, unsigned long page); +int leon_flush_needed(void); +void leon_flush_pcache_all(struct vm_area_struct *vma, unsigned long page); /* struct that hold LEON3 cache configuration registers */ struct leon3_cacheregs { @@ -218,29 +217,29 @@ struct leon3_cacheregs { struct device_node; struct task_struct; -extern unsigned int leon_build_device_irq(unsigned int real_irq, - irq_flow_handler_t flow_handler, - const char *name, int do_ack); -extern void leon_update_virq_handling(unsigned int virq, - irq_flow_handler_t flow_handler, - const char *name, int do_ack); -extern void leon_init_timers(void); -extern void leon_trans_init(struct device_node *dp); -extern void leon_node_init(struct device_node *dp, struct device_node ***nextp); -extern void init_leon(void); -extern void poke_leonsparc(void); -extern void leon3_getCacheRegs(struct leon3_cacheregs *regs); +unsigned int leon_build_device_irq(unsigned int real_irq, + irq_flow_handler_t flow_handler, + const char *name, int do_ack); +void leon_update_virq_handling(unsigned int virq, + irq_flow_handler_t flow_handler, + const char *name, int do_ack); +void leon_init_timers(void); +void leon_trans_init(struct device_node *dp); +void leon_node_init(struct device_node *dp, struct device_node ***nextp); +void init_leon(void); +void poke_leonsparc(void); +void leon3_getCacheRegs(struct leon3_cacheregs *regs); extern int leon3_ticker_irq; #ifdef CONFIG_SMP -extern int leon_smp_nrcpus(void); -extern void leon_clear_profile_irq(int cpu); -extern void leon_smp_done(void); -extern void leon_boot_cpus(void); -extern int leon_boot_one_cpu(int i, struct task_struct *); +int leon_smp_nrcpus(void); +void leon_clear_profile_irq(int cpu); +void leon_smp_done(void); +void leon_boot_cpus(void); +int leon_boot_one_cpu(int i, struct task_struct *); void leon_init_smp(void); void leon_enable_irq_cpu(unsigned int irq_nr, unsigned int cpu); -extern irqreturn_t leon_percpu_timer_interrupt(int irq, void *unused); +irqreturn_t leon_percpu_timer_interrupt(int irq, void *unused); extern unsigned int smpleon_ipi[]; extern unsigned int linux_trap_ipi15_leon[]; diff --git a/arch/sparc/include/asm/leon_amba.h b/arch/sparc/include/asm/leon_amba.h index f3034eddf46..24ec48c3ff9 100644 --- a/arch/sparc/include/asm/leon_amba.h +++ b/arch/sparc/include/asm/leon_amba.h @@ -47,6 +47,7 @@ struct amba_prom_registers { #define LEON3_GPTIMER_LD 4 #define LEON3_GPTIMER_IRQEN 8 #define LEON3_GPTIMER_SEPIRQ 8 +#define LEON3_GPTIMER_TIMERS 0x7 #define LEON23_REG_TIMER_CONTROL_EN 0x00000001 /* 1 = enable counting */ /* 0 = hold scalar and counter */ diff --git a/arch/sparc/include/asm/leon_pci.h b/arch/sparc/include/asm/leon_pci.h index f48527ebdd8..049d067ed8b 100644 --- a/arch/sparc/include/asm/leon_pci.h +++ b/arch/sparc/include/asm/leon_pci.h @@ -12,10 +12,11 @@ struct leon_pci_info { struct pci_ops *ops; struct resource io_space; struct resource mem_space; + struct resource busn; int (*map_irq)(const struct pci_dev *dev, u8 slot, u8 pin); }; -extern void leon_pci_init(struct platform_device *ofdev, - struct leon_pci_info *info); +void leon_pci_init(struct platform_device *ofdev, + struct leon_pci_info *info); #endif /* _ASM_LEON_PCI_H_ */ diff --git a/arch/sparc/include/asm/linkage.h b/arch/sparc/include/asm/linkage.h deleted file mode 100644 index 291c2d01c44..00000000000 --- a/arch/sparc/include/asm/linkage.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __ASM_LINKAGE_H -#define __ASM_LINKAGE_H - -/* Nothing to see here... */ - -#endif diff --git a/arch/sparc/include/asm/mc146818rtc.h b/arch/sparc/include/asm/mc146818rtc.h index 67ed9e3a023..d8e72f37dc4 100644 --- a/arch/sparc/include/asm/mc146818rtc.h +++ b/arch/sparc/include/asm/mc146818rtc.h @@ -1,5 +1,10 @@ #ifndef ___ASM_SPARC_MC146818RTC_H #define ___ASM_SPARC_MC146818RTC_H + +#include <linux/spinlock.h> + +extern spinlock_t rtc_lock; + #if defined(__sparc__) && defined(__arch64__) #include <asm/mc146818rtc_64.h> #else diff --git a/arch/sparc/include/asm/mdesc.h b/arch/sparc/include/asm/mdesc.h index 9faa046713f..aebeb88f70d 100644 --- a/arch/sparc/include/asm/mdesc.h +++ b/arch/sparc/include/asm/mdesc.h @@ -12,13 +12,13 @@ struct mdesc_handle; * the first argument to all of the operational calls that work * on mdescs. */ -extern struct mdesc_handle *mdesc_grab(void); -extern void mdesc_release(struct mdesc_handle *); +struct mdesc_handle *mdesc_grab(void); +void mdesc_release(struct mdesc_handle *); #define MDESC_NODE_NULL (~(u64)0) -extern u64 mdesc_node_by_name(struct mdesc_handle *handle, - u64 from_node, const char *name); +u64 mdesc_node_by_name(struct mdesc_handle *handle, + u64 from_node, const char *name); #define mdesc_for_each_node_by_name(__hdl, __node, __name) \ for (__node = mdesc_node_by_name(__hdl, MDESC_NODE_NULL, __name); \ (__node) != MDESC_NODE_NULL; \ @@ -34,9 +34,9 @@ extern u64 mdesc_node_by_name(struct mdesc_handle *handle, * * These same rules apply to mdesc_node_name(). */ -extern const void *mdesc_get_property(struct mdesc_handle *handle, - u64 node, const char *name, int *lenp); -extern const char *mdesc_node_name(struct mdesc_handle *hp, u64 node); +const void *mdesc_get_property(struct mdesc_handle *handle, + u64 node, const char *name, int *lenp); +const char *mdesc_node_name(struct mdesc_handle *hp, u64 node); /* MD arc iteration, the standard sequence is: * @@ -50,16 +50,16 @@ extern const char *mdesc_node_name(struct mdesc_handle *hp, u64 node); #define MDESC_ARC_TYPE_FWD "fwd" #define MDESC_ARC_TYPE_BACK "back" -extern u64 mdesc_next_arc(struct mdesc_handle *handle, u64 from, - const char *arc_type); +u64 mdesc_next_arc(struct mdesc_handle *handle, u64 from, + const char *arc_type); #define mdesc_for_each_arc(__arc, __hdl, __node, __type) \ for (__arc = mdesc_next_arc(__hdl, __node, __type); \ (__arc) != MDESC_NODE_NULL; \ __arc = mdesc_next_arc(__hdl, __arc, __type)) -extern u64 mdesc_arc_target(struct mdesc_handle *hp, u64 arc); +u64 mdesc_arc_target(struct mdesc_handle *hp, u64 arc); -extern void mdesc_update(void); +void mdesc_update(void); struct mdesc_notifier_client { void (*add)(struct mdesc_handle *handle, u64 node); @@ -69,11 +69,12 @@ struct mdesc_notifier_client { struct mdesc_notifier_client *next; }; -extern void mdesc_register_notifier(struct mdesc_notifier_client *client); +void mdesc_register_notifier(struct mdesc_notifier_client *client); -extern void mdesc_fill_in_cpu_data(cpumask_t *mask); -extern void mdesc_populate_present_mask(cpumask_t *mask); +void mdesc_fill_in_cpu_data(cpumask_t *mask); +void mdesc_populate_present_mask(cpumask_t *mask); +void mdesc_get_page_sizes(cpumask_t *mask, unsigned long *pgsz_mask); -extern void sun4v_mdesc_init(void); +void sun4v_mdesc_init(void); #endif diff --git a/arch/sparc/include/asm/mman.h b/arch/sparc/include/asm/mman.h index c3029ad6619..59bb5938d85 100644 --- a/arch/sparc/include/asm/mman.h +++ b/arch/sparc/include/asm/mman.h @@ -1,33 +1,10 @@ #ifndef __SPARC_MMAN_H__ #define __SPARC_MMAN_H__ -#include <asm-generic/mman-common.h> +#include <uapi/asm/mman.h> -/* SunOS'ified... */ - -#define MAP_RENAME MAP_ANONYMOUS /* In SunOS terminology */ -#define MAP_NORESERVE 0x40 /* don't reserve swap pages */ -#define MAP_INHERIT 0x80 /* SunOS doesn't do this, but... */ -#define MAP_LOCKED 0x100 /* lock the mapping */ -#define _MAP_NEW 0x80000000 /* Binary compatibility is fun... */ - -#define MAP_GROWSDOWN 0x0200 /* stack-like segment */ -#define MAP_DENYWRITE 0x0800 /* ETXTBSY */ -#define MAP_EXECUTABLE 0x1000 /* mark it as an executable */ - -#define MCL_CURRENT 0x2000 /* lock all currently mapped pages */ -#define MCL_FUTURE 0x4000 /* lock all additions to address space */ - -#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */ -#define MAP_NONBLOCK 0x10000 /* do not block on IO */ -#define MAP_STACK 0x20000 /* give out an address that is best suited for process/thread stacks */ -#define MAP_HUGETLB 0x40000 /* create a huge page mapping */ - -#ifdef __KERNEL__ #ifndef __ASSEMBLY__ #define arch_mmap_check(addr,len,flags) sparc_mmap_check(addr,len) int sparc_mmap_check(unsigned long addr, unsigned long len); #endif -#endif - #endif /* __SPARC_MMAN_H__ */ diff --git a/arch/sparc/include/asm/mmu_64.h b/arch/sparc/include/asm/mmu_64.h index 9067dc50053..70067ce184b 100644 --- a/arch/sparc/include/asm/mmu_64.h +++ b/arch/sparc/include/asm/mmu_64.h @@ -30,22 +30,8 @@ #define CTX_PGSZ_MASK ((CTX_PGSZ_BITS << CTX_PGSZ0_SHIFT) | \ (CTX_PGSZ_BITS << CTX_PGSZ1_SHIFT)) -#if defined(CONFIG_SPARC64_PAGE_SIZE_8KB) #define CTX_PGSZ_BASE CTX_PGSZ_8KB -#elif defined(CONFIG_SPARC64_PAGE_SIZE_64KB) -#define CTX_PGSZ_BASE CTX_PGSZ_64KB -#else -#error No page size specified in kernel configuration -#endif - -#if defined(CONFIG_HUGETLB_PAGE_SIZE_4MB) -#define CTX_PGSZ_HUGE CTX_PGSZ_4MB -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K) -#define CTX_PGSZ_HUGE CTX_PGSZ_512KB -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_64K) -#define CTX_PGSZ_HUGE CTX_PGSZ_64KB -#endif - +#define CTX_PGSZ_HUGE CTX_PGSZ_4MB #define CTX_PGSZ_KERN CTX_PGSZ_4MB /* Thus, when running on UltraSPARC-III+ and later, we use the following @@ -81,9 +67,9 @@ struct tsb { unsigned long pte; } __attribute__((aligned(TSB_ENTRY_ALIGNMENT))); -extern void __tsb_insert(unsigned long ent, unsigned long tag, unsigned long pte); -extern void tsb_flush(unsigned long ent, unsigned long tag); -extern void tsb_init(struct tsb *tsb, unsigned long size); +void __tsb_insert(unsigned long ent, unsigned long tag, unsigned long pte); +void tsb_flush(unsigned long ent, unsigned long tag); +void tsb_init(struct tsb *tsb, unsigned long size); struct tsb_config { struct tsb *tsb; @@ -96,7 +82,7 @@ struct tsb_config { #define MM_TSB_BASE 0 -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) #define MM_TSB_HUGE 1 #define MM_NUM_TSBS 2 #else diff --git a/arch/sparc/include/asm/mmu_context_32.h b/arch/sparc/include/asm/mmu_context_32.h index 01456c90072..2df2a9be8f6 100644 --- a/arch/sparc/include/asm/mmu_context_32.h +++ b/arch/sparc/include/asm/mmu_context_32.h @@ -9,14 +9,12 @@ static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk) { } -/* - * Initialize a new mmu context. This is invoked when a new +/* Initialize a new mmu context. This is invoked when a new * address space instance (unique or shared) is instantiated. */ -#define init_new_context(tsk, mm) (((mm)->context = NO_CONTEXT), 0) +int init_new_context(struct task_struct *tsk, struct mm_struct *mm); -/* - * Destroy a dead context. This occurs when mmput drops the +/* Destroy a dead context. This occurs when mmput drops the * mm_users count to zero, the mmaps have been released, and * all the page tables have been flushed. Our job is to destroy * any remaining processor-specific state. diff --git a/arch/sparc/include/asm/mmu_context_64.h b/arch/sparc/include/asm/mmu_context_64.h index a97fd085ceb..b84be675e50 100644 --- a/arch/sparc/include/asm/mmu_context_64.h +++ b/arch/sparc/include/asm/mmu_context_64.h @@ -17,26 +17,26 @@ extern spinlock_t ctx_alloc_lock; extern unsigned long tlb_context_cache; extern unsigned long mmu_context_bmap[]; -extern void get_new_mmu_context(struct mm_struct *mm); +void get_new_mmu_context(struct mm_struct *mm); #ifdef CONFIG_SMP -extern void smp_new_mmu_context_version(void); +void smp_new_mmu_context_version(void); #else #define smp_new_mmu_context_version() do { } while (0) #endif -extern int init_new_context(struct task_struct *tsk, struct mm_struct *mm); -extern void destroy_context(struct mm_struct *mm); +int init_new_context(struct task_struct *tsk, struct mm_struct *mm); +void destroy_context(struct mm_struct *mm); -extern void __tsb_context_switch(unsigned long pgd_pa, - struct tsb_config *tsb_base, - struct tsb_config *tsb_huge, - unsigned long tsb_descr_pa); +void __tsb_context_switch(unsigned long pgd_pa, + struct tsb_config *tsb_base, + struct tsb_config *tsb_huge, + unsigned long tsb_descr_pa); static inline void tsb_context_switch(struct mm_struct *mm) { __tsb_context_switch(__pa(mm->pgd), &mm->context.tsb_block[0], -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) (mm->context.tsb_block[1].tsb ? &mm->context.tsb_block[1] : NULL) @@ -46,9 +46,11 @@ static inline void tsb_context_switch(struct mm_struct *mm) , __pa(&mm->context.tsb_descr[0])); } -extern void tsb_grow(struct mm_struct *mm, unsigned long tsb_index, unsigned long mm_rss); +void tsb_grow(struct mm_struct *mm, + unsigned long tsb_index, + unsigned long mm_rss); #ifdef CONFIG_SMP -extern void smp_tsb_sync(struct mm_struct *mm); +void smp_tsb_sync(struct mm_struct *mm); #else #define smp_tsb_sync(__mm) do { } while (0) #endif @@ -66,9 +68,9 @@ extern void smp_tsb_sync(struct mm_struct *mm); : "r" (CTX_HWBITS((__mm)->context)), \ "r" (SECONDARY_CONTEXT), "i" (ASI_DMMU), "i" (ASI_MMU)) -extern void __flush_tlb_mm(unsigned long, unsigned long); +void __flush_tlb_mm(unsigned long, unsigned long); -/* Switch the current MM context. Interrupts are disabled. */ +/* Switch the current MM context. */ static inline void switch_mm(struct mm_struct *old_mm, struct mm_struct *mm, struct task_struct *tsk) { unsigned long ctx_valid, flags; diff --git a/arch/sparc/include/asm/module.h b/arch/sparc/include/asm/module.h deleted file mode 100644 index ff8e02d8033..00000000000 --- a/arch/sparc/include/asm/module.h +++ /dev/null @@ -1,24 +0,0 @@ -#ifndef __SPARC_MODULE_H -#define __SPARC_MODULE_H -struct mod_arch_specific { }; - -/* - * Use some preprocessor magic to define the correct symbol - * for sparc32 and sparc64. - * Elf_Addr becomes Elf32_Addr for sparc32 and Elf64_Addr for sparc64 - */ -#define ___ELF(a, b, c) a##b##c -#define __ELF(a, b, c) ___ELF(a, b, c) -#define _Elf(t) __ELF(Elf, CONFIG_BITS, t) -#define _ELF(t) __ELF(ELF, CONFIG_BITS, t) - -#define Elf_Shdr _Elf(_Shdr) -#define Elf_Sym _Elf(_Sym) -#define Elf_Ehdr _Elf(_Ehdr) -#define Elf_Rela _Elf(_Rela) -#define Elf_Addr _Elf(_Addr) - -#define ELF_R_SYM _ELF(_R_SYM) -#define ELF_R_TYPE _ELF(_R_TYPE) - -#endif /* __SPARC_MODULE_H */ diff --git a/arch/sparc/include/asm/mutex.h b/arch/sparc/include/asm/mutex.h deleted file mode 100644 index 458c1f7fbc1..00000000000 --- a/arch/sparc/include/asm/mutex.h +++ /dev/null @@ -1,9 +0,0 @@ -/* - * Pull in the generic implementation for the mutex fastpath. - * - * TODO: implement optimized primitives instead, or leave the generic - * implementation in place, or pick the atomic_xchg() based generic - * implementation. (see asm-generic/mutex-xchg.h for details) - */ - -#include <asm-generic/mutex-dec.h> diff --git a/arch/sparc/include/asm/nmi.h b/arch/sparc/include/asm/nmi.h index 72e6500e7ab..26ad2b2607c 100644 --- a/arch/sparc/include/asm/nmi.h +++ b/arch/sparc/include/asm/nmi.h @@ -1,13 +1,13 @@ #ifndef __NMI_H #define __NMI_H -extern int __init nmi_init(void); -extern void perfctr_irq(int irq, struct pt_regs *regs); -extern void nmi_adjust_hz(unsigned int new_hz); +int __init nmi_init(void); +void perfctr_irq(int irq, struct pt_regs *regs); +void nmi_adjust_hz(unsigned int new_hz); extern atomic_t nmi_active; -extern void start_nmi_watchdog(void *unused); -extern void stop_nmi_watchdog(void *unused); +void start_nmi_watchdog(void *unused); +void stop_nmi_watchdog(void *unused); #endif /* __NMI_H */ diff --git a/arch/sparc/include/asm/oplib_32.h b/arch/sparc/include/asm/oplib_32.h index 27517879a6c..56a09b9d7b1 100644 --- a/arch/sparc/include/asm/oplib_32.h +++ b/arch/sparc/include/asm/oplib_32.h @@ -43,28 +43,28 @@ extern struct linux_nodeops *prom_nodeops; /* You must call prom_init() before using any of the library services, * preferably as early as possible. Pass it the romvec pointer. */ -extern void prom_init(struct linux_romvec *rom_ptr); +void prom_init(struct linux_romvec *rom_ptr); /* Boot argument acquisition, returns the boot command line string. */ -extern char *prom_getbootargs(void); +char *prom_getbootargs(void); /* Miscellaneous routines, don't really fit in any category per se. */ /* Reboot the machine with the command line passed. */ -extern void prom_reboot(char *boot_command); +void prom_reboot(char *boot_command); /* Evaluate the forth string passed. */ -extern void prom_feval(char *forth_string); +void prom_feval(char *forth_string); /* Enter the prom, with possibility of continuation with the 'go' * command in newer proms. */ -extern void prom_cmdline(void); +void prom_cmdline(void); /* Enter the prom, with no chance of continuation for the stand-alone * which calls this. */ -extern void __noreturn prom_halt(void); +void __noreturn prom_halt(void); /* Set the PROM 'sync' callback function to the passed function pointer. * When the user gives the 'sync' command at the prom prompt while the @@ -73,37 +73,37 @@ extern void __noreturn prom_halt(void); * XXX The arguments are different on V0 vs. V2->higher proms, grrr! XXX */ typedef void (*sync_func_t)(void); -extern void prom_setsync(sync_func_t func_ptr); +void prom_setsync(sync_func_t func_ptr); /* Acquire the IDPROM of the root node in the prom device tree. This * gets passed a buffer where you would like it stuffed. The return value * is the format type of this idprom or 0xff on error. */ -extern unsigned char prom_get_idprom(char *idp_buffer, int idpbuf_size); +unsigned char prom_get_idprom(char *idp_buffer, int idpbuf_size); /* Get the prom major version. */ -extern int prom_version(void); +int prom_version(void); /* Get the prom plugin revision. */ -extern int prom_getrev(void); +int prom_getrev(void); /* Get the prom firmware revision. */ -extern int prom_getprev(void); +int prom_getprev(void); /* Write a buffer of characters to the console. */ -extern void prom_console_write_buf(const char *buf, int len); +void prom_console_write_buf(const char *buf, int len); /* Prom's internal routines, don't use in kernel/boot code. */ -extern void prom_printf(const char *fmt, ...); -extern void prom_write(const char *buf, unsigned int len); +__printf(1, 2) void prom_printf(const char *fmt, ...); +void prom_write(const char *buf, unsigned int len); /* Multiprocessor operations... */ /* Start the CPU with the given device tree node, context table, and context * at the passed program counter. */ -extern int prom_startcpu(int cpunode, struct linux_prom_registers *context_table, - int context, char *program_counter); +int prom_startcpu(int cpunode, struct linux_prom_registers *context_table, + int context, char *program_counter); /* Initialize the memory lists based upon the prom version. */ void prom_meminit(void); @@ -111,65 +111,65 @@ void prom_meminit(void); /* PROM device tree traversal functions... */ /* Get the child node of the given node, or zero if no child exists. */ -extern phandle prom_getchild(phandle parent_node); +phandle prom_getchild(phandle parent_node); /* Get the next sibling node of the given node, or zero if no further * siblings exist. */ -extern phandle prom_getsibling(phandle node); +phandle prom_getsibling(phandle node); /* Get the length, at the passed node, of the given property type. * Returns -1 on error (ie. no such property at this node). */ -extern int prom_getproplen(phandle thisnode, const char *property); +int prom_getproplen(phandle thisnode, const char *property); /* Fetch the requested property using the given buffer. Returns * the number of bytes the prom put into your buffer or -1 on error. */ -extern int __must_check prom_getproperty(phandle thisnode, const char *property, - char *prop_buffer, int propbuf_size); +int __must_check prom_getproperty(phandle thisnode, const char *property, + char *prop_buffer, int propbuf_size); /* Acquire an integer property. */ -extern int prom_getint(phandle node, char *property); +int prom_getint(phandle node, char *property); /* Acquire an integer property, with a default value. */ -extern int prom_getintdefault(phandle node, char *property, int defval); +int prom_getintdefault(phandle node, char *property, int defval); /* Acquire a boolean property, 0=FALSE 1=TRUE. */ -extern int prom_getbool(phandle node, char *prop); +int prom_getbool(phandle node, char *prop); /* Acquire a string property, null string on error. */ -extern void prom_getstring(phandle node, char *prop, char *buf, int bufsize); +void prom_getstring(phandle node, char *prop, char *buf, int bufsize); /* Search all siblings starting at the passed node for "name" matching * the given string. Returns the node on success, zero on failure. */ -extern phandle prom_searchsiblings(phandle node_start, char *name); +phandle prom_searchsiblings(phandle node_start, char *name); /* Returns the next property after the passed property for the given * node. Returns null string on failure. */ -extern char *prom_nextprop(phandle node, char *prev_property, char *buffer); +char *prom_nextprop(phandle node, char *prev_property, char *buffer); /* Returns phandle of the path specified */ -extern phandle prom_finddevice(char *name); +phandle prom_finddevice(char *name); /* Set the indicated property at the given node with the passed value. * Returns the number of bytes of your value that the prom took. */ -extern int prom_setprop(phandle node, const char *prop_name, char *prop_value, - int value_size); +int prom_setprop(phandle node, const char *prop_name, char *prop_value, + int value_size); -extern phandle prom_inst2pkg(int); +phandle prom_inst2pkg(int); /* Dorking with Bus ranges... */ /* Apply promlib probes OBIO ranges to registers. */ -extern void prom_apply_obio_ranges(struct linux_prom_registers *obioregs, int nregs); +void prom_apply_obio_ranges(struct linux_prom_registers *obioregs, int nregs); /* Apply ranges of any prom node (and optionally parent node as well) to registers. */ -extern void prom_apply_generic_ranges(phandle node, phandle parent, - struct linux_prom_registers *sbusregs, int nregs); +void prom_apply_generic_ranges(phandle node, phandle parent, + struct linux_prom_registers *sbusregs, int nregs); void prom_ranges_init(void); diff --git a/arch/sparc/include/asm/oplib_64.h b/arch/sparc/include/asm/oplib_64.h index 97a90475c31..f34682430fc 100644 --- a/arch/sparc/include/asm/oplib_64.h +++ b/arch/sparc/include/asm/oplib_64.h @@ -62,100 +62,100 @@ struct linux_mem_p1275 { /* You must call prom_init() before using any of the library services, * preferably as early as possible. Pass it the romvec pointer. */ -extern void prom_init(void *cif_handler, void *cif_stack); +void prom_init(void *cif_handler, void *cif_stack); /* Boot argument acquisition, returns the boot command line string. */ -extern char *prom_getbootargs(void); +char *prom_getbootargs(void); /* Miscellaneous routines, don't really fit in any category per se. */ /* Reboot the machine with the command line passed. */ -extern void prom_reboot(const char *boot_command); +void prom_reboot(const char *boot_command); /* Evaluate the forth string passed. */ -extern void prom_feval(const char *forth_string); +void prom_feval(const char *forth_string); /* Enter the prom, with possibility of continuation with the 'go' * command in newer proms. */ -extern void prom_cmdline(void); +void prom_cmdline(void); /* Enter the prom, with no chance of continuation for the stand-alone * which calls this. */ -extern void prom_halt(void) __attribute__ ((noreturn)); +void prom_halt(void) __attribute__ ((noreturn)); /* Halt and power-off the machine. */ -extern void prom_halt_power_off(void) __attribute__ ((noreturn)); +void prom_halt_power_off(void) __attribute__ ((noreturn)); /* Acquire the IDPROM of the root node in the prom device tree. This * gets passed a buffer where you would like it stuffed. The return value * is the format type of this idprom or 0xff on error. */ -extern unsigned char prom_get_idprom(char *idp_buffer, int idpbuf_size); +unsigned char prom_get_idprom(char *idp_buffer, int idpbuf_size); /* Write a buffer of characters to the console. */ -extern void prom_console_write_buf(const char *buf, int len); +void prom_console_write_buf(const char *buf, int len); /* Prom's internal routines, don't use in kernel/boot code. */ -extern void prom_printf(const char *fmt, ...); -extern void prom_write(const char *buf, unsigned int len); +__printf(1, 2) void prom_printf(const char *fmt, ...); +void prom_write(const char *buf, unsigned int len); /* Multiprocessor operations... */ #ifdef CONFIG_SMP /* Start the CPU with the given device tree node at the passed program * counter with the given arg passed in via register %o0. */ -extern void prom_startcpu(int cpunode, unsigned long pc, unsigned long arg); +void prom_startcpu(int cpunode, unsigned long pc, unsigned long arg); /* Start the CPU with the given cpu ID at the passed program * counter with the given arg passed in via register %o0. */ -extern void prom_startcpu_cpuid(int cpuid, unsigned long pc, unsigned long arg); +void prom_startcpu_cpuid(int cpuid, unsigned long pc, unsigned long arg); /* Stop the CPU with the given cpu ID. */ -extern void prom_stopcpu_cpuid(int cpuid); +void prom_stopcpu_cpuid(int cpuid); /* Stop the current CPU. */ -extern void prom_stopself(void); +void prom_stopself(void); /* Idle the current CPU. */ -extern void prom_idleself(void); +void prom_idleself(void); /* Resume the CPU with the passed device tree node. */ -extern void prom_resumecpu(int cpunode); +void prom_resumecpu(int cpunode); #endif /* Power management interfaces. */ /* Put the current CPU to sleep. */ -extern void prom_sleepself(void); +void prom_sleepself(void); /* Put the entire system to sleep. */ -extern int prom_sleepsystem(void); +int prom_sleepsystem(void); /* Initiate a wakeup event. */ -extern int prom_wakeupsystem(void); +int prom_wakeupsystem(void); /* MMU and memory related OBP interfaces. */ /* Get unique string identifying SIMM at given physical address. */ -extern int prom_getunumber(int syndrome_code, - unsigned long phys_addr, - char *buf, int buflen); +int prom_getunumber(int syndrome_code, + unsigned long phys_addr, + char *buf, int buflen); /* Retain physical memory to the caller across soft resets. */ -extern int prom_retain(const char *name, unsigned long size, - unsigned long align, unsigned long *paddr); +int prom_retain(const char *name, unsigned long size, + unsigned long align, unsigned long *paddr); /* Load explicit I/D TLB entries into the calling processor. */ -extern long prom_itlb_load(unsigned long index, - unsigned long tte_data, - unsigned long vaddr); +long prom_itlb_load(unsigned long index, + unsigned long tte_data, + unsigned long vaddr); -extern long prom_dtlb_load(unsigned long index, - unsigned long tte_data, - unsigned long vaddr); +long prom_dtlb_load(unsigned long index, + unsigned long tte_data, + unsigned long vaddr); /* Map/Unmap client program address ranges. First the format of * the mapping mode argument. @@ -170,81 +170,81 @@ extern long prom_dtlb_load(unsigned long index, #define PROM_MAP_IE 0x0100 /* Invert-Endianness */ #define PROM_MAP_DEFAULT (PROM_MAP_WRITE | PROM_MAP_READ | PROM_MAP_EXEC | PROM_MAP_CACHED) -extern int prom_map(int mode, unsigned long size, - unsigned long vaddr, unsigned long paddr); -extern void prom_unmap(unsigned long size, unsigned long vaddr); +int prom_map(int mode, unsigned long size, + unsigned long vaddr, unsigned long paddr); +void prom_unmap(unsigned long size, unsigned long vaddr); /* PROM device tree traversal functions... */ /* Get the child node of the given node, or zero if no child exists. */ -extern phandle prom_getchild(phandle parent_node); +phandle prom_getchild(phandle parent_node); /* Get the next sibling node of the given node, or zero if no further * siblings exist. */ -extern phandle prom_getsibling(phandle node); +phandle prom_getsibling(phandle node); /* Get the length, at the passed node, of the given property type. * Returns -1 on error (ie. no such property at this node). */ -extern int prom_getproplen(phandle thisnode, const char *property); +int prom_getproplen(phandle thisnode, const char *property); /* Fetch the requested property using the given buffer. Returns * the number of bytes the prom put into your buffer or -1 on error. */ -extern int prom_getproperty(phandle thisnode, const char *property, - char *prop_buffer, int propbuf_size); +int prom_getproperty(phandle thisnode, const char *property, + char *prop_buffer, int propbuf_size); /* Acquire an integer property. */ -extern int prom_getint(phandle node, const char *property); +int prom_getint(phandle node, const char *property); /* Acquire an integer property, with a default value. */ -extern int prom_getintdefault(phandle node, const char *property, int defval); +int prom_getintdefault(phandle node, const char *property, int defval); /* Acquire a boolean property, 0=FALSE 1=TRUE. */ -extern int prom_getbool(phandle node, const char *prop); +int prom_getbool(phandle node, const char *prop); /* Acquire a string property, null string on error. */ -extern void prom_getstring(phandle node, const char *prop, char *buf, - int bufsize); +void prom_getstring(phandle node, const char *prop, char *buf, + int bufsize); /* Does the passed node have the given "name"? YES=1 NO=0 */ -extern int prom_nodematch(phandle thisnode, const char *name); +int prom_nodematch(phandle thisnode, const char *name); /* Search all siblings starting at the passed node for "name" matching * the given string. Returns the node on success, zero on failure. */ -extern phandle prom_searchsiblings(phandle node_start, const char *name); +phandle prom_searchsiblings(phandle node_start, const char *name); /* Return the first property type, as a string, for the given node. * Returns a null string on error. Buffer should be at least 32B long. */ -extern char *prom_firstprop(phandle node, char *buffer); +char *prom_firstprop(phandle node, char *buffer); /* Returns the next property after the passed property for the given * node. Returns null string on failure. Buffer should be at least 32B long. */ -extern char *prom_nextprop(phandle node, const char *prev_property, char *buf); +char *prom_nextprop(phandle node, const char *prev_property, char *buf); /* Returns 1 if the specified node has given property. */ -extern int prom_node_has_property(phandle node, const char *property); +int prom_node_has_property(phandle node, const char *property); /* Returns phandle of the path specified */ -extern phandle prom_finddevice(const char *name); +phandle prom_finddevice(const char *name); /* Set the indicated property at the given node with the passed value. * Returns the number of bytes of your value that the prom took. */ -extern int prom_setprop(phandle node, const char *prop_name, char *prop_value, - int value_size); +int prom_setprop(phandle node, const char *prop_name, char *prop_value, + int value_size); -extern phandle prom_inst2pkg(int); -extern void prom_sun4v_guest_soft_state(void); +phandle prom_inst2pkg(int); +void prom_sun4v_guest_soft_state(void); -extern int prom_ihandle2path(int handle, char *buffer, int bufsize); +int prom_ihandle2path(int handle, char *buffer, int bufsize); /* Client interface level routines. */ -extern void p1275_cmd_direct(unsigned long *); +void p1275_cmd_direct(unsigned long *); #endif /* !(__SPARC64_OPLIB_H) */ diff --git a/arch/sparc/include/asm/page.h b/arch/sparc/include/asm/page.h index f21de034902..1be2fdec626 100644 --- a/arch/sparc/include/asm/page.h +++ b/arch/sparc/include/asm/page.h @@ -1,5 +1,8 @@ #ifndef ___ASM_SPARC_PAGE_H #define ___ASM_SPARC_PAGE_H + +#define page_to_phys(page) (page_to_pfn(page) << PAGE_SHIFT) + #if defined(__sparc__) && defined(__arch64__) #include <asm/page_64.h> #else diff --git a/arch/sparc/include/asm/page_32.h b/arch/sparc/include/asm/page_32.h index fab78a308eb..f82a1f36b65 100644 --- a/arch/sparc/include/asm/page_32.h +++ b/arch/sparc/include/asm/page_32.h @@ -107,8 +107,7 @@ typedef unsigned long iopgprot_t; typedef struct page *pgtable_t; -extern unsigned long sparc_unmapped_base; -#define TASK_UNMAPPED_BASE sparc_unmapped_base +#define TASK_UNMAPPED_BASE 0x50000000 #else /* !(__ASSEMBLY__) */ diff --git a/arch/sparc/include/asm/page_64.h b/arch/sparc/include/asm/page_64.h index f0d09b40103..bf109984a03 100644 --- a/arch/sparc/include/asm/page_64.h +++ b/arch/sparc/include/asm/page_64.h @@ -3,13 +3,7 @@ #include <linux/const.h> -#if defined(CONFIG_SPARC64_PAGE_SIZE_8KB) #define PAGE_SHIFT 13 -#elif defined(CONFIG_SPARC64_PAGE_SIZE_64KB) -#define PAGE_SHIFT 16 -#else -#error No page size specified in kernel configuration -#endif #define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) @@ -21,15 +15,12 @@ #define DCACHE_ALIASING_POSSIBLE #endif -#if defined(CONFIG_HUGETLB_PAGE_SIZE_4MB) -#define HPAGE_SHIFT 22 -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K) -#define HPAGE_SHIFT 19 -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_64K) -#define HPAGE_SHIFT 16 -#endif +#define HPAGE_SHIFT 23 +#define REAL_HPAGE_SHIFT 22 + +#define REAL_HPAGE_SIZE (_AC(1,UL) << REAL_HPAGE_SHIFT) -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) #define HPAGE_SIZE (_AC(1,UL) << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE - 1UL)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) @@ -38,14 +29,19 @@ #ifndef __ASSEMBLY__ +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) +struct pt_regs; +void hugetlb_setup(struct pt_regs *regs); +#endif + #define WANT_PAGE_VIRTUAL -extern void _clear_page(void *page); +void _clear_page(void *page); #define clear_page(X) _clear_page((void *)(X)) struct page; -extern void clear_user_page(void *addr, unsigned long vaddr, struct page *page); +void clear_user_page(void *addr, unsigned long vaddr, struct page *page); #define copy_page(X,Y) memcpy((void *)(X), (void *)(Y), PAGE_SIZE) -extern void copy_user_page(void *to, void *from, unsigned long vaddr, struct page *topage); +void copy_user_page(void *to, void *from, unsigned long vaddr, struct page *topage); /* Unlike sparc32, sparc64's parameter passing API is more * sane in that structures which as small enough are passed @@ -60,8 +56,8 @@ extern void copy_user_page(void *to, void *from, unsigned long vaddr, struct pag /* These are used to make use of C type-checking.. */ typedef struct { unsigned long pte; } pte_t; typedef struct { unsigned long iopte; } iopte_t; -typedef struct { unsigned int pmd; } pmd_t; -typedef struct { unsigned int pgd; } pgd_t; +typedef struct { unsigned long pmd; } pmd_t; +typedef struct { unsigned long pgd; } pgd_t; typedef struct { unsigned long pgprot; } pgprot_t; #define pte_val(x) ((x).pte) @@ -80,8 +76,8 @@ typedef struct { unsigned long pgprot; } pgprot_t; /* .. while these make it easier on the compiler */ typedef unsigned long pte_t; typedef unsigned long iopte_t; -typedef unsigned int pmd_t; -typedef unsigned int pgd_t; +typedef unsigned long pmd_t; +typedef unsigned long pgd_t; typedef unsigned long pgprot_t; #define pte_val(x) (x) @@ -98,20 +94,46 @@ typedef unsigned long pgprot_t; #endif /* (STRICT_MM_TYPECHECKS) */ -typedef struct page *pgtable_t; +typedef pte_t *pgtable_t; + +/* These two values define the virtual address space range in which we + * must forbid 64-bit user processes from making mappings. It used to + * represent precisely the virtual address space hole present in most + * early sparc64 chips including UltraSPARC-I. But now it also is + * further constrained by the limits of our page tables, which is + * 43-bits of virtual address. + */ +#define SPARC64_VA_HOLE_TOP _AC(0xfffffc0000000000,UL) +#define SPARC64_VA_HOLE_BOTTOM _AC(0x0000040000000000,UL) + +/* The next two defines specify the actual exclusion region we + * enforce, wherein we use a 4GB red zone on each side of the VA hole. + */ +#define VA_EXCLUDE_START (SPARC64_VA_HOLE_BOTTOM - (1UL << 32UL)) +#define VA_EXCLUDE_END (SPARC64_VA_HOLE_TOP + (1UL << 32UL)) #define TASK_UNMAPPED_BASE (test_thread_flag(TIF_32BIT) ? \ - (_AC(0x0000000070000000,UL)) : \ - (_AC(0xfffff80000000000,UL) + (1UL << 32UL))) + _AC(0x0000000070000000,UL) : \ + VA_EXCLUDE_END) #include <asm-generic/memory_model.h> +#define PAGE_OFFSET_BY_BITS(X) (-(_AC(1,UL) << (X))) +extern unsigned long PAGE_OFFSET; + #endif /* !(__ASSEMBLY__) */ -/* We used to stick this into a hard-coded global register (%g4) - * but that does not make sense anymore. +/* The maximum number of physical memory address bits we support, this + * is used to size various tables used to manage kernel TLB misses and + * also the sparsemem code. + */ +#define MAX_PHYS_ADDRESS_BITS 47 + +/* These two shift counts are used when indexing sparc64_valid_addr_bitmap + * and kpte_linear_bitmap. */ -#define PAGE_OFFSET _AC(0xFFFFF80000000000,UL) +#define ILOG2_4MB 22 +#define ILOG2_256MB 28 #ifndef __ASSEMBLY__ diff --git a/arch/sparc/include/asm/parport.h b/arch/sparc/include/asm/parport.h index cb33608cc68..c55291e5b83 100644 --- a/arch/sparc/include/asm/parport.h +++ b/arch/sparc/include/asm/parport.h @@ -103,7 +103,7 @@ static inline unsigned int get_dma_residue(unsigned int dmanr) return ebus_dma_residue(&sparc_ebus_dmas[dmanr].info); } -static int __devinit ecpp_probe(struct platform_device *op) +static int ecpp_probe(struct platform_device *op) { unsigned long base = op->resource[0].start; unsigned long config = op->resource[1].start; @@ -192,7 +192,7 @@ out_err: return err; } -static int __devexit ecpp_remove(struct platform_device *op) +static int ecpp_remove(struct platform_device *op) { struct parport *p = dev_get_drvdata(&op->dev); int slot = p->dma; @@ -242,7 +242,7 @@ static struct platform_driver ecpp_driver = { .of_match_table = ecpp_match, }, .probe = ecpp_probe, - .remove = __devexit_p(ecpp_remove), + .remove = ecpp_remove, }; static int parport_pc_find_nonpci_ports(int autoirq, int autodma) diff --git a/arch/sparc/include/asm/pci_32.h b/arch/sparc/include/asm/pci_32.h index dc503297481..53e9b4987db 100644 --- a/arch/sparc/include/asm/pci_32.h +++ b/arch/sparc/include/asm/pci_32.h @@ -16,11 +16,6 @@ #define PCI_IRQ_NONE 0xffffffff -static inline void pcibios_penalize_isa_irq(int irq, int active) -{ - /* We don't do dynamic PCI IRQ allocation */ -} - /* Dynamic DMA mapping stuff. */ #define PCI_DMA_BUS_IS_PHYS (0) diff --git a/arch/sparc/include/asm/pci_64.h b/arch/sparc/include/asm/pci_64.h index 1633b718d3b..bd00a622616 100644 --- a/arch/sparc/include/asm/pci_64.h +++ b/arch/sparc/include/asm/pci_64.h @@ -16,11 +16,6 @@ #define PCI_IRQ_NONE 0xffffffff -static inline void pcibios_penalize_isa_irq(int irq, int active) -{ - /* We don't do dynamic PCI IRQ allocation */ -} - /* The PCI address space does not equal the physical memory * address space. The networking and block device layers use * this boolean for bounce buffer decisions. @@ -57,7 +52,7 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev, /* Return the index of the PCI controller for device PDEV. */ -extern int pci_domain_nr(struct pci_bus *bus); +int pci_domain_nr(struct pci_bus *bus); static inline int pci_proc_domain(struct pci_bus *bus) { return 1; @@ -69,9 +64,9 @@ static inline int pci_proc_domain(struct pci_bus *bus) #define HAVE_ARCH_PCI_GET_UNMAPPED_AREA #define get_pci_unmapped_area get_fb_unmapped_area -extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, - enum pci_mmap_state mmap_state, - int write_combine); +int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, + enum pci_mmap_state mmap_state, + int write_combine); static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel) { @@ -79,9 +74,9 @@ static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel) } #define HAVE_ARCH_PCI_RESOURCE_TO_USER -extern void pci_resource_to_user(const struct pci_dev *dev, int bar, - const struct resource *rsrc, - resource_size_t *start, resource_size_t *end); +void pci_resource_to_user(const struct pci_dev *dev, int bar, + const struct resource *rsrc, + resource_size_t *start, resource_size_t *end); #endif /* __KERNEL__ */ #endif /* __SPARC64_PCI_H */ diff --git a/arch/sparc/include/asm/pcic.h b/arch/sparc/include/asm/pcic.h index 6676cbcc8b6..f4170679259 100644 --- a/arch/sparc/include/asm/pcic.h +++ b/arch/sparc/include/asm/pcic.h @@ -30,10 +30,10 @@ struct linux_pcic { }; #ifdef CONFIG_PCIC_PCI -extern int pcic_present(void); -extern int pcic_probe(void); -extern void pci_time_init(void); -extern void sun4m_pci_init_IRQ(void); +int pcic_present(void); +int pcic_probe(void); +void pci_time_init(void); +void sun4m_pci_init_IRQ(void); #else static inline int pcic_present(void) { return 0; } static inline int pcic_probe(void) { return 0; } diff --git a/arch/sparc/include/asm/pcr.h b/arch/sparc/include/asm/pcr.h index 288d7beba05..cdf800c3326 100644 --- a/arch/sparc/include/asm/pcr.h +++ b/arch/sparc/include/asm/pcr.h @@ -2,13 +2,18 @@ #define __PCR_H struct pcr_ops { - u64 (*read)(void); - void (*write)(u64); + u64 (*read_pcr)(unsigned long); + void (*write_pcr)(unsigned long, u64); + u64 (*read_pic)(unsigned long); + void (*write_pic)(unsigned long, u64); + u64 (*nmi_picl_value)(unsigned int nmi_hz); + u64 pcr_nmi_enable; + u64 pcr_nmi_disable; }; extern const struct pcr_ops *pcr_ops; -extern void deferred_pcr_work_irq(int irq, struct pt_regs *regs); -extern void schedule_deferred_pcr_work(void); +void deferred_pcr_work_irq(int irq, struct pt_regs *regs); +void schedule_deferred_pcr_work(void); #define PCR_PIC_PRIV 0x00000001 /* PIC access is privileged */ #define PCR_STRACE 0x00000002 /* Trace supervisor events */ @@ -27,22 +32,19 @@ extern void schedule_deferred_pcr_work(void); #define PCR_N2_SL1_SHIFT 27 #define PCR_N2_OV1 0x80000000 -extern unsigned int picl_shift; - -/* In order to commonize as much of the implementation as - * possible, we use PICH as our counter. Mostly this is - * to accommodate Niagara-1 which can only count insn cycles - * in PICH. - */ -static inline u64 picl_value(unsigned int nmi_hz) -{ - u32 delta = local_cpu_data().clock_tick / (nmi_hz << picl_shift); - - return ((u64)((0 - delta) & 0xffffffff)) << 32; -} - -extern u64 pcr_enable; - -extern int pcr_arch_init(void); +#define PCR_N4_OV 0x00000001 /* PIC overflow */ +#define PCR_N4_TOE 0x00000002 /* Trap On Event */ +#define PCR_N4_UTRACE 0x00000004 /* Trace user events */ +#define PCR_N4_STRACE 0x00000008 /* Trace supervisor events */ +#define PCR_N4_HTRACE 0x00000010 /* Trace hypervisor events */ +#define PCR_N4_MASK 0x000007e0 /* Event mask */ +#define PCR_N4_MASK_SHIFT 5 +#define PCR_N4_SL 0x0000f800 /* Event Select */ +#define PCR_N4_SL_SHIFT 11 +#define PCR_N4_PICNPT 0x00010000 /* PIC non-privileged trap */ +#define PCR_N4_PICNHT 0x00020000 /* PIC non-hypervisor trap */ +#define PCR_N4_NTC 0x00040000 /* Next-To-Commit wrap */ + +int pcr_arch_init(void); #endif /* __PCR_H */ diff --git a/arch/sparc/include/asm/pgalloc_32.h b/arch/sparc/include/asm/pgalloc_32.h index e5b169b46d2..a3890da9442 100644 --- a/arch/sparc/include/asm/pgalloc_32.h +++ b/arch/sparc/include/asm/pgalloc_32.h @@ -11,28 +11,17 @@ struct page; -extern struct pgtable_cache_struct { - unsigned long *pgd_cache; - unsigned long *pte_cache; - unsigned long pgtable_cache_sz; - unsigned long pgd_cache_sz; -} pgt_quicklists; - -unsigned long srmmu_get_nocache(int size, int align); -void srmmu_free_nocache(unsigned long vaddr, int size); - -#define pgd_quicklist (pgt_quicklists.pgd_cache) -#define pmd_quicklist ((unsigned long *)0) -#define pte_quicklist (pgt_quicklists.pte_cache) -#define pgtable_cache_size (pgt_quicklists.pgtable_cache_sz) -#define pgd_cache_size (pgt_quicklists.pgd_cache_sz) +void *srmmu_get_nocache(int size, int align); +void srmmu_free_nocache(void *addr, int size); + +extern struct resource sparc_iomap; #define check_pgt_cache() do { } while (0) pgd_t *get_pgd_fast(void); static inline void free_pgd_fast(pgd_t *pgd) { - srmmu_free_nocache((unsigned long)pgd, SRMMU_PGD_TABLE_SIZE); + srmmu_free_nocache(pgd, SRMMU_PGD_TABLE_SIZE); } #define pgd_free(mm, pgd) free_pgd_fast(pgd) @@ -50,13 +39,13 @@ static inline void pgd_set(pgd_t * pgdp, pmd_t * pmdp) static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long address) { - return (pmd_t *)srmmu_get_nocache(SRMMU_PMD_TABLE_SIZE, - SRMMU_PMD_TABLE_SIZE); + return srmmu_get_nocache(SRMMU_PMD_TABLE_SIZE, + SRMMU_PMD_TABLE_SIZE); } static inline void free_pmd_fast(pmd_t * pmd) { - srmmu_free_nocache((unsigned long)pmd, SRMMU_PMD_TABLE_SIZE); + srmmu_free_nocache(pmd, SRMMU_PMD_TABLE_SIZE); } #define pmd_free(mm, pmd) free_pmd_fast(pmd) @@ -73,13 +62,13 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address); static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { - return (pte_t *)srmmu_get_nocache(PTE_SIZE, PTE_SIZE); + return srmmu_get_nocache(PTE_SIZE, PTE_SIZE); } static inline void free_pte_fast(pte_t *pte) { - srmmu_free_nocache((unsigned long)pte, PTE_SIZE); + srmmu_free_nocache(pte, PTE_SIZE); } #define pte_free_kernel(mm, pte) free_pte_fast(pte) diff --git a/arch/sparc/include/asm/pgalloc_64.h b/arch/sparc/include/asm/pgalloc_64.h index 40b2d7a7023..39a7ac49b00 100644 --- a/arch/sparc/include/asm/pgalloc_64.h +++ b/arch/sparc/include/asm/pgalloc_64.h @@ -38,56 +38,25 @@ static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) kmem_cache_free(pgtable_cache, pmd); } -static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, - unsigned long address) -{ - return (pte_t *)__get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO); -} - -static inline pgtable_t pte_alloc_one(struct mm_struct *mm, - unsigned long address) -{ - struct page *page; - pte_t *pte; - - pte = pte_alloc_one_kernel(mm, address); - if (!pte) - return NULL; - page = virt_to_page(pte); - pgtable_page_ctor(page); - return page; -} - -static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) -{ - free_page((unsigned long)pte); -} - -static inline void pte_free(struct mm_struct *mm, pgtable_t ptepage) -{ - pgtable_page_dtor(ptepage); - __free_page(ptepage); -} +pte_t *pte_alloc_one_kernel(struct mm_struct *mm, + unsigned long address); +pgtable_t pte_alloc_one(struct mm_struct *mm, + unsigned long address); +void pte_free_kernel(struct mm_struct *mm, pte_t *pte); +void pte_free(struct mm_struct *mm, pgtable_t ptepage); -#define pmd_populate_kernel(MM, PMD, PTE) pmd_set(PMD, PTE) -#define pmd_populate(MM,PMD,PTE_PAGE) \ - pmd_populate_kernel(MM,PMD,page_address(PTE_PAGE)) -#define pmd_pgtable(pmd) pmd_page(pmd) +#define pmd_populate_kernel(MM, PMD, PTE) pmd_set(MM, PMD, PTE) +#define pmd_populate(MM, PMD, PTE) pmd_set(MM, PMD, PTE) +#define pmd_pgtable(PMD) ((pte_t *)__pmd_page(PMD)) #define check_pgt_cache() do { } while (0) -static inline void pgtable_free(void *table, bool is_page) -{ - if (is_page) - free_page((unsigned long)table); - else - kmem_cache_free(pgtable_cache, table); -} +void pgtable_free(void *table, bool is_page); #ifdef CONFIG_SMP struct mmu_gather; -extern void tlb_remove_table(struct mmu_gather *, void *); +void tlb_remove_table(struct mmu_gather *, void *); static inline void pgtable_free_tlb(struct mmu_gather *tlb, void *table, bool is_page) { @@ -113,11 +82,10 @@ static inline void pgtable_free_tlb(struct mmu_gather *tlb, void *table, bool is } #endif /* !CONFIG_SMP */ -static inline void __pte_free_tlb(struct mmu_gather *tlb, struct page *ptepage, +static inline void __pte_free_tlb(struct mmu_gather *tlb, pte_t *pte, unsigned long address) { - pgtable_page_dtor(ptepage); - pgtable_free_tlb(tlb, page_address(ptepage), true); + pgtable_free_tlb(tlb, pte, true); } #define __pmd_free_tlb(tlb, pmd, addr) \ diff --git a/arch/sparc/include/asm/pgtable_32.h b/arch/sparc/include/asm/pgtable_32.h index cbbbed5cb3a..b9b91ae19fe 100644 --- a/arch/sparc/include/asm/pgtable_32.h +++ b/arch/sparc/include/asm/pgtable_32.h @@ -25,8 +25,9 @@ struct vm_area_struct; struct page; -extern void load_mmu(void); -extern unsigned long calc_highpages(void); +void load_mmu(void); +unsigned long calc_highpages(void); +unsigned long __init bootmem_init(unsigned long *pages_avail); #define pte_ERROR(e) __builtin_trap() #define pmd_ERROR(e) __builtin_trap() @@ -52,10 +53,11 @@ extern unsigned long calc_highpages(void); #define PAGE_READONLY SRMMU_PAGE_RDONLY #define PAGE_KERNEL SRMMU_PAGE_KERNEL -/* Top-level page directory */ -extern pgd_t swapper_pg_dir[1024]; +/* Top-level page directory - dummy used by init-mm. + * srmmu.c will assign the real one (which is dynamically sized) */ +#define swapper_pg_dir NULL -extern void paging_init(void); +void paging_init(void); extern unsigned long ptr_in_current_pgd; @@ -78,8 +80,6 @@ extern unsigned long ptr_in_current_pgd; #define __S110 PAGE_SHARED #define __S111 PAGE_SHARED -extern int num_contexts; - /* First physical page can be anywhere, the following is needed so that * va-->pa and vice versa conversions work properly without performance * hit for all __pa()/__va() operations. @@ -88,18 +88,11 @@ extern unsigned long phys_base; extern unsigned long pfn_base; /* - * BAD_PAGETABLE is used when we need a bogus page-table, while - * BAD_PAGE is used for a bogus page. - * * ZERO_PAGE is a global shared page that is always zero: used * for zero-mapped memory areas etc.. */ -extern pte_t * __bad_pagetable(void); -extern pte_t __bad_page(void); extern unsigned long empty_zero_page; -#define BAD_PAGETABLE __bad_pagetable() -#define BAD_PAGE __bad_page() #define ZERO_PAGE(vaddr) (virt_to_page(&empty_zero_page)) /* @@ -398,36 +391,6 @@ static inline pte_t pgoff_to_pte(unsigned long pgoff) */ #define PTE_FILE_MAX_BITS 24 -/* - */ -struct ctx_list { - struct ctx_list *next; - struct ctx_list *prev; - unsigned int ctx_number; - struct mm_struct *ctx_mm; -}; - -extern struct ctx_list *ctx_list_pool; /* Dynamically allocated */ -extern struct ctx_list ctx_free; /* Head of free list */ -extern struct ctx_list ctx_used; /* Head of used contexts list */ - -#define NO_CONTEXT -1 - -static inline void remove_from_ctx_list(struct ctx_list *entry) -{ - entry->next->prev = entry->prev; - entry->prev->next = entry->next; -} - -static inline void add_to_ctx_list(struct ctx_list *head, struct ctx_list *entry) -{ - entry->next = head; - (entry->prev = head->prev)->next = entry; - head->prev = entry; -} -#define add_to_free_ctxlist(entry) add_to_ctx_list(&ctx_free, entry) -#define add_to_used_ctxlist(entry) add_to_ctx_list(&ctx_used, entry) - static inline unsigned long __get_phys (unsigned long addr) { @@ -466,8 +429,8 @@ extern unsigned long *sparc_valid_addr_bitmap; #define GET_IOSPACE(pfn) (pfn >> (BITS_PER_LONG - 4)) #define GET_PFN(pfn) (pfn & 0x0fffffffUL) -extern int remap_pfn_range(struct vm_area_struct *, unsigned long, unsigned long, - unsigned long, pgprot_t); +int remap_pfn_range(struct vm_area_struct *, unsigned long, unsigned long, + unsigned long, pgprot_t); static inline int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from, unsigned long pfn, @@ -481,6 +444,7 @@ static inline int io_remap_pfn_range(struct vm_area_struct *vma, return remap_pfn_range(vma, from, phys_base >> PAGE_SHIFT, size, prot); } +#define io_remap_pfn_range io_remap_pfn_range #define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS #define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \ diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 61210db139f..3770bf5c6e1 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -24,7 +24,8 @@ /* The kernel image occupies 0x4000000 to 0x6000000 (4MB --> 96MB). * The page copy blockops can use 0x6000000 to 0x8000000. - * The TSB is mapped in the 0x8000000 to 0xa000000 range. + * The 8K TSB is mapped in the 0x8000000 to 0x8400000 range. + * The 4M TSB is mapped in the 0x8400000 to 0x8800000 range. * The PROM resides in an area spanning 0xf0000000 to 0x100000000. * The vmalloc area spans 0x100000000 to 0x200000000. * Since modules need to be in the lowest 32-bits of the address space, @@ -33,7 +34,8 @@ * 0x400000000. */ #define TLBTEMP_BASE _AC(0x0000000006000000,UL) -#define TSBMAP_BASE _AC(0x0000000008000000,UL) +#define TSBMAP_8K_BASE _AC(0x0000000008000000,UL) +#define TSBMAP_4M_BASE _AC(0x0000000008400000,UL) #define MODULES_VADDR _AC(0x0000000010000000,UL) #define MODULES_LEN _AC(0x00000000e0000000,UL) #define MODULES_END _AC(0x00000000f0000000,UL) @@ -45,38 +47,49 @@ #define vmemmap ((struct page *)VMEMMAP_BASE) -/* XXX All of this needs to be rethought so we can take advantage - * XXX cheetah's full 64-bit virtual address space, ie. no more hole - * XXX in the middle like on spitfire. -DaveM - */ -/* - * Given a virtual address, the lowest PAGE_SHIFT bits determine offset - * into the page; the next higher PAGE_SHIFT-3 bits determine the pte# - * in the proper pagetable (the -3 is from the 8 byte ptes, and each page - * table is a single page long). The next higher PMD_BITS determine pmd# - * in the proper pmdtable (where we must have PMD_BITS <= (PAGE_SHIFT-2) - * since the pmd entries are 4 bytes, and each pmd page is a single page - * long). Finally, the higher few bits determine pgde#. - */ - /* PMD_SHIFT determines the size of the area a second-level page * table can map */ #define PMD_SHIFT (PAGE_SHIFT + (PAGE_SHIFT-3)) #define PMD_SIZE (_AC(1,UL) << PMD_SHIFT) #define PMD_MASK (~(PMD_SIZE-1)) -#define PMD_BITS (PAGE_SHIFT - 2) +#define PMD_BITS (PAGE_SHIFT - 3) /* PGDIR_SHIFT determines what a third-level page table entry can map */ #define PGDIR_SHIFT (PAGE_SHIFT + (PAGE_SHIFT-3) + PMD_BITS) #define PGDIR_SIZE (_AC(1,UL) << PGDIR_SHIFT) #define PGDIR_MASK (~(PGDIR_SIZE-1)) -#define PGDIR_BITS (PAGE_SHIFT - 2) +#define PGDIR_BITS (PAGE_SHIFT - 3) + +#if (PGDIR_SHIFT + PGDIR_BITS) != 43 +#error Page table parameters do not cover virtual address space properly. +#endif + +#if (PMD_SHIFT != HPAGE_SHIFT) +#error PMD_SHIFT must equal HPAGE_SHIFT for transparent huge pages. +#endif #ifndef __ASSEMBLY__ #include <linux/sched.h> +extern unsigned long sparc64_valid_addr_bitmap[]; + +/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */ +static inline bool __kern_addr_valid(unsigned long paddr) +{ + if ((paddr >> MAX_PHYS_ADDRESS_BITS) != 0UL) + return false; + return test_bit(paddr >> ILOG2_4MB, sparc64_valid_addr_bitmap); +} + +static inline bool kern_addr_valid(unsigned long addr) +{ + unsigned long paddr = __pa(addr); + + return __kern_addr_valid(paddr); +} + /* Entries per page directory level. */ #define PTRS_PER_PTE (1UL << (PAGE_SHIFT-3)) #define PTRS_PER_PMD (1UL << PMD_BITS) @@ -85,9 +98,12 @@ /* Kernel has a separate 44bit address space. */ #define FIRST_USER_ADDRESS 0 -#define pte_ERROR(e) __builtin_trap() -#define pmd_ERROR(e) __builtin_trap() -#define pgd_ERROR(e) __builtin_trap() +#define pmd_ERROR(e) \ + pr_err("%s:%d: bad pmd %p(%016lx) seen at (%pS)\n", \ + __FILE__, __LINE__, &(e), pmd_val(e), __builtin_return_address(0)) +#define pgd_ERROR(e) \ + pr_err("%s:%d: bad pgd %p(%016lx) seen at (%pS)\n", \ + __FILE__, __LINE__, &(e), pgd_val(e), __builtin_return_address(0)) #endif /* !(__ASSEMBLY__) */ @@ -95,6 +111,7 @@ #define _PAGE_VALID _AC(0x8000000000000000,UL) /* Valid TTE */ #define _PAGE_R _AC(0x8000000000000000,UL) /* Keep ref bit uptodate*/ #define _PAGE_SPECIAL _AC(0x0200000000000000,UL) /* Special page */ +#define _PAGE_PMD_HUGE _AC(0x0100000000000000,UL) /* Huge page */ /* Advertise support for _PAGE_SPECIAL */ #define __HAVE_ARCH_PTE_SPECIAL @@ -108,6 +125,7 @@ #define _PAGE_IE_4U _AC(0x0800000000000000,UL) /* Invert Endianness */ #define _PAGE_SOFT2_4U _AC(0x07FC000000000000,UL) /* Software bits, set 2 */ #define _PAGE_SPECIAL_4U _AC(0x0200000000000000,UL) /* Special page */ +#define _PAGE_PMD_HUGE_4U _AC(0x0100000000000000,UL) /* Huge page */ #define _PAGE_RES1_4U _AC(0x0002000000000000,UL) /* Reserved */ #define _PAGE_SZ32MB_4U _AC(0x0001000000000000,UL) /* (Panther) 32MB page */ #define _PAGE_SZ256MB_4U _AC(0x2001000000000000,UL) /* (Panther) 256MB page */ @@ -138,6 +156,7 @@ #define _PAGE_READ_4V _AC(0x0800000000000000,UL) /* Readable SW Bit */ #define _PAGE_WRITE_4V _AC(0x0400000000000000,UL) /* Writable SW Bit */ #define _PAGE_SPECIAL_4V _AC(0x0200000000000000,UL) /* Special page */ +#define _PAGE_PMD_HUGE_4V _AC(0x0100000000000000,UL) /* Huge page */ #define _PAGE_PADDR_4V _AC(0x00FFFFFFFFFFE000,UL) /* paddr[55:13] */ #define _PAGE_IE_4V _AC(0x0000000000001000,UL) /* Invert Endianness */ #define _PAGE_E_4V _AC(0x0000000000000800,UL) /* side-Effect */ @@ -160,26 +179,15 @@ #define _PAGE_SZ8K_4V _AC(0x0000000000000000,UL) /* 8K Page */ #define _PAGE_SZALL_4V _AC(0x0000000000000007,UL) /* All pgsz bits */ -#if PAGE_SHIFT == 13 #define _PAGE_SZBITS_4U _PAGE_SZ8K_4U #define _PAGE_SZBITS_4V _PAGE_SZ8K_4V -#elif PAGE_SHIFT == 16 -#define _PAGE_SZBITS_4U _PAGE_SZ64K_4U -#define _PAGE_SZBITS_4V _PAGE_SZ64K_4V -#else -#error Wrong PAGE_SHIFT specified + +#if REAL_HPAGE_SHIFT != 22 +#error REAL_HPAGE_SHIFT and _PAGE_SZHUGE_foo must match up #endif -#if defined(CONFIG_HUGETLB_PAGE_SIZE_4MB) #define _PAGE_SZHUGE_4U _PAGE_SZ4MB_4U #define _PAGE_SZHUGE_4V _PAGE_SZ4MB_4V -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K) -#define _PAGE_SZHUGE_4U _PAGE_SZ512K_4U -#define _PAGE_SZHUGE_4V _PAGE_SZ512K_4V -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_64K) -#define _PAGE_SZHUGE_4U _PAGE_SZ64K_4U -#define _PAGE_SZHUGE_4V _PAGE_SZ64K_4V -#endif /* These are actually filled in at boot time by sun4{u,v}_pgprot_init() */ #define __P000 __pgprot(0) @@ -202,9 +210,9 @@ #ifndef __ASSEMBLY__ -extern pte_t mk_pte_io(unsigned long, pgprot_t, int, unsigned long); +pte_t mk_pte_io(unsigned long, pgprot_t, int, unsigned long); -extern unsigned long pte_sz_bits(unsigned long size); +unsigned long pte_sz_bits(unsigned long size); extern pgprot_t PAGE_KERNEL; extern pgprot_t PAGE_KERNEL_LOCKED; @@ -218,7 +226,6 @@ extern unsigned long _PAGE_CACHE; extern unsigned long pg_iobits; extern unsigned long _PAGE_ALL_SZ_BITS; -extern unsigned long _PAGE_SZBITS; extern struct page *mem_map_zero; #define ZERO_PAGE(vaddr) (mem_map_zero) @@ -231,25 +238,22 @@ extern struct page *mem_map_zero; static inline pte_t pfn_pte(unsigned long pfn, pgprot_t prot) { unsigned long paddr = pfn << PAGE_SHIFT; - unsigned long sz_bits; - - sz_bits = 0UL; - if (_PAGE_SZBITS_4U != 0UL || _PAGE_SZBITS_4V != 0UL) { - __asm__ __volatile__( - "\n661: sethi %%uhi(%1), %0\n" - " sllx %0, 32, %0\n" - " .section .sun4v_2insn_patch, \"ax\"\n" - " .word 661b\n" - " mov %2, %0\n" - " nop\n" - " .previous\n" - : "=r" (sz_bits) - : "i" (_PAGE_SZBITS_4U), "i" (_PAGE_SZBITS_4V)); - } - return __pte(paddr | sz_bits | pgprot_val(prot)); + + BUILD_BUG_ON(_PAGE_SZBITS_4U != 0UL || _PAGE_SZBITS_4V != 0UL); + return __pte(paddr | pgprot_val(prot)); } #define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot)) +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot) +{ + pte_t pte = pfn_pte(page_nr, pgprot); + + return __pmd(pte_val(pte)); +} +#define mk_pmd(page, pgprot) pfn_pmd(page_to_pfn(page), (pgprot)) +#endif + /* This one can be done with two shifts. */ static inline unsigned long pte_pfn(pte_t pte) { @@ -276,8 +280,8 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot) { unsigned long mask, tmp; - /* SUN4U: 0x600307ffffffecb8 (negated == 0x9ffcf80000001347) - * SUN4V: 0x30ffffffffffee17 (negated == 0xcf000000000011e8) + /* SUN4U: 0x630107ffffffec38 (negated == 0x9cfef800000013c7) + * SUN4V: 0x33ffffffffffee07 (negated == 0xcc000000000011f8) * * Even if we use negation tricks the result is still a 6 * instruction sequence, so don't try to play fancy and just @@ -286,6 +290,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot) * Note: We encode this into 3 sun4v 2-insn patch sequences. */ + BUILD_BUG_ON(_PAGE_SZBITS_4U != 0UL || _PAGE_SZBITS_4V != 0UL); __asm__ __volatile__( "\n661: sethi %%uhi(%2), %1\n" " sethi %%hi(%2), %0\n" @@ -306,15 +311,26 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot) " .previous\n" : "=r" (mask), "=r" (tmp) : "i" (_PAGE_PADDR_4U | _PAGE_MODIFIED_4U | _PAGE_ACCESSED_4U | - _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U | _PAGE_PRESENT_4U | - _PAGE_SZBITS_4U | _PAGE_SPECIAL), + _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U | + _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4U), "i" (_PAGE_PADDR_4V | _PAGE_MODIFIED_4V | _PAGE_ACCESSED_4V | - _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V | _PAGE_PRESENT_4V | - _PAGE_SZBITS_4V | _PAGE_SPECIAL)); + _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V | + _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4V)); return __pte((pte_val(pte) & mask) | (pgprot_val(prot) & ~mask)); } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_modify(pte, newprot); + + return __pmd(pte_val(pte)); +} +#endif + static inline pte_t pgoff_to_pte(unsigned long off) { off <<= PAGE_SHIFT; @@ -355,7 +371,7 @@ static inline pgprot_t pgprot_noncached(pgprot_t prot) */ #define pgprot_noncached pgprot_noncached -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) static inline pte_t pte_mkhuge(pte_t pte) { unsigned long mask; @@ -373,6 +389,17 @@ static inline pte_t pte_mkhuge(pte_t pte) return __pte(pte_val(pte) | mask); } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline pmd_t pmd_mkhuge(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_mkhuge(pte); + pte_val(pte) |= _PAGE_PMD_HUGE; + + return __pmd(pte_val(pte)); +} +#endif #endif static inline pte_t pte_mkdirty(pte_t pte) @@ -613,29 +640,181 @@ static inline unsigned long pte_present(pte_t pte) return val; } +#define pte_accessible pte_accessible +static inline unsigned long pte_accessible(struct mm_struct *mm, pte_t a) +{ + return pte_val(a) & _PAGE_VALID; +} + static inline unsigned long pte_special(pte_t pte) { return pte_val(pte) & _PAGE_SPECIAL; } -#define pmd_set(pmdp, ptep) \ - (pmd_val(*(pmdp)) = (__pa((unsigned long) (ptep)) >> 11UL)) +static inline unsigned long pmd_large(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + return pte_val(pte) & _PAGE_PMD_HUGE; +} + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static inline unsigned long pmd_young(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + return pte_young(pte); +} + +static inline unsigned long pmd_write(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + return pte_write(pte); +} + +static inline unsigned long pmd_pfn(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + return pte_pfn(pte); +} + +static inline unsigned long pmd_trans_huge(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + return pte_val(pte) & _PAGE_PMD_HUGE; +} + +static inline unsigned long pmd_trans_splitting(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + return pmd_trans_huge(pmd) && pte_special(pte); +} + +#define has_transparent_hugepage() 1 + +static inline pmd_t pmd_mkold(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_mkold(pte); + + return __pmd(pte_val(pte)); +} + +static inline pmd_t pmd_wrprotect(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_wrprotect(pte); + + return __pmd(pte_val(pte)); +} + +static inline pmd_t pmd_mkdirty(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_mkdirty(pte); + + return __pmd(pte_val(pte)); +} + +static inline pmd_t pmd_mkyoung(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_mkyoung(pte); + + return __pmd(pte_val(pte)); +} + +static inline pmd_t pmd_mkwrite(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_mkwrite(pte); + + return __pmd(pte_val(pte)); +} + +static inline pmd_t pmd_mksplitting(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + + pte = pte_mkspecial(pte); + + return __pmd(pte_val(pte)); +} + +static inline pgprot_t pmd_pgprot(pmd_t entry) +{ + unsigned long val = pmd_val(entry); + + return __pgprot(val); +} +#endif + +static inline int pmd_present(pmd_t pmd) +{ + return pmd_val(pmd) != 0UL; +} + +#define pmd_none(pmd) (!pmd_val(pmd)) + +/* pmd_bad() is only called on non-trans-huge PMDs. Our encoding is + * very simple, it's just the physical address. PTE tables are of + * size PAGE_SIZE so make sure the sub-PAGE_SIZE bits are clear and + * the top bits outside of the range of any physical address size we + * support are clear as well. We also validate the physical itself. + */ +#define pmd_bad(pmd) ((pmd_val(pmd) & ~PAGE_MASK) || \ + !__kern_addr_valid(pmd_val(pmd))) + +#define pud_none(pud) (!pud_val(pud)) + +#define pud_bad(pud) ((pud_val(pud) & ~PAGE_MASK) || \ + !__kern_addr_valid(pud_val(pud))) + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t pmd); +#else +static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t pmd) +{ + *pmdp = pmd; +} +#endif + +static inline void pmd_set(struct mm_struct *mm, pmd_t *pmdp, pte_t *ptep) +{ + unsigned long val = __pa((unsigned long) (ptep)); + + pmd_val(*pmdp) = val; +} + #define pud_set(pudp, pmdp) \ - (pud_val(*(pudp)) = (__pa((unsigned long) (pmdp)) >> 11UL)) -#define __pmd_page(pmd) \ - ((unsigned long) __va((((unsigned long)pmd_val(pmd))<<11UL))) + (pud_val(*(pudp)) = (__pa((unsigned long) (pmdp)))) +static inline unsigned long __pmd_page(pmd_t pmd) +{ + pte_t pte = __pte(pmd_val(pmd)); + unsigned long pfn; + + pfn = pte_pfn(pte); + + return ((unsigned long) __va(pfn << PAGE_SHIFT)); +} #define pmd_page(pmd) virt_to_page((void *)__pmd_page(pmd)) #define pud_page_vaddr(pud) \ - ((unsigned long) __va((((unsigned long)pud_val(pud))<<11UL))) + ((unsigned long) __va(pud_val(pud))) #define pud_page(pud) virt_to_page((void *)pud_page_vaddr(pud)) -#define pmd_none(pmd) (!pmd_val(pmd)) -#define pmd_bad(pmd) (0) -#define pmd_present(pmd) (pmd_val(pmd) != 0U) -#define pmd_clear(pmdp) (pmd_val(*(pmdp)) = 0U) -#define pud_none(pud) (!pud_val(pud)) -#define pud_bad(pud) (0) +#define pmd_clear(pmdp) (pmd_val(*(pmdp)) = 0UL) #define pud_present(pud) (pud_val(pud) != 0U) -#define pud_clear(pudp) (pud_val(*(pudp)) = 0U) +#define pud_clear(pudp) (pud_val(*(pudp)) = 0UL) /* Same in both SUN4V and SUN4U. */ #define pte_none(pte) (!pte_val(pte)) @@ -661,8 +840,18 @@ static inline unsigned long pte_special(pte_t pte) #define pte_unmap(pte) do { } while (0) /* Actual page table PTE updates. */ -extern void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr, - pte_t *ptep, pte_t orig, int fullmm); +void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr, + pte_t *ptep, pte_t orig, int fullmm); + +#define __HAVE_ARCH_PMDP_GET_AND_CLEAR +static inline pmd_t pmdp_get_and_clear(struct mm_struct *mm, + unsigned long addr, + pmd_t *pmdp) +{ + pmd_t pmd = *pmdp; + set_pmd_at(mm, addr, pmdp, __pmd(0UL)); + return pmd; +} static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte, int fullmm) @@ -677,7 +866,7 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr, * SUN4V NOTE: _PAGE_VALID is the same value in both the SUN4U * and SUN4V pte layout, so this inline test is fine. */ - if (likely(mm != &init_mm) && (pte_val(orig) & _PAGE_VALID)) + if (likely(mm != &init_mm) && pte_accessible(mm, orig)) tlb_batch_add(mm, addr, ptep, orig, fullmm); } @@ -708,17 +897,32 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr, }) #endif -extern pgd_t swapper_pg_dir[2048]; -extern pmd_t swapper_low_pmd_dir[2048]; +extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; +extern pmd_t swapper_low_pmd_dir[PTRS_PER_PMD]; -extern void paging_init(void); -extern unsigned long find_ecache_flush_span(unsigned long size); +void paging_init(void); +unsigned long find_ecache_flush_span(unsigned long size); struct seq_file; -extern void mmu_info(struct seq_file *); +void mmu_info(struct seq_file *); struct vm_area_struct; -extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t *); +void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t *); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr, + pmd_t *pmd); + +#define __HAVE_ARCH_PMDP_INVALIDATE +extern void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp); + +#define __HAVE_ARCH_PGTABLE_DEPOSIT +void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp, + pgtable_t pgtable); + +#define __HAVE_ARCH_PGTABLE_WITHDRAW +pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp); +#endif /* Encode and de-code a swap entry */ #define __swp_type(entry) (((entry).val >> PAGE_SHIFT) & 0xffUL) @@ -733,24 +937,12 @@ extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t *); #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) /* File offset in PTE support. */ -extern unsigned long pte_file(pte_t); +unsigned long pte_file(pte_t); #define pte_to_pgoff(pte) (pte_val(pte) >> PAGE_SHIFT) -extern pte_t pgoff_to_pte(unsigned long); +pte_t pgoff_to_pte(unsigned long); #define PTE_FILE_MAX_BITS (64UL - PAGE_SHIFT - 1UL) -extern unsigned long sparc64_valid_addr_bitmap[]; - -/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */ -static inline bool kern_addr_valid(unsigned long addr) -{ - unsigned long paddr = __pa(addr); - - if ((paddr >> 41UL) != 0UL) - return false; - return test_bit(paddr >> 22, sparc64_valid_addr_bitmap); -} - -extern int page_in_phys_avail(unsigned long paddr); +int page_in_phys_avail(unsigned long paddr); /* * For sparc32&64, the pfn in io_remap_pfn_range() carries <iospace> in @@ -760,8 +952,8 @@ extern int page_in_phys_avail(unsigned long paddr); #define GET_IOSPACE(pfn) (pfn >> (BITS_PER_LONG - 4)) #define GET_PFN(pfn) (pfn & 0x0fffffffffffffffUL) -extern int remap_pfn_range(struct vm_area_struct *, unsigned long, unsigned long, - unsigned long, pgprot_t); +int remap_pfn_range(struct vm_area_struct *, unsigned long, unsigned long, + unsigned long, pgprot_t); static inline int io_remap_pfn_range(struct vm_area_struct *vma, unsigned long from, unsigned long pfn, @@ -775,7 +967,9 @@ static inline int io_remap_pfn_range(struct vm_area_struct *vma, return remap_pfn_range(vma, from, phys_base >> PAGE_SHIFT, size, prot); } +#define io_remap_pfn_range io_remap_pfn_range +#include <asm/tlbflush.h> #include <asm-generic/pgtable.h> /* We provide our own get_unmapped_area to cope with VA holes and @@ -787,20 +981,20 @@ static inline int io_remap_pfn_range(struct vm_area_struct *vma, /* We provide a special get_unmapped_area for framebuffer mmaps to try and use * the largest alignment possible such that larget PTEs can be used. */ -extern unsigned long get_fb_unmapped_area(struct file *filp, unsigned long, - unsigned long, unsigned long, - unsigned long); +unsigned long get_fb_unmapped_area(struct file *filp, unsigned long, + unsigned long, unsigned long, + unsigned long); #define HAVE_ARCH_FB_UNMAPPED_AREA -extern void pgtable_cache_init(void); -extern void sun4v_register_fault_status(void); -extern void sun4v_ktsb_register(void); -extern void __init cheetah_ecache_flush_init(void); -extern void sun4v_patch_tlb_handlers(void); +void pgtable_cache_init(void); +void sun4v_register_fault_status(void); +void sun4v_ktsb_register(void); +void __init cheetah_ecache_flush_init(void); +void sun4v_patch_tlb_handlers(void); extern unsigned long cmdline_memory_size; -extern asmlinkage void do_sparc64_fault(struct pt_regs *regs); +asmlinkage void do_sparc64_fault(struct pt_regs *regs); #endif /* !(__ASSEMBLY__) */ diff --git a/arch/sparc/include/asm/processor_32.h b/arch/sparc/include/asm/processor_32.h index f74ac9ee33a..a564817bbc2 100644 --- a/arch/sparc/include/asm/processor_32.h +++ b/arch/sparc/include/asm/processor_32.h @@ -74,7 +74,7 @@ struct thread_struct { } /* Return saved PC of a blocked thread. */ -extern unsigned long thread_saved_pc(struct task_struct *t); +unsigned long thread_saved_pc(struct task_struct *t); /* Do necessary setup to start up a newly executed thread. */ static inline void start_thread(struct pt_regs * regs, unsigned long pc, @@ -106,9 +106,8 @@ static inline void start_thread(struct pt_regs * regs, unsigned long pc, /* Free all resources held by a thread. */ #define release_thread(tsk) do { } while(0) -extern pid_t kernel_thread(int (*fn)(void *), void * arg, unsigned long flags); -extern unsigned long get_wchan(struct task_struct *); +unsigned long get_wchan(struct task_struct *); #define task_pt_regs(tsk) ((tsk)->thread.kregs) #define KSTK_EIP(tsk) ((tsk)->thread.kregs->pc) @@ -117,8 +116,10 @@ extern unsigned long get_wchan(struct task_struct *); #ifdef __KERNEL__ extern struct task_struct *last_task_used_math; +int do_mathemu(struct pt_regs *regs, struct task_struct *fpt); #define cpu_relax() barrier() +extern void (*sparc_idle)(void); #endif diff --git a/arch/sparc/include/asm/processor_64.h b/arch/sparc/include/asm/processor_64.h index 4e5a483122a..7028fe1a7c0 100644 --- a/arch/sparc/include/asm/processor_64.h +++ b/arch/sparc/include/asm/processor_64.h @@ -18,9 +18,6 @@ #include <asm/ptrace.h> #include <asm/page.h> -/* Don't hold the runqueue lock over context switch */ -#define __ARCH_WANT_UNLOCKED_CTXSW - /* The sparc has no problems with write protection */ #define wp_works_ok 1 #define wp_works_ok__is_a_macro /* for versions in ksyms.c */ @@ -94,10 +91,11 @@ struct thread_struct { #ifndef __ASSEMBLY__ #include <linux/types.h> +#include <asm/fpumacro.h> /* Return saved PC of a blocked thread. */ struct task_struct; -extern unsigned long thread_saved_pc(struct task_struct *); +unsigned long thread_saved_pc(struct task_struct *); /* On Uniprocessor, even in RMO processes see TSO semantics */ #ifdef CONFIG_SMP @@ -143,6 +141,10 @@ do { \ : \ : "r" (regs), "r" (sp - sizeof(struct reg_window) - STACK_BIAS), \ "i" ((const unsigned long)(&((struct pt_regs *)0)->u_regs[0]))); \ + fprs_write(0); \ + current_thread_info()->xfsr[0] = 0; \ + current_thread_info()->fpsaved[0] = 0; \ + regs->tstate &= ~TSTATE_PEF; \ } while (0) #define start_thread32(regs, pc, sp) \ @@ -183,20 +185,37 @@ do { \ : \ : "r" (regs), "r" (sp - sizeof(struct reg_window32)), \ "i" ((const unsigned long)(&((struct pt_regs *)0)->u_regs[0]))); \ + fprs_write(0); \ + current_thread_info()->xfsr[0] = 0; \ + current_thread_info()->fpsaved[0] = 0; \ + regs->tstate &= ~TSTATE_PEF; \ } while (0) /* Free all resources held by a thread. */ #define release_thread(tsk) do { } while (0) -extern pid_t kernel_thread(int (*fn)(void *), void * arg, unsigned long flags); - -extern unsigned long get_wchan(struct task_struct *task); +unsigned long get_wchan(struct task_struct *task); #define task_pt_regs(tsk) (task_thread_info(tsk)->kregs) #define KSTK_EIP(tsk) (task_pt_regs(tsk)->tpc) #define KSTK_ESP(tsk) (task_pt_regs(tsk)->u_regs[UREG_FP]) -#define cpu_relax() barrier() +/* Please see the commentary in asm/backoff.h for a description of + * what these instructions are doing and how they have been choosen. + * To make a long story short, we are trying to yield the current cpu + * strand during busy loops. + */ +#define cpu_relax() asm volatile("\n99:\n\t" \ + "rd %%ccr, %%g0\n\t" \ + "rd %%ccr, %%g0\n\t" \ + "rd %%ccr, %%g0\n\t" \ + ".section .pause_3insn_patch,\"ax\"\n\t"\ + ".word 99b\n\t" \ + "wr %%g0, 128, %%asr27\n\t" \ + "nop\n\t" \ + "nop\n\t" \ + ".previous" \ + ::: "memory") /* Prefetch support. This is tuned for UltraSPARC-III and later. * UltraSPARC-I will treat these as nops, and UltraSPARC-II has @@ -234,6 +253,8 @@ static inline void prefetchw(const void *x) #define HAVE_ARCH_PICK_MMAP_LAYOUT +int do_mathemu(struct pt_regs *regs, struct fpustate *f, bool illegal_insn_trap); + #endif /* !(__ASSEMBLY__) */ #endif /* !(__ASM_SPARC64_PROCESSOR_H) */ diff --git a/arch/sparc/include/asm/prom.h b/arch/sparc/include/asm/prom.h index c2876511070..d955c8df62d 100644 --- a/arch/sparc/include/asm/prom.h +++ b/arch/sparc/include/asm/prom.h @@ -36,32 +36,28 @@ struct of_irq_controller { void *data; }; -extern struct device_node *of_find_node_by_cpuid(int cpuid); -extern int of_set_property(struct device_node *node, const char *name, void *val, int len); +struct device_node *of_find_node_by_cpuid(int cpuid); +int of_set_property(struct device_node *node, const char *name, void *val, int len); extern struct mutex of_set_property_mutex; -extern int of_getintprop_default(struct device_node *np, - const char *name, +int of_getintprop_default(struct device_node *np, + const char *name, int def); -extern int of_find_in_proplist(const char *list, const char *match, int len); -#ifdef CONFIG_NUMA -extern int of_node_to_nid(struct device_node *dp); -#define of_node_to_nid of_node_to_nid -#endif +int of_find_in_proplist(const char *list, const char *match, int len); -extern void prom_build_devicetree(void); -extern void of_populate_present_mask(void); -extern void of_fill_in_cpu_data(void); +void prom_build_devicetree(void); +void of_populate_present_mask(void); +void of_fill_in_cpu_data(void); struct resource; -extern void __iomem *of_ioremap(struct resource *res, unsigned long offset, unsigned long size, char *name); -extern void of_iounmap(struct resource *res, void __iomem *base, unsigned long size); +void __iomem *of_ioremap(struct resource *res, unsigned long offset, unsigned long size, char *name); +void of_iounmap(struct resource *res, void __iomem *base, unsigned long size); extern struct device_node *of_console_device; extern char *of_console_path; extern char *of_console_options; -extern void irq_trans_init(struct device_node *dp); -extern char *build_path_component(struct device_node *dp); +void irq_trans_init(struct device_node *dp); +char *build_path_component(struct device_node *dp); #endif /* __KERNEL__ */ #endif /* _SPARC_PROM_H */ diff --git a/arch/sparc/include/asm/psr.h b/arch/sparc/include/asm/psr.h index cee7ed9c927..e71eb57945e 100644 --- a/arch/sparc/include/asm/psr.h +++ b/arch/sparc/include/asm/psr.h @@ -7,43 +7,11 @@ * * Copyright (C) 1994 David S. Miller (davem@caip.rutgers.edu) */ - #ifndef __LINUX_SPARC_PSR_H #define __LINUX_SPARC_PSR_H -/* The Sparc PSR fields are laid out as the following: - * - * ------------------------------------------------------------------------ - * | impl | vers | icc | resv | EC | EF | PIL | S | PS | ET | CWP | - * | 31-28 | 27-24 | 23-20 | 19-14 | 13 | 12 | 11-8 | 7 | 6 | 5 | 4-0 | - * ------------------------------------------------------------------------ - */ -#define PSR_CWP 0x0000001f /* current window pointer */ -#define PSR_ET 0x00000020 /* enable traps field */ -#define PSR_PS 0x00000040 /* previous privilege level */ -#define PSR_S 0x00000080 /* current privilege level */ -#define PSR_PIL 0x00000f00 /* processor interrupt level */ -#define PSR_EF 0x00001000 /* enable floating point */ -#define PSR_EC 0x00002000 /* enable co-processor */ -#define PSR_SYSCALL 0x00004000 /* inside of a syscall */ -#define PSR_LE 0x00008000 /* SuperSparcII little-endian */ -#define PSR_ICC 0x00f00000 /* integer condition codes */ -#define PSR_C 0x00100000 /* carry bit */ -#define PSR_V 0x00200000 /* overflow bit */ -#define PSR_Z 0x00400000 /* zero bit */ -#define PSR_N 0x00800000 /* negative bit */ -#define PSR_VERS 0x0f000000 /* cpu-version field */ -#define PSR_IMPL 0xf0000000 /* cpu-implementation field */ - -#define PSR_VERS_SHIFT 24 -#define PSR_IMPL_SHIFT 28 -#define PSR_VERS_SHIFTED_MASK 0xf -#define PSR_IMPL_SHIFTED_MASK 0xf - -#define PSR_IMPL_TI 0x4 -#define PSR_IMPL_LEON 0xf +#include <uapi/asm/psr.h> -#ifdef __KERNEL__ #ifndef __ASSEMBLY__ /* Get the %psr register. */ @@ -96,6 +64,4 @@ static inline unsigned int get_fsr(void) #endif /* !(__ASSEMBLY__) */ -#endif /* (__KERNEL__) */ - #endif /* !(__LINUX_SPARC_PSR_H) */ diff --git a/arch/sparc/include/asm/ptrace.h b/arch/sparc/include/asm/ptrace.h index fd9c3f21cbf..bac6a946ee0 100644 --- a/arch/sparc/include/asm/ptrace.h +++ b/arch/sparc/include/asm/ptrace.h @@ -1,169 +1,11 @@ #ifndef __SPARC_PTRACE_H #define __SPARC_PTRACE_H -#if defined(__sparc__) && defined(__arch64__) -/* 64 bit sparc */ -#include <asm/pstate.h> - -/* This struct defines the way the registers are stored on the - * stack during a system call and basically all traps. - */ - -/* This magic value must have the low 9 bits clear, - * as that is where we encode the %tt value, see below. - */ -#define PT_REGS_MAGIC 0x57ac6c00 - -#ifndef __ASSEMBLY__ - -#include <linux/types.h> - -struct pt_regs { - unsigned long u_regs[16]; /* globals and ins */ - unsigned long tstate; - unsigned long tpc; - unsigned long tnpc; - unsigned int y; - - /* We encode a magic number, PT_REGS_MAGIC, along - * with the %tt (trap type) register value at trap - * entry time. The magic number allows us to identify - * accurately a trap stack frame in the stack - * unwinder, and the %tt value allows us to test - * things like "in a system call" etc. for an arbitray - * process. - * - * The PT_REGS_MAGIC is chosen such that it can be - * loaded completely using just a sethi instruction. - */ - unsigned int magic; -}; - -struct pt_regs32 { - unsigned int psr; - unsigned int pc; - unsigned int npc; - unsigned int y; - unsigned int u_regs[16]; /* globals and ins */ -}; - -/* A V9 register window */ -struct reg_window { - unsigned long locals[8]; - unsigned long ins[8]; -}; - -/* A 32-bit register window. */ -struct reg_window32 { - unsigned int locals[8]; - unsigned int ins[8]; -}; - -/* A V9 Sparc stack frame */ -struct sparc_stackf { - unsigned long locals[8]; - unsigned long ins[6]; - struct sparc_stackf *fp; - unsigned long callers_pc; - char *structptr; - unsigned long xargs[6]; - unsigned long xxargs[1]; -}; - -/* A 32-bit Sparc stack frame */ -struct sparc_stackf32 { - unsigned int locals[8]; - unsigned int ins[6]; - unsigned int fp; - unsigned int callers_pc; - unsigned int structptr; - unsigned int xargs[6]; - unsigned int xxargs[1]; -}; - -struct sparc_trapf { - unsigned long locals[8]; - unsigned long ins[8]; - unsigned long _unused; - struct pt_regs *regs; -}; -#endif /* (!__ASSEMBLY__) */ -#else -/* 32 bit sparc */ - -#include <asm/psr.h> - -/* This struct defines the way the registers are stored on the - * stack during a system call and basically all traps. - */ -#ifndef __ASSEMBLY__ - -#include <linux/types.h> - -struct pt_regs { - unsigned long psr; - unsigned long pc; - unsigned long npc; - unsigned long y; - unsigned long u_regs[16]; /* globals and ins */ -}; - -/* A 32-bit register window. */ -struct reg_window32 { - unsigned long locals[8]; - unsigned long ins[8]; -}; - -/* A Sparc stack frame */ -struct sparc_stackf { - unsigned long locals[8]; - unsigned long ins[6]; - struct sparc_stackf *fp; - unsigned long callers_pc; - char *structptr; - unsigned long xargs[6]; - unsigned long xxargs[1]; -}; -#endif /* (!__ASSEMBLY__) */ - -#endif /* (defined(__sparc__) && defined(__arch64__))*/ - -#ifndef __ASSEMBLY__ - -#define TRACEREG_SZ sizeof(struct pt_regs) -#define STACKFRAME_SZ sizeof(struct sparc_stackf) - -#define TRACEREG32_SZ sizeof(struct pt_regs32) -#define STACKFRAME32_SZ sizeof(struct sparc_stackf32) - -#endif /* (!__ASSEMBLY__) */ - -#define UREG_G0 0 -#define UREG_G1 1 -#define UREG_G2 2 -#define UREG_G3 3 -#define UREG_G4 4 -#define UREG_G5 5 -#define UREG_G6 6 -#define UREG_G7 7 -#define UREG_I0 8 -#define UREG_I1 9 -#define UREG_I2 10 -#define UREG_I3 11 -#define UREG_I4 12 -#define UREG_I5 13 -#define UREG_I6 14 -#define UREG_I7 15 -#define UREG_FP UREG_I6 -#define UREG_RETPC UREG_I7 +#include <uapi/asm/ptrace.h> #if defined(__sparc__) && defined(__arch64__) -/* 64 bit sparc */ - #ifndef __ASSEMBLY__ -#ifdef __KERNEL__ - #include <linux/threads.h> #include <asm/switch_to.h> @@ -190,6 +32,9 @@ static inline bool pt_regs_clear_syscall(struct pt_regs *regs) #define arch_ptrace_stop(exit_code, info) \ synchronize_user_stack() +#define current_pt_regs() \ + ((struct pt_regs *)((unsigned long)current_thread_info() + THREAD_SIZE) - 1) + struct global_reg_snapshot { unsigned long tstate; unsigned long tpc; @@ -200,11 +45,20 @@ struct global_reg_snapshot { struct thread_info *thread; unsigned long pad1; }; -extern struct global_reg_snapshot global_reg_snapshot[NR_CPUS]; -#define force_successful_syscall_return() \ -do { current_thread_info()->syscall_noerror = 1; \ -} while (0) +struct global_pmu_snapshot { + unsigned long pcr[4]; + unsigned long pic[4]; +}; + +union global_cpu_snapshot { + struct global_reg_snapshot reg; + struct global_pmu_snapshot pmu; +}; + +extern union global_cpu_snapshot global_cpu_snapshot[NR_CPUS]; + +#define force_successful_syscall_return() set_thread_noerror(1) #define user_mode(regs) (!((regs)->tstate & TSTATE_PRIV)) #define instruction_pointer(regs) ((regs)->tpc) #define instruction_pointer_set(regs, val) ((regs)->tpc = (val)) @@ -219,28 +73,14 @@ static inline long regs_return_value(struct pt_regs *regs) return regs->u_regs[UREG_I0]; } #ifdef CONFIG_SMP -extern unsigned long profile_pc(struct pt_regs *); +unsigned long profile_pc(struct pt_regs *); #else #define profile_pc(regs) instruction_pointer(regs) #endif -#endif /* (__KERNEL__) */ - #else /* __ASSEMBLY__ */ -/* For assembly code. */ -#define TRACEREG_SZ 0xa0 -#define STACKFRAME_SZ 0xc0 - -#define TRACEREG32_SZ 0x50 -#define STACKFRAME32_SZ 0x60 #endif /* __ASSEMBLY__ */ - #else /* (defined(__sparc__) && defined(__arch64__)) */ - -/* 32 bit sparc */ - #ifndef __ASSEMBLY__ - -#ifdef __KERNEL__ #include <asm/switch_to.h> static inline bool pt_regs_is_syscall(struct pt_regs *regs) @@ -261,162 +101,17 @@ static inline bool pt_regs_clear_syscall(struct pt_regs *regs) #define arch_ptrace_stop(exit_code, info) \ synchronize_user_stack() +#define current_pt_regs() \ + ((struct pt_regs *)((unsigned long)current_thread_info() + THREAD_SIZE) - 1) + #define user_mode(regs) (!((regs)->psr & PSR_PS)) #define instruction_pointer(regs) ((regs)->pc) #define user_stack_pointer(regs) ((regs)->u_regs[UREG_FP]) unsigned long profile_pc(struct pt_regs *); -#endif /* (__KERNEL__) */ - #else /* (!__ASSEMBLY__) */ -/* For assembly code. */ -#define TRACEREG_SZ 0x50 -#define STACKFRAME_SZ 0x60 #endif /* (!__ASSEMBLY__) */ - #endif /* (defined(__sparc__) && defined(__arch64__)) */ - -#ifdef __KERNEL__ #define STACK_BIAS 2047 -#endif - -/* These are for pt_regs. */ -#define PT_V9_G0 0x00 -#define PT_V9_G1 0x08 -#define PT_V9_G2 0x10 -#define PT_V9_G3 0x18 -#define PT_V9_G4 0x20 -#define PT_V9_G5 0x28 -#define PT_V9_G6 0x30 -#define PT_V9_G7 0x38 -#define PT_V9_I0 0x40 -#define PT_V9_I1 0x48 -#define PT_V9_I2 0x50 -#define PT_V9_I3 0x58 -#define PT_V9_I4 0x60 -#define PT_V9_I5 0x68 -#define PT_V9_I6 0x70 -#define PT_V9_FP PT_V9_I6 -#define PT_V9_I7 0x78 -#define PT_V9_TSTATE 0x80 -#define PT_V9_TPC 0x88 -#define PT_V9_TNPC 0x90 -#define PT_V9_Y 0x98 -#define PT_V9_MAGIC 0x9c -#define PT_TSTATE PT_V9_TSTATE -#define PT_TPC PT_V9_TPC -#define PT_TNPC PT_V9_TNPC - -/* These for pt_regs32. */ -#define PT_PSR 0x0 -#define PT_PC 0x4 -#define PT_NPC 0x8 -#define PT_Y 0xc -#define PT_G0 0x10 -#define PT_WIM PT_G0 -#define PT_G1 0x14 -#define PT_G2 0x18 -#define PT_G3 0x1c -#define PT_G4 0x20 -#define PT_G5 0x24 -#define PT_G6 0x28 -#define PT_G7 0x2c -#define PT_I0 0x30 -#define PT_I1 0x34 -#define PT_I2 0x38 -#define PT_I3 0x3c -#define PT_I4 0x40 -#define PT_I5 0x44 -#define PT_I6 0x48 -#define PT_FP PT_I6 -#define PT_I7 0x4c - -/* Reg_window offsets */ -#define RW_V9_L0 0x00 -#define RW_V9_L1 0x08 -#define RW_V9_L2 0x10 -#define RW_V9_L3 0x18 -#define RW_V9_L4 0x20 -#define RW_V9_L5 0x28 -#define RW_V9_L6 0x30 -#define RW_V9_L7 0x38 -#define RW_V9_I0 0x40 -#define RW_V9_I1 0x48 -#define RW_V9_I2 0x50 -#define RW_V9_I3 0x58 -#define RW_V9_I4 0x60 -#define RW_V9_I5 0x68 -#define RW_V9_I6 0x70 -#define RW_V9_I7 0x78 - -#define RW_L0 0x00 -#define RW_L1 0x04 -#define RW_L2 0x08 -#define RW_L3 0x0c -#define RW_L4 0x10 -#define RW_L5 0x14 -#define RW_L6 0x18 -#define RW_L7 0x1c -#define RW_I0 0x20 -#define RW_I1 0x24 -#define RW_I2 0x28 -#define RW_I3 0x2c -#define RW_I4 0x30 -#define RW_I5 0x34 -#define RW_I6 0x38 -#define RW_I7 0x3c - -/* Stack_frame offsets */ -#define SF_V9_L0 0x00 -#define SF_V9_L1 0x08 -#define SF_V9_L2 0x10 -#define SF_V9_L3 0x18 -#define SF_V9_L4 0x20 -#define SF_V9_L5 0x28 -#define SF_V9_L6 0x30 -#define SF_V9_L7 0x38 -#define SF_V9_I0 0x40 -#define SF_V9_I1 0x48 -#define SF_V9_I2 0x50 -#define SF_V9_I3 0x58 -#define SF_V9_I4 0x60 -#define SF_V9_I5 0x68 -#define SF_V9_FP 0x70 -#define SF_V9_PC 0x78 -#define SF_V9_RETP 0x80 -#define SF_V9_XARG0 0x88 -#define SF_V9_XARG1 0x90 -#define SF_V9_XARG2 0x98 -#define SF_V9_XARG3 0xa0 -#define SF_V9_XARG4 0xa8 -#define SF_V9_XARG5 0xb0 -#define SF_V9_XXARG 0xb8 - -#define SF_L0 0x00 -#define SF_L1 0x04 -#define SF_L2 0x08 -#define SF_L3 0x0c -#define SF_L4 0x10 -#define SF_L5 0x14 -#define SF_L6 0x18 -#define SF_L7 0x1c -#define SF_I0 0x20 -#define SF_I1 0x24 -#define SF_I2 0x28 -#define SF_I3 0x2c -#define SF_I4 0x30 -#define SF_I5 0x34 -#define SF_FP 0x38 -#define SF_PC 0x3c -#define SF_RETP 0x40 -#define SF_XARG0 0x44 -#define SF_XARG1 0x48 -#define SF_XARG2 0x4c -#define SF_XARG3 0x50 -#define SF_XARG4 0x54 -#define SF_XARG5 0x58 -#define SF_XXARG 0x5c - -#ifdef __KERNEL__ /* global_reg_snapshot offsets */ #define GR_SNAP_TSTATE 0x00 @@ -428,29 +123,4 @@ unsigned long profile_pc(struct pt_regs *); #define GR_SNAP_THREAD 0x30 #define GR_SNAP_PAD1 0x38 -#endif /* __KERNEL__ */ - -/* Stuff for the ptrace system call */ -#define PTRACE_SPARC_DETACH 11 -#define PTRACE_GETREGS 12 -#define PTRACE_SETREGS 13 -#define PTRACE_GETFPREGS 14 -#define PTRACE_SETFPREGS 15 -#define PTRACE_READDATA 16 -#define PTRACE_WRITEDATA 17 -#define PTRACE_READTEXT 18 -#define PTRACE_WRITETEXT 19 -#define PTRACE_GETFPAREGS 20 -#define PTRACE_SETFPAREGS 21 - -/* There are for debugging 64-bit processes, either from a 32 or 64 bit - * parent. Thus their complements are for debugging 32-bit processes only. - */ - -#define PTRACE_GETREGS64 22 -#define PTRACE_SETREGS64 23 -/* PTRACE_SYSCALL is 24 */ -#define PTRACE_GETFPREGS64 25 -#define PTRACE_SETFPREGS64 26 - #endif /* !(__SPARC_PTRACE_H) */ diff --git a/arch/sparc/include/asm/serial.h b/arch/sparc/include/asm/serial.h deleted file mode 100644 index f90d61c2805..00000000000 --- a/arch/sparc/include/asm/serial.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __SPARC_SERIAL_H -#define __SPARC_SERIAL_H - -#define BASE_BAUD ( 1843200 / 16 ) - -#endif /* __SPARC_SERIAL_H */ diff --git a/arch/sparc/include/asm/setup.h b/arch/sparc/include/asm/setup.h index 8a83699a550..f5fffd84d0d 100644 --- a/arch/sparc/include/asm/setup.h +++ b/arch/sparc/include/asm/setup.h @@ -1,17 +1,12 @@ /* * Just a place holder. */ - #ifndef _SPARC_SETUP_H #define _SPARC_SETUP_H -#if defined(__sparc__) && defined(__arch64__) -# define COMMAND_LINE_SIZE 2048 -#else -# define COMMAND_LINE_SIZE 256 -#endif +#include <linux/interrupt.h> -#ifdef __KERNEL__ +#include <uapi/asm/setup.h> extern char reboot_command[]; @@ -28,12 +23,44 @@ static inline int con_is_present(void) { return serial_console ? 0 : 1; } + +/* from irq_32.c */ +extern volatile unsigned char *fdc_status; +extern char *pdma_vaddr; +extern unsigned long pdma_size; +extern volatile int doing_pdma; + +/* This is software state */ +extern char *pdma_base; +extern unsigned long pdma_areasize; + +int sparc_floppy_request_irq(unsigned int irq, irq_handler_t irq_handler); + +/* setup_32.c */ +extern unsigned long cmdline_memory_size; + +/* devices.c */ +void __init device_scan(void); + +/* unaligned_32.c */ +unsigned long safe_compute_effective_address(struct pt_regs *, unsigned int); + #endif -extern void sun_do_break(void); +#ifdef CONFIG_SPARC64 +/* unaligned_64.c */ +int handle_ldf_stq(u32 insn, struct pt_regs *regs); +void handle_ld_nf(u32 insn, struct pt_regs *regs); + +/* init_64.c */ +extern atomic_t dcpage_flushes; +extern atomic_t dcpage_flushes_xcall; + +extern int sysctl_tsb_ratio; +#endif + +void sun_do_break(void); extern int stop_a_enabled; extern int scons_pwroff; -#endif /* __KERNEL__ */ - #endif /* _SPARC_SETUP_H */ diff --git a/arch/sparc/include/asm/sfp-machine_32.h b/arch/sparc/include/asm/sfp-machine_32.h index 01d9c3b5a73..838c9d58f3b 100644 --- a/arch/sparc/include/asm/sfp-machine_32.h +++ b/arch/sparc/include/asm/sfp-machine_32.h @@ -79,9 +79,9 @@ __asm__ ("addcc %r7,%8,%2\n\t" \ "addxcc %r5,%6,%1\n\t" \ "addx %r3,%4,%0\n" \ - : "=r" ((USItype)(r2)), \ - "=&r" ((USItype)(r1)), \ - "=&r" ((USItype)(r0)) \ + : "=r" (r2), \ + "=&r" (r1), \ + "=&r" (r0) \ : "%rJ" ((USItype)(x2)), \ "rI" ((USItype)(y2)), \ "%rJ" ((USItype)(x1)), \ @@ -94,9 +94,9 @@ __asm__ ("subcc %r7,%8,%2\n\t" \ "subxcc %r5,%6,%1\n\t" \ "subx %r3,%4,%0\n" \ - : "=r" ((USItype)(r2)), \ - "=&r" ((USItype)(r1)), \ - "=&r" ((USItype)(r0)) \ + : "=r" (r2), \ + "=&r" (r1), \ + "=&r" (r0) \ : "%rJ" ((USItype)(x2)), \ "rI" ((USItype)(y2)), \ "%rJ" ((USItype)(x1)), \ @@ -115,8 +115,8 @@ "addxcc %r6,%7,%0\n\t" \ "addxcc %r4,%5,%%g2\n\t" \ "addx %r2,%3,%%g1\n\t" \ - : "=&r" ((USItype)(r1)), \ - "=&r" ((USItype)(r0)) \ + : "=&r" (r1), \ + "=&r" (r0) \ : "%rJ" ((USItype)(x3)), \ "rI" ((USItype)(y3)), \ "%rJ" ((USItype)(x2)), \ @@ -140,8 +140,8 @@ "subxcc %r6,%7,%0\n\t" \ "subxcc %r4,%5,%%g2\n\t" \ "subx %r2,%3,%%g1\n\t" \ - : "=&r" ((USItype)(r1)), \ - "=&r" ((USItype)(r0)) \ + : "=&r" (r1), \ + "=&r" (r0) \ : "%rJ" ((USItype)(x3)), \ "rI" ((USItype)(y3)), \ "%rJ" ((USItype)(x2)), \ @@ -164,10 +164,10 @@ "addxcc %2,%%g0,%2\n\t" \ "addxcc %1,%%g0,%1\n\t" \ "addx %0,%%g0,%0\n\t" \ - : "=&r" ((USItype)(x3)), \ - "=&r" ((USItype)(x2)), \ - "=&r" ((USItype)(x1)), \ - "=&r" ((USItype)(x0)) \ + : "=&r" (x3), \ + "=&r" (x2), \ + "=&r" (x1), \ + "=&r" (x0) \ : "rI" ((USItype)(i)), \ "0" ((USItype)(x3)), \ "1" ((USItype)(x2)), \ diff --git a/arch/sparc/include/asm/sigcontext.h b/arch/sparc/include/asm/sigcontext.h index 69914d74813..fc2df1e892c 100644 --- a/arch/sparc/include/asm/sigcontext.h +++ b/arch/sparc/include/asm/sigcontext.h @@ -1,8 +1,8 @@ #ifndef __SPARC_SIGCONTEXT_H #define __SPARC_SIGCONTEXT_H -#ifdef __KERNEL__ #include <asm/ptrace.h> +#include <uapi/asm/sigcontext.h> #ifndef __ASSEMBLY__ @@ -105,6 +105,4 @@ typedef struct { #endif /* !(__ASSEMBLY__) */ -#endif /* (__KERNEL__) */ - #endif /* !(__SPARC_SIGCONTEXT_H) */ diff --git a/arch/sparc/include/asm/siginfo.h b/arch/sparc/include/asm/siginfo.h index 215900fce21..48c34c19f81 100644 --- a/arch/sparc/include/asm/siginfo.h +++ b/arch/sparc/include/asm/siginfo.h @@ -1,20 +1,8 @@ #ifndef __SPARC_SIGINFO_H #define __SPARC_SIGINFO_H -#if defined(__sparc__) && defined(__arch64__) +#include <uapi/asm/siginfo.h> -#define SI_PAD_SIZE32 ((SI_MAX_SIZE/sizeof(int)) - 3) -#define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) -#define __ARCH_SI_BAND_T int - -#endif /* defined(__sparc__) && defined(__arch64__) */ - - -#define __ARCH_SI_TRAPNO - -#include <asm-generic/siginfo.h> - -#ifdef __KERNEL__ #ifdef CONFIG_COMPAT @@ -22,14 +10,4 @@ struct compat_siginfo; #endif /* CONFIG_COMPAT */ -#endif /* __KERNEL__ */ - -#define SI_NOINFO 32767 /* no information in siginfo_t */ - -/* - * SIGEMT si_codes - */ -#define EMT_TAGOVF (__SI_FAULT|1) /* tag overflow */ -#define NSIGEMT 1 - #endif /* !(__SPARC_SIGINFO_H) */ diff --git a/arch/sparc/include/asm/signal.h b/arch/sparc/include/asm/signal.h index aa42fe30d5b..c33ce3f2ba8 100644 --- a/arch/sparc/include/asm/signal.h +++ b/arch/sparc/include/asm/signal.h @@ -1,168 +1,13 @@ #ifndef __SPARC_SIGNAL_H #define __SPARC_SIGNAL_H -#include <asm/sigcontext.h> -#include <linux/compiler.h> - -#ifdef __KERNEL__ #ifndef __ASSEMBLY__ #include <linux/personality.h> #include <linux/types.h> #endif -#endif - -/* On the Sparc the signal handlers get passed a 'sub-signal' code - * for certain signal types, which we document here. - */ -#define SIGHUP 1 -#define SIGINT 2 -#define SIGQUIT 3 -#define SIGILL 4 -#define SUBSIG_STACK 0 -#define SUBSIG_ILLINST 2 -#define SUBSIG_PRIVINST 3 -#define SUBSIG_BADTRAP(t) (0x80 + (t)) - -#define SIGTRAP 5 -#define SIGABRT 6 -#define SIGIOT 6 - -#define SIGEMT 7 -#define SUBSIG_TAG 10 - -#define SIGFPE 8 -#define SUBSIG_FPDISABLED 0x400 -#define SUBSIG_FPERROR 0x404 -#define SUBSIG_FPINTOVFL 0x001 -#define SUBSIG_FPSTSIG 0x002 -#define SUBSIG_IDIVZERO 0x014 -#define SUBSIG_FPINEXACT 0x0c4 -#define SUBSIG_FPDIVZERO 0x0c8 -#define SUBSIG_FPUNFLOW 0x0cc -#define SUBSIG_FPOPERROR 0x0d0 -#define SUBSIG_FPOVFLOW 0x0d4 - -#define SIGKILL 9 -#define SIGBUS 10 -#define SUBSIG_BUSTIMEOUT 1 -#define SUBSIG_ALIGNMENT 2 -#define SUBSIG_MISCERROR 5 - -#define SIGSEGV 11 -#define SUBSIG_NOMAPPING 3 -#define SUBSIG_PROTECTION 4 -#define SUBSIG_SEGERROR 5 - -#define SIGSYS 12 - -#define SIGPIPE 13 -#define SIGALRM 14 -#define SIGTERM 15 -#define SIGURG 16 - -/* SunOS values which deviate from the Linux/i386 ones */ -#define SIGSTOP 17 -#define SIGTSTP 18 -#define SIGCONT 19 -#define SIGCHLD 20 -#define SIGTTIN 21 -#define SIGTTOU 22 -#define SIGIO 23 -#define SIGPOLL SIGIO /* SysV name for SIGIO */ -#define SIGXCPU 24 -#define SIGXFSZ 25 -#define SIGVTALRM 26 -#define SIGPROF 27 -#define SIGWINCH 28 -#define SIGLOST 29 -#define SIGPWR SIGLOST -#define SIGUSR1 30 -#define SIGUSR2 31 - -/* Most things should be clean enough to redefine this at will, if care - is taken to make libc match. */ - -#define __OLD_NSIG 32 -#define __NEW_NSIG 64 -#ifdef __arch64__ -#define _NSIG_BPW 64 -#else -#define _NSIG_BPW 32 -#endif -#define _NSIG_WORDS (__NEW_NSIG / _NSIG_BPW) - -#define SIGRTMIN 32 -#define SIGRTMAX __NEW_NSIG - -#if defined(__KERNEL__) || defined(__WANT_POSIX1B_SIGNALS__) -#define _NSIG __NEW_NSIG -#define __new_sigset_t sigset_t -#define __new_sigaction sigaction -#define __new_sigaction32 sigaction32 -#define __old_sigset_t old_sigset_t -#define __old_sigaction old_sigaction -#define __old_sigaction32 old_sigaction32 -#else -#define _NSIG __OLD_NSIG -#define NSIG _NSIG -#define __old_sigset_t sigset_t -#define __old_sigaction sigaction -#define __old_sigaction32 sigaction32 -#endif +#include <uapi/asm/signal.h> #ifndef __ASSEMBLY__ - -typedef unsigned long __old_sigset_t; /* at least 32 bits */ - -typedef struct { - unsigned long sig[_NSIG_WORDS]; -} __new_sigset_t; - -/* A SunOS sigstack */ -struct sigstack { - /* XXX 32-bit pointers pinhead XXX */ - char *the_stack; - int cur_status; -}; - -/* Sigvec flags */ -#define _SV_SSTACK 1u /* This signal handler should use sig-stack */ -#define _SV_INTR 2u /* Sig return should not restart system call */ -#define _SV_RESET 4u /* Set handler to SIG_DFL upon taken signal */ -#define _SV_IGNCHILD 8u /* Do not send SIGCHLD */ - -/* - * sa_flags values: SA_STACK is not currently supported, but will allow the - * usage of signal stacks by using the (now obsolete) sa_restorer field in - * the sigaction structure as a stack pointer. This is now possible due to - * the changes in signal handling. LBT 010493. - * SA_RESTART flag to get restarting signals (which were the default long ago) - */ -#define SA_NOCLDSTOP _SV_IGNCHILD -#define SA_STACK _SV_SSTACK -#define SA_ONSTACK _SV_SSTACK -#define SA_RESTART _SV_INTR -#define SA_ONESHOT _SV_RESET -#define SA_NODEFER 0x20u -#define SA_NOCLDWAIT 0x100u -#define SA_SIGINFO 0x200u - -#define SA_NOMASK SA_NODEFER - -#define SIG_BLOCK 0x01 /* for blocking signals */ -#define SIG_UNBLOCK 0x02 /* for unblocking signals */ -#define SIG_SETMASK 0x04 /* for setting the signal mask */ - -/* - * sigaltstack controls - */ -#define SS_ONSTACK 1 -#define SS_DISABLE 2 - -#define MINSIGSTKSZ 4096 -#define SIGSTKSZ 16384 - -#ifdef __KERNEL__ /* * DJHR * SA_STATIC_ALLOC is used for the sparc32 system to indicate that this @@ -175,41 +20,9 @@ struct sigstack { * */ #define SA_STATIC_ALLOC 0x8000 -#endif - -#include <asm-generic/signal-defs.h> - -struct __new_sigaction { - __sighandler_t sa_handler; - unsigned long sa_flags; - __sigrestore_t sa_restorer; /* not used by Linux/SPARC yet */ - __new_sigset_t sa_mask; -}; - -struct __old_sigaction { - __sighandler_t sa_handler; - __old_sigset_t sa_mask; - unsigned long sa_flags; - void (*sa_restorer)(void); /* not used by Linux/SPARC yet */ -}; -typedef struct sigaltstack { - void __user *ss_sp; - int ss_flags; - size_t ss_size; -} stack_t; - -#ifdef __KERNEL__ - -struct k_sigaction { - struct __new_sigaction sa; - void __user *ka_restorer; -}; - -#define ptrace_signal_deliver(regs, cookie) do { } while (0) - -#endif /* !(__KERNEL__) */ +#define __ARCH_HAS_KA_RESTORER +#define __ARCH_HAS_SA_RESTORER #endif /* !(__ASSEMBLY__) */ - #endif /* !(__SPARC_SIGNAL_H) */ diff --git a/arch/sparc/include/asm/smp_32.h b/arch/sparc/include/asm/smp_32.h index b73da3c5f10..7c24e08a88d 100644 --- a/arch/sparc/include/asm/smp_32.h +++ b/arch/sparc/include/asm/smp_32.h @@ -36,7 +36,6 @@ typedef void (*smpfunc_t)(unsigned long, unsigned long, unsigned long, unsigned long, unsigned long); void cpu_panic(void); -extern void smp4m_irq_rotate(int cpu); /* * General functions that each host system must provide. @@ -46,7 +45,6 @@ void sun4m_init_smp(void); void sun4d_init_smp(void); void smp_callin(void); -void smp_boot_cpus(void); void smp_store_cpu_info(int); void smp_resched_interrupt(void); @@ -95,21 +93,18 @@ static inline void xc4(smpfunc_t func, unsigned long arg1, unsigned long arg2, arg1, arg2, arg3, arg4); } -extern void arch_send_call_function_single_ipi(int cpu); -extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); +void arch_send_call_function_single_ipi(int cpu); +void arch_send_call_function_ipi_mask(const struct cpumask *mask); static inline int cpu_logical_map(int cpu) { return cpu; } -extern int hard_smp_processor_id(void); +int hard_smp_processor_id(void); #define raw_smp_processor_id() (current_thread_info()->cpu) -#define prof_multiplier(__cpu) cpu_data(__cpu).multiplier -#define prof_counter(__cpu) cpu_data(__cpu).counter - void smp_setup_cpu_possible_map(void); #endif /* !(__ASSEMBLY__) */ diff --git a/arch/sparc/include/asm/smp_64.h b/arch/sparc/include/asm/smp_64.h index 29862a9e906..26d9e772686 100644 --- a/arch/sparc/include/asm/smp_64.h +++ b/arch/sparc/include/asm/smp_64.h @@ -32,30 +32,36 @@ DECLARE_PER_CPU(cpumask_t, cpu_sibling_map); extern cpumask_t cpu_core_map[NR_CPUS]; -extern int sparc64_multi_core; -extern void arch_send_call_function_single_ipi(int cpu); -extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); +void arch_send_call_function_single_ipi(int cpu); +void arch_send_call_function_ipi_mask(const struct cpumask *mask); /* * General functions that each host system must provide. */ -extern int hard_smp_processor_id(void); +int hard_smp_processor_id(void); #define raw_smp_processor_id() (current_thread_info()->cpu) -extern void smp_fill_in_sib_core_maps(void); -extern void cpu_play_dead(void); +void smp_fill_in_sib_core_maps(void); +void cpu_play_dead(void); -extern void smp_fetch_global_regs(void); +void smp_fetch_global_regs(void); +void smp_fetch_global_pmu(void); struct seq_file; void smp_bogo(struct seq_file *); void smp_info(struct seq_file *); +void smp_callin(void); +void cpu_panic(void); +void smp_synchronize_tick_client(void); +void smp_capture(void); +void smp_release(void); + #ifdef CONFIG_HOTPLUG_CPU -extern int __cpu_disable(void); -extern void __cpu_die(unsigned int cpu); +int __cpu_disable(void); +void __cpu_die(unsigned int cpu); #endif #endif /* !(__ASSEMBLY__) */ @@ -65,6 +71,7 @@ extern void __cpu_die(unsigned int cpu); #define hard_smp_processor_id() 0 #define smp_fill_in_sib_core_maps() do { } while (0) #define smp_fetch_global_regs() do { } while (0) +#define smp_fetch_global_pmu() do { } while (0) #endif /* !(CONFIG_SMP) */ diff --git a/arch/sparc/include/asm/sparsemem.h b/arch/sparc/include/asm/sparsemem.h index b99d4e4b6d2..e5e1752d5d7 100644 --- a/arch/sparc/include/asm/sparsemem.h +++ b/arch/sparc/include/asm/sparsemem.h @@ -3,9 +3,11 @@ #ifdef __KERNEL__ +#include <asm/page.h> + #define SECTION_SIZE_BITS 30 -#define MAX_PHYSADDR_BITS 42 -#define MAX_PHYSMEM_BITS 42 +#define MAX_PHYSADDR_BITS MAX_PHYS_ADDRESS_BITS +#define MAX_PHYSMEM_BITS MAX_PHYS_ADDRESS_BITS #endif /* !(__KERNEL__) */ diff --git a/arch/sparc/include/asm/spitfire.h b/arch/sparc/include/asm/spitfire.h index d06a2660175..3fc58691dbd 100644 --- a/arch/sparc/include/asm/spitfire.h +++ b/arch/sparc/include/asm/spitfire.h @@ -45,6 +45,7 @@ #define SUN4V_CHIP_NIAGARA3 0x03 #define SUN4V_CHIP_NIAGARA4 0x04 #define SUN4V_CHIP_NIAGARA5 0x05 +#define SUN4V_CHIP_SPARC64X 0x8a #define SUN4V_CHIP_UNKNOWN 0xff #ifndef __ASSEMBLY__ @@ -61,7 +62,7 @@ extern enum ultra_tlb_layout tlb_type; extern int sun4v_chip_type; extern int cheetah_pcache_forced_on; -extern void cheetah_enable_pcache(void); +void cheetah_enable_pcache(void); #define sparc64_highest_locked_tlbent() \ (tlb_type == spitfire ? \ diff --git a/arch/sparc/include/asm/stacktrace.h b/arch/sparc/include/asm/stacktrace.h index 6cee39adf6d..c30d066f304 100644 --- a/arch/sparc/include/asm/stacktrace.h +++ b/arch/sparc/include/asm/stacktrace.h @@ -1,6 +1,6 @@ #ifndef _SPARC64_STACKTRACE_H #define _SPARC64_STACKTRACE_H -extern void stack_trace_flush(void); +void stack_trace_flush(void); #endif /* _SPARC64_STACKTRACE_H */ diff --git a/arch/sparc/include/asm/starfire.h b/arch/sparc/include/asm/starfire.h index d56ce60a599..c100dc27a0a 100644 --- a/arch/sparc/include/asm/starfire.h +++ b/arch/sparc/include/asm/starfire.h @@ -11,10 +11,10 @@ extern int this_is_starfire; -extern void check_if_starfire(void); -extern int starfire_hard_smp_processor_id(void); -extern void starfire_hookup(int); -extern unsigned int starfire_translate(unsigned long imap, unsigned int upaid); +void check_if_starfire(void); +int starfire_hard_smp_processor_id(void); +void starfire_hookup(int); +unsigned int starfire_translate(unsigned long imap, unsigned int upaid); #endif #endif diff --git a/arch/sparc/include/asm/string_32.h b/arch/sparc/include/asm/string_32.h index 12f67857152..69974e92461 100644 --- a/arch/sparc/include/asm/string_32.h +++ b/arch/sparc/include/asm/string_32.h @@ -15,7 +15,7 @@ #ifdef __KERNEL__ -extern void __memmove(void *,const void *,__kernel_size_t); +void __memmove(void *,const void *,__kernel_size_t); #ifndef EXPORT_SYMTAB_STROPS @@ -40,8 +40,8 @@ extern void __memmove(void *,const void *,__kernel_size_t); #undef memscan #define memscan(__arg0, __char, __arg2) \ ({ \ - extern void *__memscan_zero(void *, size_t); \ - extern void *__memscan_generic(void *, int, size_t); \ + void *__memscan_zero(void *, size_t); \ + void *__memscan_generic(void *, int, size_t); \ void *__retval, *__addr = (__arg0); \ size_t __size = (__arg2); \ \ @@ -54,14 +54,14 @@ extern void __memmove(void *,const void *,__kernel_size_t); }) #define __HAVE_ARCH_MEMCMP -extern int memcmp(const void *,const void *,__kernel_size_t); +int memcmp(const void *,const void *,__kernel_size_t); /* Now the str*() stuff... */ #define __HAVE_ARCH_STRLEN -extern __kernel_size_t strlen(const char *); +__kernel_size_t strlen(const char *); #define __HAVE_ARCH_STRNCMP -extern int strncmp(const char *, const char *, __kernel_size_t); +int strncmp(const char *, const char *, __kernel_size_t); #endif /* !EXPORT_SYMTAB_STROPS */ diff --git a/arch/sparc/include/asm/string_64.h b/arch/sparc/include/asm/string_64.h index 9623bc21315..5936b8ff3c0 100644 --- a/arch/sparc/include/asm/string_64.h +++ b/arch/sparc/include/asm/string_64.h @@ -19,7 +19,7 @@ /* First the mem*() things. */ #define __HAVE_ARCH_MEMMOVE -extern void *memmove(void *, const void *, __kernel_size_t); +void *memmove(void *, const void *, __kernel_size_t); #define __HAVE_ARCH_MEMCPY #define memcpy(t, f, n) __builtin_memcpy(t, f, n) @@ -32,8 +32,8 @@ extern void *memmove(void *, const void *, __kernel_size_t); #undef memscan #define memscan(__arg0, __char, __arg2) \ ({ \ - extern void *__memscan_zero(void *, size_t); \ - extern void *__memscan_generic(void *, int, size_t); \ + void *__memscan_zero(void *, size_t); \ + void *__memscan_generic(void *, int, size_t); \ void *__retval, *__addr = (__arg0); \ size_t __size = (__arg2); \ \ @@ -46,14 +46,14 @@ extern void *memmove(void *, const void *, __kernel_size_t); }) #define __HAVE_ARCH_MEMCMP -extern int memcmp(const void *,const void *,__kernel_size_t); +int memcmp(const void *,const void *,__kernel_size_t); /* Now the str*() stuff... */ #define __HAVE_ARCH_STRLEN -extern __kernel_size_t strlen(const char *); +__kernel_size_t strlen(const char *); #define __HAVE_ARCH_STRNCMP -extern int strncmp(const char *, const char *, __kernel_size_t); +int strncmp(const char *, const char *, __kernel_size_t); #endif /* !EXPORT_SYMTAB_STROPS */ diff --git a/arch/sparc/include/asm/switch_to_32.h b/arch/sparc/include/asm/switch_to_32.h index e32e82b76ee..16f10374feb 100644 --- a/arch/sparc/include/asm/switch_to_32.h +++ b/arch/sparc/include/asm/switch_to_32.h @@ -99,8 +99,8 @@ extern struct thread_info *current_set[NR_CPUS]; "o0", "o1", "o2", "o3", "o7"); \ } while(0) -extern void fpsave(unsigned long *fpregs, unsigned long *fsr, - void *fpqueue, unsigned long *fpqdepth); -extern void synchronize_user_stack(void); +void fpsave(unsigned long *fpregs, unsigned long *fsr, + void *fpqueue, unsigned long *fpqdepth); +void synchronize_user_stack(void); #endif /* __SPARC_SWITCH_TO_H */ diff --git a/arch/sparc/include/asm/switch_to_64.h b/arch/sparc/include/asm/switch_to_64.h index 7923c4a2be3..10e76332dc9 100644 --- a/arch/sparc/include/asm/switch_to_64.h +++ b/arch/sparc/include/asm/switch_to_64.h @@ -18,12 +18,11 @@ do { \ * and 2 stores in this critical code path. -DaveM */ #define switch_to(prev, next, last) \ -do { flush_tlb_pending(); \ - save_and_clear_fpu(); \ +do { save_and_clear_fpu(); \ /* If you are tempted to conditionalize the following */ \ /* so that ASI is only written if it changes, think again. */ \ __asm__ __volatile__("wr %%g0, %0, %%asi" \ - : : "r" (__thread_flag_byte_ptr(task_thread_info(next))[TI_FLAG_BYTE_CURRENT_DS]));\ + : : "r" (task_thread_info(next)->current_ds));\ trap_block[current_thread_info()->cpu].thread = \ task_thread_info(next); \ __asm__ __volatile__( \ @@ -49,8 +48,8 @@ do { flush_tlb_pending(); \ "wrpr %%g0, 14, %%pil\n\t" \ "brz,pt %%o7, switch_to_pc\n\t" \ " mov %%g7, %0\n\t" \ - "sethi %%hi(ret_from_syscall), %%g1\n\t" \ - "jmpl %%g1 + %%lo(ret_from_syscall), %%g0\n\t" \ + "sethi %%hi(ret_from_fork), %%g1\n\t" \ + "jmpl %%g1 + %%lo(ret_from_fork), %%g0\n\t" \ " nop\n\t" \ ".globl switch_to_pc\n\t" \ "switch_to_pc:\n\t" \ @@ -66,7 +65,7 @@ do { flush_tlb_pending(); \ "o0", "o1", "o2", "o3", "o4", "o5", "o7"); \ } while(0) -extern void synchronize_user_stack(void); -extern void fault_in_user_windows(void); +void synchronize_user_stack(void); +void fault_in_user_windows(void); #endif /* __SPARC64_SWITCH_TO_64_H */ diff --git a/arch/sparc/include/asm/syscalls.h b/arch/sparc/include/asm/syscalls.h index 45a43f637a1..b0a0db8ea61 100644 --- a/arch/sparc/include/asm/syscalls.h +++ b/arch/sparc/include/asm/syscalls.h @@ -3,11 +3,9 @@ struct pt_regs; -extern asmlinkage long sparc_do_fork(unsigned long clone_flags, - unsigned long stack_start, - struct pt_regs *regs, - unsigned long stack_size); - -extern asmlinkage int sparc_execve(struct pt_regs *regs); +asmlinkage long sparc_do_fork(unsigned long clone_flags, + unsigned long stack_start, + struct pt_regs *regs, + unsigned long stack_size); #endif /* _SPARC64_SYSCALLS_H */ diff --git a/arch/sparc/include/asm/termbits.h b/arch/sparc/include/asm/termbits.h index 23b10ff08df..948067065ac 100644 --- a/arch/sparc/include/asm/termbits.h +++ b/arch/sparc/include/asm/termbits.h @@ -1,266 +1,8 @@ #ifndef _SPARC_TERMBITS_H #define _SPARC_TERMBITS_H -#include <linux/posix_types.h> +#include <uapi/asm/termbits.h> -typedef unsigned char cc_t; -typedef unsigned int speed_t; - -#if defined(__sparc__) && defined(__arch64__) -typedef unsigned int tcflag_t; -#else -typedef unsigned long tcflag_t; -#endif - -#define NCC 8 -struct termio { - unsigned short c_iflag; /* input mode flags */ - unsigned short c_oflag; /* output mode flags */ - unsigned short c_cflag; /* control mode flags */ - unsigned short c_lflag; /* local mode flags */ - unsigned char c_line; /* line discipline */ - unsigned char c_cc[NCC]; /* control characters */ -}; - -#define NCCS 17 -struct termios { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ -#ifndef __KERNEL__ - cc_t c_cc[NCCS]; /* control characters */ -#else - cc_t c_cc[NCCS+2]; /* kernel needs 2 more to hold vmin/vtime */ -#define SIZEOF_USER_TERMIOS sizeof (struct termios) - (2*sizeof (cc_t)) -#endif -}; - -struct termios2 { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS+2]; /* control characters */ - speed_t c_ispeed; /* input speed */ - speed_t c_ospeed; /* output speed */ -}; - -struct ktermios { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS+2]; /* control characters */ - speed_t c_ispeed; /* input speed */ - speed_t c_ospeed; /* output speed */ -}; - -/* c_cc characters */ -#define VINTR 0 -#define VQUIT 1 -#define VERASE 2 -#define VKILL 3 -#define VEOF 4 -#define VEOL 5 -#define VEOL2 6 -#define VSWTC 7 -#define VSTART 8 -#define VSTOP 9 - - - -#define VSUSP 10 -#define VDSUSP 11 /* SunOS POSIX nicety I do believe... */ -#define VREPRINT 12 -#define VDISCARD 13 -#define VWERASE 14 -#define VLNEXT 15 - -/* Kernel keeps vmin/vtime separated, user apps assume vmin/vtime is - * shared with eof/eol - */ -#ifdef __KERNEL__ #define VMIN 16 #define VTIME 17 -#else -#define VMIN VEOF -#define VTIME VEOL -#endif - -/* c_iflag bits */ -#define IGNBRK 0x00000001 -#define BRKINT 0x00000002 -#define IGNPAR 0x00000004 -#define PARMRK 0x00000008 -#define INPCK 0x00000010 -#define ISTRIP 0x00000020 -#define INLCR 0x00000040 -#define IGNCR 0x00000080 -#define ICRNL 0x00000100 -#define IUCLC 0x00000200 -#define IXON 0x00000400 -#define IXANY 0x00000800 -#define IXOFF 0x00001000 -#define IMAXBEL 0x00002000 -#define IUTF8 0x00004000 - -/* c_oflag bits */ -#define OPOST 0x00000001 -#define OLCUC 0x00000002 -#define ONLCR 0x00000004 -#define OCRNL 0x00000008 -#define ONOCR 0x00000010 -#define ONLRET 0x00000020 -#define OFILL 0x00000040 -#define OFDEL 0x00000080 -#define NLDLY 0x00000100 -#define NL0 0x00000000 -#define NL1 0x00000100 -#define CRDLY 0x00000600 -#define CR0 0x00000000 -#define CR1 0x00000200 -#define CR2 0x00000400 -#define CR3 0x00000600 -#define TABDLY 0x00001800 -#define TAB0 0x00000000 -#define TAB1 0x00000800 -#define TAB2 0x00001000 -#define TAB3 0x00001800 -#define XTABS 0x00001800 -#define BSDLY 0x00002000 -#define BS0 0x00000000 -#define BS1 0x00002000 -#define VTDLY 0x00004000 -#define VT0 0x00000000 -#define VT1 0x00004000 -#define FFDLY 0x00008000 -#define FF0 0x00000000 -#define FF1 0x00008000 -#define PAGEOUT 0x00010000 /* SUNOS specific */ -#define WRAP 0x00020000 /* SUNOS specific */ - -/* c_cflag bit meaning */ -#define CBAUD 0x0000100f -#define B0 0x00000000 /* hang up */ -#define B50 0x00000001 -#define B75 0x00000002 -#define B110 0x00000003 -#define B134 0x00000004 -#define B150 0x00000005 -#define B200 0x00000006 -#define B300 0x00000007 -#define B600 0x00000008 -#define B1200 0x00000009 -#define B1800 0x0000000a -#define B2400 0x0000000b -#define B4800 0x0000000c -#define B9600 0x0000000d -#define B19200 0x0000000e -#define B38400 0x0000000f -#define EXTA B19200 -#define EXTB B38400 -#define CSIZE 0x00000030 -#define CS5 0x00000000 -#define CS6 0x00000010 -#define CS7 0x00000020 -#define CS8 0x00000030 -#define CSTOPB 0x00000040 -#define CREAD 0x00000080 -#define PARENB 0x00000100 -#define PARODD 0x00000200 -#define HUPCL 0x00000400 -#define CLOCAL 0x00000800 -#define CBAUDEX 0x00001000 -/* We'll never see these speeds with the Zilogs, but for completeness... */ -#define BOTHER 0x00001000 -#define B57600 0x00001001 -#define B115200 0x00001002 -#define B230400 0x00001003 -#define B460800 0x00001004 -/* This is what we can do with the Zilogs. */ -#define B76800 0x00001005 -/* This is what we can do with the SAB82532. */ -#define B153600 0x00001006 -#define B307200 0x00001007 -#define B614400 0x00001008 -#define B921600 0x00001009 -/* And these are the rest... */ -#define B500000 0x0000100a -#define B576000 0x0000100b -#define B1000000 0x0000100c -#define B1152000 0x0000100d -#define B1500000 0x0000100e -#define B2000000 0x0000100f -/* These have totally bogus values and nobody uses them - so far. Later on we'd have to use say 0x10000x and - adjust CBAUD constant and drivers accordingly. -#define B2500000 0x00001010 -#define B3000000 0x00001011 -#define B3500000 0x00001012 -#define B4000000 0x00001013 */ -#define CIBAUD 0x100f0000 /* input baud rate (not used) */ -#define CMSPAR 0x40000000 /* mark or space (stick) parity */ -#define CRTSCTS 0x80000000 /* flow control */ - -#define IBSHIFT 16 /* Shift from CBAUD to CIBAUD */ - -/* c_lflag bits */ -#define ISIG 0x00000001 -#define ICANON 0x00000002 -#define XCASE 0x00000004 -#define ECHO 0x00000008 -#define ECHOE 0x00000010 -#define ECHOK 0x00000020 -#define ECHONL 0x00000040 -#define NOFLSH 0x00000080 -#define TOSTOP 0x00000100 -#define ECHOCTL 0x00000200 -#define ECHOPRT 0x00000400 -#define ECHOKE 0x00000800 -#define DEFECHO 0x00001000 /* SUNOS thing, what is it? */ -#define FLUSHO 0x00002000 -#define PENDIN 0x00004000 -#define IEXTEN 0x00008000 -#define EXTPROC 0x00010000 - -/* modem lines */ -#define TIOCM_LE 0x001 -#define TIOCM_DTR 0x002 -#define TIOCM_RTS 0x004 -#define TIOCM_ST 0x008 -#define TIOCM_SR 0x010 -#define TIOCM_CTS 0x020 -#define TIOCM_CAR 0x040 -#define TIOCM_RNG 0x080 -#define TIOCM_DSR 0x100 -#define TIOCM_CD TIOCM_CAR -#define TIOCM_RI TIOCM_RNG -#define TIOCM_OUT1 0x2000 -#define TIOCM_OUT2 0x4000 -#define TIOCM_LOOP 0x8000 - -/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */ -#define TIOCSER_TEMT 0x01 /* Transmitter physically empty */ - - -/* tcflow() and TCXONC use these */ -#define TCOOFF 0 -#define TCOON 1 -#define TCIOFF 2 -#define TCION 3 - -/* tcflush() and TCFLSH use these */ -#define TCIFLUSH 0 -#define TCOFLUSH 1 -#define TCIOFLUSH 2 - -/* tcsetattr uses these */ -#define TCSANOW 0 -#define TCSADRAIN 1 -#define TCSAFLUSH 2 - #endif /* !(_SPARC_TERMBITS_H) */ diff --git a/arch/sparc/include/asm/termios.h b/arch/sparc/include/asm/termios.h index e2f46705a21..0c2414ddd52 100644 --- a/arch/sparc/include/asm/termios.h +++ b/arch/sparc/include/asm/termios.h @@ -1,45 +1,8 @@ #ifndef _SPARC_TERMIOS_H #define _SPARC_TERMIOS_H -#include <asm/ioctls.h> -#include <asm/termbits.h> +#include <uapi/asm/termios.h> -#if defined(__KERNEL__) || defined(__DEFINE_BSD_TERMIOS) -struct sgttyb { - char sg_ispeed; - char sg_ospeed; - char sg_erase; - char sg_kill; - short sg_flags; -}; - -struct tchars { - char t_intrc; - char t_quitc; - char t_startc; - char t_stopc; - char t_eofc; - char t_brkc; -}; - -struct ltchars { - char t_suspc; - char t_dsuspc; - char t_rprntc; - char t_flushc; - char t_werasc; - char t_lnextc; -}; -#endif /* __KERNEL__ */ - -struct winsize { - unsigned short ws_row; - unsigned short ws_col; - unsigned short ws_xpixel; - unsigned short ws_ypixel; -}; - -#ifdef __KERNEL__ /* * c_cc characters in the termio structure. Oh, how I love being @@ -180,6 +143,4 @@ struct winsize { err; \ }) -#endif /* __KERNEL__ */ - #endif /* _SPARC_TERMIOS_H */ diff --git a/arch/sparc/include/asm/thread_info_32.h b/arch/sparc/include/asm/thread_info_32.h index e6cd224506a..96efa7adc22 100644 --- a/arch/sparc/include/asm/thread_info_32.h +++ b/arch/sparc/include/asm/thread_info_32.h @@ -105,8 +105,6 @@ register struct thread_info *current_thread_info_reg asm("g6"); #define TI_W_SAVED 0x250 /* #define TI_RESTART_BLOCK 0x25n */ /* Nobody cares */ -#define PREEMPT_ACTIVE 0x4000000 - /* * thread information flag bit numbers */ @@ -126,7 +124,6 @@ register struct thread_info *current_thread_info_reg asm("g6"); #define _TIF_NOTIFY_RESUME (1<<TIF_NOTIFY_RESUME) #define _TIF_SIGPENDING (1<<TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED) -#define _TIF_RESTORE_SIGMASK (1<<TIF_RESTORE_SIGMASK) #define _TIF_USEDFPU (1<<TIF_USEDFPU) #define _TIF_POLLING_NRFLAG (1<<TIF_POLLING_NRFLAG) diff --git a/arch/sparc/include/asm/thread_info_64.h b/arch/sparc/include/asm/thread_info_64.h index cfa8c38fb9c..a5f01ac6d0f 100644 --- a/arch/sparc/include/asm/thread_info_64.h +++ b/arch/sparc/include/asm/thread_info_64.h @@ -14,12 +14,12 @@ #define TI_FLAG_FAULT_CODE_SHIFT 56 #define TI_FLAG_BYTE_WSTATE 1 #define TI_FLAG_WSTATE_SHIFT 48 -#define TI_FLAG_BYTE_CWP 2 -#define TI_FLAG_CWP_SHIFT 40 -#define TI_FLAG_BYTE_CURRENT_DS 3 -#define TI_FLAG_CURRENT_DS_SHIFT 32 -#define TI_FLAG_BYTE_FPDEPTH 4 -#define TI_FLAG_FPDEPTH_SHIFT 24 +#define TI_FLAG_BYTE_NOERROR 2 +#define TI_FLAG_BYTE_NOERROR_SHIFT 40 +#define TI_FLAG_BYTE_FPDEPTH 3 +#define TI_FLAG_FPDEPTH_SHIFT 32 +#define TI_FLAG_BYTE_CWP 4 +#define TI_FLAG_CWP_SHIFT 24 #define TI_FLAG_BYTE_WSAVED 5 #define TI_FLAG_WSAVED_SHIFT 16 @@ -47,7 +47,7 @@ struct thread_info { struct exec_domain *exec_domain; int preempt_count; /* 0 => preemptable, <0 => BUG */ __u8 new_child; - __u8 syscall_noerror; + __u8 current_ds; __u16 cpu; unsigned long *utraps; @@ -74,9 +74,9 @@ struct thread_info { #define TI_FAULT_CODE (TI_FLAGS + TI_FLAG_BYTE_FAULT_CODE) #define TI_WSTATE (TI_FLAGS + TI_FLAG_BYTE_WSTATE) #define TI_CWP (TI_FLAGS + TI_FLAG_BYTE_CWP) -#define TI_CURRENT_DS (TI_FLAGS + TI_FLAG_BYTE_CURRENT_DS) #define TI_FPDEPTH (TI_FLAGS + TI_FLAG_BYTE_FPDEPTH) #define TI_WSAVED (TI_FLAGS + TI_FLAG_BYTE_WSAVED) +#define TI_SYS_NOERROR (TI_FLAGS + TI_FLAG_BYTE_NOERROR) #define TI_FPSAVED 0x00000010 #define TI_KSP 0x00000018 #define TI_FAULT_ADDR 0x00000020 @@ -84,7 +84,7 @@ struct thread_info { #define TI_EXEC_DOMAIN 0x00000030 #define TI_PRE_COUNT 0x00000038 #define TI_NEW_CHILD 0x0000003c -#define TI_SYS_NOERROR 0x0000003d +#define TI_CURRENT_DS 0x0000003d #define TI_CPU 0x0000003e #define TI_UTRAPS 0x00000040 #define TI_REG_WINDOW 0x00000048 @@ -111,8 +111,6 @@ struct thread_info { #define THREAD_SHIFT PAGE_SHIFT #endif /* PAGE_SHIFT == 13 */ -#define PREEMPT_ACTIVE 0x10000000 - /* * macros/functions for gaining access to the thread information structure */ @@ -121,7 +119,7 @@ struct thread_info { #define INIT_THREAD_INFO(tsk) \ { \ .task = &tsk, \ - .flags = ((unsigned long)ASI_P) << TI_FLAG_CURRENT_DS_SHIFT, \ + .current_ds = ASI_P, \ .exec_domain = &default_exec_domain, \ .preempt_count = INIT_PREEMPT_COUNT, \ .restart_block = { \ @@ -153,13 +151,12 @@ register struct thread_info *current_thread_info_reg asm("g6"); #define set_thread_wstate(val) (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_WSTATE] = (val)) #define get_thread_cwp() (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_CWP]) #define set_thread_cwp(val) (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_CWP] = (val)) -#define get_thread_current_ds() (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_CURRENT_DS]) -#define set_thread_current_ds(val) (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_CURRENT_DS] = (val)) +#define get_thread_noerror() (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_NOERROR]) +#define set_thread_noerror(val) (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_NOERROR] = (val)) #define get_thread_fpdepth() (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_FPDEPTH]) #define set_thread_fpdepth(val) (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_FPDEPTH] = (val)) #define get_thread_wsaved() (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_WSAVED]) #define set_thread_wsaved(val) (__cur_thread_flag_byte_ptr[TI_FLAG_BYTE_WSAVED] = (val)) - #endif /* !(__ASSEMBLY__) */ /* @@ -193,7 +190,7 @@ register struct thread_info *current_thread_info_reg asm("g6"); #define TIF_UNALIGNED 5 /* allowed to do unaligned accesses */ /* flag bit 6 is available */ #define TIF_32BIT 7 /* 32-bit binary */ -/* flag bit 8 is available */ +#define TIF_NOHZ 8 /* in adaptive nohz mode */ #define TIF_SECCOMP 9 /* secure computing */ #define TIF_SYSCALL_AUDIT 10 /* syscall auditing active */ #define TIF_SYSCALL_TRACEPOINT 11 /* syscall tracepoint instrumentation */ @@ -211,6 +208,7 @@ register struct thread_info *current_thread_info_reg asm("g6"); #define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED) #define _TIF_UNALIGNED (1<<TIF_UNALIGNED) #define _TIF_32BIT (1<<TIF_32BIT) +#define _TIF_NOHZ (1<<TIF_NOHZ) #define _TIF_SECCOMP (1<<TIF_SECCOMP) #define _TIF_SYSCALL_AUDIT (1<<TIF_SYSCALL_AUDIT) #define _TIF_SYSCALL_TRACEPOINT (1<<TIF_SYSCALL_TRACEPOINT) @@ -256,6 +254,12 @@ static inline bool test_and_clear_restore_sigmask(void) ti->status &= ~TS_RESTORE_SIGMASK; return true; } + +#define thread32_stack_is_64bit(__SP) (((__SP) & 0x1) != 0) +#define test_thread_64bit_stack(__SP) \ + ((test_thread_flag(TIF_32BIT) && !thread32_stack_is_64bit(__SP)) ? \ + false : true) + #endif /* !__ASSEMBLY__ */ #endif /* __KERNEL__ */ diff --git a/arch/sparc/include/asm/timer_32.h b/arch/sparc/include/asm/timer_32.h index 72f40a546de..f8e708a0aa5 100644 --- a/arch/sparc/include/asm/timer_32.h +++ b/arch/sparc/include/asm/timer_32.h @@ -32,13 +32,13 @@ static inline unsigned int timer_value(unsigned int value) return (value + 1) << TIMER_VALUE_SHIFT; } -extern __volatile__ unsigned int *master_l10_counter; +extern volatile u32 __iomem *master_l10_counter; -extern irqreturn_t notrace timer_interrupt(int dummy, void *dev_id); +irqreturn_t notrace timer_interrupt(int dummy, void *dev_id); #ifdef CONFIG_SMP DECLARE_PER_CPU(struct clock_event_device, sparc32_clockevent); -extern void register_percpu_ce(int cpu); +void register_percpu_ce(int cpu); #endif #endif /* !(_SPARC_TIMER_H) */ diff --git a/arch/sparc/include/asm/timer_64.h b/arch/sparc/include/asm/timer_64.h index ef3c3682deb..fce41503400 100644 --- a/arch/sparc/include/asm/timer_64.h +++ b/arch/sparc/include/asm/timer_64.h @@ -23,8 +23,8 @@ struct sparc64_tick_ops { extern struct sparc64_tick_ops *tick_ops; -extern unsigned long sparc64_get_clock_tick(unsigned int cpu); -extern void __devinit setup_sparc64_timer(void); -extern void __init time_init(void); +unsigned long sparc64_get_clock_tick(unsigned int cpu); +void setup_sparc64_timer(void); +void __init time_init(void); #endif /* _SPARC64_TIMER_H */ diff --git a/arch/sparc/include/asm/tlb_64.h b/arch/sparc/include/asm/tlb_64.h index 190e18913cc..4cb392f75d2 100644 --- a/arch/sparc/include/asm/tlb_64.h +++ b/arch/sparc/include/asm/tlb_64.h @@ -8,19 +8,19 @@ #include <asm/mmu_context.h> #ifdef CONFIG_SMP -extern void smp_flush_tlb_pending(struct mm_struct *, +void smp_flush_tlb_pending(struct mm_struct *, unsigned long, unsigned long *); #endif #ifdef CONFIG_SMP -extern void smp_flush_tlb_mm(struct mm_struct *mm); +void smp_flush_tlb_mm(struct mm_struct *mm); #define do_flush_tlb_mm(mm) smp_flush_tlb_mm(mm) #else #define do_flush_tlb_mm(mm) __flush_tlb_mm(CTX_HWBITS(mm->context), SECONDARY_CONTEXT) #endif -extern void __flush_tlb_pending(unsigned long, unsigned long, unsigned long *); -extern void flush_tlb_pending(void); +void __flush_tlb_pending(unsigned long, unsigned long, unsigned long *); +void flush_tlb_pending(void); #define tlb_start_vma(tlb, vma) do { } while (0) #define tlb_end_vma(tlb, vma) do { } while (0) diff --git a/arch/sparc/include/asm/tlbflush_64.h b/arch/sparc/include/asm/tlbflush_64.h index 2ef46349415..816d8202fa0 100644 --- a/arch/sparc/include/asm/tlbflush_64.h +++ b/arch/sparc/include/asm/tlbflush_64.h @@ -1,7 +1,6 @@ #ifndef _SPARC64_TLBFLUSH_H #define _SPARC64_TLBFLUSH_H -#include <linux/mm.h> #include <asm/mmu_context.h> /* TSB flush operations. */ @@ -11,25 +10,41 @@ struct tlb_batch { struct mm_struct *mm; unsigned long tlb_nr; + unsigned long active; unsigned long vaddrs[TLB_BATCH_NR]; }; -extern void flush_tsb_kernel_range(unsigned long start, unsigned long end); -extern void flush_tsb_user(struct tlb_batch *tb); +void flush_tsb_kernel_range(unsigned long start, unsigned long end); +void flush_tsb_user(struct tlb_batch *tb); +void flush_tsb_user_page(struct mm_struct *mm, unsigned long vaddr); /* TLB flush operations. */ -extern void flush_tlb_pending(void); +static inline void flush_tlb_mm(struct mm_struct *mm) +{ +} -#define flush_tlb_range(vma,start,end) \ - do { (void)(start); flush_tlb_pending(); } while (0) -#define flush_tlb_page(vma,addr) flush_tlb_pending() -#define flush_tlb_mm(mm) flush_tlb_pending() +static inline void flush_tlb_page(struct vm_area_struct *vma, + unsigned long vmaddr) +{ +} -/* Local cpu only. */ -extern void __flush_tlb_all(void); +static inline void flush_tlb_range(struct vm_area_struct *vma, + unsigned long start, unsigned long end) +{ +} + +#define __HAVE_ARCH_ENTER_LAZY_MMU_MODE + +void flush_tlb_pending(void); +void arch_enter_lazy_mmu_mode(void); +void arch_leave_lazy_mmu_mode(void); +#define arch_flush_lazy_mmu_mode() do {} while (0) -extern void __flush_tlb_kernel_range(unsigned long start, unsigned long end); +/* Local cpu only. */ +void __flush_tlb_all(void); +void __flush_tlb_page(unsigned long context, unsigned long vaddr); +void __flush_tlb_kernel_range(unsigned long start, unsigned long end); #ifndef CONFIG_SMP @@ -38,15 +53,24 @@ do { flush_tsb_kernel_range(start,end); \ __flush_tlb_kernel_range(start,end); \ } while (0) +static inline void global_flush_tlb_page(struct mm_struct *mm, unsigned long vaddr) +{ + __flush_tlb_page(CTX_HWBITS(mm->context), vaddr); +} + #else /* CONFIG_SMP */ -extern void smp_flush_tlb_kernel_range(unsigned long start, unsigned long end); +void smp_flush_tlb_kernel_range(unsigned long start, unsigned long end); +void smp_flush_tlb_page(struct mm_struct *mm, unsigned long vaddr); #define flush_tlb_kernel_range(start, end) \ do { flush_tsb_kernel_range(start,end); \ smp_flush_tlb_kernel_range(start, end); \ } while (0) +#define global_flush_tlb_page(mm, vaddr) \ + smp_flush_tlb_page(mm, vaddr) + #endif /* ! CONFIG_SMP */ #endif /* _SPARC64_TLBFLUSH_H */ diff --git a/arch/sparc/include/asm/topology_64.h b/arch/sparc/include/asm/topology_64.h index 1754390a426..ed8f071132e 100644 --- a/arch/sparc/include/asm/topology_64.h +++ b/arch/sparc/include/asm/topology_64.h @@ -18,7 +18,7 @@ static inline int cpu_to_node(int cpu) struct pci_bus; #ifdef CONFIG_PCI -extern int pcibus_to_node(struct pci_bus *pbus); +int pcibus_to_node(struct pci_bus *pbus); #else static inline int pcibus_to_node(struct pci_bus *pbus) { @@ -42,8 +42,6 @@ static inline int pcibus_to_node(struct pci_bus *pbus) #define topology_core_id(cpu) (cpu_data(cpu).core_id) #define topology_core_cpumask(cpu) (&cpu_core_map[cpu]) #define topology_thread_cpumask(cpu) (&per_cpu(cpu_sibling_map, cpu)) -#define mc_capable() (sparc64_multi_core) -#define smt_capable() (sparc64_multi_core) #endif /* CONFIG_SMP */ extern cpumask_t cpu_core_map[NR_CPUS]; diff --git a/arch/sparc/include/asm/trap_block.h b/arch/sparc/include/asm/trap_block.h index 7e26b2db621..6fd4436d32f 100644 --- a/arch/sparc/include/asm/trap_block.h +++ b/arch/sparc/include/asm/trap_block.h @@ -51,11 +51,11 @@ struct trap_per_cpu { unsigned long __per_cpu_base; } __attribute__((aligned(64))); extern struct trap_per_cpu trap_block[NR_CPUS]; -extern void init_cur_cpu_trap(struct thread_info *); -extern void setup_tba(void); +void init_cur_cpu_trap(struct thread_info *); +void setup_tba(void); extern int ncpus_probed; -extern unsigned long real_hard_smp_processor_id(void); +unsigned long real_hard_smp_processor_id(void); struct cpuid_patch_entry { unsigned int addr; diff --git a/arch/sparc/include/asm/traps.h b/arch/sparc/include/asm/traps.h index 3aa62dde343..51abcb1f9b3 100644 --- a/arch/sparc/include/asm/traps.h +++ b/arch/sparc/include/asm/traps.h @@ -3,14 +3,12 @@ * * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu) */ - #ifndef _SPARC_TRAPS_H #define _SPARC_TRAPS_H -#define NUM_SPARC_TRAPS 255 +#include <uapi/asm/traps.h> #ifndef __ASSEMBLY__ -#ifdef __KERNEL__ /* This is for V8 compliant Sparc CPUS */ struct tt_entry { unsigned long inst_one; @@ -22,112 +20,5 @@ struct tt_entry { /* We set this to _start in system setup. */ extern struct tt_entry *sparc_ttable; -#endif /* (__KERNEL__) */ #endif /* !(__ASSEMBLY__) */ - -/* For patching the trap table at boot time, we need to know how to - * form various common Sparc instructions. Thus these macros... - */ - -#define SPARC_MOV_CONST_L3(const) (0xa6102000 | (const&0xfff)) - -/* The following assumes that the branch lies before the place we - * are branching to. This is the case for a trap vector... - * You have been warned. - */ -#define SPARC_BRANCH(dest_addr, inst_addr) \ - (0x10800000 | (((dest_addr-inst_addr)>>2)&0x3fffff)) - -#define SPARC_RD_PSR_L0 (0xa1480000) -#define SPARC_RD_WIM_L3 (0xa7500000) -#define SPARC_NOP (0x01000000) - -/* Various interesting trap levels. */ -/* First, hardware traps. */ -#define SP_TRAP_TFLT 0x1 /* Text fault */ -#define SP_TRAP_II 0x2 /* Illegal Instruction */ -#define SP_TRAP_PI 0x3 /* Privileged Instruction */ -#define SP_TRAP_FPD 0x4 /* Floating Point Disabled */ -#define SP_TRAP_WOVF 0x5 /* Window Overflow */ -#define SP_TRAP_WUNF 0x6 /* Window Underflow */ -#define SP_TRAP_MNA 0x7 /* Memory Address Unaligned */ -#define SP_TRAP_FPE 0x8 /* Floating Point Exception */ -#define SP_TRAP_DFLT 0x9 /* Data Fault */ -#define SP_TRAP_TOF 0xa /* Tag Overflow */ -#define SP_TRAP_WDOG 0xb /* Watchpoint Detected */ -#define SP_TRAP_IRQ1 0x11 /* IRQ level 1 */ -#define SP_TRAP_IRQ2 0x12 /* IRQ level 2 */ -#define SP_TRAP_IRQ3 0x13 /* IRQ level 3 */ -#define SP_TRAP_IRQ4 0x14 /* IRQ level 4 */ -#define SP_TRAP_IRQ5 0x15 /* IRQ level 5 */ -#define SP_TRAP_IRQ6 0x16 /* IRQ level 6 */ -#define SP_TRAP_IRQ7 0x17 /* IRQ level 7 */ -#define SP_TRAP_IRQ8 0x18 /* IRQ level 8 */ -#define SP_TRAP_IRQ9 0x19 /* IRQ level 9 */ -#define SP_TRAP_IRQ10 0x1a /* IRQ level 10 */ -#define SP_TRAP_IRQ11 0x1b /* IRQ level 11 */ -#define SP_TRAP_IRQ12 0x1c /* IRQ level 12 */ -#define SP_TRAP_IRQ13 0x1d /* IRQ level 13 */ -#define SP_TRAP_IRQ14 0x1e /* IRQ level 14 */ -#define SP_TRAP_IRQ15 0x1f /* IRQ level 15 Non-maskable */ -#define SP_TRAP_RACC 0x20 /* Register Access Error ??? */ -#define SP_TRAP_IACC 0x21 /* Instruction Access Error */ -#define SP_TRAP_CPDIS 0x24 /* Co-Processor Disabled */ -#define SP_TRAP_BADFL 0x25 /* Unimplemented Flush Instruction */ -#define SP_TRAP_CPEXP 0x28 /* Co-Processor Exception */ -#define SP_TRAP_DACC 0x29 /* Data Access Error */ -#define SP_TRAP_DIVZ 0x2a /* Divide By Zero */ -#define SP_TRAP_DSTORE 0x2b /* Data Store Error ??? */ -#define SP_TRAP_DMM 0x2c /* Data Access MMU Miss ??? */ -#define SP_TRAP_IMM 0x3c /* Instruction Access MMU Miss ??? */ - -/* Now the Software Traps... */ -#define SP_TRAP_SUNOS 0x80 /* SunOS System Call */ -#define SP_TRAP_SBPT 0x81 /* Software Breakpoint */ -#define SP_TRAP_SDIVZ 0x82 /* Software Divide-by-Zero trap */ -#define SP_TRAP_FWIN 0x83 /* Flush Windows */ -#define SP_TRAP_CWIN 0x84 /* Clean Windows */ -#define SP_TRAP_RCHK 0x85 /* Range Check */ -#define SP_TRAP_FUNA 0x86 /* Fix Unaligned Access */ -#define SP_TRAP_IOWFL 0x87 /* Integer Overflow */ -#define SP_TRAP_SOLARIS 0x88 /* Solaris System Call */ -#define SP_TRAP_NETBSD 0x89 /* NetBSD System Call */ -#define SP_TRAP_LINUX 0x90 /* Linux System Call */ - -/* Names used for compatibility with SunOS */ -#define ST_SYSCALL 0x00 -#define ST_BREAKPOINT 0x01 -#define ST_DIV0 0x02 -#define ST_FLUSH_WINDOWS 0x03 -#define ST_CLEAN_WINDOWS 0x04 -#define ST_RANGE_CHECK 0x05 -#define ST_FIX_ALIGN 0x06 -#define ST_INT_OVERFLOW 0x07 - -/* Special traps... */ -#define SP_TRAP_KBPT1 0xfe /* KADB/PROM Breakpoint one */ -#define SP_TRAP_KBPT2 0xff /* KADB/PROM Breakpoint two */ - -/* Handy Macros */ -/* Is this a trap we never expect to get? */ -#define BAD_TRAP_P(level) \ - ((level > SP_TRAP_WDOG && level < SP_TRAP_IRQ1) || \ - (level > SP_TRAP_IACC && level < SP_TRAP_CPDIS) || \ - (level > SP_TRAP_BADFL && level < SP_TRAP_CPEXP) || \ - (level > SP_TRAP_DMM && level < SP_TRAP_IMM) || \ - (level > SP_TRAP_IMM && level < SP_TRAP_SUNOS) || \ - (level > SP_TRAP_LINUX && level < SP_TRAP_KBPT1)) - -/* Is this a Hardware trap? */ -#define HW_TRAP_P(level) ((level > 0) && (level < SP_TRAP_SUNOS)) - -/* Is this a Software trap? */ -#define SW_TRAP_P(level) ((level >= SP_TRAP_SUNOS) && (level <= SP_TRAP_KBPT2)) - -/* Is this a system call for some OS we know about? */ -#define SCALL_TRAP_P(level) ((level == SP_TRAP_SUNOS) || \ - (level == SP_TRAP_SOLARIS) || \ - (level == SP_TRAP_NETBSD) || \ - (level == SP_TRAP_LINUX)) - #endif /* !(_SPARC_TRAPS_H) */ diff --git a/arch/sparc/include/asm/tsb.h b/arch/sparc/include/asm/tsb.h index 1a8afd1ad04..90916f955ca 100644 --- a/arch/sparc/include/asm/tsb.h +++ b/arch/sparc/include/asm/tsb.h @@ -142,45 +142,75 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end; or REG1, %lo(swapper_pg_dir), REG1; \ sllx VADDR, 64 - (PGDIR_SHIFT + PGDIR_BITS), REG2; \ srlx REG2, 64 - PAGE_SHIFT, REG2; \ - andn REG2, 0x3, REG2; \ - lduw [REG1 + REG2], REG1; \ + andn REG2, 0x7, REG2; \ + ldx [REG1 + REG2], REG1; \ brz,pn REG1, FAIL_LABEL; \ sllx VADDR, 64 - (PMD_SHIFT + PMD_BITS), REG2; \ srlx REG2, 64 - PAGE_SHIFT, REG2; \ - sllx REG1, 11, REG1; \ - andn REG2, 0x3, REG2; \ - lduwa [REG1 + REG2] ASI_PHYS_USE_EC, REG1; \ + andn REG2, 0x7, REG2; \ + ldxa [REG1 + REG2] ASI_PHYS_USE_EC, REG1; \ brz,pn REG1, FAIL_LABEL; \ sllx VADDR, 64 - PMD_SHIFT, REG2; \ srlx REG2, 64 - PAGE_SHIFT, REG2; \ - sllx REG1, 11, REG1; \ andn REG2, 0x7, REG2; \ add REG1, REG2, REG1; - /* Do a user page table walk in MMU globals. Leaves physical PTE - * pointer in REG1. Jumps to FAIL_LABEL on early page table walk - * termination. Physical base of page tables is in PHYS_PGD which - * will not be modified. + /* PMD has been loaded into REG1, interpret the value, seeing + * if it is a HUGE PMD or a normal one. If it is not valid + * then jump to FAIL_LABEL. If it is a HUGE PMD, and it + * translates to a valid PTE, branch to PTE_LABEL. + * + * We have to propagate the 4MB bit of the virtual address + * because we are fabricating 8MB pages using 4MB hw pages. + */ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +#define USER_PGTABLE_CHECK_PMD_HUGE(VADDR, REG1, REG2, FAIL_LABEL, PTE_LABEL) \ + brz,pn REG1, FAIL_LABEL; \ + sethi %uhi(_PAGE_PMD_HUGE), REG2; \ + sllx REG2, 32, REG2; \ + andcc REG1, REG2, %g0; \ + be,pt %xcc, 700f; \ + sethi %hi(4 * 1024 * 1024), REG2; \ + brgez,pn REG1, FAIL_LABEL; \ + andn REG1, REG2, REG1; \ + and VADDR, REG2, REG2; \ + brlz,pt REG1, PTE_LABEL; \ + or REG1, REG2, REG1; \ +700: +#else +#define USER_PGTABLE_CHECK_PMD_HUGE(VADDR, REG1, REG2, FAIL_LABEL, PTE_LABEL) \ + brz,pn REG1, FAIL_LABEL; \ + nop; +#endif + + /* Do a user page table walk in MMU globals. Leaves final, + * valid, PTE value in REG1. Jumps to FAIL_LABEL on early + * page table walk termination or if the PTE is not valid. + * + * Physical base of page tables is in PHYS_PGD which will not + * be modified. * * VADDR will not be clobbered, but REG1 and REG2 will. */ #define USER_PGTABLE_WALK_TL1(VADDR, PHYS_PGD, REG1, REG2, FAIL_LABEL) \ sllx VADDR, 64 - (PGDIR_SHIFT + PGDIR_BITS), REG2; \ srlx REG2, 64 - PAGE_SHIFT, REG2; \ - andn REG2, 0x3, REG2; \ - lduwa [PHYS_PGD + REG2] ASI_PHYS_USE_EC, REG1; \ + andn REG2, 0x7, REG2; \ + ldxa [PHYS_PGD + REG2] ASI_PHYS_USE_EC, REG1; \ brz,pn REG1, FAIL_LABEL; \ sllx VADDR, 64 - (PMD_SHIFT + PMD_BITS), REG2; \ srlx REG2, 64 - PAGE_SHIFT, REG2; \ - sllx REG1, 11, REG1; \ - andn REG2, 0x3, REG2; \ - lduwa [REG1 + REG2] ASI_PHYS_USE_EC, REG1; \ - brz,pn REG1, FAIL_LABEL; \ - sllx VADDR, 64 - PMD_SHIFT, REG2; \ + andn REG2, 0x7, REG2; \ + ldxa [REG1 + REG2] ASI_PHYS_USE_EC, REG1; \ + USER_PGTABLE_CHECK_PMD_HUGE(VADDR, REG1, REG2, FAIL_LABEL, 800f) \ + sllx VADDR, 64 - PMD_SHIFT, REG2; \ srlx REG2, 64 - PAGE_SHIFT, REG2; \ - sllx REG1, 11, REG1; \ andn REG2, 0x7, REG2; \ - add REG1, REG2, REG1; + add REG1, REG2, REG1; \ + ldxa [REG1] ASI_PHYS_USE_EC, REG1; \ + brgez,pn REG1, FAIL_LABEL; \ + nop; \ +800: /* Lookup a OBP mapping on VADDR in the prom_trans[] table at TL>0. * If no entry is found, FAIL_LABEL will be branched to. On success diff --git a/arch/sparc/include/asm/ttable.h b/arch/sparc/include/asm/ttable.h index 48f2807d326..71b5a67522a 100644 --- a/arch/sparc/include/asm/ttable.h +++ b/arch/sparc/include/asm/ttable.h @@ -372,7 +372,9 @@ etrap_spill_fixup_64bit: \ /* Normal 32bit spill */ #define SPILL_2_GENERIC(ASI) \ - srl %sp, 0, %sp; \ + and %sp, 1, %g3; \ + brnz,pn %g3, (. - (128 + 4)); \ + srl %sp, 0, %sp; \ stwa %l0, [%sp + %g0] ASI; \ mov 0x04, %g3; \ stwa %l1, [%sp + %g3] ASI; \ @@ -398,14 +400,16 @@ etrap_spill_fixup_64bit: \ stwa %i6, [%g1 + %g0] ASI; \ stwa %i7, [%g1 + %g3] ASI; \ saved; \ - retry; nop; nop; \ + retry; \ b,a,pt %xcc, spill_fixup_dax; \ b,a,pt %xcc, spill_fixup_mna; \ b,a,pt %xcc, spill_fixup; #define SPILL_2_GENERIC_ETRAP \ etrap_user_spill_32bit: \ - srl %sp, 0, %sp; \ + and %sp, 1, %g3; \ + brnz,pn %g3, etrap_user_spill_64bit; \ + srl %sp, 0, %sp; \ stwa %l0, [%sp + 0x00] %asi; \ stwa %l1, [%sp + 0x04] %asi; \ stwa %l2, [%sp + 0x08] %asi; \ @@ -427,7 +431,7 @@ etrap_user_spill_32bit: \ ba,pt %xcc, etrap_save; \ wrpr %g1, %cwp; \ nop; nop; nop; nop; \ - nop; nop; nop; nop; \ + nop; nop; \ ba,a,pt %xcc, etrap_spill_fixup_32bit; \ ba,a,pt %xcc, etrap_spill_fixup_32bit; \ ba,a,pt %xcc, etrap_spill_fixup_32bit; @@ -592,7 +596,9 @@ user_rtt_fill_64bit: \ /* Normal 32bit fill */ #define FILL_2_GENERIC(ASI) \ - srl %sp, 0, %sp; \ + and %sp, 1, %g3; \ + brnz,pn %g3, (. - (128 + 4)); \ + srl %sp, 0, %sp; \ lduwa [%sp + %g0] ASI, %l0; \ mov 0x04, %g2; \ mov 0x08, %g3; \ @@ -616,14 +622,16 @@ user_rtt_fill_64bit: \ lduwa [%g1 + %g3] ASI, %i6; \ lduwa [%g1 + %g5] ASI, %i7; \ restored; \ - retry; nop; nop; nop; nop; \ + retry; nop; nop; \ b,a,pt %xcc, fill_fixup_dax; \ b,a,pt %xcc, fill_fixup_mna; \ b,a,pt %xcc, fill_fixup; #define FILL_2_GENERIC_RTRAP \ user_rtt_fill_32bit: \ - srl %sp, 0, %sp; \ + and %sp, 1, %g3; \ + brnz,pn %g3, user_rtt_fill_64bit; \ + srl %sp, 0, %sp; \ lduwa [%sp + 0x00] %asi, %l0; \ lduwa [%sp + 0x04] %asi, %l1; \ lduwa [%sp + 0x08] %asi, %l2; \ @@ -643,7 +651,7 @@ user_rtt_fill_32bit: \ ba,pt %xcc, user_rtt_pre_restore; \ restored; \ nop; nop; nop; nop; nop; \ - nop; nop; nop; nop; nop; \ + nop; nop; nop; \ ba,a,pt %xcc, user_rtt_fill_fixup; \ ba,a,pt %xcc, user_rtt_fill_fixup; \ ba,a,pt %xcc, user_rtt_fill_fixup; diff --git a/arch/sparc/include/asm/types.h b/arch/sparc/include/asm/types.h deleted file mode 100644 index 383d156cde9..00000000000 --- a/arch/sparc/include/asm/types.h +++ /dev/null @@ -1,17 +0,0 @@ -#ifndef _SPARC_TYPES_H -#define _SPARC_TYPES_H -/* - * This file is never included by application software unless - * explicitly requested (e.g., via linux/types.h) in which case the - * application is Linux specific so (user-) name space pollution is - * not a major issue. However, for interoperability, libraries still - * need to be careful to avoid a name clashes. - */ - -#if defined(__sparc__) - -#include <asm-generic/int-ll64.h> - -#endif /* defined(__sparc__) */ - -#endif /* defined(_SPARC_TYPES_H) */ diff --git a/arch/sparc/include/asm/uaccess.h b/arch/sparc/include/asm/uaccess.h index 0167d26d0d1..bd56c28fff9 100644 --- a/arch/sparc/include/asm/uaccess.h +++ b/arch/sparc/include/asm/uaccess.h @@ -9,6 +9,6 @@ #define user_addr_max() \ (segment_eq(get_fs(), USER_DS) ? TASK_SIZE : ~0UL) -extern long strncpy_from_user(char *dest, const char __user *src, long count); +long strncpy_from_user(char *dest, const char __user *src, long count); #endif diff --git a/arch/sparc/include/asm/uaccess_32.h b/arch/sparc/include/asm/uaccess_32.h index 53a28dd59f5..9634d086fc5 100644 --- a/arch/sparc/include/asm/uaccess_32.h +++ b/arch/sparc/include/asm/uaccess_32.h @@ -78,9 +78,9 @@ struct exception_table_entry }; /* Returns 0 if exception not found and fixup otherwise. */ -extern unsigned long search_extables_range(unsigned long addr, unsigned long *g2); +unsigned long search_extables_range(unsigned long addr, unsigned long *g2); -extern void __ret_efault(void); +void __ret_efault(void); /* Uh, these should become the main single-value transfer routines.. * They automatically use the right size if we just have the right @@ -152,7 +152,7 @@ __asm__ __volatile__( \ : "=&r" (ret) : "r" (x), "m" (*__m(addr)), \ "i" (-EFAULT)) -extern int __put_user_bad(void); +int __put_user_bad(void); #define __get_user_check(x,addr,size,type) ({ \ register int __gu_ret; \ @@ -244,9 +244,9 @@ __asm__ __volatile__( \ ".previous\n\t" \ : "=&r" (x) : "m" (*__m(addr)), "i" (retval)) -extern int __get_user_bad(void); +int __get_user_bad(void); -extern unsigned long __copy_user(void __user *to, const void __user *from, unsigned long size); +unsigned long __copy_user(void __user *to, const void __user *from, unsigned long size); static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) { @@ -306,8 +306,8 @@ static inline unsigned long clear_user(void __user *addr, unsigned long n) return n; } -extern __must_check long strlen_user(const char __user *str); -extern __must_check long strnlen_user(const char __user *str, long n); +__must_check long strlen_user(const char __user *str); +__must_check long strnlen_user(const char __user *str, long n); #endif /* __ASSEMBLY__ */ diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h index 7c831d848b4..c990a5e577f 100644 --- a/arch/sparc/include/asm/uaccess_64.h +++ b/arch/sparc/include/asm/uaccess_64.h @@ -38,14 +38,14 @@ #define VERIFY_READ 0 #define VERIFY_WRITE 1 -#define get_fs() ((mm_segment_t) { get_thread_current_ds() }) +#define get_fs() ((mm_segment_t){(current_thread_info()->current_ds)}) #define get_ds() (KERNEL_DS) #define segment_eq(a,b) ((a).seg == (b).seg) #define set_fs(val) \ do { \ - set_thread_current_ds((val).seg); \ + current_thread_info()->current_ds =(val).seg; \ __asm__ __volatile__ ("wr %%g0, %0, %%asi" : : "r" ((val).seg)); \ } while(0) @@ -76,8 +76,8 @@ struct exception_table_entry { unsigned int insn, fixup; }; -extern void __ret_efault(void); -extern void __retl_efault(void); +void __ret_efault(void); +void __retl_efault(void); /* Uh, these should become the main single-value transfer routines.. * They automatically use the right size if we just have the right @@ -134,7 +134,7 @@ __asm__ __volatile__( \ : "=r" (ret) : "r" (x), "r" (__m(addr)), \ "i" (-EFAULT)) -extern int __put_user_bad(void); +int __put_user_bad(void); #define __get_user_nocheck(data,addr,size,type) ({ \ register int __gu_ret; \ @@ -204,13 +204,13 @@ __asm__ __volatile__( \ ".previous\n\t" \ : "=r" (x) : "r" (__m(addr)), "i" (retval)) -extern int __get_user_bad(void); +int __get_user_bad(void); -extern unsigned long __must_check ___copy_from_user(void *to, - const void __user *from, - unsigned long size); -extern unsigned long copy_from_user_fixup(void *to, const void __user *from, - unsigned long size); +unsigned long __must_check ___copy_from_user(void *to, + const void __user *from, + unsigned long size); +unsigned long copy_from_user_fixup(void *to, const void __user *from, + unsigned long size); static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long size) { @@ -223,11 +223,11 @@ copy_from_user(void *to, const void __user *from, unsigned long size) } #define __copy_from_user copy_from_user -extern unsigned long __must_check ___copy_to_user(void __user *to, - const void *from, - unsigned long size); -extern unsigned long copy_to_user_fixup(void __user *to, const void *from, - unsigned long size); +unsigned long __must_check ___copy_to_user(void __user *to, + const void *from, + unsigned long size); +unsigned long copy_to_user_fixup(void __user *to, const void *from, + unsigned long size); static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long size) { @@ -239,11 +239,11 @@ copy_to_user(void __user *to, const void *from, unsigned long size) } #define __copy_to_user copy_to_user -extern unsigned long __must_check ___copy_in_user(void __user *to, - const void __user *from, - unsigned long size); -extern unsigned long copy_in_user_fixup(void __user *to, void __user *from, - unsigned long size); +unsigned long __must_check ___copy_in_user(void __user *to, + const void __user *from, + unsigned long size); +unsigned long copy_in_user_fixup(void __user *to, void __user *from, + unsigned long size); static inline unsigned long __must_check copy_in_user(void __user *to, void __user *from, unsigned long size) { @@ -255,15 +255,20 @@ copy_in_user(void __user *to, void __user *from, unsigned long size) } #define __copy_in_user copy_in_user -extern unsigned long __must_check __clear_user(void __user *, unsigned long); +unsigned long __must_check __clear_user(void __user *, unsigned long); #define clear_user __clear_user -extern __must_check long strlen_user(const char __user *str); -extern __must_check long strnlen_user(const char __user *str, long n); +__must_check long strlen_user(const char __user *str); +__must_check long strnlen_user(const char __user *str, long n); -#define __copy_to_user_inatomic ___copy_to_user -#define __copy_from_user_inatomic ___copy_from_user +#define __copy_to_user_inatomic __copy_to_user +#define __copy_from_user_inatomic __copy_from_user + +struct pt_regs; +unsigned long compute_effective_address(struct pt_regs *, + unsigned int insn, + unsigned int rd); #endif /* __ASSEMBLY__ */ diff --git a/arch/sparc/include/asm/unistd.h b/arch/sparc/include/asm/unistd.h index c7cb0af0eb5..0aac1e8f296 100644 --- a/arch/sparc/include/asm/unistd.h +++ b/arch/sparc/include/asm/unistd.h @@ -1,6 +1,3 @@ -#ifndef _SPARC_UNISTD_H -#define _SPARC_UNISTD_H - /* * System calls under the Sparc. * @@ -14,422 +11,20 @@ * * Copyright (C) 1995 Adrian M. Rodriguez (adrian@remus.rutgers.edu) */ -#ifndef __32bit_syscall_numbers__ -#ifndef __arch64__ -#define __32bit_syscall_numbers__ -#endif -#endif +#ifndef _SPARC_UNISTD_H +#define _SPARC_UNISTD_H + +#include <uapi/asm/unistd.h> -#define __NR_restart_syscall 0 /* Linux Specific */ -#define __NR_exit 1 /* Common */ -#define __NR_fork 2 /* Common */ -#define __NR_read 3 /* Common */ -#define __NR_write 4 /* Common */ -#define __NR_open 5 /* Common */ -#define __NR_close 6 /* Common */ -#define __NR_wait4 7 /* Common */ -#define __NR_creat 8 /* Common */ -#define __NR_link 9 /* Common */ -#define __NR_unlink 10 /* Common */ -#define __NR_execv 11 /* SunOS Specific */ -#define __NR_chdir 12 /* Common */ -#define __NR_chown 13 /* Common */ -#define __NR_mknod 14 /* Common */ -#define __NR_chmod 15 /* Common */ -#define __NR_lchown 16 /* Common */ -#define __NR_brk 17 /* Common */ -#define __NR_perfctr 18 /* Performance counter operations */ -#define __NR_lseek 19 /* Common */ -#define __NR_getpid 20 /* Common */ -#define __NR_capget 21 /* Linux Specific */ -#define __NR_capset 22 /* Linux Specific */ -#define __NR_setuid 23 /* Implemented via setreuid in SunOS */ -#define __NR_getuid 24 /* Common */ -#define __NR_vmsplice 25 /* ENOSYS under SunOS */ -#define __NR_ptrace 26 /* Common */ -#define __NR_alarm 27 /* Implemented via setitimer in SunOS */ -#define __NR_sigaltstack 28 /* Common */ -#define __NR_pause 29 /* Is sigblock(0)->sigpause() in SunOS */ -#define __NR_utime 30 /* Implemented via utimes() under SunOS */ -#ifdef __32bit_syscall_numbers__ -#define __NR_lchown32 31 /* Linux sparc32 specific */ -#define __NR_fchown32 32 /* Linux sparc32 specific */ -#endif -#define __NR_access 33 /* Common */ -#define __NR_nice 34 /* Implemented via get/setpriority() in SunOS */ -#ifdef __32bit_syscall_numbers__ -#define __NR_chown32 35 /* Linux sparc32 specific */ -#endif -#define __NR_sync 36 /* Common */ -#define __NR_kill 37 /* Common */ -#define __NR_stat 38 /* Common */ -#define __NR_sendfile 39 /* Linux Specific */ -#define __NR_lstat 40 /* Common */ -#define __NR_dup 41 /* Common */ -#define __NR_pipe 42 /* Common */ -#define __NR_times 43 /* Implemented via getrusage() in SunOS */ -#ifdef __32bit_syscall_numbers__ -#define __NR_getuid32 44 /* Linux sparc32 specific */ -#endif -#define __NR_umount2 45 /* Linux Specific */ -#define __NR_setgid 46 /* Implemented via setregid() in SunOS */ -#define __NR_getgid 47 /* Common */ -#define __NR_signal 48 /* Implemented via sigvec() in SunOS */ -#define __NR_geteuid 49 /* SunOS calls getuid() */ -#define __NR_getegid 50 /* SunOS calls getgid() */ -#define __NR_acct 51 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_getgid32 53 /* Linux sparc32 specific */ -#else -#define __NR_memory_ordering 52 /* Linux Specific */ -#endif -#define __NR_ioctl 54 /* Common */ -#define __NR_reboot 55 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_mmap2 56 /* Linux sparc32 Specific */ -#endif -#define __NR_symlink 57 /* Common */ -#define __NR_readlink 58 /* Common */ -#define __NR_execve 59 /* Common */ -#define __NR_umask 60 /* Common */ -#define __NR_chroot 61 /* Common */ -#define __NR_fstat 62 /* Common */ -#define __NR_fstat64 63 /* Linux Specific */ -#define __NR_getpagesize 64 /* Common */ -#define __NR_msync 65 /* Common in newer 1.3.x revs... */ -#define __NR_vfork 66 /* Common */ -#define __NR_pread64 67 /* Linux Specific */ -#define __NR_pwrite64 68 /* Linux Specific */ -#ifdef __32bit_syscall_numbers__ -#define __NR_geteuid32 69 /* Linux sparc32, sbrk under SunOS */ -#define __NR_getegid32 70 /* Linux sparc32, sstk under SunOS */ -#endif -#define __NR_mmap 71 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_setreuid32 72 /* Linux sparc32, vadvise under SunOS */ -#endif -#define __NR_munmap 73 /* Common */ -#define __NR_mprotect 74 /* Common */ -#define __NR_madvise 75 /* Common */ -#define __NR_vhangup 76 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_truncate64 77 /* Linux sparc32 Specific */ -#endif -#define __NR_mincore 78 /* Common */ -#define __NR_getgroups 79 /* Common */ -#define __NR_setgroups 80 /* Common */ -#define __NR_getpgrp 81 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_setgroups32 82 /* Linux sparc32, setpgrp under SunOS */ -#endif -#define __NR_setitimer 83 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_ftruncate64 84 /* Linux sparc32 Specific */ -#endif -#define __NR_swapon 85 /* Common */ -#define __NR_getitimer 86 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_setuid32 87 /* Linux sparc32, gethostname under SunOS */ -#endif -#define __NR_sethostname 88 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_setgid32 89 /* Linux sparc32, getdtablesize under SunOS */ -#endif -#define __NR_dup2 90 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_setfsuid32 91 /* Linux sparc32, getdopt under SunOS */ -#endif -#define __NR_fcntl 92 /* Common */ -#define __NR_select 93 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_setfsgid32 94 /* Linux sparc32, setdopt under SunOS */ -#endif -#define __NR_fsync 95 /* Common */ -#define __NR_setpriority 96 /* Common */ -#define __NR_socket 97 /* Common */ -#define __NR_connect 98 /* Common */ -#define __NR_accept 99 /* Common */ -#define __NR_getpriority 100 /* Common */ -#define __NR_rt_sigreturn 101 /* Linux Specific */ -#define __NR_rt_sigaction 102 /* Linux Specific */ -#define __NR_rt_sigprocmask 103 /* Linux Specific */ -#define __NR_rt_sigpending 104 /* Linux Specific */ -#define __NR_rt_sigtimedwait 105 /* Linux Specific */ -#define __NR_rt_sigqueueinfo 106 /* Linux Specific */ -#define __NR_rt_sigsuspend 107 /* Linux Specific */ -#ifdef __32bit_syscall_numbers__ -#define __NR_setresuid32 108 /* Linux Specific, sigvec under SunOS */ -#define __NR_getresuid32 109 /* Linux Specific, sigblock under SunOS */ -#define __NR_setresgid32 110 /* Linux Specific, sigsetmask under SunOS */ -#define __NR_getresgid32 111 /* Linux Specific, sigpause under SunOS */ -#define __NR_setregid32 112 /* Linux sparc32, sigstack under SunOS */ -#else -#define __NR_setresuid 108 /* Linux Specific, sigvec under SunOS */ -#define __NR_getresuid 109 /* Linux Specific, sigblock under SunOS */ -#define __NR_setresgid 110 /* Linux Specific, sigsetmask under SunOS */ -#define __NR_getresgid 111 /* Linux Specific, sigpause under SunOS */ -#endif -#define __NR_recvmsg 113 /* Common */ -#define __NR_sendmsg 114 /* Common */ -#ifdef __32bit_syscall_numbers__ -#define __NR_getgroups32 115 /* Linux sparc32, vtrace under SunOS */ -#endif -#define __NR_gettimeofday 116 /* Common */ -#define __NR_getrusage 117 /* Common */ -#define __NR_getsockopt 118 /* Common */ -#define __NR_getcwd 119 /* Linux Specific */ -#define __NR_readv 120 /* Common */ -#define __NR_writev 121 /* Common */ -#define __NR_settimeofday 122 /* Common */ -#define __NR_fchown 123 /* Common */ -#define __NR_fchmod 124 /* Common */ -#define __NR_recvfrom 125 /* Common */ -#define __NR_setreuid 126 /* Common */ -#define __NR_setregid 127 /* Common */ -#define __NR_rename 128 /* Common */ -#define __NR_truncate 129 /* Common */ -#define __NR_ftruncate 130 /* Common */ -#define __NR_flock 131 /* Common */ -#define __NR_lstat64 132 /* Linux Specific */ -#define __NR_sendto 133 /* Common */ -#define __NR_shutdown 134 /* Common */ -#define __NR_socketpair 135 /* Common */ -#define __NR_mkdir 136 /* Common */ -#define __NR_rmdir 137 /* Common */ -#define __NR_utimes 138 /* SunOS Specific */ -#define __NR_stat64 139 /* Linux Specific */ -#define __NR_sendfile64 140 /* adjtime under SunOS */ -#define __NR_getpeername 141 /* Common */ -#define __NR_futex 142 /* gethostid under SunOS */ -#define __NR_gettid 143 /* ENOSYS under SunOS */ -#define __NR_getrlimit 144 /* Common */ -#define __NR_setrlimit 145 /* Common */ -#define __NR_pivot_root 146 /* Linux Specific, killpg under SunOS */ -#define __NR_prctl 147 /* ENOSYS under SunOS */ -#define __NR_pciconfig_read 148 /* ENOSYS under SunOS */ -#define __NR_pciconfig_write 149 /* ENOSYS under SunOS */ -#define __NR_getsockname 150 /* Common */ -#define __NR_inotify_init 151 /* Linux specific */ -#define __NR_inotify_add_watch 152 /* Linux specific */ -#define __NR_poll 153 /* Common */ -#define __NR_getdents64 154 /* Linux specific */ -#ifdef __32bit_syscall_numbers__ -#define __NR_fcntl64 155 /* Linux sparc32 Specific */ -#endif -#define __NR_inotify_rm_watch 156 /* Linux specific */ -#define __NR_statfs 157 /* Common */ -#define __NR_fstatfs 158 /* Common */ -#define __NR_umount 159 /* Common */ -#define __NR_sched_set_affinity 160 /* Linux specific, async_daemon under SunOS */ -#define __NR_sched_get_affinity 161 /* Linux specific, getfh under SunOS */ -#define __NR_getdomainname 162 /* SunOS Specific */ -#define __NR_setdomainname 163 /* Common */ -#ifndef __32bit_syscall_numbers__ -#define __NR_utrap_install 164 /* SYSV ABI/v9 required */ -#endif -#define __NR_quotactl 165 /* Common */ -#define __NR_set_tid_address 166 /* Linux specific, exportfs under SunOS */ -#define __NR_mount 167 /* Common */ -#define __NR_ustat 168 /* Common */ -#define __NR_setxattr 169 /* SunOS: semsys */ -#define __NR_lsetxattr 170 /* SunOS: msgsys */ -#define __NR_fsetxattr 171 /* SunOS: shmsys */ -#define __NR_getxattr 172 /* SunOS: auditsys */ -#define __NR_lgetxattr 173 /* SunOS: rfssys */ -#define __NR_getdents 174 /* Common */ -#define __NR_setsid 175 /* Common */ -#define __NR_fchdir 176 /* Common */ -#define __NR_fgetxattr 177 /* SunOS: fchroot */ -#define __NR_listxattr 178 /* SunOS: vpixsys */ -#define __NR_llistxattr 179 /* SunOS: aioread */ -#define __NR_flistxattr 180 /* SunOS: aiowrite */ -#define __NR_removexattr 181 /* SunOS: aiowait */ -#define __NR_lremovexattr 182 /* SunOS: aiocancel */ -#define __NR_sigpending 183 /* Common */ -#define __NR_query_module 184 /* Linux Specific */ -#define __NR_setpgid 185 /* Common */ -#define __NR_fremovexattr 186 /* SunOS: pathconf */ -#define __NR_tkill 187 /* SunOS: fpathconf */ -#define __NR_exit_group 188 /* Linux specific, sysconf undef SunOS */ -#define __NR_uname 189 /* Linux Specific */ -#define __NR_init_module 190 /* Linux Specific */ -#define __NR_personality 191 /* Linux Specific */ -#define __NR_remap_file_pages 192 /* Linux Specific */ -#define __NR_epoll_create 193 /* Linux Specific */ -#define __NR_epoll_ctl 194 /* Linux Specific */ -#define __NR_epoll_wait 195 /* Linux Specific */ -#define __NR_ioprio_set 196 /* Linux Specific */ -#define __NR_getppid 197 /* Linux Specific */ -#define __NR_sigaction 198 /* Linux Specific */ -#define __NR_sgetmask 199 /* Linux Specific */ -#define __NR_ssetmask 200 /* Linux Specific */ -#define __NR_sigsuspend 201 /* Linux Specific */ -#define __NR_oldlstat 202 /* Linux Specific */ -#define __NR_uselib 203 /* Linux Specific */ -#define __NR_readdir 204 /* Linux Specific */ -#define __NR_readahead 205 /* Linux Specific */ -#define __NR_socketcall 206 /* Linux Specific */ -#define __NR_syslog 207 /* Linux Specific */ -#define __NR_lookup_dcookie 208 /* Linux Specific */ -#define __NR_fadvise64 209 /* Linux Specific */ -#define __NR_fadvise64_64 210 /* Linux Specific */ -#define __NR_tgkill 211 /* Linux Specific */ -#define __NR_waitpid 212 /* Linux Specific */ -#define __NR_swapoff 213 /* Linux Specific */ -#define __NR_sysinfo 214 /* Linux Specific */ -#define __NR_ipc 215 /* Linux Specific */ -#define __NR_sigreturn 216 /* Linux Specific */ -#define __NR_clone 217 /* Linux Specific */ -#define __NR_ioprio_get 218 /* Linux Specific */ -#define __NR_adjtimex 219 /* Linux Specific */ -#define __NR_sigprocmask 220 /* Linux Specific */ -#define __NR_create_module 221 /* Linux Specific */ -#define __NR_delete_module 222 /* Linux Specific */ -#define __NR_get_kernel_syms 223 /* Linux Specific */ -#define __NR_getpgid 224 /* Linux Specific */ -#define __NR_bdflush 225 /* Linux Specific */ -#define __NR_sysfs 226 /* Linux Specific */ -#define __NR_afs_syscall 227 /* Linux Specific */ -#define __NR_setfsuid 228 /* Linux Specific */ -#define __NR_setfsgid 229 /* Linux Specific */ -#define __NR__newselect 230 /* Linux Specific */ #ifdef __32bit_syscall_numbers__ -#define __NR_time 231 /* Linux Specific */ #else -#ifdef __KERNEL__ #define __NR_time 231 /* Linux sparc32 */ #endif -#endif -#define __NR_splice 232 /* Linux Specific */ -#define __NR_stime 233 /* Linux Specific */ -#define __NR_statfs64 234 /* Linux Specific */ -#define __NR_fstatfs64 235 /* Linux Specific */ -#define __NR__llseek 236 /* Linux Specific */ -#define __NR_mlock 237 -#define __NR_munlock 238 -#define __NR_mlockall 239 -#define __NR_munlockall 240 -#define __NR_sched_setparam 241 -#define __NR_sched_getparam 242 -#define __NR_sched_setscheduler 243 -#define __NR_sched_getscheduler 244 -#define __NR_sched_yield 245 -#define __NR_sched_get_priority_max 246 -#define __NR_sched_get_priority_min 247 -#define __NR_sched_rr_get_interval 248 -#define __NR_nanosleep 249 -#define __NR_mremap 250 -#define __NR__sysctl 251 -#define __NR_getsid 252 -#define __NR_fdatasync 253 -#define __NR_nfsservctl 254 -#define __NR_sync_file_range 255 -#define __NR_clock_settime 256 -#define __NR_clock_gettime 257 -#define __NR_clock_getres 258 -#define __NR_clock_nanosleep 259 -#define __NR_sched_getaffinity 260 -#define __NR_sched_setaffinity 261 -#define __NR_timer_settime 262 -#define __NR_timer_gettime 263 -#define __NR_timer_getoverrun 264 -#define __NR_timer_delete 265 -#define __NR_timer_create 266 -/* #define __NR_vserver 267 Reserved for VSERVER */ -#define __NR_io_setup 268 -#define __NR_io_destroy 269 -#define __NR_io_submit 270 -#define __NR_io_cancel 271 -#define __NR_io_getevents 272 -#define __NR_mq_open 273 -#define __NR_mq_unlink 274 -#define __NR_mq_timedsend 275 -#define __NR_mq_timedreceive 276 -#define __NR_mq_notify 277 -#define __NR_mq_getsetattr 278 -#define __NR_waitid 279 -#define __NR_tee 280 -#define __NR_add_key 281 -#define __NR_request_key 282 -#define __NR_keyctl 283 -#define __NR_openat 284 -#define __NR_mkdirat 285 -#define __NR_mknodat 286 -#define __NR_fchownat 287 -#define __NR_futimesat 288 -#define __NR_fstatat64 289 -#define __NR_unlinkat 290 -#define __NR_renameat 291 -#define __NR_linkat 292 -#define __NR_symlinkat 293 -#define __NR_readlinkat 294 -#define __NR_fchmodat 295 -#define __NR_faccessat 296 -#define __NR_pselect6 297 -#define __NR_ppoll 298 -#define __NR_unshare 299 -#define __NR_set_robust_list 300 -#define __NR_get_robust_list 301 -#define __NR_migrate_pages 302 -#define __NR_mbind 303 -#define __NR_get_mempolicy 304 -#define __NR_set_mempolicy 305 -#define __NR_kexec_load 306 -#define __NR_move_pages 307 -#define __NR_getcpu 308 -#define __NR_epoll_pwait 309 -#define __NR_utimensat 310 -#define __NR_signalfd 311 -#define __NR_timerfd_create 312 -#define __NR_eventfd 313 -#define __NR_fallocate 314 -#define __NR_timerfd_settime 315 -#define __NR_timerfd_gettime 316 -#define __NR_signalfd4 317 -#define __NR_eventfd2 318 -#define __NR_epoll_create1 319 -#define __NR_dup3 320 -#define __NR_pipe2 321 -#define __NR_inotify_init1 322 -#define __NR_accept4 323 -#define __NR_preadv 324 -#define __NR_pwritev 325 -#define __NR_rt_tgsigqueueinfo 326 -#define __NR_perf_event_open 327 -#define __NR_recvmmsg 328 -#define __NR_fanotify_init 329 -#define __NR_fanotify_mark 330 -#define __NR_prlimit64 331 -#define __NR_name_to_handle_at 332 -#define __NR_open_by_handle_at 333 -#define __NR_clock_adjtime 334 -#define __NR_syncfs 335 -#define __NR_sendmmsg 336 -#define __NR_setns 337 -#define __NR_process_vm_readv 338 -#define __NR_process_vm_writev 339 - -#define NR_syscalls 340 - -#ifdef __32bit_syscall_numbers__ -/* Sparc 32-bit only has the "setresuid32", "getresuid32" variants, - * it never had the plain ones and there is no value to adding those - * old versions into the syscall table. - */ -#define __IGNORE_setresuid -#define __IGNORE_getresuid -#define __IGNORE_setresgid -#define __IGNORE_getresgid -#endif - -#ifdef __KERNEL__ -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_ALARM #define __ARCH_WANT_SYS_GETHOSTNAME #define __ARCH_WANT_SYS_PAUSE -#define __ARCH_WANT_SYS_SGETMASK #define __ARCH_WANT_SYS_SIGNAL #define __ARCH_WANT_SYS_TIME #define __ARCH_WANT_SYS_UTIME @@ -442,21 +37,11 @@ #define __ARCH_WANT_SYS_OLDUMOUNT #define __ARCH_WANT_SYS_SIGPENDING #define __ARCH_WANT_SYS_SIGPROCMASK -#define __ARCH_WANT_SYS_RT_SIGSUSPEND #ifdef __32bit_syscall_numbers__ #define __ARCH_WANT_SYS_IPC #else #define __ARCH_WANT_COMPAT_SYS_TIME -#define __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND +#define __ARCH_WANT_COMPAT_SYS_SENDFILE #endif -/* - * "Conditional" syscalls - * - * What we want is __attribute__((weak,alias("sys_ni_syscall"))), - * but it doesn't work on all toolchains, so we just do it by hand - */ -#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall") - -#endif /* __KERNEL__ */ #endif /* _SPARC_UNISTD_H */ diff --git a/arch/sparc/include/asm/vaddrs.h b/arch/sparc/include/asm/vaddrs.h index da6535d88a7..c3dbcf90203 100644 --- a/arch/sparc/include/asm/vaddrs.h +++ b/arch/sparc/include/asm/vaddrs.h @@ -30,6 +30,28 @@ */ #define SRMMU_NOCACHE_ALCRATIO 64 /* 256 pages per 64MB of system RAM */ +#ifndef __ASSEMBLY__ +#include <asm/kmap_types.h> + +enum fixed_addresses { + FIX_HOLE, +#ifdef CONFIG_HIGHMEM + FIX_KMAP_BEGIN, + FIX_KMAP_END = (KM_TYPE_NR * NR_CPUS), +#endif + __end_of_fixed_addresses +}; +#endif + +/* Leave one empty page between IO pages at 0xfd000000 and + * the top of the fixmap. + */ +#define FIXADDR_TOP (0xfcfff000UL) +#define FIXADDR_SIZE ((FIX_KMAP_END + 1) << PAGE_SHIFT) +#define FIXADDR_START (FIXADDR_TOP - FIXADDR_SIZE) + +#define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT)) + #define SUN4M_IOBASE_VADDR 0xfd000000 /* Base for mapping pages */ #define IOBASE_VADDR 0xfe000000 #define IOBASE_END 0xfe600000 diff --git a/arch/sparc/include/asm/vio.h b/arch/sparc/include/asm/vio.h index 432afa83886..e0f6c399f1d 100644 --- a/arch/sparc/include/asm/vio.h +++ b/arch/sparc/include/asm/vio.h @@ -372,14 +372,14 @@ do { if (vio->debug & VIO_DEBUG_##TYPE) \ vio->vdev->channel_id, ## a); \ } while (0) -extern int __vio_register_driver(struct vio_driver *drv, struct module *owner, +int __vio_register_driver(struct vio_driver *drv, struct module *owner, const char *mod_name); /* * vio_register_driver must be a macro so that KBUILD_MODNAME can be expanded */ #define vio_register_driver(driver) \ __vio_register_driver(driver, THIS_MODULE, KBUILD_MODNAME) -extern void vio_unregister_driver(struct vio_driver *drv); +void vio_unregister_driver(struct vio_driver *drv); static inline struct vio_driver *to_vio_driver(struct device_driver *drv) { @@ -391,21 +391,21 @@ static inline struct vio_dev *to_vio_dev(struct device *dev) return container_of(dev, struct vio_dev, dev); } -extern int vio_ldc_send(struct vio_driver_state *vio, void *data, int len); -extern void vio_link_state_change(struct vio_driver_state *vio, int event); -extern void vio_conn_reset(struct vio_driver_state *vio); -extern int vio_control_pkt_engine(struct vio_driver_state *vio, void *pkt); -extern int vio_validate_sid(struct vio_driver_state *vio, - struct vio_msg_tag *tp); -extern u32 vio_send_sid(struct vio_driver_state *vio); -extern int vio_ldc_alloc(struct vio_driver_state *vio, - struct ldc_channel_config *base_cfg, void *event_arg); -extern void vio_ldc_free(struct vio_driver_state *vio); -extern int vio_driver_init(struct vio_driver_state *vio, struct vio_dev *vdev, - u8 dev_class, struct vio_version *ver_table, - int ver_table_size, struct vio_driver_ops *ops, - char *name); - -extern void vio_port_up(struct vio_driver_state *vio); +int vio_ldc_send(struct vio_driver_state *vio, void *data, int len); +void vio_link_state_change(struct vio_driver_state *vio, int event); +void vio_conn_reset(struct vio_driver_state *vio); +int vio_control_pkt_engine(struct vio_driver_state *vio, void *pkt); +int vio_validate_sid(struct vio_driver_state *vio, + struct vio_msg_tag *tp); +u32 vio_send_sid(struct vio_driver_state *vio); +int vio_ldc_alloc(struct vio_driver_state *vio, + struct ldc_channel_config *base_cfg, void *event_arg); +void vio_ldc_free(struct vio_driver_state *vio); +int vio_driver_init(struct vio_driver_state *vio, struct vio_dev *vdev, + u8 dev_class, struct vio_version *ver_table, + int ver_table_size, struct vio_driver_ops *ops, + char *name); + +void vio_port_up(struct vio_driver_state *vio); #endif /* _SPARC64_VIO_H */ diff --git a/arch/sparc/include/asm/visasm.h b/arch/sparc/include/asm/visasm.h index 39ca301920d..b2667375928 100644 --- a/arch/sparc/include/asm/visasm.h +++ b/arch/sparc/include/asm/visasm.h @@ -57,7 +57,8 @@ static inline void save_and_clear_fpu(void) { " " : : "i" (FPRS_FEF|FPRS_DU) : "o5", "g1", "g2", "g3", "g7", "cc"); } -extern int vis_emul(struct pt_regs *, unsigned int); + +int vis_emul(struct pt_regs *, unsigned int); #endif #endif /* _SPARC64_ASI_H */ diff --git a/arch/sparc/include/asm/xor_64.h b/arch/sparc/include/asm/xor_64.h index ee8edc68423..50c88285603 100644 --- a/arch/sparc/include/asm/xor_64.h +++ b/arch/sparc/include/asm/xor_64.h @@ -20,13 +20,13 @@ #include <asm/spitfire.h> -extern void xor_vis_2(unsigned long, unsigned long *, unsigned long *); -extern void xor_vis_3(unsigned long, unsigned long *, unsigned long *, - unsigned long *); -extern void xor_vis_4(unsigned long, unsigned long *, unsigned long *, - unsigned long *, unsigned long *); -extern void xor_vis_5(unsigned long, unsigned long *, unsigned long *, - unsigned long *, unsigned long *, unsigned long *); +void xor_vis_2(unsigned long, unsigned long *, unsigned long *); +void xor_vis_3(unsigned long, unsigned long *, unsigned long *, + unsigned long *); +void xor_vis_4(unsigned long, unsigned long *, unsigned long *, + unsigned long *, unsigned long *); +void xor_vis_5(unsigned long, unsigned long *, unsigned long *, + unsigned long *, unsigned long *, unsigned long *); /* XXX Ugh, write cheetah versions... -DaveM */ @@ -38,13 +38,13 @@ static struct xor_block_template xor_block_VIS = { .do_5 = xor_vis_5, }; -extern void xor_niagara_2(unsigned long, unsigned long *, unsigned long *); -extern void xor_niagara_3(unsigned long, unsigned long *, unsigned long *, - unsigned long *); -extern void xor_niagara_4(unsigned long, unsigned long *, unsigned long *, - unsigned long *, unsigned long *); -extern void xor_niagara_5(unsigned long, unsigned long *, unsigned long *, - unsigned long *, unsigned long *, unsigned long *); +void xor_niagara_2(unsigned long, unsigned long *, unsigned long *); +void xor_niagara_3(unsigned long, unsigned long *, unsigned long *, + unsigned long *); +void xor_niagara_4(unsigned long, unsigned long *, unsigned long *, + unsigned long *, unsigned long *); +void xor_niagara_5(unsigned long, unsigned long *, unsigned long *, + unsigned long *, unsigned long *, unsigned long *); static struct xor_block_template xor_block_niagara = { .name = "Niagara", diff --git a/arch/sparc/include/uapi/asm/Kbuild b/arch/sparc/include/uapi/asm/Kbuild new file mode 100644 index 00000000000..b5843ee09fb --- /dev/null +++ b/arch/sparc/include/uapi/asm/Kbuild @@ -0,0 +1,50 @@ +# UAPI Header export list +# User exported sparc header files + +include include/uapi/asm-generic/Kbuild.asm + +header-y += apc.h +header-y += asi.h +header-y += auxvec.h +header-y += bitsperlong.h +header-y += byteorder.h +header-y += display7seg.h +header-y += envctrl.h +header-y += errno.h +header-y += fbio.h +header-y += fcntl.h +header-y += ioctl.h +header-y += ioctls.h +header-y += ipcbuf.h +header-y += jsflash.h +header-y += kvm_para.h +header-y += mman.h +header-y += msgbuf.h +header-y += openpromio.h +header-y += param.h +header-y += perfctr.h +header-y += poll.h +header-y += posix_types.h +header-y += psr.h +header-y += psrcompat.h +header-y += pstate.h +header-y += ptrace.h +header-y += resource.h +header-y += sembuf.h +header-y += setup.h +header-y += shmbuf.h +header-y += sigcontext.h +header-y += siginfo.h +header-y += signal.h +header-y += socket.h +header-y += sockios.h +header-y += stat.h +header-y += statfs.h +header-y += swab.h +header-y += termbits.h +header-y += termios.h +header-y += traps.h +header-y += uctx.h +header-y += unistd.h +header-y += utrap.h +header-y += watchdog.h diff --git a/arch/sparc/include/asm/apc.h b/arch/sparc/include/uapi/asm/apc.h index 24e9a7d4d97..24e9a7d4d97 100644 --- a/arch/sparc/include/asm/apc.h +++ b/arch/sparc/include/uapi/asm/apc.h diff --git a/arch/sparc/include/asm/asi.h b/arch/sparc/include/uapi/asm/asi.h index 61ebe7411ce..aace6f31371 100644 --- a/arch/sparc/include/asm/asi.h +++ b/arch/sparc/include/uapi/asm/asi.h @@ -141,7 +141,8 @@ /* SpitFire and later extended ASIs. The "(III)" marker designates * UltraSparc-III and later specific ASIs. The "(CMT)" marker designates * Chip Multi Threading specific ASIs. "(NG)" designates Niagara specific - * ASIs, "(4V)" designates SUN4V specific ASIs. + * ASIs, "(4V)" designates SUN4V specific ASIs. "(NG4)" designates SPARC-T4 + * and later ASIs. */ #define ASI_PHYS_USE_EC 0x14 /* PADDR, E-cachable */ #define ASI_PHYS_BYPASS_EC_E 0x15 /* PADDR, E-bit */ @@ -243,6 +244,7 @@ #define ASI_UDBL_CONTROL_R 0x7f /* External UDB control regs rd low*/ #define ASI_INTR_R 0x7f /* IRQ vector dispatch read */ #define ASI_INTR_DATAN_R 0x7f /* (III) In irq vector data reg N */ +#define ASI_PIC 0xb0 /* (NG4) PIC registers */ #define ASI_PST8_P 0xc0 /* Primary, 8 8-bit, partial */ #define ASI_PST8_S 0xc1 /* Secondary, 8 8-bit, partial */ #define ASI_PST16_P 0xc2 /* Primary, 4 16-bit, partial */ @@ -268,9 +270,28 @@ #define ASI_BLK_INIT_QUAD_LDD_P 0xe2 /* (NG) init-store, twin load, * primary, implicit */ +#define ASI_BLK_INIT_QUAD_LDD_S 0xe3 /* (NG) init-store, twin load, + * secondary, implicit + */ #define ASI_BLK_P 0xf0 /* Primary, blk ld/st */ #define ASI_BLK_S 0xf1 /* Secondary, blk ld/st */ +#define ASI_ST_BLKINIT_MRU_P 0xf2 /* (NG4) init-store, twin load, + * Most-Recently-Used, primary, + * implicit + */ +#define ASI_ST_BLKINIT_MRU_S 0xf2 /* (NG4) init-store, twin load, + * Most-Recently-Used, secondary, + * implicit + */ #define ASI_BLK_PL 0xf8 /* Primary, blk ld/st, little */ #define ASI_BLK_SL 0xf9 /* Secondary, blk ld/st, little */ +#define ASI_ST_BLKINIT_MRU_PL 0xfa /* (NG4) init-store, twin load, + * Most-Recently-Used, primary, + * implicit, little-endian + */ +#define ASI_ST_BLKINIT_MRU_SL 0xfb /* (NG4) init-store, twin load, + * Most-Recently-Used, secondary, + * implicit, little-endian + */ #endif /* _SPARC_ASI_H */ diff --git a/arch/sparc/include/asm/auxvec.h b/arch/sparc/include/uapi/asm/auxvec.h index ad6f360261f..ad6f360261f 100644 --- a/arch/sparc/include/asm/auxvec.h +++ b/arch/sparc/include/uapi/asm/auxvec.h diff --git a/arch/sparc/include/asm/bitsperlong.h b/arch/sparc/include/uapi/asm/bitsperlong.h index 40dcaa3aaa5..40dcaa3aaa5 100644 --- a/arch/sparc/include/asm/bitsperlong.h +++ b/arch/sparc/include/uapi/asm/bitsperlong.h diff --git a/arch/sparc/include/asm/byteorder.h b/arch/sparc/include/uapi/asm/byteorder.h index ccc1b6b7de6..ccc1b6b7de6 100644 --- a/arch/sparc/include/asm/byteorder.h +++ b/arch/sparc/include/uapi/asm/byteorder.h diff --git a/arch/sparc/include/asm/display7seg.h b/arch/sparc/include/uapi/asm/display7seg.h index 86d4a901df2..86d4a901df2 100644 --- a/arch/sparc/include/asm/display7seg.h +++ b/arch/sparc/include/uapi/asm/display7seg.h diff --git a/arch/sparc/include/asm/envctrl.h b/arch/sparc/include/uapi/asm/envctrl.h index 624fa7e2da8..624fa7e2da8 100644 --- a/arch/sparc/include/asm/envctrl.h +++ b/arch/sparc/include/uapi/asm/envctrl.h diff --git a/arch/sparc/include/asm/errno.h b/arch/sparc/include/uapi/asm/errno.h index c351aba997b..20423e17285 100644 --- a/arch/sparc/include/asm/errno.h +++ b/arch/sparc/include/uapi/asm/errno.h @@ -40,7 +40,7 @@ #define EPROCLIM 67 /* SUNOS: Too many processes */ #define EUSERS 68 /* Too many users */ #define EDQUOT 69 /* Quota exceeded */ -#define ESTALE 70 /* Stale NFS file handle */ +#define ESTALE 70 /* Stale file handle */ #define EREMOTE 71 /* Object is remote */ #define ENOSTR 72 /* Device not a stream */ #define ETIME 73 /* Timer expired */ diff --git a/arch/sparc/include/uapi/asm/fbio.h b/arch/sparc/include/uapi/asm/fbio.h new file mode 100644 index 00000000000..d6cea07afb6 --- /dev/null +++ b/arch/sparc/include/uapi/asm/fbio.h @@ -0,0 +1,259 @@ +#ifndef _UAPI__LINUX_FBIO_H +#define _UAPI__LINUX_FBIO_H + +#include <linux/compiler.h> +#include <linux/types.h> + +/* Constants used for fbio SunOS compatibility */ +/* (C) 1996 Miguel de Icaza */ + +/* Frame buffer types */ +#define FBTYPE_NOTYPE -1 +#define FBTYPE_SUN1BW 0 /* mono */ +#define FBTYPE_SUN1COLOR 1 +#define FBTYPE_SUN2BW 2 +#define FBTYPE_SUN2COLOR 3 +#define FBTYPE_SUN2GP 4 +#define FBTYPE_SUN5COLOR 5 +#define FBTYPE_SUN3COLOR 6 +#define FBTYPE_MEMCOLOR 7 +#define FBTYPE_SUN4COLOR 8 + +#define FBTYPE_NOTSUN1 9 +#define FBTYPE_NOTSUN2 10 +#define FBTYPE_NOTSUN3 11 + +#define FBTYPE_SUNFAST_COLOR 12 /* cg6 */ +#define FBTYPE_SUNROP_COLOR 13 +#define FBTYPE_SUNFB_VIDEO 14 +#define FBTYPE_SUNGIFB 15 +#define FBTYPE_SUNGPLAS 16 +#define FBTYPE_SUNGP3 17 +#define FBTYPE_SUNGT 18 +#define FBTYPE_SUNLEO 19 /* zx Leo card */ +#define FBTYPE_MDICOLOR 20 /* cg14 */ +#define FBTYPE_TCXCOLOR 21 /* SUNW,tcx card */ + +#define FBTYPE_LASTPLUSONE 21 /* This is not last + 1 in fact... */ + +/* Does not seem to be listed in the Sun file either */ +#define FBTYPE_CREATOR 22 +#define FBTYPE_PCI_IGA1682 23 +#define FBTYPE_P9100COLOR 24 + +#define FBTYPE_PCI_GENERIC 1000 +#define FBTYPE_PCI_MACH64 1001 + +/* fbio ioctls */ +/* Returned by FBIOGTYPE */ +struct fbtype { + int fb_type; /* fb type, see above */ + int fb_height; /* pixels */ + int fb_width; /* pixels */ + int fb_depth; + int fb_cmsize; /* color map entries */ + int fb_size; /* fb size in bytes */ +}; +#define FBIOGTYPE _IOR('F', 0, struct fbtype) + +struct fbcmap { + int index; /* first element (0 origin) */ + int count; + unsigned char __user *red; + unsigned char __user *green; + unsigned char __user *blue; +}; + +#ifndef __KERNEL__ +#define FBIOPUTCMAP _IOW('F', 3, struct fbcmap) +#define FBIOGETCMAP _IOW('F', 4, struct fbcmap) +#endif + +/* # of device specific values */ +#define FB_ATTR_NDEVSPECIFIC 8 +/* # of possible emulations */ +#define FB_ATTR_NEMUTYPES 4 + +struct fbsattr { + int flags; + int emu_type; /* -1 if none */ + int dev_specific[FB_ATTR_NDEVSPECIFIC]; +}; + +struct fbgattr { + int real_type; /* real frame buffer type */ + int owner; /* unknown */ + struct fbtype fbtype; /* real frame buffer fbtype */ + struct fbsattr sattr; + int emu_types[FB_ATTR_NEMUTYPES]; /* supported emulations */ +}; +#define FBIOSATTR _IOW('F', 5, struct fbgattr) /* Unsupported: */ +#define FBIOGATTR _IOR('F', 6, struct fbgattr) /* supported */ + +#define FBIOSVIDEO _IOW('F', 7, int) +#define FBIOGVIDEO _IOR('F', 8, int) + +struct fbcursor { + short set; /* what to set, choose from the list above */ + short enable; /* cursor on/off */ + struct fbcurpos pos; /* cursor position */ + struct fbcurpos hot; /* cursor hot spot */ + struct fbcmap cmap; /* color map info */ + struct fbcurpos size; /* cursor bit map size */ + char __user *image; /* cursor image bits */ + char __user *mask; /* cursor mask bits */ +}; + +/* set/get cursor attributes/shape */ +#define FBIOSCURSOR _IOW('F', 24, struct fbcursor) +#define FBIOGCURSOR _IOWR('F', 25, struct fbcursor) + +/* set/get cursor position */ +#define FBIOSCURPOS _IOW('F', 26, struct fbcurpos) +#define FBIOGCURPOS _IOW('F', 27, struct fbcurpos) + +/* get max cursor size */ +#define FBIOGCURMAX _IOR('F', 28, struct fbcurpos) + +/* wid manipulation */ +struct fb_wid_alloc { +#define FB_WID_SHARED_8 0 +#define FB_WID_SHARED_24 1 +#define FB_WID_DBL_8 2 +#define FB_WID_DBL_24 3 + __u32 wa_type; + __s32 wa_index; /* Set on return */ + __u32 wa_count; +}; +struct fb_wid_item { + __u32 wi_type; + __s32 wi_index; + __u32 wi_attrs; + __u32 wi_values[32]; +}; +struct fb_wid_list { + __u32 wl_flags; + __u32 wl_count; + struct fb_wid_item *wl_list; +}; + +#define FBIO_WID_ALLOC _IOWR('F', 30, struct fb_wid_alloc) +#define FBIO_WID_FREE _IOW('F', 31, struct fb_wid_alloc) +#define FBIO_WID_PUT _IOW('F', 32, struct fb_wid_list) +#define FBIO_WID_GET _IOWR('F', 33, struct fb_wid_list) + +/* Creator ioctls */ +#define FFB_IOCTL ('F'<<8) +#define FFB_SYS_INFO (FFB_IOCTL|80) +#define FFB_CLUTREAD (FFB_IOCTL|81) +#define FFB_CLUTPOST (FFB_IOCTL|82) +#define FFB_SETDIAGMODE (FFB_IOCTL|83) +#define FFB_GETMONITORID (FFB_IOCTL|84) +#define FFB_GETVIDEOMODE (FFB_IOCTL|85) +#define FFB_SETVIDEOMODE (FFB_IOCTL|86) +#define FFB_SETSERVER (FFB_IOCTL|87) +#define FFB_SETOVCTL (FFB_IOCTL|88) +#define FFB_GETOVCTL (FFB_IOCTL|89) +#define FFB_GETSAXNUM (FFB_IOCTL|90) +#define FFB_FBDEBUG (FFB_IOCTL|91) + +/* Cg14 ioctls */ +#define MDI_IOCTL ('M'<<8) +#define MDI_RESET (MDI_IOCTL|1) +#define MDI_GET_CFGINFO (MDI_IOCTL|2) +#define MDI_SET_PIXELMODE (MDI_IOCTL|3) +# define MDI_32_PIX 32 +# define MDI_16_PIX 16 +# define MDI_8_PIX 8 + +struct mdi_cfginfo { + int mdi_ncluts; /* Number of implemented CLUTs in this MDI */ + int mdi_type; /* FBTYPE name */ + int mdi_height; /* height */ + int mdi_width; /* width */ + int mdi_size; /* available ram */ + int mdi_mode; /* 8bpp, 16bpp or 32bpp */ + int mdi_pixfreq; /* pixel clock (from PROM) */ +}; + +/* SparcLinux specific ioctl for the MDI, should be replaced for + * the SET_XLUT/SET_CLUTn ioctls instead + */ +#define MDI_CLEAR_XLUT (MDI_IOCTL|9) + +/* leo & ffb ioctls */ +struct fb_clut_alloc { + __u32 clutid; /* Set on return */ + __u32 flag; + __u32 index; +}; + +struct fb_clut { +#define FB_CLUT_WAIT 0x00000001 /* Not yet implemented */ + __u32 flag; + __u32 clutid; + __u32 offset; + __u32 count; + char * red; + char * green; + char * blue; +}; + +struct fb_clut32 { + __u32 flag; + __u32 clutid; + __u32 offset; + __u32 count; + __u32 red; + __u32 green; + __u32 blue; +}; + +#define LEO_CLUTALLOC _IOWR('L', 53, struct fb_clut_alloc) +#define LEO_CLUTFREE _IOW('L', 54, struct fb_clut_alloc) +#define LEO_CLUTREAD _IOW('L', 55, struct fb_clut) +#define LEO_CLUTPOST _IOW('L', 56, struct fb_clut) +#define LEO_SETGAMMA _IOW('L', 68, int) /* Not yet implemented */ +#define LEO_GETGAMMA _IOR('L', 69, int) /* Not yet implemented */ + + +/* These are exported to userland for applications to use */ +/* Mappable offsets for the cg14: control registers */ +#define MDI_DIRECT_MAP 0x10000000 +#define MDI_CTLREG_MAP 0x20000000 +#define MDI_CURSOR_MAP 0x30000000 +#define MDI_SHDW_VRT_MAP 0x40000000 + +/* Mappable offsets for the cg14: frame buffer resolutions */ +/* 32 bits */ +#define MDI_CHUNKY_XBGR_MAP 0x50000000 +#define MDI_CHUNKY_BGR_MAP 0x60000000 + +/* 16 bits */ +#define MDI_PLANAR_X16_MAP 0x70000000 +#define MDI_PLANAR_C16_MAP 0x80000000 + +/* 8 bit is done as CG3 MMAP offset */ +/* 32 bits, planar */ +#define MDI_PLANAR_X32_MAP 0x90000000 +#define MDI_PLANAR_B32_MAP 0xa0000000 +#define MDI_PLANAR_G32_MAP 0xb0000000 +#define MDI_PLANAR_R32_MAP 0xc0000000 + +/* Mappable offsets on leo */ +#define LEO_SS0_MAP 0x00000000 +#define LEO_LC_SS0_USR_MAP 0x00800000 +#define LEO_LD_SS0_MAP 0x00801000 +#define LEO_LX_CURSOR_MAP 0x00802000 +#define LEO_SS1_MAP 0x00803000 +#define LEO_LC_SS1_USR_MAP 0x01003000 +#define LEO_LD_SS1_MAP 0x01004000 +#define LEO_UNK_MAP 0x01005000 +#define LEO_LX_KRN_MAP 0x01006000 +#define LEO_LC_SS0_KRN_MAP 0x01007000 +#define LEO_LC_SS1_KRN_MAP 0x01008000 +#define LEO_LD_GBL_MAP 0x01009000 +#define LEO_UNK2_MAP 0x0100a000 + + +#endif /* _UAPI__LINUX_FBIO_H */ diff --git a/arch/sparc/include/asm/fcntl.h b/arch/sparc/include/uapi/asm/fcntl.h index d0b83f66f35..7e8ace5bf76 100644 --- a/arch/sparc/include/asm/fcntl.h +++ b/arch/sparc/include/uapi/asm/fcntl.h @@ -35,6 +35,7 @@ #define O_SYNC (__O_SYNC|O_DSYNC) #define O_PATH 0x1000000 +#define __O_TMPFILE 0x2000000 #define F_GETOWN 5 /* for sockets. */ #define F_SETOWN 6 /* for sockets. */ diff --git a/arch/sparc/include/asm/ioctl.h b/arch/sparc/include/uapi/asm/ioctl.h index 7d6bd51321b..7d6bd51321b 100644 --- a/arch/sparc/include/asm/ioctl.h +++ b/arch/sparc/include/uapi/asm/ioctl.h diff --git a/arch/sparc/include/uapi/asm/ioctls.h b/arch/sparc/include/uapi/asm/ioctls.h new file mode 100644 index 00000000000..897d1723fa1 --- /dev/null +++ b/arch/sparc/include/uapi/asm/ioctls.h @@ -0,0 +1,134 @@ +#ifndef _UAPI_ASM_SPARC_IOCTLS_H +#define _UAPI_ASM_SPARC_IOCTLS_H + +#include <asm/ioctl.h> + +/* Big T */ +#define TCGETA _IOR('T', 1, struct termio) +#define TCSETA _IOW('T', 2, struct termio) +#define TCSETAW _IOW('T', 3, struct termio) +#define TCSETAF _IOW('T', 4, struct termio) +#define TCSBRK _IO('T', 5) +#define TCXONC _IO('T', 6) +#define TCFLSH _IO('T', 7) +#define TCGETS _IOR('T', 8, struct termios) +#define TCSETS _IOW('T', 9, struct termios) +#define TCSETSW _IOW('T', 10, struct termios) +#define TCSETSF _IOW('T', 11, struct termios) +#define TCGETS2 _IOR('T', 12, struct termios2) +#define TCSETS2 _IOW('T', 13, struct termios2) +#define TCSETSW2 _IOW('T', 14, struct termios2) +#define TCSETSF2 _IOW('T', 15, struct termios2) +#define TIOCGDEV _IOR('T',0x32, unsigned int) /* Get primary device node of /dev/console */ +#define TIOCVHANGUP _IO('T', 0x37) +#define TIOCGPKT _IOR('T', 0x38, int) /* Get packet mode state */ +#define TIOCGPTLCK _IOR('T', 0x39, int) /* Get Pty lock state */ +#define TIOCGEXCL _IOR('T', 0x40, int) /* Get exclusive mode state */ + +/* Note that all the ioctls that are not available in Linux have a + * double underscore on the front to: a) avoid some programs to + * think we support some ioctls under Linux (autoconfiguration stuff) + */ +/* Little t */ +#define TIOCGETD _IOR('t', 0, int) +#define TIOCSETD _IOW('t', 1, int) +#define __TIOCHPCL _IO('t', 2) /* SunOS Specific */ +#define __TIOCMODG _IOR('t', 3, int) /* SunOS Specific */ +#define __TIOCMODS _IOW('t', 4, int) /* SunOS Specific */ +#define __TIOCGETP _IOR('t', 8, struct sgttyb) /* SunOS Specific */ +#define __TIOCSETP _IOW('t', 9, struct sgttyb) /* SunOS Specific */ +#define __TIOCSETN _IOW('t', 10, struct sgttyb) /* SunOS Specific */ +#define TIOCEXCL _IO('t', 13) +#define TIOCNXCL _IO('t', 14) +#define __TIOCFLUSH _IOW('t', 16, int) /* SunOS Specific */ +#define __TIOCSETC _IOW('t', 17, struct tchars) /* SunOS Specific */ +#define __TIOCGETC _IOR('t', 18, struct tchars) /* SunOS Specific */ +#define __TIOCTCNTL _IOW('t', 32, int) /* SunOS Specific */ +#define __TIOCSIGNAL _IOW('t', 33, int) /* SunOS Specific */ +#define __TIOCSETX _IOW('t', 34, int) /* SunOS Specific */ +#define __TIOCGETX _IOR('t', 35, int) /* SunOS Specific */ +#define TIOCCONS _IO('t', 36) +#define TIOCGSOFTCAR _IOR('t', 100, int) +#define TIOCSSOFTCAR _IOW('t', 101, int) +#define __TIOCUCNTL _IOW('t', 102, int) /* SunOS Specific */ +#define TIOCSWINSZ _IOW('t', 103, struct winsize) +#define TIOCGWINSZ _IOR('t', 104, struct winsize) +#define __TIOCREMOTE _IOW('t', 105, int) /* SunOS Specific */ +#define TIOCMGET _IOR('t', 106, int) +#define TIOCMBIC _IOW('t', 107, int) +#define TIOCMBIS _IOW('t', 108, int) +#define TIOCMSET _IOW('t', 109, int) +#define TIOCSTART _IO('t', 110) +#define TIOCSTOP _IO('t', 111) +#define TIOCPKT _IOW('t', 112, int) +#define TIOCNOTTY _IO('t', 113) +#define TIOCSTI _IOW('t', 114, char) +#define TIOCOUTQ _IOR('t', 115, int) +#define __TIOCGLTC _IOR('t', 116, struct ltchars) /* SunOS Specific */ +#define __TIOCSLTC _IOW('t', 117, struct ltchars) /* SunOS Specific */ +/* 118 is the non-posix setpgrp tty ioctl */ +/* 119 is the non-posix getpgrp tty ioctl */ +#define __TIOCCDTR _IO('t', 120) /* SunOS Specific */ +#define __TIOCSDTR _IO('t', 121) /* SunOS Specific */ +#define TIOCCBRK _IO('t', 122) +#define TIOCSBRK _IO('t', 123) +#define __TIOCLGET _IOW('t', 124, int) /* SunOS Specific */ +#define __TIOCLSET _IOW('t', 125, int) /* SunOS Specific */ +#define __TIOCLBIC _IOW('t', 126, int) /* SunOS Specific */ +#define __TIOCLBIS _IOW('t', 127, int) /* SunOS Specific */ +#define __TIOCISPACE _IOR('t', 128, int) /* SunOS Specific */ +#define __TIOCISIZE _IOR('t', 129, int) /* SunOS Specific */ +#define TIOCSPGRP _IOW('t', 130, int) +#define TIOCGPGRP _IOR('t', 131, int) +#define TIOCSCTTY _IO('t', 132) +#define TIOCGSID _IOR('t', 133, int) +/* Get minor device of a pty master's FD -- Solaris equiv is ISPTM */ +#define TIOCGPTN _IOR('t', 134, unsigned int) /* Get Pty Number */ +#define TIOCSPTLCK _IOW('t', 135, int) /* Lock/unlock PTY */ +#define TIOCSIG _IOW('t', 136, int) /* Generate signal on Pty slave */ + +/* Little f */ +#define FIOCLEX _IO('f', 1) +#define FIONCLEX _IO('f', 2) +#define FIOASYNC _IOW('f', 125, int) +#define FIONBIO _IOW('f', 126, int) +#define FIONREAD _IOR('f', 127, int) +#define TIOCINQ FIONREAD +#define FIOQSIZE _IOR('f', 128, loff_t) + +/* SCARY Rutgers local SunOS kernel hackery, perhaps I will support it + * someday. This is completely bogus, I know... + */ +#define __TCGETSTAT _IO('T', 200) /* Rutgers specific */ +#define __TCSETSTAT _IO('T', 201) /* Rutgers specific */ + +/* Linux specific, no SunOS equivalent. */ +#define TIOCLINUX 0x541C +#define TIOCGSERIAL 0x541E +#define TIOCSSERIAL 0x541F +#define TCSBRKP 0x5425 +#define TIOCSERCONFIG 0x5453 +#define TIOCSERGWILD 0x5454 +#define TIOCSERSWILD 0x5455 +#define TIOCGLCKTRMIOS 0x5456 +#define TIOCSLCKTRMIOS 0x5457 +#define TIOCSERGSTRUCT 0x5458 /* For debugging only */ +#define TIOCSERGETLSR 0x5459 /* Get line status register */ +#define TIOCSERGETMULTI 0x545A /* Get multiport config */ +#define TIOCSERSETMULTI 0x545B /* Set multiport config */ +#define TIOCMIWAIT 0x545C /* Wait for change on serial input line(s) */ +#define TIOCGICOUNT 0x545D /* Read serial port inline interrupt counts */ + +/* Kernel definitions */ + +/* Used for packet mode */ +#define TIOCPKT_DATA 0 +#define TIOCPKT_FLUSHREAD 1 +#define TIOCPKT_FLUSHWRITE 2 +#define TIOCPKT_STOP 4 +#define TIOCPKT_START 8 +#define TIOCPKT_NOSTOP 16 +#define TIOCPKT_DOSTOP 32 +#define TIOCPKT_IOCTL 64 + +#endif /* _UAPI_ASM_SPARC_IOCTLS_H */ diff --git a/arch/sparc/include/asm/ipcbuf.h b/arch/sparc/include/uapi/asm/ipcbuf.h index 66013b4fe10..66013b4fe10 100644 --- a/arch/sparc/include/asm/ipcbuf.h +++ b/arch/sparc/include/uapi/asm/ipcbuf.h diff --git a/arch/sparc/include/asm/jsflash.h b/arch/sparc/include/uapi/asm/jsflash.h index 0717d9e39d2..0717d9e39d2 100644 --- a/arch/sparc/include/asm/jsflash.h +++ b/arch/sparc/include/uapi/asm/jsflash.h diff --git a/arch/sparc/include/asm/kvm_para.h b/arch/sparc/include/uapi/asm/kvm_para.h index 14fab8f0b95..14fab8f0b95 100644 --- a/arch/sparc/include/asm/kvm_para.h +++ b/arch/sparc/include/uapi/asm/kvm_para.h diff --git a/arch/sparc/include/uapi/asm/mman.h b/arch/sparc/include/uapi/asm/mman.h new file mode 100644 index 00000000000..0b14df33cff --- /dev/null +++ b/arch/sparc/include/uapi/asm/mman.h @@ -0,0 +1,27 @@ +#ifndef _UAPI__SPARC_MMAN_H__ +#define _UAPI__SPARC_MMAN_H__ + +#include <asm-generic/mman-common.h> + +/* SunOS'ified... */ + +#define MAP_RENAME MAP_ANONYMOUS /* In SunOS terminology */ +#define MAP_NORESERVE 0x40 /* don't reserve swap pages */ +#define MAP_INHERIT 0x80 /* SunOS doesn't do this, but... */ +#define MAP_LOCKED 0x100 /* lock the mapping */ +#define _MAP_NEW 0x80000000 /* Binary compatibility is fun... */ + +#define MAP_GROWSDOWN 0x0200 /* stack-like segment */ +#define MAP_DENYWRITE 0x0800 /* ETXTBSY */ +#define MAP_EXECUTABLE 0x1000 /* mark it as an executable */ + +#define MCL_CURRENT 0x2000 /* lock all currently mapped pages */ +#define MCL_FUTURE 0x4000 /* lock all additions to address space */ + +#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */ +#define MAP_NONBLOCK 0x10000 /* do not block on IO */ +#define MAP_STACK 0x20000 /* give out an address that is best suited for process/thread stacks */ +#define MAP_HUGETLB 0x40000 /* create a huge page mapping */ + + +#endif /* _UAPI__SPARC_MMAN_H__ */ diff --git a/arch/sparc/include/asm/msgbuf.h b/arch/sparc/include/uapi/asm/msgbuf.h index efc7cbe9788..efc7cbe9788 100644 --- a/arch/sparc/include/asm/msgbuf.h +++ b/arch/sparc/include/uapi/asm/msgbuf.h diff --git a/arch/sparc/include/asm/openpromio.h b/arch/sparc/include/uapi/asm/openpromio.h index 917fb8e9c63..917fb8e9c63 100644 --- a/arch/sparc/include/asm/openpromio.h +++ b/arch/sparc/include/uapi/asm/openpromio.h diff --git a/arch/sparc/include/asm/param.h b/arch/sparc/include/uapi/asm/param.h index 0bc356bf8c5..0bc356bf8c5 100644 --- a/arch/sparc/include/asm/param.h +++ b/arch/sparc/include/uapi/asm/param.h diff --git a/arch/sparc/include/asm/perfctr.h b/arch/sparc/include/uapi/asm/perfctr.h index 3332d2cba6c..214feefa577 100644 --- a/arch/sparc/include/asm/perfctr.h +++ b/arch/sparc/include/uapi/asm/perfctr.h @@ -54,11 +54,6 @@ enum perfctr_opcode { PERFCTR_GETPCR }; -/* I don't want the kernel's namespace to be polluted with this - * stuff when this file is included. --DaveM - */ -#ifndef __KERNEL__ - #define PRIV 0x00000001 #define SYS 0x00000002 #define USR 0x00000004 @@ -168,29 +163,4 @@ struct vcounter_struct { unsigned long long vcnt1; }; -#else /* !(__KERNEL__) */ - -#ifndef CONFIG_SPARC32 - -/* Performance counter register access. */ -#define read_pcr(__p) __asm__ __volatile__("rd %%pcr, %0" : "=r" (__p)) -#define write_pcr(__p) __asm__ __volatile__("wr %0, 0x0, %%pcr" : : "r" (__p)) -#define read_pic(__p) __asm__ __volatile__("rd %%pic, %0" : "=r" (__p)) - -/* Blackbird errata workaround. See commentary in - * arch/sparc64/kernel/smp.c:smp_percpu_timer_interrupt() - * for more information. - */ -#define write_pic(__p) \ - __asm__ __volatile__("ba,pt %%xcc, 99f\n\t" \ - " nop\n\t" \ - ".align 64\n" \ - "99:wr %0, 0x0, %%pic\n\t" \ - "rd %%pic, %%g0" : : "r" (__p)) -#define reset_pic() write_pic(0) - -#endif /* !CONFIG_SPARC32 */ - -#endif /* !(__KERNEL__) */ - #endif /* !(PERF_COUNTER_API) */ diff --git a/arch/sparc/include/asm/poll.h b/arch/sparc/include/uapi/asm/poll.h index 091d3ad2e83..091d3ad2e83 100644 --- a/arch/sparc/include/asm/poll.h +++ b/arch/sparc/include/uapi/asm/poll.h diff --git a/arch/sparc/include/asm/posix_types.h b/arch/sparc/include/uapi/asm/posix_types.h index 156220ed99e..156220ed99e 100644 --- a/arch/sparc/include/asm/posix_types.h +++ b/arch/sparc/include/uapi/asm/posix_types.h diff --git a/arch/sparc/include/uapi/asm/psr.h b/arch/sparc/include/uapi/asm/psr.h new file mode 100644 index 00000000000..2f0ed856530 --- /dev/null +++ b/arch/sparc/include/uapi/asm/psr.h @@ -0,0 +1,47 @@ +/* + * psr.h: This file holds the macros for masking off various parts of + * the processor status register on the Sparc. This is valid + * for Version 8. On the V9 this is renamed to the PSTATE + * register and its members are accessed as fields like + * PSTATE.PRIV for the current CPU privilege level. + * + * Copyright (C) 1994 David S. Miller (davem@caip.rutgers.edu) + */ + +#ifndef _UAPI__LINUX_SPARC_PSR_H +#define _UAPI__LINUX_SPARC_PSR_H + +/* The Sparc PSR fields are laid out as the following: + * + * ------------------------------------------------------------------------ + * | impl | vers | icc | resv | EC | EF | PIL | S | PS | ET | CWP | + * | 31-28 | 27-24 | 23-20 | 19-14 | 13 | 12 | 11-8 | 7 | 6 | 5 | 4-0 | + * ------------------------------------------------------------------------ + */ +#define PSR_CWP 0x0000001f /* current window pointer */ +#define PSR_ET 0x00000020 /* enable traps field */ +#define PSR_PS 0x00000040 /* previous privilege level */ +#define PSR_S 0x00000080 /* current privilege level */ +#define PSR_PIL 0x00000f00 /* processor interrupt level */ +#define PSR_EF 0x00001000 /* enable floating point */ +#define PSR_EC 0x00002000 /* enable co-processor */ +#define PSR_SYSCALL 0x00004000 /* inside of a syscall */ +#define PSR_LE 0x00008000 /* SuperSparcII little-endian */ +#define PSR_ICC 0x00f00000 /* integer condition codes */ +#define PSR_C 0x00100000 /* carry bit */ +#define PSR_V 0x00200000 /* overflow bit */ +#define PSR_Z 0x00400000 /* zero bit */ +#define PSR_N 0x00800000 /* negative bit */ +#define PSR_VERS 0x0f000000 /* cpu-version field */ +#define PSR_IMPL 0xf0000000 /* cpu-implementation field */ + +#define PSR_VERS_SHIFT 24 +#define PSR_IMPL_SHIFT 28 +#define PSR_VERS_SHIFTED_MASK 0xf +#define PSR_IMPL_SHIFTED_MASK 0xf + +#define PSR_IMPL_TI 0x4 +#define PSR_IMPL_LEON 0xf + + +#endif /* _UAPI__LINUX_SPARC_PSR_H */ diff --git a/arch/sparc/include/asm/psrcompat.h b/arch/sparc/include/uapi/asm/psrcompat.h index 44b6327dbbf..44b6327dbbf 100644 --- a/arch/sparc/include/asm/psrcompat.h +++ b/arch/sparc/include/uapi/asm/psrcompat.h diff --git a/arch/sparc/include/asm/pstate.h b/arch/sparc/include/uapi/asm/pstate.h index a26a53777bb..4b6b998afd9 100644 --- a/arch/sparc/include/asm/pstate.h +++ b/arch/sparc/include/uapi/asm/pstate.h @@ -88,4 +88,18 @@ #define VERS_MAXTL _AC(0x000000000000ff00,UL) /* Max Trap Level. */ #define VERS_MAXWIN _AC(0x000000000000001f,UL) /* Max RegWindow Idx.*/ +/* Compatability Feature Register (%asr26), SPARC-T4 and later */ +#define CFR_AES _AC(0x0000000000000001,UL) /* Supports AES opcodes */ +#define CFR_DES _AC(0x0000000000000002,UL) /* Supports DES opcodes */ +#define CFR_KASUMI _AC(0x0000000000000004,UL) /* Supports KASUMI opcodes */ +#define CFR_CAMELLIA _AC(0x0000000000000008,UL) /* Supports CAMELLIA opcodes*/ +#define CFR_MD5 _AC(0x0000000000000010,UL) /* Supports MD5 opcodes */ +#define CFR_SHA1 _AC(0x0000000000000020,UL) /* Supports SHA1 opcodes */ +#define CFR_SHA256 _AC(0x0000000000000040,UL) /* Supports SHA256 opcodes */ +#define CFR_SHA512 _AC(0x0000000000000080,UL) /* Supports SHA512 opcodes */ +#define CFR_MPMUL _AC(0x0000000000000100,UL) /* Supports MPMUL opcodes */ +#define CFR_MONTMUL _AC(0x0000000000000200,UL) /* Supports MONTMUL opcodes */ +#define CFR_MONTSQR _AC(0x0000000000000400,UL) /* Supports MONTSQR opcodes */ +#define CFR_CRC32C _AC(0x0000000000000800,UL) /* Supports CRC32C opcodes */ + #endif /* !(_SPARC64_PSTATE_H) */ diff --git a/arch/sparc/include/uapi/asm/ptrace.h b/arch/sparc/include/uapi/asm/ptrace.h new file mode 100644 index 00000000000..56fe4ea73fe --- /dev/null +++ b/arch/sparc/include/uapi/asm/ptrace.h @@ -0,0 +1,352 @@ +#ifndef _UAPI__SPARC_PTRACE_H +#define _UAPI__SPARC_PTRACE_H + +#if defined(__sparc__) && defined(__arch64__) +/* 64 bit sparc */ +#include <asm/pstate.h> + +/* This struct defines the way the registers are stored on the + * stack during a system call and basically all traps. + */ + +/* This magic value must have the low 9 bits clear, + * as that is where we encode the %tt value, see below. + */ +#define PT_REGS_MAGIC 0x57ac6c00 + +#ifndef __ASSEMBLY__ + +#include <linux/types.h> + +struct pt_regs { + unsigned long u_regs[16]; /* globals and ins */ + unsigned long tstate; + unsigned long tpc; + unsigned long tnpc; + unsigned int y; + + /* We encode a magic number, PT_REGS_MAGIC, along + * with the %tt (trap type) register value at trap + * entry time. The magic number allows us to identify + * accurately a trap stack frame in the stack + * unwinder, and the %tt value allows us to test + * things like "in a system call" etc. for an arbitray + * process. + * + * The PT_REGS_MAGIC is chosen such that it can be + * loaded completely using just a sethi instruction. + */ + unsigned int magic; +}; + +struct pt_regs32 { + unsigned int psr; + unsigned int pc; + unsigned int npc; + unsigned int y; + unsigned int u_regs[16]; /* globals and ins */ +}; + +/* A V9 register window */ +struct reg_window { + unsigned long locals[8]; + unsigned long ins[8]; +}; + +/* A 32-bit register window. */ +struct reg_window32 { + unsigned int locals[8]; + unsigned int ins[8]; +}; + +/* A V9 Sparc stack frame */ +struct sparc_stackf { + unsigned long locals[8]; + unsigned long ins[6]; + struct sparc_stackf *fp; + unsigned long callers_pc; + char *structptr; + unsigned long xargs[6]; + unsigned long xxargs[1]; +}; + +/* A 32-bit Sparc stack frame */ +struct sparc_stackf32 { + unsigned int locals[8]; + unsigned int ins[6]; + unsigned int fp; + unsigned int callers_pc; + unsigned int structptr; + unsigned int xargs[6]; + unsigned int xxargs[1]; +}; + +struct sparc_trapf { + unsigned long locals[8]; + unsigned long ins[8]; + unsigned long _unused; + struct pt_regs *regs; +}; +#endif /* (!__ASSEMBLY__) */ +#else +/* 32 bit sparc */ + +#include <asm/psr.h> + +/* This struct defines the way the registers are stored on the + * stack during a system call and basically all traps. + */ +#ifndef __ASSEMBLY__ + +#include <linux/types.h> + +struct pt_regs { + unsigned long psr; + unsigned long pc; + unsigned long npc; + unsigned long y; + unsigned long u_regs[16]; /* globals and ins */ +}; + +/* A 32-bit register window. */ +struct reg_window32 { + unsigned long locals[8]; + unsigned long ins[8]; +}; + +/* A Sparc stack frame */ +struct sparc_stackf { + unsigned long locals[8]; + unsigned long ins[6]; + struct sparc_stackf *fp; + unsigned long callers_pc; + char *structptr; + unsigned long xargs[6]; + unsigned long xxargs[1]; +}; +#endif /* (!__ASSEMBLY__) */ + +#endif /* (defined(__sparc__) && defined(__arch64__))*/ + +#ifndef __ASSEMBLY__ + +#define TRACEREG_SZ sizeof(struct pt_regs) +#define STACKFRAME_SZ sizeof(struct sparc_stackf) + +#define TRACEREG32_SZ sizeof(struct pt_regs32) +#define STACKFRAME32_SZ sizeof(struct sparc_stackf32) + +#endif /* (!__ASSEMBLY__) */ + +#define UREG_G0 0 +#define UREG_G1 1 +#define UREG_G2 2 +#define UREG_G3 3 +#define UREG_G4 4 +#define UREG_G5 5 +#define UREG_G6 6 +#define UREG_G7 7 +#define UREG_I0 8 +#define UREG_I1 9 +#define UREG_I2 10 +#define UREG_I3 11 +#define UREG_I4 12 +#define UREG_I5 13 +#define UREG_I6 14 +#define UREG_I7 15 +#define UREG_FP UREG_I6 +#define UREG_RETPC UREG_I7 + +#if defined(__sparc__) && defined(__arch64__) +/* 64 bit sparc */ + +#ifndef __ASSEMBLY__ + + +#else /* __ASSEMBLY__ */ +/* For assembly code. */ +#define TRACEREG_SZ 0xa0 +#define STACKFRAME_SZ 0xc0 + +#define TRACEREG32_SZ 0x50 +#define STACKFRAME32_SZ 0x60 +#endif /* __ASSEMBLY__ */ + +#else /* (defined(__sparc__) && defined(__arch64__)) */ + +/* 32 bit sparc */ + +#ifndef __ASSEMBLY__ + + +#else /* (!__ASSEMBLY__) */ +/* For assembly code. */ +#define TRACEREG_SZ 0x50 +#define STACKFRAME_SZ 0x60 +#endif /* (!__ASSEMBLY__) */ + +#endif /* (defined(__sparc__) && defined(__arch64__)) */ + + +/* These are for pt_regs. */ +#define PT_V9_G0 0x00 +#define PT_V9_G1 0x08 +#define PT_V9_G2 0x10 +#define PT_V9_G3 0x18 +#define PT_V9_G4 0x20 +#define PT_V9_G5 0x28 +#define PT_V9_G6 0x30 +#define PT_V9_G7 0x38 +#define PT_V9_I0 0x40 +#define PT_V9_I1 0x48 +#define PT_V9_I2 0x50 +#define PT_V9_I3 0x58 +#define PT_V9_I4 0x60 +#define PT_V9_I5 0x68 +#define PT_V9_I6 0x70 +#define PT_V9_FP PT_V9_I6 +#define PT_V9_I7 0x78 +#define PT_V9_TSTATE 0x80 +#define PT_V9_TPC 0x88 +#define PT_V9_TNPC 0x90 +#define PT_V9_Y 0x98 +#define PT_V9_MAGIC 0x9c +#define PT_TSTATE PT_V9_TSTATE +#define PT_TPC PT_V9_TPC +#define PT_TNPC PT_V9_TNPC + +/* These for pt_regs32. */ +#define PT_PSR 0x0 +#define PT_PC 0x4 +#define PT_NPC 0x8 +#define PT_Y 0xc +#define PT_G0 0x10 +#define PT_WIM PT_G0 +#define PT_G1 0x14 +#define PT_G2 0x18 +#define PT_G3 0x1c +#define PT_G4 0x20 +#define PT_G5 0x24 +#define PT_G6 0x28 +#define PT_G7 0x2c +#define PT_I0 0x30 +#define PT_I1 0x34 +#define PT_I2 0x38 +#define PT_I3 0x3c +#define PT_I4 0x40 +#define PT_I5 0x44 +#define PT_I6 0x48 +#define PT_FP PT_I6 +#define PT_I7 0x4c + +/* Reg_window offsets */ +#define RW_V9_L0 0x00 +#define RW_V9_L1 0x08 +#define RW_V9_L2 0x10 +#define RW_V9_L3 0x18 +#define RW_V9_L4 0x20 +#define RW_V9_L5 0x28 +#define RW_V9_L6 0x30 +#define RW_V9_L7 0x38 +#define RW_V9_I0 0x40 +#define RW_V9_I1 0x48 +#define RW_V9_I2 0x50 +#define RW_V9_I3 0x58 +#define RW_V9_I4 0x60 +#define RW_V9_I5 0x68 +#define RW_V9_I6 0x70 +#define RW_V9_I7 0x78 + +#define RW_L0 0x00 +#define RW_L1 0x04 +#define RW_L2 0x08 +#define RW_L3 0x0c +#define RW_L4 0x10 +#define RW_L5 0x14 +#define RW_L6 0x18 +#define RW_L7 0x1c +#define RW_I0 0x20 +#define RW_I1 0x24 +#define RW_I2 0x28 +#define RW_I3 0x2c +#define RW_I4 0x30 +#define RW_I5 0x34 +#define RW_I6 0x38 +#define RW_I7 0x3c + +/* Stack_frame offsets */ +#define SF_V9_L0 0x00 +#define SF_V9_L1 0x08 +#define SF_V9_L2 0x10 +#define SF_V9_L3 0x18 +#define SF_V9_L4 0x20 +#define SF_V9_L5 0x28 +#define SF_V9_L6 0x30 +#define SF_V9_L7 0x38 +#define SF_V9_I0 0x40 +#define SF_V9_I1 0x48 +#define SF_V9_I2 0x50 +#define SF_V9_I3 0x58 +#define SF_V9_I4 0x60 +#define SF_V9_I5 0x68 +#define SF_V9_FP 0x70 +#define SF_V9_PC 0x78 +#define SF_V9_RETP 0x80 +#define SF_V9_XARG0 0x88 +#define SF_V9_XARG1 0x90 +#define SF_V9_XARG2 0x98 +#define SF_V9_XARG3 0xa0 +#define SF_V9_XARG4 0xa8 +#define SF_V9_XARG5 0xb0 +#define SF_V9_XXARG 0xb8 + +#define SF_L0 0x00 +#define SF_L1 0x04 +#define SF_L2 0x08 +#define SF_L3 0x0c +#define SF_L4 0x10 +#define SF_L5 0x14 +#define SF_L6 0x18 +#define SF_L7 0x1c +#define SF_I0 0x20 +#define SF_I1 0x24 +#define SF_I2 0x28 +#define SF_I3 0x2c +#define SF_I4 0x30 +#define SF_I5 0x34 +#define SF_FP 0x38 +#define SF_PC 0x3c +#define SF_RETP 0x40 +#define SF_XARG0 0x44 +#define SF_XARG1 0x48 +#define SF_XARG2 0x4c +#define SF_XARG3 0x50 +#define SF_XARG4 0x54 +#define SF_XARG5 0x58 +#define SF_XXARG 0x5c + + +/* Stuff for the ptrace system call */ +#define PTRACE_SPARC_DETACH 11 +#define PTRACE_GETREGS 12 +#define PTRACE_SETREGS 13 +#define PTRACE_GETFPREGS 14 +#define PTRACE_SETFPREGS 15 +#define PTRACE_READDATA 16 +#define PTRACE_WRITEDATA 17 +#define PTRACE_READTEXT 18 +#define PTRACE_WRITETEXT 19 +#define PTRACE_GETFPAREGS 20 +#define PTRACE_SETFPAREGS 21 + +/* There are for debugging 64-bit processes, either from a 32 or 64 bit + * parent. Thus their complements are for debugging 32-bit processes only. + */ + +#define PTRACE_GETREGS64 22 +#define PTRACE_SETREGS64 23 +/* PTRACE_SYSCALL is 24 */ +#define PTRACE_GETFPREGS64 25 +#define PTRACE_SETFPREGS64 26 + +#endif /* _UAPI__SPARC_PTRACE_H */ diff --git a/arch/sparc/include/asm/resource.h b/arch/sparc/include/uapi/asm/resource.h index fe163cafb4c..fe163cafb4c 100644 --- a/arch/sparc/include/asm/resource.h +++ b/arch/sparc/include/uapi/asm/resource.h diff --git a/arch/sparc/include/asm/sembuf.h b/arch/sparc/include/uapi/asm/sembuf.h index faee1be08d6..faee1be08d6 100644 --- a/arch/sparc/include/asm/sembuf.h +++ b/arch/sparc/include/uapi/asm/sembuf.h diff --git a/arch/sparc/include/uapi/asm/setup.h b/arch/sparc/include/uapi/asm/setup.h new file mode 100644 index 00000000000..53376845087 --- /dev/null +++ b/arch/sparc/include/uapi/asm/setup.h @@ -0,0 +1,15 @@ +/* + * Just a place holder. + */ + +#ifndef _UAPI_SPARC_SETUP_H +#define _UAPI_SPARC_SETUP_H + +#if defined(__sparc__) && defined(__arch64__) +# define COMMAND_LINE_SIZE 2048 +#else +# define COMMAND_LINE_SIZE 256 +#endif + + +#endif /* _UAPI_SPARC_SETUP_H */ diff --git a/arch/sparc/include/asm/shmbuf.h b/arch/sparc/include/uapi/asm/shmbuf.h index 83a16055363..83a16055363 100644 --- a/arch/sparc/include/asm/shmbuf.h +++ b/arch/sparc/include/uapi/asm/shmbuf.h diff --git a/arch/sparc/include/uapi/asm/sigcontext.h b/arch/sparc/include/uapi/asm/sigcontext.h new file mode 100644 index 00000000000..ae5704fa77a --- /dev/null +++ b/arch/sparc/include/uapi/asm/sigcontext.h @@ -0,0 +1,4 @@ +/* + * There isn't anything here anymore, but the file must not be empty or patch + * will delete it. + */ diff --git a/arch/sparc/include/uapi/asm/siginfo.h b/arch/sparc/include/uapi/asm/siginfo.h new file mode 100644 index 00000000000..2d9b79ccaa5 --- /dev/null +++ b/arch/sparc/include/uapi/asm/siginfo.h @@ -0,0 +1,25 @@ +#ifndef _UAPI__SPARC_SIGINFO_H +#define _UAPI__SPARC_SIGINFO_H + +#if defined(__sparc__) && defined(__arch64__) + +#define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) +#define __ARCH_SI_BAND_T int + +#endif /* defined(__sparc__) && defined(__arch64__) */ + + +#define __ARCH_SI_TRAPNO + +#include <asm-generic/siginfo.h> + + +#define SI_NOINFO 32767 /* no information in siginfo_t */ + +/* + * SIGEMT si_codes + */ +#define EMT_TAGOVF (__SI_FAULT|1) /* tag overflow */ +#define NSIGEMT 1 + +#endif /* _UAPI__SPARC_SIGINFO_H */ diff --git a/arch/sparc/include/uapi/asm/signal.h b/arch/sparc/include/uapi/asm/signal.h new file mode 100644 index 00000000000..f387400fcfd --- /dev/null +++ b/arch/sparc/include/uapi/asm/signal.h @@ -0,0 +1,181 @@ +#ifndef _UAPI__SPARC_SIGNAL_H +#define _UAPI__SPARC_SIGNAL_H + +#include <asm/sigcontext.h> +#include <linux/compiler.h> + + +/* On the Sparc the signal handlers get passed a 'sub-signal' code + * for certain signal types, which we document here. + */ +#define SIGHUP 1 +#define SIGINT 2 +#define SIGQUIT 3 +#define SIGILL 4 +#define SUBSIG_STACK 0 +#define SUBSIG_ILLINST 2 +#define SUBSIG_PRIVINST 3 +#define SUBSIG_BADTRAP(t) (0x80 + (t)) + +#define SIGTRAP 5 +#define SIGABRT 6 +#define SIGIOT 6 + +#define SIGEMT 7 +#define SUBSIG_TAG 10 + +#define SIGFPE 8 +#define SUBSIG_FPDISABLED 0x400 +#define SUBSIG_FPERROR 0x404 +#define SUBSIG_FPINTOVFL 0x001 +#define SUBSIG_FPSTSIG 0x002 +#define SUBSIG_IDIVZERO 0x014 +#define SUBSIG_FPINEXACT 0x0c4 +#define SUBSIG_FPDIVZERO 0x0c8 +#define SUBSIG_FPUNFLOW 0x0cc +#define SUBSIG_FPOPERROR 0x0d0 +#define SUBSIG_FPOVFLOW 0x0d4 + +#define SIGKILL 9 +#define SIGBUS 10 +#define SUBSIG_BUSTIMEOUT 1 +#define SUBSIG_ALIGNMENT 2 +#define SUBSIG_MISCERROR 5 + +#define SIGSEGV 11 +#define SUBSIG_NOMAPPING 3 +#define SUBSIG_PROTECTION 4 +#define SUBSIG_SEGERROR 5 + +#define SIGSYS 12 + +#define SIGPIPE 13 +#define SIGALRM 14 +#define SIGTERM 15 +#define SIGURG 16 + +/* SunOS values which deviate from the Linux/i386 ones */ +#define SIGSTOP 17 +#define SIGTSTP 18 +#define SIGCONT 19 +#define SIGCHLD 20 +#define SIGTTIN 21 +#define SIGTTOU 22 +#define SIGIO 23 +#define SIGPOLL SIGIO /* SysV name for SIGIO */ +#define SIGXCPU 24 +#define SIGXFSZ 25 +#define SIGVTALRM 26 +#define SIGPROF 27 +#define SIGWINCH 28 +#define SIGLOST 29 +#define SIGPWR SIGLOST +#define SIGUSR1 30 +#define SIGUSR2 31 + +/* Most things should be clean enough to redefine this at will, if care + is taken to make libc match. */ + +#define __OLD_NSIG 32 +#define __NEW_NSIG 64 +#ifdef __arch64__ +#define _NSIG_BPW 64 +#else +#define _NSIG_BPW 32 +#endif +#define _NSIG_WORDS (__NEW_NSIG / _NSIG_BPW) + +#define SIGRTMIN 32 +#define SIGRTMAX __NEW_NSIG + +#if defined(__KERNEL__) || defined(__WANT_POSIX1B_SIGNALS__) +#define _NSIG __NEW_NSIG +#define __new_sigset_t sigset_t +#define __new_sigaction sigaction +#define __new_sigaction32 sigaction32 +#define __old_sigset_t old_sigset_t +#define __old_sigaction old_sigaction +#define __old_sigaction32 old_sigaction32 +#else +#define _NSIG __OLD_NSIG +#define NSIG _NSIG +#define __old_sigset_t sigset_t +#define __old_sigaction sigaction +#define __old_sigaction32 sigaction32 +#endif + +#ifndef __ASSEMBLY__ + +typedef unsigned long __old_sigset_t; /* at least 32 bits */ + +typedef struct { + unsigned long sig[_NSIG_WORDS]; +} __new_sigset_t; + +/* A SunOS sigstack */ +struct sigstack { + /* XXX 32-bit pointers pinhead XXX */ + char *the_stack; + int cur_status; +}; + +/* Sigvec flags */ +#define _SV_SSTACK 1u /* This signal handler should use sig-stack */ +#define _SV_INTR 2u /* Sig return should not restart system call */ +#define _SV_RESET 4u /* Set handler to SIG_DFL upon taken signal */ +#define _SV_IGNCHILD 8u /* Do not send SIGCHLD */ + +/* + * sa_flags values: SA_STACK is not currently supported, but will allow the + * usage of signal stacks by using the (now obsolete) sa_restorer field in + * the sigaction structure as a stack pointer. This is now possible due to + * the changes in signal handling. LBT 010493. + * SA_RESTART flag to get restarting signals (which were the default long ago) + */ +#define SA_NOCLDSTOP _SV_IGNCHILD +#define SA_STACK _SV_SSTACK +#define SA_ONSTACK _SV_SSTACK +#define SA_RESTART _SV_INTR +#define SA_ONESHOT _SV_RESET +#define SA_NODEFER 0x20u +#define SA_NOCLDWAIT 0x100u +#define SA_SIGINFO 0x200u + +#define SA_NOMASK SA_NODEFER + +#define SIG_BLOCK 0x01 /* for blocking signals */ +#define SIG_UNBLOCK 0x02 /* for unblocking signals */ +#define SIG_SETMASK 0x04 /* for setting the signal mask */ + +#define MINSIGSTKSZ 4096 +#define SIGSTKSZ 16384 + + +#include <asm-generic/signal-defs.h> + +#ifndef __KERNEL__ +struct __new_sigaction { + __sighandler_t sa_handler; + unsigned long sa_flags; + __sigrestore_t sa_restorer; /* not used by Linux/SPARC yet */ + __new_sigset_t sa_mask; +}; + +struct __old_sigaction { + __sighandler_t sa_handler; + __old_sigset_t sa_mask; + unsigned long sa_flags; + void (*sa_restorer)(void); /* not used by Linux/SPARC yet */ +}; +#endif + +typedef struct sigaltstack { + void __user *ss_sp; + int ss_flags; + size_t ss_size; +} stack_t; + + +#endif /* !(__ASSEMBLY__) */ + +#endif /* _UAPI__SPARC_SIGNAL_H */ diff --git a/arch/sparc/include/asm/socket.h b/arch/sparc/include/uapi/asm/socket.h index bea1568ae4a..54d9608681b 100644 --- a/arch/sparc/include/asm/socket.h +++ b/arch/sparc/include/uapi/asm/socket.h @@ -15,7 +15,7 @@ #define SO_PEERCRED 0x0040 #define SO_LINGER 0x0080 #define SO_OOBINLINE 0x0100 -/* To add :#define SO_REUSEPORT 0x0200 */ +#define SO_REUSEPORT 0x0200 #define SO_BSDCOMPAT 0x0400 #define SO_RCVLOWAT 0x0800 #define SO_SNDLOWAT 0x1000 @@ -41,6 +41,7 @@ #define SO_ATTACH_FILTER 0x001a #define SO_DETACH_FILTER 0x001b +#define SO_GET_FILTER SO_ATTACH_FILTER #define SO_PEERNAME 0x001c #define SO_TIMESTAMP 0x001d @@ -65,6 +66,15 @@ /* Instruct lower device to use last 4-bytes of skb data as FCS */ #define SO_NOFCS 0x0027 +#define SO_LOCK_FILTER 0x0028 + +#define SO_SELECT_ERR_QUEUE 0x0029 + +#define SO_BUSY_POLL 0x0030 + +#define SO_MAX_PACING_RATE 0x0031 + +#define SO_BPF_EXTENSIONS 0x0032 /* Security levels - as per NRL IPv6 - don't actually do anything */ #define SO_SECURITY_AUTHENTICATION 0x5001 diff --git a/arch/sparc/include/asm/sockios.h b/arch/sparc/include/uapi/asm/sockios.h index 990ea746486..990ea746486 100644 --- a/arch/sparc/include/asm/sockios.h +++ b/arch/sparc/include/uapi/asm/sockios.h diff --git a/arch/sparc/include/asm/stat.h b/arch/sparc/include/uapi/asm/stat.h index a232e9e1f4e..a232e9e1f4e 100644 --- a/arch/sparc/include/asm/stat.h +++ b/arch/sparc/include/uapi/asm/stat.h diff --git a/arch/sparc/include/asm/statfs.h b/arch/sparc/include/uapi/asm/statfs.h index 55e607ad461..55e607ad461 100644 --- a/arch/sparc/include/asm/statfs.h +++ b/arch/sparc/include/uapi/asm/statfs.h diff --git a/arch/sparc/include/asm/swab.h b/arch/sparc/include/uapi/asm/swab.h index a34ad079487..a34ad079487 100644 --- a/arch/sparc/include/asm/swab.h +++ b/arch/sparc/include/uapi/asm/swab.h diff --git a/arch/sparc/include/uapi/asm/termbits.h b/arch/sparc/include/uapi/asm/termbits.h new file mode 100644 index 00000000000..dd91642fcca --- /dev/null +++ b/arch/sparc/include/uapi/asm/termbits.h @@ -0,0 +1,263 @@ +#ifndef _UAPI_SPARC_TERMBITS_H +#define _UAPI_SPARC_TERMBITS_H + +#include <linux/posix_types.h> + +typedef unsigned char cc_t; +typedef unsigned int speed_t; + +#if defined(__sparc__) && defined(__arch64__) +typedef unsigned int tcflag_t; +#else +typedef unsigned long tcflag_t; +#endif + +#define NCC 8 +struct termio { + unsigned short c_iflag; /* input mode flags */ + unsigned short c_oflag; /* output mode flags */ + unsigned short c_cflag; /* control mode flags */ + unsigned short c_lflag; /* local mode flags */ + unsigned char c_line; /* line discipline */ + unsigned char c_cc[NCC]; /* control characters */ +}; + +#define NCCS 17 +struct termios { + tcflag_t c_iflag; /* input mode flags */ + tcflag_t c_oflag; /* output mode flags */ + tcflag_t c_cflag; /* control mode flags */ + tcflag_t c_lflag; /* local mode flags */ + cc_t c_line; /* line discipline */ +#ifndef __KERNEL__ + cc_t c_cc[NCCS]; /* control characters */ +#else + cc_t c_cc[NCCS+2]; /* kernel needs 2 more to hold vmin/vtime */ +#define SIZEOF_USER_TERMIOS sizeof (struct termios) - (2*sizeof (cc_t)) +#endif +}; + +struct termios2 { + tcflag_t c_iflag; /* input mode flags */ + tcflag_t c_oflag; /* output mode flags */ + tcflag_t c_cflag; /* control mode flags */ + tcflag_t c_lflag; /* local mode flags */ + cc_t c_line; /* line discipline */ + cc_t c_cc[NCCS+2]; /* control characters */ + speed_t c_ispeed; /* input speed */ + speed_t c_ospeed; /* output speed */ +}; + +struct ktermios { + tcflag_t c_iflag; /* input mode flags */ + tcflag_t c_oflag; /* output mode flags */ + tcflag_t c_cflag; /* control mode flags */ + tcflag_t c_lflag; /* local mode flags */ + cc_t c_line; /* line discipline */ + cc_t c_cc[NCCS+2]; /* control characters */ + speed_t c_ispeed; /* input speed */ + speed_t c_ospeed; /* output speed */ +}; + +/* c_cc characters */ +#define VINTR 0 +#define VQUIT 1 +#define VERASE 2 +#define VKILL 3 +#define VEOF 4 +#define VEOL 5 +#define VEOL2 6 +#define VSWTC 7 +#define VSTART 8 +#define VSTOP 9 + + + +#define VSUSP 10 +#define VDSUSP 11 /* SunOS POSIX nicety I do believe... */ +#define VREPRINT 12 +#define VDISCARD 13 +#define VWERASE 14 +#define VLNEXT 15 + +/* Kernel keeps vmin/vtime separated, user apps assume vmin/vtime is + * shared with eof/eol + */ +#ifndef __KERNEL__ +#define VMIN VEOF +#define VTIME VEOL +#endif + +/* c_iflag bits */ +#define IGNBRK 0x00000001 +#define BRKINT 0x00000002 +#define IGNPAR 0x00000004 +#define PARMRK 0x00000008 +#define INPCK 0x00000010 +#define ISTRIP 0x00000020 +#define INLCR 0x00000040 +#define IGNCR 0x00000080 +#define ICRNL 0x00000100 +#define IUCLC 0x00000200 +#define IXON 0x00000400 +#define IXANY 0x00000800 +#define IXOFF 0x00001000 +#define IMAXBEL 0x00002000 +#define IUTF8 0x00004000 + +/* c_oflag bits */ +#define OPOST 0x00000001 +#define OLCUC 0x00000002 +#define ONLCR 0x00000004 +#define OCRNL 0x00000008 +#define ONOCR 0x00000010 +#define ONLRET 0x00000020 +#define OFILL 0x00000040 +#define OFDEL 0x00000080 +#define NLDLY 0x00000100 +#define NL0 0x00000000 +#define NL1 0x00000100 +#define CRDLY 0x00000600 +#define CR0 0x00000000 +#define CR1 0x00000200 +#define CR2 0x00000400 +#define CR3 0x00000600 +#define TABDLY 0x00001800 +#define TAB0 0x00000000 +#define TAB1 0x00000800 +#define TAB2 0x00001000 +#define TAB3 0x00001800 +#define XTABS 0x00001800 +#define BSDLY 0x00002000 +#define BS0 0x00000000 +#define BS1 0x00002000 +#define VTDLY 0x00004000 +#define VT0 0x00000000 +#define VT1 0x00004000 +#define FFDLY 0x00008000 +#define FF0 0x00000000 +#define FF1 0x00008000 +#define PAGEOUT 0x00010000 /* SUNOS specific */ +#define WRAP 0x00020000 /* SUNOS specific */ + +/* c_cflag bit meaning */ +#define CBAUD 0x0000100f +#define B0 0x00000000 /* hang up */ +#define B50 0x00000001 +#define B75 0x00000002 +#define B110 0x00000003 +#define B134 0x00000004 +#define B150 0x00000005 +#define B200 0x00000006 +#define B300 0x00000007 +#define B600 0x00000008 +#define B1200 0x00000009 +#define B1800 0x0000000a +#define B2400 0x0000000b +#define B4800 0x0000000c +#define B9600 0x0000000d +#define B19200 0x0000000e +#define B38400 0x0000000f +#define EXTA B19200 +#define EXTB B38400 +#define CSIZE 0x00000030 +#define CS5 0x00000000 +#define CS6 0x00000010 +#define CS7 0x00000020 +#define CS8 0x00000030 +#define CSTOPB 0x00000040 +#define CREAD 0x00000080 +#define PARENB 0x00000100 +#define PARODD 0x00000200 +#define HUPCL 0x00000400 +#define CLOCAL 0x00000800 +#define CBAUDEX 0x00001000 +/* We'll never see these speeds with the Zilogs, but for completeness... */ +#define BOTHER 0x00001000 +#define B57600 0x00001001 +#define B115200 0x00001002 +#define B230400 0x00001003 +#define B460800 0x00001004 +/* This is what we can do with the Zilogs. */ +#define B76800 0x00001005 +/* This is what we can do with the SAB82532. */ +#define B153600 0x00001006 +#define B307200 0x00001007 +#define B614400 0x00001008 +#define B921600 0x00001009 +/* And these are the rest... */ +#define B500000 0x0000100a +#define B576000 0x0000100b +#define B1000000 0x0000100c +#define B1152000 0x0000100d +#define B1500000 0x0000100e +#define B2000000 0x0000100f +/* These have totally bogus values and nobody uses them + so far. Later on we'd have to use say 0x10000x and + adjust CBAUD constant and drivers accordingly. +#define B2500000 0x00001010 +#define B3000000 0x00001011 +#define B3500000 0x00001012 +#define B4000000 0x00001013 */ +#define CIBAUD 0x100f0000 /* input baud rate (not used) */ +#define CMSPAR 0x40000000 /* mark or space (stick) parity */ +#define CRTSCTS 0x80000000 /* flow control */ + +#define IBSHIFT 16 /* Shift from CBAUD to CIBAUD */ + +/* c_lflag bits */ +#define ISIG 0x00000001 +#define ICANON 0x00000002 +#define XCASE 0x00000004 +#define ECHO 0x00000008 +#define ECHOE 0x00000010 +#define ECHOK 0x00000020 +#define ECHONL 0x00000040 +#define NOFLSH 0x00000080 +#define TOSTOP 0x00000100 +#define ECHOCTL 0x00000200 +#define ECHOPRT 0x00000400 +#define ECHOKE 0x00000800 +#define DEFECHO 0x00001000 /* SUNOS thing, what is it? */ +#define FLUSHO 0x00002000 +#define PENDIN 0x00004000 +#define IEXTEN 0x00008000 +#define EXTPROC 0x00010000 + +/* modem lines */ +#define TIOCM_LE 0x001 +#define TIOCM_DTR 0x002 +#define TIOCM_RTS 0x004 +#define TIOCM_ST 0x008 +#define TIOCM_SR 0x010 +#define TIOCM_CTS 0x020 +#define TIOCM_CAR 0x040 +#define TIOCM_RNG 0x080 +#define TIOCM_DSR 0x100 +#define TIOCM_CD TIOCM_CAR +#define TIOCM_RI TIOCM_RNG +#define TIOCM_OUT1 0x2000 +#define TIOCM_OUT2 0x4000 +#define TIOCM_LOOP 0x8000 + +/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */ +#define TIOCSER_TEMT 0x01 /* Transmitter physically empty */ + + +/* tcflow() and TCXONC use these */ +#define TCOOFF 0 +#define TCOON 1 +#define TCIOFF 2 +#define TCION 3 + +/* tcflush() and TCFLSH use these */ +#define TCIFLUSH 0 +#define TCOFLUSH 1 +#define TCIOFLUSH 2 + +/* tcsetattr uses these */ +#define TCSANOW 0 +#define TCSADRAIN 1 +#define TCSAFLUSH 2 + +#endif /* _UAPI_SPARC_TERMBITS_H */ diff --git a/arch/sparc/include/uapi/asm/termios.h b/arch/sparc/include/uapi/asm/termios.h new file mode 100644 index 00000000000..ea6f09e51e5 --- /dev/null +++ b/arch/sparc/include/uapi/asm/termios.h @@ -0,0 +1,43 @@ +#ifndef _UAPI_SPARC_TERMIOS_H +#define _UAPI_SPARC_TERMIOS_H + +#include <asm/ioctls.h> +#include <asm/termbits.h> + +#if defined(__KERNEL__) || defined(__DEFINE_BSD_TERMIOS) +struct sgttyb { + char sg_ispeed; + char sg_ospeed; + char sg_erase; + char sg_kill; + short sg_flags; +}; + +struct tchars { + char t_intrc; + char t_quitc; + char t_startc; + char t_stopc; + char t_eofc; + char t_brkc; +}; + +struct ltchars { + char t_suspc; + char t_dsuspc; + char t_rprntc; + char t_flushc; + char t_werasc; + char t_lnextc; +}; +#endif /* __KERNEL__ */ + +struct winsize { + unsigned short ws_row; + unsigned short ws_col; + unsigned short ws_xpixel; + unsigned short ws_ypixel; +}; + + +#endif /* _UAPI_SPARC_TERMIOS_H */ diff --git a/arch/sparc/include/uapi/asm/traps.h b/arch/sparc/include/uapi/asm/traps.h new file mode 100644 index 00000000000..a4eceace6cc --- /dev/null +++ b/arch/sparc/include/uapi/asm/traps.h @@ -0,0 +1,120 @@ +/* + * traps.h: Format of entries for the Sparc trap table. + * + * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu) + */ + +#ifndef _UAPI_SPARC_TRAPS_H +#define _UAPI_SPARC_TRAPS_H + +#define NUM_SPARC_TRAPS 255 + +#ifndef __ASSEMBLY__ +#endif /* !(__ASSEMBLY__) */ + +/* For patching the trap table at boot time, we need to know how to + * form various common Sparc instructions. Thus these macros... + */ + +#define SPARC_MOV_CONST_L3(const) (0xa6102000 | (const&0xfff)) + +/* The following assumes that the branch lies before the place we + * are branching to. This is the case for a trap vector... + * You have been warned. + */ +#define SPARC_BRANCH(dest_addr, inst_addr) \ + (0x10800000 | (((dest_addr-inst_addr)>>2)&0x3fffff)) + +#define SPARC_RD_PSR_L0 (0xa1480000) +#define SPARC_RD_WIM_L3 (0xa7500000) +#define SPARC_NOP (0x01000000) + +/* Various interesting trap levels. */ +/* First, hardware traps. */ +#define SP_TRAP_TFLT 0x1 /* Text fault */ +#define SP_TRAP_II 0x2 /* Illegal Instruction */ +#define SP_TRAP_PI 0x3 /* Privileged Instruction */ +#define SP_TRAP_FPD 0x4 /* Floating Point Disabled */ +#define SP_TRAP_WOVF 0x5 /* Window Overflow */ +#define SP_TRAP_WUNF 0x6 /* Window Underflow */ +#define SP_TRAP_MNA 0x7 /* Memory Address Unaligned */ +#define SP_TRAP_FPE 0x8 /* Floating Point Exception */ +#define SP_TRAP_DFLT 0x9 /* Data Fault */ +#define SP_TRAP_TOF 0xa /* Tag Overflow */ +#define SP_TRAP_WDOG 0xb /* Watchpoint Detected */ +#define SP_TRAP_IRQ1 0x11 /* IRQ level 1 */ +#define SP_TRAP_IRQ2 0x12 /* IRQ level 2 */ +#define SP_TRAP_IRQ3 0x13 /* IRQ level 3 */ +#define SP_TRAP_IRQ4 0x14 /* IRQ level 4 */ +#define SP_TRAP_IRQ5 0x15 /* IRQ level 5 */ +#define SP_TRAP_IRQ6 0x16 /* IRQ level 6 */ +#define SP_TRAP_IRQ7 0x17 /* IRQ level 7 */ +#define SP_TRAP_IRQ8 0x18 /* IRQ level 8 */ +#define SP_TRAP_IRQ9 0x19 /* IRQ level 9 */ +#define SP_TRAP_IRQ10 0x1a /* IRQ level 10 */ +#define SP_TRAP_IRQ11 0x1b /* IRQ level 11 */ +#define SP_TRAP_IRQ12 0x1c /* IRQ level 12 */ +#define SP_TRAP_IRQ13 0x1d /* IRQ level 13 */ +#define SP_TRAP_IRQ14 0x1e /* IRQ level 14 */ +#define SP_TRAP_IRQ15 0x1f /* IRQ level 15 Non-maskable */ +#define SP_TRAP_RACC 0x20 /* Register Access Error ??? */ +#define SP_TRAP_IACC 0x21 /* Instruction Access Error */ +#define SP_TRAP_CPDIS 0x24 /* Co-Processor Disabled */ +#define SP_TRAP_BADFL 0x25 /* Unimplemented Flush Instruction */ +#define SP_TRAP_CPEXP 0x28 /* Co-Processor Exception */ +#define SP_TRAP_DACC 0x29 /* Data Access Error */ +#define SP_TRAP_DIVZ 0x2a /* Divide By Zero */ +#define SP_TRAP_DSTORE 0x2b /* Data Store Error ??? */ +#define SP_TRAP_DMM 0x2c /* Data Access MMU Miss ??? */ +#define SP_TRAP_IMM 0x3c /* Instruction Access MMU Miss ??? */ + +/* Now the Software Traps... */ +#define SP_TRAP_SUNOS 0x80 /* SunOS System Call */ +#define SP_TRAP_SBPT 0x81 /* Software Breakpoint */ +#define SP_TRAP_SDIVZ 0x82 /* Software Divide-by-Zero trap */ +#define SP_TRAP_FWIN 0x83 /* Flush Windows */ +#define SP_TRAP_CWIN 0x84 /* Clean Windows */ +#define SP_TRAP_RCHK 0x85 /* Range Check */ +#define SP_TRAP_FUNA 0x86 /* Fix Unaligned Access */ +#define SP_TRAP_IOWFL 0x87 /* Integer Overflow */ +#define SP_TRAP_SOLARIS 0x88 /* Solaris System Call */ +#define SP_TRAP_NETBSD 0x89 /* NetBSD System Call */ +#define SP_TRAP_LINUX 0x90 /* Linux System Call */ + +/* Names used for compatibility with SunOS */ +#define ST_SYSCALL 0x00 +#define ST_BREAKPOINT 0x01 +#define ST_DIV0 0x02 +#define ST_FLUSH_WINDOWS 0x03 +#define ST_CLEAN_WINDOWS 0x04 +#define ST_RANGE_CHECK 0x05 +#define ST_FIX_ALIGN 0x06 +#define ST_INT_OVERFLOW 0x07 + +/* Special traps... */ +#define SP_TRAP_KBPT1 0xfe /* KADB/PROM Breakpoint one */ +#define SP_TRAP_KBPT2 0xff /* KADB/PROM Breakpoint two */ + +/* Handy Macros */ +/* Is this a trap we never expect to get? */ +#define BAD_TRAP_P(level) \ + ((level > SP_TRAP_WDOG && level < SP_TRAP_IRQ1) || \ + (level > SP_TRAP_IACC && level < SP_TRAP_CPDIS) || \ + (level > SP_TRAP_BADFL && level < SP_TRAP_CPEXP) || \ + (level > SP_TRAP_DMM && level < SP_TRAP_IMM) || \ + (level > SP_TRAP_IMM && level < SP_TRAP_SUNOS) || \ + (level > SP_TRAP_LINUX && level < SP_TRAP_KBPT1)) + +/* Is this a Hardware trap? */ +#define HW_TRAP_P(level) ((level > 0) && (level < SP_TRAP_SUNOS)) + +/* Is this a Software trap? */ +#define SW_TRAP_P(level) ((level >= SP_TRAP_SUNOS) && (level <= SP_TRAP_KBPT2)) + +/* Is this a system call for some OS we know about? */ +#define SCALL_TRAP_P(level) ((level == SP_TRAP_SUNOS) || \ + (level == SP_TRAP_SOLARIS) || \ + (level == SP_TRAP_NETBSD) || \ + (level == SP_TRAP_LINUX)) + +#endif /* _UAPI_SPARC_TRAPS_H */ diff --git a/arch/sparc/include/asm/uctx.h b/arch/sparc/include/uapi/asm/uctx.h index dc937c75ffd..dc937c75ffd 100644 --- a/arch/sparc/include/asm/uctx.h +++ b/arch/sparc/include/uapi/asm/uctx.h diff --git a/arch/sparc/include/uapi/asm/unistd.h b/arch/sparc/include/uapi/asm/unistd.h new file mode 100644 index 00000000000..42f2bca1d33 --- /dev/null +++ b/arch/sparc/include/uapi/asm/unistd.h @@ -0,0 +1,431 @@ +/* + * System calls under the Sparc. + * + * Don't be scared by the ugly clobbers, it is the only way I can + * think of right now to force the arguments into fixed registers + * before the trap into the system call with gcc 'asm' statements. + * + * Copyright (C) 1995, 2007 David S. Miller (davem@davemloft.net) + * + * SunOS compatibility based upon preliminary work which is: + * + * Copyright (C) 1995 Adrian M. Rodriguez (adrian@remus.rutgers.edu) + */ +#ifndef _UAPI_SPARC_UNISTD_H +#define _UAPI_SPARC_UNISTD_H + +#ifndef __32bit_syscall_numbers__ +#ifndef __arch64__ +#define __32bit_syscall_numbers__ +#endif +#endif + +#define __NR_restart_syscall 0 /* Linux Specific */ +#define __NR_exit 1 /* Common */ +#define __NR_fork 2 /* Common */ +#define __NR_read 3 /* Common */ +#define __NR_write 4 /* Common */ +#define __NR_open 5 /* Common */ +#define __NR_close 6 /* Common */ +#define __NR_wait4 7 /* Common */ +#define __NR_creat 8 /* Common */ +#define __NR_link 9 /* Common */ +#define __NR_unlink 10 /* Common */ +#define __NR_execv 11 /* SunOS Specific */ +#define __NR_chdir 12 /* Common */ +#define __NR_chown 13 /* Common */ +#define __NR_mknod 14 /* Common */ +#define __NR_chmod 15 /* Common */ +#define __NR_lchown 16 /* Common */ +#define __NR_brk 17 /* Common */ +#define __NR_perfctr 18 /* Performance counter operations */ +#define __NR_lseek 19 /* Common */ +#define __NR_getpid 20 /* Common */ +#define __NR_capget 21 /* Linux Specific */ +#define __NR_capset 22 /* Linux Specific */ +#define __NR_setuid 23 /* Implemented via setreuid in SunOS */ +#define __NR_getuid 24 /* Common */ +#define __NR_vmsplice 25 /* ENOSYS under SunOS */ +#define __NR_ptrace 26 /* Common */ +#define __NR_alarm 27 /* Implemented via setitimer in SunOS */ +#define __NR_sigaltstack 28 /* Common */ +#define __NR_pause 29 /* Is sigblock(0)->sigpause() in SunOS */ +#define __NR_utime 30 /* Implemented via utimes() under SunOS */ +#ifdef __32bit_syscall_numbers__ +#define __NR_lchown32 31 /* Linux sparc32 specific */ +#define __NR_fchown32 32 /* Linux sparc32 specific */ +#endif +#define __NR_access 33 /* Common */ +#define __NR_nice 34 /* Implemented via get/setpriority() in SunOS */ +#ifdef __32bit_syscall_numbers__ +#define __NR_chown32 35 /* Linux sparc32 specific */ +#endif +#define __NR_sync 36 /* Common */ +#define __NR_kill 37 /* Common */ +#define __NR_stat 38 /* Common */ +#define __NR_sendfile 39 /* Linux Specific */ +#define __NR_lstat 40 /* Common */ +#define __NR_dup 41 /* Common */ +#define __NR_pipe 42 /* Common */ +#define __NR_times 43 /* Implemented via getrusage() in SunOS */ +#ifdef __32bit_syscall_numbers__ +#define __NR_getuid32 44 /* Linux sparc32 specific */ +#endif +#define __NR_umount2 45 /* Linux Specific */ +#define __NR_setgid 46 /* Implemented via setregid() in SunOS */ +#define __NR_getgid 47 /* Common */ +#define __NR_signal 48 /* Implemented via sigvec() in SunOS */ +#define __NR_geteuid 49 /* SunOS calls getuid() */ +#define __NR_getegid 50 /* SunOS calls getgid() */ +#define __NR_acct 51 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_getgid32 53 /* Linux sparc32 specific */ +#else +#define __NR_memory_ordering 52 /* Linux Specific */ +#endif +#define __NR_ioctl 54 /* Common */ +#define __NR_reboot 55 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_mmap2 56 /* Linux sparc32 Specific */ +#endif +#define __NR_symlink 57 /* Common */ +#define __NR_readlink 58 /* Common */ +#define __NR_execve 59 /* Common */ +#define __NR_umask 60 /* Common */ +#define __NR_chroot 61 /* Common */ +#define __NR_fstat 62 /* Common */ +#define __NR_fstat64 63 /* Linux Specific */ +#define __NR_getpagesize 64 /* Common */ +#define __NR_msync 65 /* Common in newer 1.3.x revs... */ +#define __NR_vfork 66 /* Common */ +#define __NR_pread64 67 /* Linux Specific */ +#define __NR_pwrite64 68 /* Linux Specific */ +#ifdef __32bit_syscall_numbers__ +#define __NR_geteuid32 69 /* Linux sparc32, sbrk under SunOS */ +#define __NR_getegid32 70 /* Linux sparc32, sstk under SunOS */ +#endif +#define __NR_mmap 71 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_setreuid32 72 /* Linux sparc32, vadvise under SunOS */ +#endif +#define __NR_munmap 73 /* Common */ +#define __NR_mprotect 74 /* Common */ +#define __NR_madvise 75 /* Common */ +#define __NR_vhangup 76 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_truncate64 77 /* Linux sparc32 Specific */ +#endif +#define __NR_mincore 78 /* Common */ +#define __NR_getgroups 79 /* Common */ +#define __NR_setgroups 80 /* Common */ +#define __NR_getpgrp 81 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_setgroups32 82 /* Linux sparc32, setpgrp under SunOS */ +#endif +#define __NR_setitimer 83 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_ftruncate64 84 /* Linux sparc32 Specific */ +#endif +#define __NR_swapon 85 /* Common */ +#define __NR_getitimer 86 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_setuid32 87 /* Linux sparc32, gethostname under SunOS */ +#endif +#define __NR_sethostname 88 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_setgid32 89 /* Linux sparc32, getdtablesize under SunOS */ +#endif +#define __NR_dup2 90 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_setfsuid32 91 /* Linux sparc32, getdopt under SunOS */ +#endif +#define __NR_fcntl 92 /* Common */ +#define __NR_select 93 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_setfsgid32 94 /* Linux sparc32, setdopt under SunOS */ +#endif +#define __NR_fsync 95 /* Common */ +#define __NR_setpriority 96 /* Common */ +#define __NR_socket 97 /* Common */ +#define __NR_connect 98 /* Common */ +#define __NR_accept 99 /* Common */ +#define __NR_getpriority 100 /* Common */ +#define __NR_rt_sigreturn 101 /* Linux Specific */ +#define __NR_rt_sigaction 102 /* Linux Specific */ +#define __NR_rt_sigprocmask 103 /* Linux Specific */ +#define __NR_rt_sigpending 104 /* Linux Specific */ +#define __NR_rt_sigtimedwait 105 /* Linux Specific */ +#define __NR_rt_sigqueueinfo 106 /* Linux Specific */ +#define __NR_rt_sigsuspend 107 /* Linux Specific */ +#ifdef __32bit_syscall_numbers__ +#define __NR_setresuid32 108 /* Linux Specific, sigvec under SunOS */ +#define __NR_getresuid32 109 /* Linux Specific, sigblock under SunOS */ +#define __NR_setresgid32 110 /* Linux Specific, sigsetmask under SunOS */ +#define __NR_getresgid32 111 /* Linux Specific, sigpause under SunOS */ +#define __NR_setregid32 112 /* Linux sparc32, sigstack under SunOS */ +#else +#define __NR_setresuid 108 /* Linux Specific, sigvec under SunOS */ +#define __NR_getresuid 109 /* Linux Specific, sigblock under SunOS */ +#define __NR_setresgid 110 /* Linux Specific, sigsetmask under SunOS */ +#define __NR_getresgid 111 /* Linux Specific, sigpause under SunOS */ +#endif +#define __NR_recvmsg 113 /* Common */ +#define __NR_sendmsg 114 /* Common */ +#ifdef __32bit_syscall_numbers__ +#define __NR_getgroups32 115 /* Linux sparc32, vtrace under SunOS */ +#endif +#define __NR_gettimeofday 116 /* Common */ +#define __NR_getrusage 117 /* Common */ +#define __NR_getsockopt 118 /* Common */ +#define __NR_getcwd 119 /* Linux Specific */ +#define __NR_readv 120 /* Common */ +#define __NR_writev 121 /* Common */ +#define __NR_settimeofday 122 /* Common */ +#define __NR_fchown 123 /* Common */ +#define __NR_fchmod 124 /* Common */ +#define __NR_recvfrom 125 /* Common */ +#define __NR_setreuid 126 /* Common */ +#define __NR_setregid 127 /* Common */ +#define __NR_rename 128 /* Common */ +#define __NR_truncate 129 /* Common */ +#define __NR_ftruncate 130 /* Common */ +#define __NR_flock 131 /* Common */ +#define __NR_lstat64 132 /* Linux Specific */ +#define __NR_sendto 133 /* Common */ +#define __NR_shutdown 134 /* Common */ +#define __NR_socketpair 135 /* Common */ +#define __NR_mkdir 136 /* Common */ +#define __NR_rmdir 137 /* Common */ +#define __NR_utimes 138 /* SunOS Specific */ +#define __NR_stat64 139 /* Linux Specific */ +#define __NR_sendfile64 140 /* adjtime under SunOS */ +#define __NR_getpeername 141 /* Common */ +#define __NR_futex 142 /* gethostid under SunOS */ +#define __NR_gettid 143 /* ENOSYS under SunOS */ +#define __NR_getrlimit 144 /* Common */ +#define __NR_setrlimit 145 /* Common */ +#define __NR_pivot_root 146 /* Linux Specific, killpg under SunOS */ +#define __NR_prctl 147 /* ENOSYS under SunOS */ +#define __NR_pciconfig_read 148 /* ENOSYS under SunOS */ +#define __NR_pciconfig_write 149 /* ENOSYS under SunOS */ +#define __NR_getsockname 150 /* Common */ +#define __NR_inotify_init 151 /* Linux specific */ +#define __NR_inotify_add_watch 152 /* Linux specific */ +#define __NR_poll 153 /* Common */ +#define __NR_getdents64 154 /* Linux specific */ +#ifdef __32bit_syscall_numbers__ +#define __NR_fcntl64 155 /* Linux sparc32 Specific */ +#endif +#define __NR_inotify_rm_watch 156 /* Linux specific */ +#define __NR_statfs 157 /* Common */ +#define __NR_fstatfs 158 /* Common */ +#define __NR_umount 159 /* Common */ +#define __NR_sched_set_affinity 160 /* Linux specific, async_daemon under SunOS */ +#define __NR_sched_get_affinity 161 /* Linux specific, getfh under SunOS */ +#define __NR_getdomainname 162 /* SunOS Specific */ +#define __NR_setdomainname 163 /* Common */ +#ifndef __32bit_syscall_numbers__ +#define __NR_utrap_install 164 /* SYSV ABI/v9 required */ +#endif +#define __NR_quotactl 165 /* Common */ +#define __NR_set_tid_address 166 /* Linux specific, exportfs under SunOS */ +#define __NR_mount 167 /* Common */ +#define __NR_ustat 168 /* Common */ +#define __NR_setxattr 169 /* SunOS: semsys */ +#define __NR_lsetxattr 170 /* SunOS: msgsys */ +#define __NR_fsetxattr 171 /* SunOS: shmsys */ +#define __NR_getxattr 172 /* SunOS: auditsys */ +#define __NR_lgetxattr 173 /* SunOS: rfssys */ +#define __NR_getdents 174 /* Common */ +#define __NR_setsid 175 /* Common */ +#define __NR_fchdir 176 /* Common */ +#define __NR_fgetxattr 177 /* SunOS: fchroot */ +#define __NR_listxattr 178 /* SunOS: vpixsys */ +#define __NR_llistxattr 179 /* SunOS: aioread */ +#define __NR_flistxattr 180 /* SunOS: aiowrite */ +#define __NR_removexattr 181 /* SunOS: aiowait */ +#define __NR_lremovexattr 182 /* SunOS: aiocancel */ +#define __NR_sigpending 183 /* Common */ +#define __NR_query_module 184 /* Linux Specific */ +#define __NR_setpgid 185 /* Common */ +#define __NR_fremovexattr 186 /* SunOS: pathconf */ +#define __NR_tkill 187 /* SunOS: fpathconf */ +#define __NR_exit_group 188 /* Linux specific, sysconf undef SunOS */ +#define __NR_uname 189 /* Linux Specific */ +#define __NR_init_module 190 /* Linux Specific */ +#define __NR_personality 191 /* Linux Specific */ +#define __NR_remap_file_pages 192 /* Linux Specific */ +#define __NR_epoll_create 193 /* Linux Specific */ +#define __NR_epoll_ctl 194 /* Linux Specific */ +#define __NR_epoll_wait 195 /* Linux Specific */ +#define __NR_ioprio_set 196 /* Linux Specific */ +#define __NR_getppid 197 /* Linux Specific */ +#define __NR_sigaction 198 /* Linux Specific */ +#define __NR_sgetmask 199 /* Linux Specific */ +#define __NR_ssetmask 200 /* Linux Specific */ +#define __NR_sigsuspend 201 /* Linux Specific */ +#define __NR_oldlstat 202 /* Linux Specific */ +#define __NR_uselib 203 /* Linux Specific */ +#define __NR_readdir 204 /* Linux Specific */ +#define __NR_readahead 205 /* Linux Specific */ +#define __NR_socketcall 206 /* Linux Specific */ +#define __NR_syslog 207 /* Linux Specific */ +#define __NR_lookup_dcookie 208 /* Linux Specific */ +#define __NR_fadvise64 209 /* Linux Specific */ +#define __NR_fadvise64_64 210 /* Linux Specific */ +#define __NR_tgkill 211 /* Linux Specific */ +#define __NR_waitpid 212 /* Linux Specific */ +#define __NR_swapoff 213 /* Linux Specific */ +#define __NR_sysinfo 214 /* Linux Specific */ +#define __NR_ipc 215 /* Linux Specific */ +#define __NR_sigreturn 216 /* Linux Specific */ +#define __NR_clone 217 /* Linux Specific */ +#define __NR_ioprio_get 218 /* Linux Specific */ +#define __NR_adjtimex 219 /* Linux Specific */ +#define __NR_sigprocmask 220 /* Linux Specific */ +#define __NR_create_module 221 /* Linux Specific */ +#define __NR_delete_module 222 /* Linux Specific */ +#define __NR_get_kernel_syms 223 /* Linux Specific */ +#define __NR_getpgid 224 /* Linux Specific */ +#define __NR_bdflush 225 /* Linux Specific */ +#define __NR_sysfs 226 /* Linux Specific */ +#define __NR_afs_syscall 227 /* Linux Specific */ +#define __NR_setfsuid 228 /* Linux Specific */ +#define __NR_setfsgid 229 /* Linux Specific */ +#define __NR__newselect 230 /* Linux Specific */ +#ifdef __32bit_syscall_numbers__ +#define __NR_time 231 /* Linux Specific */ +#else +#endif +#define __NR_splice 232 /* Linux Specific */ +#define __NR_stime 233 /* Linux Specific */ +#define __NR_statfs64 234 /* Linux Specific */ +#define __NR_fstatfs64 235 /* Linux Specific */ +#define __NR__llseek 236 /* Linux Specific */ +#define __NR_mlock 237 +#define __NR_munlock 238 +#define __NR_mlockall 239 +#define __NR_munlockall 240 +#define __NR_sched_setparam 241 +#define __NR_sched_getparam 242 +#define __NR_sched_setscheduler 243 +#define __NR_sched_getscheduler 244 +#define __NR_sched_yield 245 +#define __NR_sched_get_priority_max 246 +#define __NR_sched_get_priority_min 247 +#define __NR_sched_rr_get_interval 248 +#define __NR_nanosleep 249 +#define __NR_mremap 250 +#define __NR__sysctl 251 +#define __NR_getsid 252 +#define __NR_fdatasync 253 +#define __NR_nfsservctl 254 +#define __NR_sync_file_range 255 +#define __NR_clock_settime 256 +#define __NR_clock_gettime 257 +#define __NR_clock_getres 258 +#define __NR_clock_nanosleep 259 +#define __NR_sched_getaffinity 260 +#define __NR_sched_setaffinity 261 +#define __NR_timer_settime 262 +#define __NR_timer_gettime 263 +#define __NR_timer_getoverrun 264 +#define __NR_timer_delete 265 +#define __NR_timer_create 266 +/* #define __NR_vserver 267 Reserved for VSERVER */ +#define __NR_io_setup 268 +#define __NR_io_destroy 269 +#define __NR_io_submit 270 +#define __NR_io_cancel 271 +#define __NR_io_getevents 272 +#define __NR_mq_open 273 +#define __NR_mq_unlink 274 +#define __NR_mq_timedsend 275 +#define __NR_mq_timedreceive 276 +#define __NR_mq_notify 277 +#define __NR_mq_getsetattr 278 +#define __NR_waitid 279 +#define __NR_tee 280 +#define __NR_add_key 281 +#define __NR_request_key 282 +#define __NR_keyctl 283 +#define __NR_openat 284 +#define __NR_mkdirat 285 +#define __NR_mknodat 286 +#define __NR_fchownat 287 +#define __NR_futimesat 288 +#define __NR_fstatat64 289 +#define __NR_unlinkat 290 +#define __NR_renameat 291 +#define __NR_linkat 292 +#define __NR_symlinkat 293 +#define __NR_readlinkat 294 +#define __NR_fchmodat 295 +#define __NR_faccessat 296 +#define __NR_pselect6 297 +#define __NR_ppoll 298 +#define __NR_unshare 299 +#define __NR_set_robust_list 300 +#define __NR_get_robust_list 301 +#define __NR_migrate_pages 302 +#define __NR_mbind 303 +#define __NR_get_mempolicy 304 +#define __NR_set_mempolicy 305 +#define __NR_kexec_load 306 +#define __NR_move_pages 307 +#define __NR_getcpu 308 +#define __NR_epoll_pwait 309 +#define __NR_utimensat 310 +#define __NR_signalfd 311 +#define __NR_timerfd_create 312 +#define __NR_eventfd 313 +#define __NR_fallocate 314 +#define __NR_timerfd_settime 315 +#define __NR_timerfd_gettime 316 +#define __NR_signalfd4 317 +#define __NR_eventfd2 318 +#define __NR_epoll_create1 319 +#define __NR_dup3 320 +#define __NR_pipe2 321 +#define __NR_inotify_init1 322 +#define __NR_accept4 323 +#define __NR_preadv 324 +#define __NR_pwritev 325 +#define __NR_rt_tgsigqueueinfo 326 +#define __NR_perf_event_open 327 +#define __NR_recvmmsg 328 +#define __NR_fanotify_init 329 +#define __NR_fanotify_mark 330 +#define __NR_prlimit64 331 +#define __NR_name_to_handle_at 332 +#define __NR_open_by_handle_at 333 +#define __NR_clock_adjtime 334 +#define __NR_syncfs 335 +#define __NR_sendmmsg 336 +#define __NR_setns 337 +#define __NR_process_vm_readv 338 +#define __NR_process_vm_writev 339 +#define __NR_kern_features 340 +#define __NR_kcmp 341 +#define __NR_finit_module 342 +#define __NR_sched_setattr 343 +#define __NR_sched_getattr 344 +#define __NR_renameat2 345 + +#define NR_syscalls 346 + +/* Bitmask values returned from kern_features system call. */ +#define KERN_FEATURE_MIXED_MODE_STACK 0x00000001 + +#ifdef __32bit_syscall_numbers__ +/* Sparc 32-bit only has the "setresuid32", "getresuid32" variants, + * it never had the plain ones and there is no value to adding those + * old versions into the syscall table. + */ +#define __IGNORE_setresuid +#define __IGNORE_getresuid +#define __IGNORE_setresgid +#define __IGNORE_getresgid +#endif + +#endif /* _UAPI_SPARC_UNISTD_H */ diff --git a/arch/sparc/include/asm/utrap.h b/arch/sparc/include/uapi/asm/utrap.h index b10e527c22d..b10e527c22d 100644 --- a/arch/sparc/include/asm/utrap.h +++ b/arch/sparc/include/uapi/asm/utrap.h diff --git a/arch/sparc/include/asm/watchdog.h b/arch/sparc/include/uapi/asm/watchdog.h index 5baf2d3919c..5baf2d3919c 100644 --- a/arch/sparc/include/asm/watchdog.h +++ b/arch/sparc/include/uapi/asm/watchdog.h diff --git a/arch/sparc/kernel/Makefile b/arch/sparc/kernel/Makefile index 6cf591b7e1c..7cf9c6ea3f1 100644 --- a/arch/sparc/kernel/Makefile +++ b/arch/sparc/kernel/Makefile @@ -1,3 +1,4 @@ + # # Makefile for the linux kernel. # @@ -41,7 +42,6 @@ obj-y += time_$(BITS).o obj-$(CONFIG_SPARC32) += windows.o obj-y += cpu.o obj-$(CONFIG_SPARC32) += devices.o -obj-$(CONFIG_SPARC32) += tadpole.o obj-y += ptrace_$(BITS).o obj-y += unaligned_$(BITS).o obj-y += una_asm_$(BITS).o @@ -74,7 +74,8 @@ obj-y += dma.o obj-$(CONFIG_PCIC_PCI) += pcic.o obj-$(CONFIG_LEON_PCI) += leon_pci.o -obj-$(CONFIG_GRPCI2) += leon_pci_grpci2.o +obj-$(CONFIG_SPARC_GRPCI2)+= leon_pci_grpci2.o +obj-$(CONFIG_SPARC_GRPCI1)+= leon_pci_grpci1.o obj-$(CONFIG_SMP) += trampoline_$(BITS).o smp_$(BITS).o obj-$(CONFIG_SPARC32_SMP) += sun4m_smp.o sun4d_smp.o leon_smp.o @@ -98,13 +99,10 @@ obj-$(CONFIG_STACKTRACE) += stacktrace.o obj-$(CONFIG_SPARC64_PCI) += pci.o pci_common.o psycho_common.o obj-$(CONFIG_SPARC64_PCI) += pci_psycho.o pci_sabre.o pci_schizo.o obj-$(CONFIG_SPARC64_PCI) += pci_sun4v.o pci_sun4v_asm.o pci_fire.o -obj-$(CONFIG_PCI_MSI) += pci_msi.o +obj-$(CONFIG_SPARC64_PCI_MSI) += pci_msi.o obj-$(CONFIG_COMPAT) += sys32.o sys_sparc32.o signal32.o -# sparc64 cpufreq -obj-$(CONFIG_US3_FREQ) += us3_cpufreq.o -obj-$(CONFIG_US2E_FREQ) += us2e_cpufreq.o obj-$(CONFIG_US3_MC) += chmc.o obj-$(CONFIG_KPROBES) += kprobes.o diff --git a/arch/sparc/kernel/apc.c b/arch/sparc/kernel/apc.c index 812e10bbb0b..eefda32b595 100644 --- a/arch/sparc/kernel/apc.c +++ b/arch/sparc/kernel/apc.c @@ -20,6 +20,7 @@ #include <asm/uaccess.h> #include <asm/auxio.h> #include <asm/apc.h> +#include <asm/processor.h> /* Debugging * @@ -31,7 +32,7 @@ #define APC_DEVNAME "apc" static u8 __iomem *regs; -static int apc_no_idle __devinitdata = 0; +static int apc_no_idle = 0; #define apc_readb(offs) (sbus_readb(regs+offs)) #define apc_writeb(val, offs) (sbus_writeb(val, regs+offs)) @@ -138,7 +139,7 @@ static const struct file_operations apc_fops = { static struct miscdevice apc_miscdev = { APC_MINOR, APC_DEVNAME, &apc_fops }; -static int __devinit apc_probe(struct platform_device *op) +static int apc_probe(struct platform_device *op) { int err; @@ -158,7 +159,7 @@ static int __devinit apc_probe(struct platform_device *op) /* Assign power management IDLE handler */ if (!apc_no_idle) - pm_idle = apc_swift_idle; + sparc_idle = apc_swift_idle; printk(KERN_INFO "%s: power management initialized%s\n", APC_DEVNAME, apc_no_idle ? " (CPU idle disabled)" : ""); diff --git a/arch/sparc/kernel/asm-offsets.c b/arch/sparc/kernel/asm-offsets.c index 68f7e1118e9..f76389a3234 100644 --- a/arch/sparc/kernel/asm-offsets.c +++ b/arch/sparc/kernel/asm-offsets.c @@ -14,6 +14,8 @@ // #include <linux/mm.h> #include <linux/kbuild.h> +#include <asm/hibernate.h> + #ifdef CONFIG_SPARC32 int sparc32_foo(void) { @@ -24,6 +26,19 @@ int sparc32_foo(void) #else int sparc64_foo(void) { +#ifdef CONFIG_HIBERNATION + BLANK(); + OFFSET(SC_REG_FP, saved_context, fp); + OFFSET(SC_REG_CWP, saved_context, cwp); + OFFSET(SC_REG_WSTATE, saved_context, wstate); + + OFFSET(SC_REG_TICK, saved_context, tick); + OFFSET(SC_REG_PSTATE, saved_context, pstate); + + OFFSET(SC_REG_G4, saved_context, g4); + OFFSET(SC_REG_G5, saved_context, g5); + OFFSET(SC_REG_G6, saved_context, g6); +#endif return 0; } #endif @@ -34,6 +49,8 @@ int foo(void) DEFINE(AOFF_task_thread, offsetof(struct task_struct, thread)); BLANK(); DEFINE(AOFF_mm_context, offsetof(struct mm_struct, context)); + BLANK(); + DEFINE(VMA_VM_MM, offsetof(struct vm_area_struct, vm_mm)); /* DEFINE(NUM_USER_SEGMENTS, TASK_SIZE>>28); */ return 0; diff --git a/arch/sparc/kernel/audit.c b/arch/sparc/kernel/audit.c index 8fff0ac63d5..24361b494a9 100644 --- a/arch/sparc/kernel/audit.c +++ b/arch/sparc/kernel/audit.c @@ -3,6 +3,8 @@ #include <linux/audit.h> #include <asm/unistd.h> +#include "kernel.h" + static unsigned dir_class[] = { #include <asm-generic/audit_dir_write.h> ~0U @@ -40,7 +42,6 @@ int audit_classify_arch(int arch) int audit_classify_syscall(int abi, unsigned syscall) { #ifdef CONFIG_COMPAT - extern int sparc32_classify_syscall(unsigned); if (abi == AUDIT_ARCH_SPARC) return sparc32_classify_syscall(syscall); #endif @@ -61,11 +62,6 @@ int audit_classify_syscall(int abi, unsigned syscall) static int __init audit_classes_init(void) { #ifdef CONFIG_COMPAT - extern __u32 sparc32_dir_class[]; - extern __u32 sparc32_write_class[]; - extern __u32 sparc32_read_class[]; - extern __u32 sparc32_chattr_class[]; - extern __u32 sparc32_signal_class[]; audit_register_class(AUDIT_CLASS_WRITE_32, sparc32_write_class); audit_register_class(AUDIT_CLASS_READ_32, sparc32_read_class); audit_register_class(AUDIT_CLASS_DIR_WRITE_32, sparc32_dir_class); diff --git a/arch/sparc/kernel/auxio_32.c b/arch/sparc/kernel/auxio_32.c index e20cc55fb76..ae88c223e4d 100644 --- a/arch/sparc/kernel/auxio_32.c +++ b/arch/sparc/kernel/auxio_32.c @@ -9,12 +9,15 @@ #include <linux/of.h> #include <linux/of_device.h> #include <linux/export.h> + #include <asm/oplib.h> #include <asm/io.h> #include <asm/auxio.h> #include <asm/string.h> /* memset(), Linux has no bzero() */ #include <asm/cpu_type.h> +#include "kernel.h" + /* Probe and map in the Auxiliary I/O register */ /* auxio_register is not static because it is referenced @@ -103,7 +106,7 @@ EXPORT_SYMBOL(set_auxio); /* sun4m power control register (AUXIO2) */ -volatile unsigned char * auxio_power_register = NULL; +volatile u8 __iomem *auxio_power_register = NULL; void __init auxio_power_probe(void) { @@ -127,8 +130,8 @@ void __init auxio_power_probe(void) r.flags = regs.which_io & 0xF; r.start = regs.phys_addr; r.end = regs.phys_addr + regs.reg_size - 1; - auxio_power_register = (unsigned char *) of_ioremap(&r, 0, - regs.reg_size, "auxpower"); + auxio_power_register = + (u8 __iomem *)of_ioremap(&r, 0, regs.reg_size, "auxpower"); /* Display a quick message on the console. */ if (auxio_power_register) diff --git a/arch/sparc/kernel/auxio_64.c b/arch/sparc/kernel/auxio_64.c index 773091ac71a..86e55778e4a 100644 --- a/arch/sparc/kernel/auxio_64.c +++ b/arch/sparc/kernel/auxio_64.c @@ -102,7 +102,7 @@ static const struct of_device_id auxio_match[] = { MODULE_DEVICE_TABLE(of, auxio_match); -static int __devinit auxio_probe(struct platform_device *dev) +static int auxio_probe(struct platform_device *dev) { struct device_node *dp = dev->dev.of_node; unsigned long size; diff --git a/arch/sparc/kernel/btext.c b/arch/sparc/kernel/btext.c index 57073e56ba9..987f7ec497c 100644 --- a/arch/sparc/kernel/btext.c +++ b/arch/sparc/kernel/btext.c @@ -137,7 +137,7 @@ static void scrollscreen(void) } #endif /* ndef NO_SCROLL */ -void btext_drawchar(char c) +static void btext_drawchar(char c) { int cline = 0; #ifdef NO_SCROLL diff --git a/arch/sparc/kernel/central.c b/arch/sparc/kernel/central.c index 9708851a8b9..052b5a44318 100644 --- a/arch/sparc/kernel/central.c +++ b/arch/sparc/kernel/central.c @@ -33,7 +33,7 @@ struct fhc { struct platform_device leds_pdev; }; -static int __devinit clock_board_calc_nslots(struct clock_board *p) +static int clock_board_calc_nslots(struct clock_board *p) { u8 reg = upa_readb(p->clock_regs + CLOCK_STAT1) & 0xc0; @@ -60,7 +60,7 @@ static int __devinit clock_board_calc_nslots(struct clock_board *p) } } -static int __devinit clock_board_probe(struct platform_device *op) +static int clock_board_probe(struct platform_device *op) { struct clock_board *p = kzalloc(sizeof(*p), GFP_KERNEL); int err = -ENOMEM; @@ -157,7 +157,7 @@ static struct platform_driver clock_board_driver = { }, }; -static int __devinit fhc_probe(struct platform_device *op) +static int fhc_probe(struct platform_device *op) { struct fhc *p = kzalloc(sizeof(*p), GFP_KERNEL); int err = -ENOMEM; diff --git a/arch/sparc/kernel/chmc.c b/arch/sparc/kernel/chmc.c index 5f450260981..dbb210d74e2 100644 --- a/arch/sparc/kernel/chmc.c +++ b/arch/sparc/kernel/chmc.c @@ -336,9 +336,9 @@ static int jbusmc_print_dimm(int syndrome_code, return 0; } -static u64 __devinit jbusmc_dimm_group_size(u64 base, - const struct linux_prom64_registers *mem_regs, - int num_mem_regs) +static u64 jbusmc_dimm_group_size(u64 base, + const struct linux_prom64_registers *mem_regs, + int num_mem_regs) { u64 max = base + (8UL * 1024 * 1024 * 1024); u64 max_seen = base; @@ -363,10 +363,10 @@ static u64 __devinit jbusmc_dimm_group_size(u64 base, return max_seen - base; } -static void __devinit jbusmc_construct_one_dimm_group(struct jbusmc *p, - unsigned long index, - const struct linux_prom64_registers *mem_regs, - int num_mem_regs) +static void jbusmc_construct_one_dimm_group(struct jbusmc *p, + unsigned long index, + const struct linux_prom64_registers *mem_regs, + int num_mem_regs) { struct jbusmc_dimm_group *dp = &p->dimm_groups[index]; @@ -378,9 +378,9 @@ static void __devinit jbusmc_construct_one_dimm_group(struct jbusmc *p, dp->size = jbusmc_dimm_group_size(dp->base_addr, mem_regs, num_mem_regs); } -static void __devinit jbusmc_construct_dimm_groups(struct jbusmc *p, - const struct linux_prom64_registers *mem_regs, - int num_mem_regs) +static void jbusmc_construct_dimm_groups(struct jbusmc *p, + const struct linux_prom64_registers *mem_regs, + int num_mem_regs) { if (p->mc_reg_1 & JB_MC_REG1_DIMM1_BANK0) { jbusmc_construct_one_dimm_group(p, 0, mem_regs, num_mem_regs); @@ -392,7 +392,7 @@ static void __devinit jbusmc_construct_dimm_groups(struct jbusmc *p, } } -static int __devinit jbusmc_probe(struct platform_device *op) +static int jbusmc_probe(struct platform_device *op) { const struct linux_prom64_registers *mem_regs; struct device_node *mem_node; @@ -689,7 +689,7 @@ static void chmc_fetch_decode_regs(struct chmc *p) chmc_read_mcreg(p, CHMCTRL_DECODE4)); } -static int __devinit chmc_probe(struct platform_device *op) +static int chmc_probe(struct platform_device *op) { struct device_node *dp = op->dev.of_node; unsigned long ver; @@ -763,7 +763,7 @@ out_free: goto out; } -static int __devinit us3mc_probe(struct platform_device *op) +static int us3mc_probe(struct platform_device *op) { if (mc_type == MC_TYPE_SAFARI) return chmc_probe(op); @@ -772,21 +772,21 @@ static int __devinit us3mc_probe(struct platform_device *op) return -ENODEV; } -static void __devexit chmc_destroy(struct platform_device *op, struct chmc *p) +static void chmc_destroy(struct platform_device *op, struct chmc *p) { list_del(&p->list); of_iounmap(&op->resource[0], p->regs, 0x48); kfree(p); } -static void __devexit jbusmc_destroy(struct platform_device *op, struct jbusmc *p) +static void jbusmc_destroy(struct platform_device *op, struct jbusmc *p) { mc_list_del(&p->list); of_iounmap(&op->resource[0], p->regs, JBUSMC_REGS_SIZE); kfree(p); } -static int __devexit us3mc_remove(struct platform_device *op) +static int us3mc_remove(struct platform_device *op) { void *p = dev_get_drvdata(&op->dev); @@ -814,7 +814,7 @@ static struct platform_driver us3mc_driver = { .of_match_table = us3mc_match, }, .probe = us3mc_probe, - .remove = __devexit_p(us3mc_remove), + .remove = us3mc_remove, }; static inline bool us3mc_platform(void) diff --git a/arch/sparc/kernel/compat_audit.c b/arch/sparc/kernel/compat_audit.c index d865575b25b..7062263d09c 100644 --- a/arch/sparc/kernel/compat_audit.c +++ b/arch/sparc/kernel/compat_audit.c @@ -1,5 +1,6 @@ #define __32bit_syscall_numbers__ #include <asm/unistd.h> +#include "kernel.h" unsigned sparc32_dir_class[] = { #include <asm-generic/audit_dir_write.h> diff --git a/arch/sparc/kernel/cpu.c b/arch/sparc/kernel/cpu.c index a6c94a2bf9d..82a3a71c451 100644 --- a/arch/sparc/kernel/cpu.c +++ b/arch/sparc/kernel/cpu.c @@ -22,6 +22,7 @@ #include <asm/cpudata.h> #include "kernel.h" +#include "entry.h" DEFINE_PER_CPU(cpuinfo_sparc, __cpu_data) = { 0 }; EXPORT_PER_CPU_SYMBOL(__cpu_data); @@ -493,6 +494,12 @@ static void __init sun4v_cpu_probe(void) sparc_pmu_type = "niagara5"; break; + case SUN4V_CHIP_SPARC64X: + sparc_cpu_type = "SPARC64-X"; + sparc_fpu_type = "SPARC64-X integrated FPU"; + sparc_pmu_type = "sparc64-x"; + break; + default: printk(KERN_WARNING "CPU: Unknown sun4v cpu type [%s]\n", prom_cpu_compatible); diff --git a/arch/sparc/kernel/cpumap.c b/arch/sparc/kernel/cpumap.c index e4de74c2c9b..de1c844dfab 100644 --- a/arch/sparc/kernel/cpumap.c +++ b/arch/sparc/kernel/cpumap.c @@ -6,7 +6,6 @@ #include <linux/export.h> #include <linux/slab.h> #include <linux/kernel.h> -#include <linux/init.h> #include <linux/cpumask.h> #include <linux/spinlock.h> #include <asm/cpudata.h> @@ -327,6 +326,7 @@ static int iterate_cpu(struct cpuinfo_tree *t, unsigned int root_index) case SUN4V_CHIP_NIAGARA3: case SUN4V_CHIP_NIAGARA4: case SUN4V_CHIP_NIAGARA5: + case SUN4V_CHIP_SPARC64X: rover_inc_table = niagara_iterate_method; break; default: diff --git a/arch/sparc/kernel/cpumap.h b/arch/sparc/kernel/cpumap.h index e639880ab86..9dac398c434 100644 --- a/arch/sparc/kernel/cpumap.h +++ b/arch/sparc/kernel/cpumap.h @@ -2,8 +2,8 @@ #define _CPUMAP_H #ifdef CONFIG_SMP -extern void cpu_map_rebuild(void); -extern int map_to_cpu(unsigned int index); +void cpu_map_rebuild(void); +int map_to_cpu(unsigned int index); #define cpu_map_init() cpu_map_rebuild() #else #define cpu_map_init() do {} while (0) diff --git a/arch/sparc/kernel/devices.c b/arch/sparc/kernel/devices.c index 3d465e87f7e..8d5d09f09ca 100644 --- a/arch/sparc/kernel/devices.c +++ b/arch/sparc/kernel/devices.c @@ -19,8 +19,9 @@ #include <asm/smp.h> #include <asm/cpudata.h> #include <asm/cpu_type.h> +#include <asm/setup.h> -extern void clock_stop_probe(void); /* tadpole.c */ +#include "kernel.h" static char *cpu_mid_prop(void) { @@ -131,11 +132,6 @@ void __init device_scan(void) } #endif /* !CONFIG_SMP */ - { - extern void auxio_probe(void); - extern void auxio_power_probe(void); - auxio_probe(); - auxio_power_probe(); - } - clock_stop_probe(); + auxio_probe(); + auxio_power_probe(); } diff --git a/arch/sparc/kernel/ds.c b/arch/sparc/kernel/ds.c index f09257c8610..dff60abbea0 100644 --- a/arch/sparc/kernel/ds.c +++ b/arch/sparc/kernel/ds.c @@ -29,7 +29,7 @@ #define DRV_MODULE_VERSION "1.0" #define DRV_MODULE_RELDATE "Jul 11, 2007" -static char version[] __devinitdata = +static char version[] = DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; MODULE_AUTHOR("David S. Miller (davem@davemloft.net)"); MODULE_DESCRIPTION("Sun LDOM domain services driver"); @@ -528,10 +528,8 @@ static void dr_cpu_mark(struct ds_data *resp, int cpu, int ncpus, } } -static int __cpuinit dr_cpu_configure(struct ds_info *dp, - struct ds_cap_state *cp, - u64 req_num, - cpumask_t *mask) +static int dr_cpu_configure(struct ds_info *dp, struct ds_cap_state *cp, + u64 req_num, cpumask_t *mask) { struct ds_data *resp; int resp_len, ncpus, cpu; @@ -627,9 +625,8 @@ static int dr_cpu_unconfigure(struct ds_info *dp, return 0; } -static void __cpuinit dr_cpu_data(struct ds_info *dp, - struct ds_cap_state *cp, - void *buf, int len) +static void dr_cpu_data(struct ds_info *dp, struct ds_cap_state *cp, void *buf, + int len) { struct ds_data *data = buf; struct dr_cpu_tag *tag = (struct dr_cpu_tag *) (data + 1); @@ -783,6 +780,16 @@ void ldom_set_var(const char *var, const char *value) char *base, *p; int msg_len, loops; + if (strlen(var) + strlen(value) + 2 > + sizeof(pkt) - sizeof(pkt.header)) { + printk(KERN_ERR PFX + "contents length: %zu, which more than max: %lu," + "so could not set (%s) variable to (%s).\n", + strlen(var) + strlen(value) + 2, + sizeof(pkt) - sizeof(pkt.header), var, value); + return; + } + memset(&pkt, 0, sizeof(pkt)); pkt.header.data.tag.type = DS_DATA; pkt.header.data.handle = cp->handle; @@ -842,8 +849,8 @@ void ldom_reboot(const char *boot_command) if (boot_command && strlen(boot_command)) { unsigned long len; - strcpy(full_boot_str, "boot "); - strcpy(full_boot_str + strlen("boot "), boot_command); + snprintf(full_boot_str, sizeof(full_boot_str), "boot %s", + boot_command); len = strlen(full_boot_str); if (reboot_data_supported) { @@ -1146,8 +1153,7 @@ static void ds_event(void *arg, int event) spin_unlock_irqrestore(&ds_lock, flags); } -static int __devinit ds_probe(struct vio_dev *vdev, - const struct vio_device_id *id) +static int ds_probe(struct vio_dev *vdev, const struct vio_device_id *id) { static int ds_version_printed; struct ldc_channel_config ds_cfg = { diff --git a/arch/sparc/kernel/ebus.c b/arch/sparc/kernel/ebus.c index e306fb08ee5..acf8314cec4 100644 --- a/arch/sparc/kernel/ebus.c +++ b/arch/sparc/kernel/ebus.c @@ -7,7 +7,6 @@ #include <linux/export.h> #include <linux/kernel.h> #include <linux/types.h> -#include <linux/init.h> #include <linux/interrupt.h> #include <linux/delay.h> diff --git a/arch/sparc/kernel/entry.S b/arch/sparc/kernel/entry.S index dcaa1cf0de4..33c02b15f47 100644 --- a/arch/sparc/kernel/entry.S +++ b/arch/sparc/kernel/entry.S @@ -806,23 +806,10 @@ sys_nis_syscall: call c_sys_nis_syscall mov %l5, %o7 - .align 4 - .globl sys_execve -sys_execve: - mov %o7, %l5 - add %sp, STACKFRAME_SZ, %o0 ! pt_regs *regs arg - call sparc_execve - mov %l5, %o7 - - .globl sunos_execv sunos_execv: - st %g0, [%sp + STACKFRAME_SZ + PT_I2] - - call sparc_execve - add %sp, STACKFRAME_SZ, %o0 - - b ret_sys_call - ld [%sp + STACKFRAME_SZ + PT_I0], %o0 + .globl sunos_execv + b sys_execve + clr %i2 .align 4 .globl sys_sparc_pipe @@ -833,14 +820,6 @@ sys_sparc_pipe: mov %l5, %o7 .align 4 - .globl sys_sigaltstack -sys_sigaltstack: - mov %o7, %l5 - mov %fp, %o2 - call do_sigaltstack - mov %l5, %o7 - - .align 4 .globl sys_sigstack sys_sigstack: mov %o7, %l5 @@ -860,7 +839,7 @@ sys_sigreturn: nop call syscall_trace - nop + mov 1, %o1 1: /* We don't want to muck with user registers like a @@ -959,17 +938,9 @@ flush_patch_four: .align 4 linux_sparc_ni_syscall: sethi %hi(sys_ni_syscall), %l7 - b syscall_is_too_hard + b do_syscall or %l7, %lo(sys_ni_syscall), %l7 -linux_fast_syscall: - andn %l7, 3, %l7 - mov %i0, %o0 - mov %i1, %o1 - mov %i2, %o2 - jmpl %l7 + %g0, %g0 - mov %i3, %o3 - linux_syscall_trace: add %sp, STACKFRAME_SZ, %o0 call syscall_trace @@ -991,6 +962,23 @@ ret_from_fork: b ret_sys_call ld [%sp + STACKFRAME_SZ + PT_I0], %o0 + .globl ret_from_kernel_thread +ret_from_kernel_thread: + call schedule_tail + ld [%g3 + TI_TASK], %o0 + ld [%sp + STACKFRAME_SZ + PT_G1], %l0 + call %l0 + ld [%sp + STACKFRAME_SZ + PT_G2], %o0 + rd %psr, %l1 + ld [%sp + STACKFRAME_SZ + PT_PSR], %l0 + andn %l0, PSR_CWP, %l0 + nop + and %l1, PSR_CWP, %l1 + or %l0, %l1, %l0 + st %l0, [%sp + STACKFRAME_SZ + PT_PSR] + b ret_sys_call + mov 0, %o0 + /* Linux native system calls enter here... */ .align 4 .globl linux_sparc_syscall @@ -1002,11 +990,8 @@ linux_sparc_syscall: bgeu linux_sparc_ni_syscall sll %g1, 2, %l4 ld [%l7 + %l4], %l7 - andcc %l7, 1, %g0 - bne linux_fast_syscall - /* Just do first insn from SAVE_ALL in the delay slot */ -syscall_is_too_hard: +do_syscall: SAVE_ALL_HEAD rd %wim, %l3 diff --git a/arch/sparc/kernel/entry.h b/arch/sparc/kernel/entry.h index 0c218e4c088..ebaba6167dd 100644 --- a/arch/sparc/kernel/entry.h +++ b/arch/sparc/kernel/entry.h @@ -6,40 +6,39 @@ #include <linux/init.h> /* irq */ -extern void handler_irq(int irq, struct pt_regs *regs); +void handler_irq(int irq, struct pt_regs *regs); #ifdef CONFIG_SPARC32 /* traps */ -extern void do_hw_interrupt(struct pt_regs *regs, unsigned long type); -extern void do_illegal_instruction(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); - -extern void do_priv_instruction(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); -extern void do_memaccess_unaligned(struct pt_regs *regs, unsigned long pc, - unsigned long npc, - unsigned long psr); -extern void do_fpd_trap(struct pt_regs *regs, unsigned long pc, +void do_hw_interrupt(struct pt_regs *regs, unsigned long type); +void do_illegal_instruction(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); + +void do_priv_instruction(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); +void do_memaccess_unaligned(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); +void do_fpd_trap(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); +void do_fpe_trap(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); +void handle_tag_overflow(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); +void handle_watchpoint(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); +void handle_reg_access(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); +void handle_cp_disabled(struct pt_regs *regs, unsigned long pc, unsigned long npc, unsigned long psr); -extern void do_fpe_trap(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); -extern void handle_tag_overflow(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); -extern void handle_watchpoint(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); -extern void handle_reg_access(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); -extern void handle_cp_disabled(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); -extern void handle_cp_exception(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); +void handle_cp_exception(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); /* entry.S */ -extern void fpsave(unsigned long *fpregs, unsigned long *fsr, - void *fpqueue, unsigned long *fpqdepth); -extern void fpload(unsigned long *fpregs, unsigned long *fsr); +void fpsave(unsigned long *fpregs, unsigned long *fsr, + void *fpqueue, unsigned long *fpqdepth); +void fpload(unsigned long *fpregs, unsigned long *fsr); #else /* CONFIG_SPARC32 */ @@ -59,124 +58,130 @@ struct popc_6insn_patch_entry { extern struct popc_6insn_patch_entry __popc_6insn_patch, __popc_6insn_patch_end; -extern void __init per_cpu_patch(void); -extern void sun4v_patch_1insn_range(struct sun4v_1insn_patch_entry *, - struct sun4v_1insn_patch_entry *); -extern void sun4v_patch_2insn_range(struct sun4v_2insn_patch_entry *, - struct sun4v_2insn_patch_entry *); -extern void __init sun4v_patch(void); -extern void __init boot_cpu_id_too_large(int cpu); +struct pause_patch_entry { + unsigned int addr; + unsigned int insns[3]; +}; +extern struct pause_patch_entry __pause_3insn_patch, + __pause_3insn_patch_end; + +void __init per_cpu_patch(void); +void sun4v_patch_1insn_range(struct sun4v_1insn_patch_entry *, + struct sun4v_1insn_patch_entry *); +void sun4v_patch_2insn_range(struct sun4v_2insn_patch_entry *, + struct sun4v_2insn_patch_entry *); +void __init sun4v_patch(void); +void __init boot_cpu_id_too_large(int cpu); extern unsigned int dcache_parity_tl1_occurred; extern unsigned int icache_parity_tl1_occurred; -extern asmlinkage void sparc_breakpoint(struct pt_regs *regs); -extern void timer_interrupt(int irq, struct pt_regs *regs); - -extern void do_notify_resume(struct pt_regs *regs, - unsigned long orig_i0, - unsigned long thread_info_flags); - -extern asmlinkage int syscall_trace_enter(struct pt_regs *regs); -extern asmlinkage void syscall_trace_leave(struct pt_regs *regs); - -extern void bad_trap_tl1(struct pt_regs *regs, long lvl); - -extern void do_fpe_common(struct pt_regs *regs); -extern void do_fpieee(struct pt_regs *regs); -extern void do_fpother(struct pt_regs *regs); -extern void do_tof(struct pt_regs *regs); -extern void do_div0(struct pt_regs *regs); -extern void do_illegal_instruction(struct pt_regs *regs); -extern void mem_address_unaligned(struct pt_regs *regs, - unsigned long sfar, - unsigned long sfsr); -extern void sun4v_do_mna(struct pt_regs *regs, - unsigned long addr, - unsigned long type_ctx); -extern void do_privop(struct pt_regs *regs); -extern void do_privact(struct pt_regs *regs); -extern void do_cee(struct pt_regs *regs); -extern void do_cee_tl1(struct pt_regs *regs); -extern void do_dae_tl1(struct pt_regs *regs); -extern void do_iae_tl1(struct pt_regs *regs); -extern void do_div0_tl1(struct pt_regs *regs); -extern void do_fpdis_tl1(struct pt_regs *regs); -extern void do_fpieee_tl1(struct pt_regs *regs); -extern void do_fpother_tl1(struct pt_regs *regs); -extern void do_ill_tl1(struct pt_regs *regs); -extern void do_irq_tl1(struct pt_regs *regs); -extern void do_lddfmna_tl1(struct pt_regs *regs); -extern void do_stdfmna_tl1(struct pt_regs *regs); -extern void do_paw(struct pt_regs *regs); -extern void do_paw_tl1(struct pt_regs *regs); -extern void do_vaw(struct pt_regs *regs); -extern void do_vaw_tl1(struct pt_regs *regs); -extern void do_tof_tl1(struct pt_regs *regs); -extern void do_getpsr(struct pt_regs *regs); - -extern void spitfire_insn_access_exception(struct pt_regs *regs, - unsigned long sfsr, - unsigned long sfar); -extern void spitfire_insn_access_exception_tl1(struct pt_regs *regs, - unsigned long sfsr, - unsigned long sfar); -extern void spitfire_data_access_exception(struct pt_regs *regs, - unsigned long sfsr, - unsigned long sfar); -extern void spitfire_data_access_exception_tl1(struct pt_regs *regs, - unsigned long sfsr, - unsigned long sfar); -extern void spitfire_access_error(struct pt_regs *regs, - unsigned long status_encoded, - unsigned long afar); - -extern void cheetah_fecc_handler(struct pt_regs *regs, - unsigned long afsr, - unsigned long afar); -extern void cheetah_cee_handler(struct pt_regs *regs, - unsigned long afsr, - unsigned long afar); -extern void cheetah_deferred_handler(struct pt_regs *regs, - unsigned long afsr, - unsigned long afar); -extern void cheetah_plus_parity_error(int type, struct pt_regs *regs); - -extern void sun4v_insn_access_exception(struct pt_regs *regs, - unsigned long addr, - unsigned long type_ctx); -extern void sun4v_insn_access_exception_tl1(struct pt_regs *regs, - unsigned long addr, - unsigned long type_ctx); -extern void sun4v_data_access_exception(struct pt_regs *regs, - unsigned long addr, - unsigned long type_ctx); -extern void sun4v_data_access_exception_tl1(struct pt_regs *regs, - unsigned long addr, - unsigned long type_ctx); -extern void sun4v_resum_error(struct pt_regs *regs, - unsigned long offset); -extern void sun4v_resum_overflow(struct pt_regs *regs); -extern void sun4v_nonresum_error(struct pt_regs *regs, - unsigned long offset); -extern void sun4v_nonresum_overflow(struct pt_regs *regs); +asmlinkage void sparc_breakpoint(struct pt_regs *regs); +void timer_interrupt(int irq, struct pt_regs *regs); + +void do_notify_resume(struct pt_regs *regs, + unsigned long orig_i0, + unsigned long thread_info_flags); + +asmlinkage int syscall_trace_enter(struct pt_regs *regs); +asmlinkage void syscall_trace_leave(struct pt_regs *regs); + +void bad_trap_tl1(struct pt_regs *regs, long lvl); + +void do_fpieee(struct pt_regs *regs); +void do_fpother(struct pt_regs *regs); +void do_tof(struct pt_regs *regs); +void do_div0(struct pt_regs *regs); +void do_illegal_instruction(struct pt_regs *regs); +void mem_address_unaligned(struct pt_regs *regs, + unsigned long sfar, + unsigned long sfsr); +void sun4v_do_mna(struct pt_regs *regs, + unsigned long addr, + unsigned long type_ctx); +void do_privop(struct pt_regs *regs); +void do_privact(struct pt_regs *regs); +void do_cee(struct pt_regs *regs); +void do_cee_tl1(struct pt_regs *regs); +void do_dae_tl1(struct pt_regs *regs); +void do_iae_tl1(struct pt_regs *regs); +void do_div0_tl1(struct pt_regs *regs); +void do_fpdis_tl1(struct pt_regs *regs); +void do_fpieee_tl1(struct pt_regs *regs); +void do_fpother_tl1(struct pt_regs *regs); +void do_ill_tl1(struct pt_regs *regs); +void do_irq_tl1(struct pt_regs *regs); +void do_lddfmna_tl1(struct pt_regs *regs); +void do_stdfmna_tl1(struct pt_regs *regs); +void do_paw(struct pt_regs *regs); +void do_paw_tl1(struct pt_regs *regs); +void do_vaw(struct pt_regs *regs); +void do_vaw_tl1(struct pt_regs *regs); +void do_tof_tl1(struct pt_regs *regs); +void do_getpsr(struct pt_regs *regs); + +void spitfire_insn_access_exception(struct pt_regs *regs, + unsigned long sfsr, + unsigned long sfar); +void spitfire_insn_access_exception_tl1(struct pt_regs *regs, + unsigned long sfsr, + unsigned long sfar); +void spitfire_data_access_exception(struct pt_regs *regs, + unsigned long sfsr, + unsigned long sfar); +void spitfire_data_access_exception_tl1(struct pt_regs *regs, + unsigned long sfsr, + unsigned long sfar); +void spitfire_access_error(struct pt_regs *regs, + unsigned long status_encoded, + unsigned long afar); + +void cheetah_fecc_handler(struct pt_regs *regs, + unsigned long afsr, + unsigned long afar); +void cheetah_cee_handler(struct pt_regs *regs, + unsigned long afsr, + unsigned long afar); +void cheetah_deferred_handler(struct pt_regs *regs, + unsigned long afsr, + unsigned long afar); +void cheetah_plus_parity_error(int type, struct pt_regs *regs); + +void sun4v_insn_access_exception(struct pt_regs *regs, + unsigned long addr, + unsigned long type_ctx); +void sun4v_insn_access_exception_tl1(struct pt_regs *regs, + unsigned long addr, + unsigned long type_ctx); +void sun4v_data_access_exception(struct pt_regs *regs, + unsigned long addr, + unsigned long type_ctx); +void sun4v_data_access_exception_tl1(struct pt_regs *regs, + unsigned long addr, + unsigned long type_ctx); +void sun4v_resum_error(struct pt_regs *regs, + unsigned long offset); +void sun4v_resum_overflow(struct pt_regs *regs); +void sun4v_nonresum_error(struct pt_regs *regs, + unsigned long offset); +void sun4v_nonresum_overflow(struct pt_regs *regs); extern unsigned long sun4v_err_itlb_vaddr; extern unsigned long sun4v_err_itlb_ctx; extern unsigned long sun4v_err_itlb_pte; extern unsigned long sun4v_err_itlb_error; -extern void sun4v_itlb_error_report(struct pt_regs *regs, int tl); +void sun4v_itlb_error_report(struct pt_regs *regs, int tl); extern unsigned long sun4v_err_dtlb_vaddr; extern unsigned long sun4v_err_dtlb_ctx; extern unsigned long sun4v_err_dtlb_pte; extern unsigned long sun4v_err_dtlb_error; -extern void sun4v_dtlb_error_report(struct pt_regs *regs, int tl); -extern void hypervisor_tlbop_error(unsigned long err, - unsigned long op); -extern void hypervisor_tlbop_error_xcall(unsigned long err, - unsigned long op); +void sun4v_dtlb_error_report(struct pt_regs *regs, int tl); +void hypervisor_tlbop_error(unsigned long err, + unsigned long op); +void hypervisor_tlbop_error_xcall(unsigned long err, + unsigned long op); /* WARNING: The error trap handlers in assembly know the precise * layout of the following structure. @@ -242,8 +247,8 @@ struct ino_bucket { extern struct ino_bucket *ivector_table; extern unsigned long ivector_table_pa; -extern void init_irqwork_curcpu(void); -extern void __cpuinit sun4v_register_mondo_queues(int this_cpu); +void init_irqwork_curcpu(void); +void sun4v_register_mondo_queues(int this_cpu); #endif /* CONFIG_SPARC32 */ #endif /* _ENTRY_H */ diff --git a/arch/sparc/kernel/etrap_64.S b/arch/sparc/kernel/etrap_64.S index 786b185e6e3..1276ca2567b 100644 --- a/arch/sparc/kernel/etrap_64.S +++ b/arch/sparc/kernel/etrap_64.S @@ -92,8 +92,10 @@ etrap_save: save %g2, -STACK_BIAS, %sp rdpr %wstate, %g2 wrpr %g0, 0, %canrestore sll %g2, 3, %g2 + + /* Set TI_SYS_FPDEPTH to 1 and clear TI_SYS_NOERROR. */ mov 1, %l5 - stb %l5, [%l6 + TI_FPDEPTH] + sth %l5, [%l6 + TI_SYS_NOERROR] wrpr %g3, 0, %otherwin wrpr %g2, 0, %wstate @@ -152,7 +154,9 @@ etrap_save: save %g2, -STACK_BIAS, %sp add %l6, TI_FPSAVED + 1, %l4 srl %l5, 1, %l3 add %l5, 2, %l5 - stb %l5, [%l6 + TI_FPDEPTH] + + /* Set TI_SYS_FPDEPTH to %l5 and clear TI_SYS_NOERROR. */ + sth %l5, [%l6 + TI_SYS_NOERROR] ba,pt %xcc, 2b stb %g0, [%l4 + %l3] nop diff --git a/arch/sparc/kernel/ftrace.c b/arch/sparc/kernel/ftrace.c index 03ab022e51c..0a2d2ddff54 100644 --- a/arch/sparc/kernel/ftrace.c +++ b/arch/sparc/kernel/ftrace.c @@ -82,12 +82,8 @@ int ftrace_update_ftrace_func(ftrace_func_t func) return ftrace_modify_code(ip, old, new); } -int __init ftrace_dyn_arch_init(void *data) +int __init ftrace_dyn_arch_init(void) { - unsigned long *p = data; - - *p = 0; - return 0; } #endif diff --git a/arch/sparc/kernel/head_32.S b/arch/sparc/kernel/head_32.S index afeb1d77030..3d92c0a8f6c 100644 --- a/arch/sparc/kernel/head_32.S +++ b/arch/sparc/kernel/head_32.S @@ -58,8 +58,6 @@ sun4e_notsup: /* This was the only reasonable way I could think of to properly align * these page-table data structures. */ - .globl swapper_pg_dir -swapper_pg_dir: .skip PAGE_SIZE .globl empty_zero_page empty_zero_page: .skip PAGE_SIZE diff --git a/arch/sparc/kernel/head_64.S b/arch/sparc/kernel/head_64.S index b42ddbf9651..452f04fe8da 100644 --- a/arch/sparc/kernel/head_64.S +++ b/arch/sparc/kernel/head_64.S @@ -134,6 +134,8 @@ prom_niagara_prefix: .asciz "SUNW,UltraSPARC-T" prom_sparc_prefix: .asciz "SPARC-" +prom_sparc64x_prefix: + .asciz "SPARC64-X" .align 4 prom_root_compatible: .skip 64 @@ -280,8 +282,8 @@ sun4v_chip_type: stx %l2, [%l4 + 0x0] ldx [%sp + 2047 + 128 + 0x50], %l3 ! physaddr low /* 4MB align */ - srlx %l3, 22, %l3 - sllx %l3, 22, %l3 + srlx %l3, ILOG2_4MB, %l3 + sllx %l3, ILOG2_4MB, %l3 stx %l3, [%l4 + 0x8] /* Leave service as-is, "call-method" */ @@ -412,7 +414,7 @@ sun4v_chip_type: cmp %g2, 'T' be,pt %xcc, 70f cmp %g2, 'M' - bne,pn %xcc, 4f + bne,pn %xcc, 49f nop 70: ldub [%g1 + 7], %g2 @@ -425,7 +427,7 @@ sun4v_chip_type: cmp %g2, '5' be,pt %xcc, 5f mov SUN4V_CHIP_NIAGARA5, %g4 - ba,pt %xcc, 4f + ba,pt %xcc, 49f nop 91: sethi %hi(prom_cpu_compatible), %g1 @@ -439,6 +441,25 @@ sun4v_chip_type: mov SUN4V_CHIP_NIAGARA2, %g4 4: + /* Athena */ + sethi %hi(prom_cpu_compatible), %g1 + or %g1, %lo(prom_cpu_compatible), %g1 + sethi %hi(prom_sparc64x_prefix), %g7 + or %g7, %lo(prom_sparc64x_prefix), %g7 + mov 9, %g3 +41: ldub [%g7], %g2 + ldub [%g1], %g4 + cmp %g2, %g4 + bne,pn %icc, 49f + add %g7, 1, %g7 + subcc %g3, 1, %g3 + bne,pt %xcc, 41b + add %g1, 1, %g1 + mov SUN4V_CHIP_SPARC64X, %g4 + ba,pt %xcc, 5f + nop + +49: mov SUN4V_CHIP_UNKNOWN, %g4 5: sethi %hi(sun4v_chip_type), %g2 or %g2, %lo(sun4v_chip_type), %g2 @@ -559,10 +580,10 @@ niagara_tlb_fixup: be,pt %xcc, niagara2_patch nop cmp %g1, SUN4V_CHIP_NIAGARA4 - be,pt %xcc, niagara2_patch + be,pt %xcc, niagara4_patch nop cmp %g1, SUN4V_CHIP_NIAGARA5 - be,pt %xcc, niagara2_patch + be,pt %xcc, niagara4_patch nop call generic_patch_copyops @@ -573,6 +594,16 @@ niagara_tlb_fixup: nop ba,a,pt %xcc, 80f +niagara4_patch: + call niagara4_patch_copyops + nop + call niagara4_patch_bzero + nop + call niagara4_patch_pageops + nop + + ba,a,pt %xcc, 80f + niagara2_patch: call niagara2_patch_copyops nop diff --git a/arch/sparc/kernel/hvapi.c b/arch/sparc/kernel/hvapi.c index 8593672838f..c0a2de0fd62 100644 --- a/arch/sparc/kernel/hvapi.c +++ b/arch/sparc/kernel/hvapi.c @@ -45,6 +45,7 @@ static struct api_info api_table[] = { { .group = HV_GRP_NIU, }, { .group = HV_GRP_VF_CPU, }, { .group = HV_GRP_KT_CPU, }, + { .group = HV_GRP_VT_CPU, }, { .group = HV_GRP_DIAG, .flags = FLAG_PRE_API }, }; @@ -193,7 +194,7 @@ void __init sun4v_hvapi_init(void) bad: prom_printf("HVAPI: Cannot register API group " - "%lx with major(%u) minor(%u)\n", + "%lx with major(%lu) minor(%lu)\n", group, major, minor); prom_halt(); } diff --git a/arch/sparc/kernel/hvcalls.S b/arch/sparc/kernel/hvcalls.S index 58d60de4d65..f3ab509b76a 100644 --- a/arch/sparc/kernel/hvcalls.S +++ b/arch/sparc/kernel/hvcalls.S @@ -805,3 +805,19 @@ ENTRY(sun4v_reboot_data_set) retl nop ENDPROC(sun4v_reboot_data_set) + +ENTRY(sun4v_vt_get_perfreg) + mov %o1, %o4 + mov HV_FAST_VT_GET_PERFREG, %o5 + ta HV_FAST_TRAP + stx %o1, [%o4] + retl + nop +ENDPROC(sun4v_vt_get_perfreg) + +ENTRY(sun4v_vt_set_perfreg) + mov HV_FAST_VT_SET_PERFREG, %o5 + ta HV_FAST_TRAP + retl + nop +ENDPROC(sun4v_vt_set_perfreg) diff --git a/arch/sparc/kernel/hvtramp.S b/arch/sparc/kernel/hvtramp.S index 9365432904d..b7ddcdd1dea 100644 --- a/arch/sparc/kernel/hvtramp.S +++ b/arch/sparc/kernel/hvtramp.S @@ -3,7 +3,6 @@ * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net> */ -#include <linux/init.h> #include <asm/thread_info.h> #include <asm/hypervisor.h> @@ -16,7 +15,6 @@ #include <asm/asi.h> #include <asm/pil.h> - __CPUINIT .align 8 .globl hv_cpu_startup, hv_cpu_startup_end @@ -128,8 +126,7 @@ hv_cpu_startup: call smp_callin nop - call cpu_idle - mov 0, %o0 + call cpu_panic nop diff --git a/arch/sparc/kernel/iommu.c b/arch/sparc/kernel/iommu.c index 070ed141aac..bfa4d0c2df4 100644 --- a/arch/sparc/kernel/iommu.c +++ b/arch/sparc/kernel/iommu.c @@ -21,6 +21,7 @@ #include <asm/iommu.h> #include "iommu_common.h" +#include "kernel.h" #define STC_CTXMATCH_ADDR(STC, CTX) \ ((STC)->strbuf_ctxmatch_base + ((CTX) << 3)) @@ -840,8 +841,6 @@ static struct dma_map_ops sun4u_dma_ops = { struct dma_map_ops *dma_ops = &sun4u_dma_ops; EXPORT_SYMBOL(dma_ops); -extern int pci64_dma_supported(struct pci_dev *pdev, u64 device_mask); - int dma_supported(struct device *dev, u64 device_mask) { struct iommu *iommu = dev->archdata.iommu; @@ -854,7 +853,7 @@ int dma_supported(struct device *dev, u64 device_mask) return 1; #ifdef CONFIG_PCI - if (dev->bus == &pci_bus_type) + if (dev_is_pci(dev)) return pci64_dma_supported(to_pci_dev(dev), device_mask); #endif diff --git a/arch/sparc/kernel/iommu_common.h b/arch/sparc/kernel/iommu_common.h index 591f5879039..1ec0de4156e 100644 --- a/arch/sparc/kernel/iommu_common.h +++ b/arch/sparc/kernel/iommu_common.h @@ -48,12 +48,12 @@ static inline int is_span_boundary(unsigned long entry, return iommu_is_span_boundary(entry, nr, shift, boundary_size); } -extern unsigned long iommu_range_alloc(struct device *dev, - struct iommu *iommu, - unsigned long npages, - unsigned long *handle); -extern void iommu_range_free(struct iommu *iommu, - dma_addr_t dma_addr, - unsigned long npages); +unsigned long iommu_range_alloc(struct device *dev, + struct iommu *iommu, + unsigned long npages, + unsigned long *handle); +void iommu_range_free(struct iommu *iommu, + dma_addr_t dma_addr, + unsigned long npages); #endif /* _IOMMU_COMMON_H */ diff --git a/arch/sparc/kernel/ioport.c b/arch/sparc/kernel/ioport.c index 0f094db918c..7f08ec8a7c6 100644 --- a/arch/sparc/kernel/ioport.c +++ b/arch/sparc/kernel/ioport.c @@ -186,7 +186,7 @@ static void __iomem *_sparc_alloc_io(unsigned int busno, unsigned long phys, if (name == NULL) name = "???"; - if ((xres = xres_alloc()) != 0) { + if ((xres = xres_alloc()) != NULL) { tack = xres->xname; res = &xres->xres; } else { @@ -400,7 +400,7 @@ static void sbus_sync_sg_for_device(struct device *dev, struct scatterlist *sg, BUG(); } -struct dma_map_ops sbus_dma_ops = { +static struct dma_map_ops sbus_dma_ops = { .alloc = sbus_alloc_coherent, .free = sbus_free_coherent, .map_page = sbus_map_page, @@ -666,10 +666,9 @@ EXPORT_SYMBOL(dma_ops); */ int dma_supported(struct device *dev, u64 mask) { -#ifdef CONFIG_PCI - if (dev->bus == &pci_bus_type) + if (dev_is_pci(dev)) return 1; -#endif + return 0; } EXPORT_SYMBOL(dma_supported); @@ -682,7 +681,7 @@ static int sparc_io_proc_show(struct seq_file *m, void *v) const char *nm; for (r = root->child; r != NULL; r = r->sibling) { - if ((nm = r->name) == 0) nm = "???"; + if ((nm = r->name) == NULL) nm = "???"; seq_printf(m, "%016llx-%016llx: %s\n", (unsigned long long)r->start, (unsigned long long)r->end, nm); @@ -693,7 +692,7 @@ static int sparc_io_proc_show(struct seq_file *m, void *v) static int sparc_io_proc_open(struct inode *inode, struct file *file) { - return single_open(file, sparc_io_proc_show, PDE(inode)->data); + return single_open(file, sparc_io_proc_show, PDE_DATA(inode)); } static const struct file_operations sparc_io_proc_fops = { diff --git a/arch/sparc/kernel/irq.h b/arch/sparc/kernel/irq.h index b66b6aad1d6..70a0b8ddd0b 100644 --- a/arch/sparc/kernel/irq.h +++ b/arch/sparc/kernel/irq.h @@ -82,11 +82,20 @@ void handler_irq(unsigned int pil, struct pt_regs *regs); unsigned long leon_get_irqmask(unsigned int irq); +/* irq_32.c */ +void sparc_floppy_irq(int irq, void *dev_id, struct pt_regs *regs); + +/* sun4m_irq.c */ +void sun4m_nmi(struct pt_regs *regs); + +/* sun4d_irq.c */ +void sun4d_handler_irq(unsigned int pil, struct pt_regs *regs); + #ifdef CONFIG_SMP /* All SUN4D IPIs are sent on this IRQ, may be shared with hard IRQs */ #define SUN4D_IPI_IRQ 13 -extern void sun4d_ipi_interrupt(void); +void sun4d_ipi_interrupt(void); #endif diff --git a/arch/sparc/kernel/irq_32.c b/arch/sparc/kernel/irq_32.c index c145f6fd123..a979e99f875 100644 --- a/arch/sparc/kernel/irq_32.c +++ b/arch/sparc/kernel/irq_32.c @@ -17,6 +17,7 @@ #include <asm/cacheflush.h> #include <asm/cpudata.h> +#include <asm/setup.h> #include <asm/pcic.h> #include <asm/leon.h> diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c index 9bcbbe2c4e7..666193f4e8b 100644 --- a/arch/sparc/kernel/irq_64.c +++ b/arch/sparc/kernel/irq_64.c @@ -698,30 +698,19 @@ void __irq_entry handler_irq(int pil, struct pt_regs *regs) set_irq_regs(old_regs); } -void do_softirq(void) +void do_softirq_own_stack(void) { - unsigned long flags; - - if (in_interrupt()) - return; - - local_irq_save(flags); + void *orig_sp, *sp = softirq_stack[smp_processor_id()]; - if (local_softirq_pending()) { - void *orig_sp, *sp = softirq_stack[smp_processor_id()]; - - sp += THREAD_SIZE - 192 - STACK_BIAS; - - __asm__ __volatile__("mov %%sp, %0\n\t" - "mov %1, %%sp" - : "=&r" (orig_sp) - : "r" (sp)); - __do_softirq(); - __asm__ __volatile__("mov %0, %%sp" - : : "r" (orig_sp)); - } + sp += THREAD_SIZE - 192 - STACK_BIAS; - local_irq_restore(flags); + __asm__ __volatile__("mov %%sp, %0\n\t" + "mov %1, %%sp" + : "=&r" (orig_sp) + : "r" (sp)); + __do_softirq(); + __asm__ __volatile__("mov %0, %%sp" + : : "r" (orig_sp)); } #ifdef CONFIG_HOTPLUG_CPU @@ -835,7 +824,8 @@ void notrace init_irqwork_curcpu(void) * Therefore you cannot make any OBP calls, not even prom_printf, * from these two routines. */ -static void __cpuinit notrace register_one_mondo(unsigned long paddr, unsigned long type, unsigned long qmask) +static void notrace register_one_mondo(unsigned long paddr, unsigned long type, + unsigned long qmask) { unsigned long num_entries = (qmask + 1) / 64; unsigned long status; @@ -848,7 +838,7 @@ static void __cpuinit notrace register_one_mondo(unsigned long paddr, unsigned l } } -void __cpuinit notrace sun4v_register_mondo_queues(int this_cpu) +void notrace sun4v_register_mondo_queues(int this_cpu) { struct trap_per_cpu *tb = &trap_block[this_cpu]; diff --git a/arch/sparc/kernel/kernel.h b/arch/sparc/kernel/kernel.h index 291bb5de9ce..e7f652be9e6 100644 --- a/arch/sparc/kernel/kernel.h +++ b/arch/sparc/kernel/kernel.h @@ -2,6 +2,7 @@ #define __SPARC_KERNEL_H #include <linux/interrupt.h> +#include <linux/ftrace.h> #include <asm/traps.h> #include <asm/head.h> @@ -15,50 +16,111 @@ extern int ncpus_probed; #ifdef CONFIG_SPARC64 /* setup_64.c */ struct seq_file; -extern void cpucap_info(struct seq_file *); +void cpucap_info(struct seq_file *); -static inline unsigned long kimage_addr_to_ra(const char *p) +static inline unsigned long kimage_addr_to_ra(const void *p) { unsigned long val = (unsigned long) p; return kern_base + (val - KERNBASE); } + +/* sys_sparc_64.c */ +asmlinkage long sys_kern_features(void); + +/* unaligned_64.c */ +asmlinkage void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn); +int handle_popc(u32 insn, struct pt_regs *regs); +void handle_lddfmna(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr); +void handle_stdfmna(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr); + +/* smp_64.c */ +void __irq_entry smp_call_function_client(int irq, struct pt_regs *regs); +void __irq_entry smp_call_function_single_client(int irq, struct pt_regs *regs); +void __irq_entry smp_new_mmu_context_version_client(int irq, struct pt_regs *regs); +void __irq_entry smp_penguin_jailcell(int irq, struct pt_regs *regs); +void __irq_entry smp_receive_signal_client(int irq, struct pt_regs *regs); + +/* kgdb_64.c */ +void __irq_entry smp_kgdb_capture_client(int irq, struct pt_regs *regs); + +/* pci.c */ +int pci64_dma_supported(struct pci_dev *pdev, u64 device_mask); + +/* signal32.c */ +void do_sigreturn32(struct pt_regs *regs); +asmlinkage void do_rt_sigreturn32(struct pt_regs *regs); +void do_signal32(struct pt_regs * regs); +asmlinkage int do_sys32_sigstack(u32 u_ssptr, u32 u_ossptr, unsigned long sp); + +/* compat_audit.c */ +extern unsigned sparc32_dir_class[]; +extern unsigned sparc32_chattr_class[]; +extern unsigned sparc32_write_class[]; +extern unsigned sparc32_read_class[]; +extern unsigned sparc32_signal_class[]; +int sparc32_classify_syscall(unsigned syscall); #endif #ifdef CONFIG_SPARC32 /* setup_32.c */ +struct linux_romvec; void sparc32_start_kernel(struct linux_romvec *rp); /* cpu.c */ -extern void cpu_probe(void); +void cpu_probe(void); /* traps_32.c */ -extern void handle_hw_divzero(struct pt_regs *regs, unsigned long pc, - unsigned long npc, unsigned long psr); +void handle_hw_divzero(struct pt_regs *regs, unsigned long pc, + unsigned long npc, unsigned long psr); /* irq_32.c */ extern struct irqaction static_irqaction[]; extern int static_irq_count; extern spinlock_t irq_action_lock; -extern void unexpected_irq(int irq, void *dev_id, struct pt_regs * regs); -extern void init_IRQ(void); +void unexpected_irq(int irq, void *dev_id, struct pt_regs * regs); +void init_IRQ(void); /* sun4m_irq.c */ -extern void sun4m_init_IRQ(void); -extern void sun4m_unmask_profile_irq(void); -extern void sun4m_clear_profile_irq(int cpu); +void sun4m_init_IRQ(void); +void sun4m_unmask_profile_irq(void); +void sun4m_clear_profile_irq(int cpu); + +/* sun4m_smp.c */ +void sun4m_cpu_pre_starting(void *arg); +void sun4m_cpu_pre_online(void *arg); +void __init smp4m_boot_cpus(void); +int smp4m_boot_one_cpu(int i, struct task_struct *idle); +void __init smp4m_smp_done(void); +void smp4m_cross_call_irq(void); +void smp4m_percpu_timer_interrupt(struct pt_regs *regs); /* sun4d_irq.c */ extern spinlock_t sun4d_imsk_lock; -extern void sun4d_init_IRQ(void); -extern int sun4d_request_irq(unsigned int irq, - irq_handler_t handler, - unsigned long irqflags, - const char *devname, void *dev_id); -extern int show_sun4d_interrupts(struct seq_file *, void *); -extern void sun4d_distribute_irqs(void); -extern void sun4d_free_irq(unsigned int irq, void *dev_id); +void sun4d_init_IRQ(void); +int sun4d_request_irq(unsigned int irq, + irq_handler_t handler, + unsigned long irqflags, + const char *devname, void *dev_id); +int show_sun4d_interrupts(struct seq_file *, void *); +void sun4d_distribute_irqs(void); +void sun4d_free_irq(unsigned int irq, void *dev_id); + +/* sun4d_smp.c */ +void sun4d_cpu_pre_starting(void *arg); +void sun4d_cpu_pre_online(void *arg); +void __init smp4d_boot_cpus(void); +int smp4d_boot_one_cpu(int i, struct task_struct *idle); +void __init smp4d_smp_done(void); +void smp4d_cross_call_irq(void); +void smp4d_percpu_timer_interrupt(struct pt_regs *regs); + +/* leon_smp.c */ +void leon_cpu_pre_starting(void *arg); +void leon_cpu_pre_online(void *arg); +void leonsmp_ipi_interrupt(void); +void leon_cross_call_irq(void); /* head_32.S */ extern unsigned int t_nmi[]; @@ -77,12 +139,48 @@ extern unsigned int real_irq_entry[]; extern unsigned int smp4d_ticker[]; extern unsigned int patchme_maybe_smp_msg[]; -extern void floppy_hardint(void); +void floppy_hardint(void); /* trampoline_32.S */ extern unsigned long sun4m_cpu_startup; extern unsigned long sun4d_cpu_startup; +/* process_32.c */ +asmlinkage int sparc_do_fork(unsigned long clone_flags, + unsigned long stack_start, + struct pt_regs *regs, + unsigned long stack_size); + +/* signal_32.c */ +asmlinkage void do_sigreturn(struct pt_regs *regs); +asmlinkage void do_rt_sigreturn(struct pt_regs *regs); +void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0, + unsigned long thread_info_flags); +asmlinkage int do_sys_sigstack(struct sigstack __user *ssptr, + struct sigstack __user *ossptr, + unsigned long sp); + +/* ptrace_32.c */ +asmlinkage int syscall_trace(struct pt_regs *regs, int syscall_exit_p); + +/* unaligned_32.c */ +asmlinkage void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn); +asmlinkage void user_unaligned_trap(struct pt_regs *regs, unsigned int insn); + +/* windows.c */ +void try_to_clear_window_buffer(struct pt_regs *regs, int who); + +/* auxio_32.c */ +void __init auxio_probe(void); +void __init auxio_power_probe(void); + +/* pcic.c */ +extern void __iomem *pcic_regs; +void pcic_nmi(unsigned int pend, struct pt_regs *regs); + +/* time_32.c */ +void __init time_init(void); + #else /* CONFIG_SPARC32 */ #endif /* CONFIG_SPARC32 */ #endif /* !(__SPARC_KERNEL_H) */ diff --git a/arch/sparc/kernel/kgdb_32.c b/arch/sparc/kernel/kgdb_32.c index 2e424a576a3..dcf210811af 100644 --- a/arch/sparc/kernel/kgdb_32.c +++ b/arch/sparc/kernel/kgdb_32.c @@ -5,6 +5,7 @@ #include <linux/kgdb.h> #include <linux/kdebug.h> +#include <linux/sched.h> #include <asm/kdebug.h> #include <asm/ptrace.h> diff --git a/arch/sparc/kernel/kgdb_64.c b/arch/sparc/kernel/kgdb_64.c index c8759550799..cbf21d0870e 100644 --- a/arch/sparc/kernel/kgdb_64.c +++ b/arch/sparc/kernel/kgdb_64.c @@ -6,12 +6,15 @@ #include <linux/kgdb.h> #include <linux/kdebug.h> #include <linux/ftrace.h> +#include <linux/context_tracking.h> #include <asm/cacheflush.h> #include <asm/kdebug.h> #include <asm/ptrace.h> #include <asm/irq.h> +#include "kernel.h" + void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs) { struct reg_window *win; @@ -42,7 +45,7 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) { struct thread_info *t = task_thread_info(p); extern unsigned int switch_to_pc; - extern unsigned int ret_from_syscall; + extern unsigned int ret_from_fork; struct reg_window *win; unsigned long pc, cwp; int i; @@ -66,7 +69,7 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p) gdb_regs[i] = 0; if (t->new_child) - pc = (unsigned long) &ret_from_syscall; + pc = (unsigned long) &ret_from_fork; else pc = (unsigned long) &switch_to_pc; @@ -159,11 +162,12 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code, asmlinkage void kgdb_trap(unsigned long trap_level, struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); unsigned long flags; if (user_mode(regs)) { bad_trap(regs, trap_level); - return; + goto out; } flushw_all(); @@ -171,6 +175,8 @@ asmlinkage void kgdb_trap(unsigned long trap_level, struct pt_regs *regs) local_irq_save(flags); kgdb_handle_exception(0x172, SIGTRAP, 0, regs); local_irq_restore(flags); +out: + exception_exit(prev_state); } int kgdb_arch_init(void) diff --git a/arch/sparc/kernel/kprobes.c b/arch/sparc/kernel/kprobes.c index a39d1ba5a11..98d71284341 100644 --- a/arch/sparc/kernel/kprobes.c +++ b/arch/sparc/kernel/kprobes.c @@ -8,6 +8,7 @@ #include <linux/module.h> #include <linux/kdebug.h> #include <linux/slab.h> +#include <linux/context_tracking.h> #include <asm/signal.h> #include <asm/cacheflush.h> #include <asm/uaccess.h> @@ -349,7 +350,7 @@ int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr) case KPROBE_HIT_SSDONE: /* * We increment the nmissed count for accounting, - * we can also use npre/npostfault count for accouting + * we can also use npre/npostfault count for accounting * these specific fault cases. */ kprobes_inc_nmissed_count(cur); @@ -418,12 +419,14 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, asmlinkage void __kprobes kprobe_trap(unsigned long trap_level, struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); + BUG_ON(trap_level != 0x170 && trap_level != 0x171); if (user_mode(regs)) { local_irq_enable(); bad_trap(regs, trap_level); - return; + goto out; } /* trap_level == 0x170 --> ta 0x70 @@ -433,6 +436,8 @@ asmlinkage void __kprobes kprobe_trap(unsigned long trap_level, (trap_level == 0x170) ? "debug" : "debug_2", regs, 0, trap_level, SIGTRAP) != NOTIFY_STOP) bad_trap(regs, trap_level); +out: + exception_exit(prev_state); } /* Jprobes support. */ @@ -507,11 +512,12 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, /* * Called when the probe at kretprobe trampoline is hit */ -int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) +static int __kprobes trampoline_probe_handler(struct kprobe *p, + struct pt_regs *regs) { struct kretprobe_instance *ri = NULL; struct hlist_head *head, empty_rp; - struct hlist_node *node, *tmp; + struct hlist_node *tmp; unsigned long flags, orig_ret_address = 0; unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline; @@ -531,7 +537,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) * real return address, and all the rest will point to * kretprobe_trampoline */ - hlist_for_each_entry_safe(ri, node, tmp, head, hlist) { + hlist_for_each_entry_safe(ri, tmp, head, hlist) { if (ri->task != current) /* another task is sharing our hash bucket */ continue; @@ -559,7 +565,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) kretprobe_hash_unlock(current, &flags); preempt_enable_no_resched(); - hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) { + hlist_for_each_entry_safe(ri, tmp, &empty_rp, hlist) { hlist_del(&ri->hlist); kfree(ri); } @@ -571,7 +577,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs) return 1; } -void kretprobe_trampoline_holder(void) +static void __used kretprobe_trampoline_holder(void) { asm volatile(".global kretprobe_trampoline\n" "kretprobe_trampoline:\n" diff --git a/arch/sparc/kernel/ktlb.S b/arch/sparc/kernel/ktlb.S index 79f31036484..605d4920458 100644 --- a/arch/sparc/kernel/ktlb.S +++ b/arch/sparc/kernel/ktlb.S @@ -25,11 +25,10 @@ kvmap_itlb: */ kvmap_itlb_4v: -kvmap_itlb_nonlinear: /* Catch kernel NULL pointer calls. */ sethi %hi(PAGE_SIZE), %g5 cmp %g4, %g5 - bleu,pn %xcc, kvmap_dtlb_longpath + blu,pn %xcc, kvmap_itlb_longpath nop KERN_TSB_LOOKUP_TL1(%g4, %g6, %g5, %g1, %g2, %g3, kvmap_itlb_load) @@ -154,12 +153,19 @@ kvmap_dtlb_tsb4m_miss: /* Clear the PAGE_OFFSET top virtual bits, shift * down to get PFN, and make sure PFN is in range. */ - sllx %g4, 21, %g5 +661: sllx %g4, 0, %g5 + .section .page_offset_shift_patch, "ax" + .word 661b + .previous /* Check to see if we know about valid memory at the 4MB * chunk this physical address will reside within. */ - srlx %g5, 21 + 41, %g2 +661: srlx %g5, MAX_PHYS_ADDRESS_BITS, %g2 + .section .page_offset_shift_patch, "ax" + .word 661b + .previous + brnz,pn %g2, kvmap_dtlb_longpath nop @@ -177,7 +183,11 @@ valid_addr_bitmap_patch: or %g7, %lo(sparc64_valid_addr_bitmap), %g7 .previous - srlx %g5, 21 + 22, %g2 +661: srlx %g5, ILOG2_4MB, %g2 + .section .page_offset_shift_patch, "ax" + .word 661b + .previous + srlx %g2, 6, %g5 and %g2, 63, %g2 sllx %g5, 3, %g5 @@ -188,31 +198,35 @@ valid_addr_bitmap_patch: be,pn %xcc, kvmap_dtlb_longpath 2: sethi %hi(kpte_linear_bitmap), %g2 - or %g2, %lo(kpte_linear_bitmap), %g2 /* Get the 256MB physical address index. */ - sllx %g4, 21, %g5 - mov 1, %g7 - srlx %g5, 21 + 28, %g5 +661: sllx %g4, 0, %g5 + .section .page_offset_shift_patch, "ax" + .word 661b + .previous - /* Don't try this at home kids... this depends upon srlx - * only taking the low 6 bits of the shift count in %g5. - */ - sllx %g7, %g5, %g7 + or %g2, %lo(kpte_linear_bitmap), %g2 - /* Divide by 64 to get the offset into the bitmask. */ - srlx %g5, 6, %g5 +661: srlx %g5, ILOG2_256MB, %g5 + .section .page_offset_shift_patch, "ax" + .word 661b + .previous + + and %g5, (32 - 1), %g7 + + /* Divide by 32 to get the offset into the bitmask. */ + srlx %g5, 5, %g5 + add %g7, %g7, %g7 sllx %g5, 3, %g5 - /* kern_linear_pte_xor[((mask & bit) ? 1 : 0)] */ + /* kern_linear_pte_xor[(mask >> shift) & 3)] */ ldx [%g2 + %g5], %g2 - andcc %g2, %g7, %g0 + srlx %g2, %g7, %g7 sethi %hi(kern_linear_pte_xor), %g5 + and %g7, 3, %g7 or %g5, %lo(kern_linear_pte_xor), %g5 - bne,a,pt %xcc, 1f - add %g5, 8, %g5 - -1: ldx [%g5], %g2 + sllx %g7, 3, %g7 + ldx [%g5 + %g7], %g2 .globl kvmap_linear_patch kvmap_linear_patch: @@ -263,7 +277,7 @@ kvmap_dtlb_load: #ifdef CONFIG_SPARSEMEM_VMEMMAP kvmap_vmemmap: sub %g4, %g5, %g5 - srlx %g5, 22, %g5 + srlx %g5, ILOG2_4MB, %g5 sethi %hi(vmemmap_table), %g1 sllx %g5, 3, %g5 or %g1, %lo(vmemmap_table), %g1 diff --git a/arch/sparc/kernel/ldc.c b/arch/sparc/kernel/ldc.c index 435e406fdec..e01d75d4032 100644 --- a/arch/sparc/kernel/ldc.c +++ b/arch/sparc/kernel/ldc.c @@ -27,7 +27,7 @@ #define DRV_MODULE_VERSION "1.1" #define DRV_MODULE_RELDATE "July 22, 2008" -static char version[] __devinitdata = +static char version[] = DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; #define LDC_PACKET_SIZE 64 @@ -953,9 +953,8 @@ static HLIST_HEAD(ldc_channel_list); static int __ldc_channel_exists(unsigned long id) { struct ldc_channel *lp; - struct hlist_node *n; - hlist_for_each_entry(lp, n, &ldc_channel_list, list) { + hlist_for_each_entry(lp, &ldc_channel_list, list) { if (lp->id == id) return 1; } @@ -1250,14 +1249,12 @@ int ldc_bind(struct ldc_channel *lp, const char *name) snprintf(lp->rx_irq_name, LDC_IRQ_NAME_MAX, "%s RX", name); snprintf(lp->tx_irq_name, LDC_IRQ_NAME_MAX, "%s TX", name); - err = request_irq(lp->cfg.rx_irq, ldc_rx, - IRQF_SAMPLE_RANDOM | IRQF_DISABLED, + err = request_irq(lp->cfg.rx_irq, ldc_rx, 0, lp->rx_irq_name, lp); if (err) return err; - err = request_irq(lp->cfg.tx_irq, ldc_tx, - IRQF_SAMPLE_RANDOM | IRQF_DISABLED, + err = request_irq(lp->cfg.tx_irq, ldc_tx, 0, lp->tx_irq_name, lp); if (err) { free_irq(lp->cfg.rx_irq, lp); diff --git a/arch/sparc/kernel/leon_kernel.c b/arch/sparc/kernel/leon_kernel.c index e34e2c40c06..683c4af999d 100644 --- a/arch/sparc/kernel/leon_kernel.c +++ b/arch/sparc/kernel/leon_kernel.c @@ -32,13 +32,12 @@ struct leon3_gptimer_regs_map *leon3_gptimer_regs; /* timer controller base addr int leondebug_irq_disable; int leon_debug_irqout; -static int dummy_master_l10_counter; +static volatile u32 dummy_master_l10_counter; unsigned long amba_system_id; static DEFINE_SPINLOCK(leon_irq_lock); +static unsigned long leon3_gptimer_idx; /* Timer Index (0..6) within Timer Core */ unsigned long leon3_gptimer_irq; /* interrupt controller irq number */ -unsigned long leon3_gptimer_idx; /* Timer Index (0..6) within Timer Core */ -int leon3_ticker_irq; /* Timer ticker IRQ */ unsigned int sparc_leon_eirq; #define LEON_IMASK(cpu) (&leon3_irqctrl_regs->mask[cpu]) #define LEON_IACK (&leon3_irqctrl_regs->iclear) @@ -56,15 +55,17 @@ static inline unsigned int leon_eirq_get(int cpu) static void leon_handle_ext_irq(unsigned int irq, struct irq_desc *desc) { unsigned int eirq; + struct irq_bucket *p; int cpu = sparc_leon3_cpuid(); eirq = leon_eirq_get(cpu); - if ((eirq & 0x10) && irq_map[eirq]->irq) /* bit4 tells if IRQ happened */ - generic_handle_irq(irq_map[eirq]->irq); + p = irq_map[eirq]; + if ((eirq & 0x10) && p && p->irq) /* bit4 tells if IRQ happened */ + generic_handle_irq(p->irq); } /* The extended IRQ controller has been found, this function registers it */ -void leon_eirq_setup(unsigned int eirq) +static void leon_eirq_setup(unsigned int eirq) { unsigned long mask, oldmask; unsigned int veirq; @@ -211,6 +212,7 @@ unsigned int leon_build_device_irq(unsigned int real_irq, { unsigned int irq; unsigned long mask; + struct irq_desc *desc; irq = 0; mask = leon_get_irqmask(real_irq); @@ -224,9 +226,12 @@ unsigned int leon_build_device_irq(unsigned int real_irq, if (do_ack) mask |= LEON_DO_ACK_HW; - irq_set_chip_and_handler_name(irq, &leon_irq, - flow_handler, name); - irq_set_chip_data(irq, (void *)mask); + desc = irq_to_desc(irq); + if (!desc || !desc->handle_irq || desc->handle_irq == handle_bad_irq) { + irq_set_chip_and_handler_name(irq, &leon_irq, + flow_handler, name); + irq_set_chip_data(irq, (void *)mask); + } out: return irq; @@ -265,13 +270,16 @@ static u32 leon_cycles_offset(void) #ifdef CONFIG_SMP /* smp clockevent irq */ -irqreturn_t leon_percpu_timer_ce_interrupt(int irq, void *unused) +static irqreturn_t leon_percpu_timer_ce_interrupt(int irq, void *unused) { struct clock_event_device *ce; int cpu = smp_processor_id(); leon_clear_profile_irq(cpu); + if (cpu == boot_cpu_id) + timer_interrupt(irq, NULL); + ce = &per_cpu(sparc32_clockevent, cpu); irq_enter(); @@ -293,6 +301,7 @@ void __init leon_init_timers(void) int icsel; int ampopts; int err; + u32 config; sparc_config.get_cycles_offset = leon_cycles_offset; sparc_config.cs_period = 1000000 / HZ; @@ -304,7 +313,7 @@ void __init leon_init_timers(void) leondebug_irq_disable = 0; leon_debug_irqout = 0; - master_l10_counter = (unsigned int *)&dummy_master_l10_counter; + master_l10_counter = (u32 __iomem *)&dummy_master_l10_counter; dummy_master_l10_counter = 0; rootnp = of_find_node_by_path("/ambapp0"); @@ -371,23 +380,6 @@ void __init leon_init_timers(void) LEON3_BYPASS_STORE_PA( &leon3_gptimer_regs->e[leon3_gptimer_idx].ctrl, 0); -#ifdef CONFIG_SMP - leon3_ticker_irq = leon3_gptimer_irq + 1 + leon3_gptimer_idx; - - if (!(LEON3_BYPASS_LOAD_PA(&leon3_gptimer_regs->config) & - (1<<LEON3_GPTIMER_SEPIRQ))) { - printk(KERN_ERR "timer not configured with separate irqs\n"); - BUG(); - } - - LEON3_BYPASS_STORE_PA(&leon3_gptimer_regs->e[leon3_gptimer_idx+1].val, - 0); - LEON3_BYPASS_STORE_PA(&leon3_gptimer_regs->e[leon3_gptimer_idx+1].rld, - (((1000000/HZ) - 1))); - LEON3_BYPASS_STORE_PA(&leon3_gptimer_regs->e[leon3_gptimer_idx+1].ctrl, - 0); -#endif - /* * The IRQ controller may (if implemented) consist of multiple * IRQ controllers, each mapped on a 4Kb boundary. @@ -410,13 +402,6 @@ void __init leon_init_timers(void) if (eirq != 0) leon_eirq_setup(eirq); - irq = _leon_build_device_irq(NULL, leon3_gptimer_irq+leon3_gptimer_idx); - err = request_irq(irq, timer_interrupt, IRQF_TIMER, "timer", NULL); - if (err) { - printk(KERN_ERR "unable to attach timer IRQ%d\n", irq); - prom_halt(); - } - #ifdef CONFIG_SMP { unsigned long flags; @@ -433,30 +418,31 @@ void __init leon_init_timers(void) } #endif - LEON3_BYPASS_STORE_PA(&leon3_gptimer_regs->e[leon3_gptimer_idx].ctrl, - LEON3_GPTIMER_EN | - LEON3_GPTIMER_RL | - LEON3_GPTIMER_LD | - LEON3_GPTIMER_IRQEN); + config = LEON3_BYPASS_LOAD_PA(&leon3_gptimer_regs->config); + if (config & (1 << LEON3_GPTIMER_SEPIRQ)) + leon3_gptimer_irq += leon3_gptimer_idx; + else if ((config & LEON3_GPTIMER_TIMERS) > 1) + pr_warn("GPTIMER uses shared irqs, using other timers of the same core will fail.\n"); #ifdef CONFIG_SMP /* Install per-cpu IRQ handler for broadcasted ticker */ - irq = leon_build_device_irq(leon3_ticker_irq, handle_percpu_irq, + irq = leon_build_device_irq(leon3_gptimer_irq, handle_percpu_irq, "per-cpu", 0); err = request_irq(irq, leon_percpu_timer_ce_interrupt, - IRQF_PERCPU | IRQF_TIMER, "ticker", - NULL); + IRQF_PERCPU | IRQF_TIMER, "timer", NULL); +#else + irq = _leon_build_device_irq(NULL, leon3_gptimer_irq); + err = request_irq(irq, timer_interrupt, IRQF_TIMER, "timer", NULL); +#endif if (err) { - printk(KERN_ERR "unable to attach ticker IRQ%d\n", irq); + pr_err("Unable to attach timer IRQ%d\n", irq); prom_halt(); } - - LEON3_BYPASS_STORE_PA(&leon3_gptimer_regs->e[leon3_gptimer_idx+1].ctrl, + LEON3_BYPASS_STORE_PA(&leon3_gptimer_regs->e[leon3_gptimer_idx].ctrl, LEON3_GPTIMER_EN | LEON3_GPTIMER_RL | LEON3_GPTIMER_LD | LEON3_GPTIMER_IRQEN); -#endif return; bad: printk(KERN_ERR "No Timer/irqctrl found\n"); @@ -486,17 +472,6 @@ void __init leon_trans_init(struct device_node *dp) } } -void __initdata (*prom_amba_init)(struct device_node *dp, struct device_node ***nextp) = 0; - -void __init leon_node_init(struct device_node *dp, struct device_node ***nextp) -{ - if (prom_amba_init && - strcmp(dp->type, "ambapp") == 0 && - strcmp(dp->name, "ambapp0") == 0) { - prom_amba_init(dp, nextp); - } -} - #ifdef CONFIG_SMP void leon_clear_profile_irq(int cpu) { @@ -522,8 +497,3 @@ void __init leon_init_IRQ(void) sparc_config.clear_clock_irq = leon_clear_clock_irq; sparc_config.load_profile_irq = leon_load_profile_irq; } - -void __init leon_init(void) -{ - of_pdt_build_more = &leon_node_init; -} diff --git a/arch/sparc/kernel/leon_pci.c b/arch/sparc/kernel/leon_pci.c index 19f56058742..899b7203a4e 100644 --- a/arch/sparc/kernel/leon_pci.c +++ b/arch/sparc/kernel/leon_pci.c @@ -29,6 +29,8 @@ void leon_pci_init(struct platform_device *ofdev, struct leon_pci_info *info) pci_add_resource_offset(&resources, &info->io_space, info->io_space.start - 0x1000); pci_add_resource(&resources, &info->mem_space); + info->busn.flags = IORESOURCE_BUS; + pci_add_resource(&resources, &info->busn); root_bus = pci_scan_root_bus(&ofdev->dev, 0, info->ops, info, &resources); @@ -43,7 +45,7 @@ void leon_pci_init(struct platform_device *ofdev, struct leon_pci_info *info) } } -void __devinit pcibios_fixup_bus(struct pci_bus *pbus) +void pcibios_fixup_bus(struct pci_bus *pbus) { struct pci_dev *dev; int i, has_io, has_mem; @@ -91,109 +93,8 @@ void __devinit pcibios_fixup_bus(struct pci_bus *pbus) } } -/* - * Other archs parse arguments here. - */ -char * __devinit pcibios_setup(char *str) -{ - return str; -} - resource_size_t pcibios_align_resource(void *data, const struct resource *res, resource_size_t size, resource_size_t align) { return res->start; } - -int pcibios_enable_device(struct pci_dev *dev, int mask) -{ - return pci_enable_resources(dev, mask); -} - -void __devinit pcibios_update_irq(struct pci_dev *dev, int irq) -{ -#ifdef CONFIG_PCI_DEBUG - printk(KERN_DEBUG "LEONPCI: Assigning IRQ %02d to %s\n", irq, - pci_name(dev)); -#endif - pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq); -} - -/* in/out routines taken from pcic.c - * - * This probably belongs here rather than ioport.c because - * we do not want this crud linked into SBus kernels. - * Also, think for a moment about likes of floppy.c that - * include architecture specific parts. They may want to redefine ins/outs. - * - * We do not use horrible macros here because we want to - * advance pointer by sizeof(size). - */ -void outsb(unsigned long addr, const void *src, unsigned long count) -{ - while (count) { - count -= 1; - outb(*(const char *)src, addr); - src += 1; - /* addr += 1; */ - } -} -EXPORT_SYMBOL(outsb); - -void outsw(unsigned long addr, const void *src, unsigned long count) -{ - while (count) { - count -= 2; - outw(*(const short *)src, addr); - src += 2; - /* addr += 2; */ - } -} -EXPORT_SYMBOL(outsw); - -void outsl(unsigned long addr, const void *src, unsigned long count) -{ - while (count) { - count -= 4; - outl(*(const long *)src, addr); - src += 4; - /* addr += 4; */ - } -} -EXPORT_SYMBOL(outsl); - -void insb(unsigned long addr, void *dst, unsigned long count) -{ - while (count) { - count -= 1; - *(unsigned char *)dst = inb(addr); - dst += 1; - /* addr += 1; */ - } -} -EXPORT_SYMBOL(insb); - -void insw(unsigned long addr, void *dst, unsigned long count) -{ - while (count) { - count -= 2; - *(unsigned short *)dst = inw(addr); - dst += 2; - /* addr += 2; */ - } -} -EXPORT_SYMBOL(insw); - -void insl(unsigned long addr, void *dst, unsigned long count) -{ - while (count) { - count -= 4; - /* - * XXX I am sure we are in for an unaligned trap here. - */ - *(unsigned long *)dst = inl(addr); - dst += 4; - /* addr += 4; */ - } -} -EXPORT_SYMBOL(insl); diff --git a/arch/sparc/kernel/leon_pci_grpci1.c b/arch/sparc/kernel/leon_pci_grpci1.c new file mode 100644 index 00000000000..c8bf26edfa7 --- /dev/null +++ b/arch/sparc/kernel/leon_pci_grpci1.c @@ -0,0 +1,722 @@ +/* + * leon_pci_grpci1.c: GRPCI1 Host PCI driver + * + * Copyright (C) 2013 Aeroflex Gaisler AB + * + * This GRPCI1 driver does not support PCI interrupts taken from + * GPIO pins. Interrupt generation at PCI parity and system error + * detection is by default turned off since some GRPCI1 cores does + * not support detection. It can be turned on from the bootloader + * using the all_pci_errors property. + * + * Contributors: Daniel Hellstrom <daniel@gaisler.com> + */ + +#include <linux/of_device.h> +#include <linux/export.h> +#include <linux/kernel.h> +#include <linux/of_irq.h> +#include <linux/delay.h> +#include <linux/pci.h> + +#include <asm/leon_pci.h> +#include <asm/sections.h> +#include <asm/vaddrs.h> +#include <asm/leon.h> +#include <asm/io.h> + +#include "irq.h" + +/* Enable/Disable Debugging Configuration Space Access */ +#undef GRPCI1_DEBUG_CFGACCESS + +/* + * GRPCI1 APB Register MAP + */ +struct grpci1_regs { + unsigned int cfg_stat; /* 0x00 Configuration / Status */ + unsigned int bar0; /* 0x04 BAR0 (RO) */ + unsigned int page0; /* 0x08 PAGE0 (RO) */ + unsigned int bar1; /* 0x0C BAR1 (RO) */ + unsigned int page1; /* 0x10 PAGE1 */ + unsigned int iomap; /* 0x14 IO Map */ + unsigned int stat_cmd; /* 0x18 PCI Status & Command (RO) */ + unsigned int irq; /* 0x1C Interrupt register */ +}; + +#define REGLOAD(a) (be32_to_cpu(__raw_readl(&(a)))) +#define REGSTORE(a, v) (__raw_writel(cpu_to_be32(v), &(a))) + +#define PAGE0_BTEN_BIT 0 +#define PAGE0_BTEN (1 << PAGE0_BTEN_BIT) + +#define CFGSTAT_HOST_BIT 13 +#define CFGSTAT_CTO_BIT 8 +#define CFGSTAT_HOST (1 << CFGSTAT_HOST_BIT) +#define CFGSTAT_CTO (1 << CFGSTAT_CTO_BIT) + +#define IRQ_DPE (1 << 9) +#define IRQ_SSE (1 << 8) +#define IRQ_RMA (1 << 7) +#define IRQ_RTA (1 << 6) +#define IRQ_STA (1 << 5) +#define IRQ_DPED (1 << 4) +#define IRQ_INTD (1 << 3) +#define IRQ_INTC (1 << 2) +#define IRQ_INTB (1 << 1) +#define IRQ_INTA (1 << 0) +#define IRQ_DEF_ERRORS (IRQ_RMA | IRQ_RTA | IRQ_STA) +#define IRQ_ALL_ERRORS (IRQ_DPED | IRQ_DEF_ERRORS | IRQ_SSE | IRQ_DPE) +#define IRQ_INTX (IRQ_INTA | IRQ_INTB | IRQ_INTC | IRQ_INTD) +#define IRQ_MASK_BIT 16 + +#define DEF_PCI_ERRORS (PCI_STATUS_SIG_TARGET_ABORT | \ + PCI_STATUS_REC_TARGET_ABORT | \ + PCI_STATUS_REC_MASTER_ABORT) +#define ALL_PCI_ERRORS (PCI_STATUS_PARITY | PCI_STATUS_DETECTED_PARITY | \ + PCI_STATUS_SIG_SYSTEM_ERROR | DEF_PCI_ERRORS) + +#define TGT 256 + +struct grpci1_priv { + struct leon_pci_info info; /* must be on top of this structure */ + struct grpci1_regs __iomem *regs; /* GRPCI register map */ + struct device *dev; + int pci_err_mask; /* STATUS register error mask */ + int irq; /* LEON irqctrl GRPCI IRQ */ + unsigned char irq_map[4]; /* GRPCI nexus PCI INTX# IRQs */ + unsigned int irq_err; /* GRPCI nexus Virt Error IRQ */ + + /* AHB PCI Windows */ + unsigned long pci_area; /* MEMORY */ + unsigned long pci_area_end; + unsigned long pci_io; /* I/O */ + unsigned long pci_conf; /* CONFIGURATION */ + unsigned long pci_conf_end; + unsigned long pci_io_va; +}; + +static struct grpci1_priv *grpci1priv; + +static int grpci1_cfg_w32(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val); + +static int grpci1_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) +{ + struct grpci1_priv *priv = dev->bus->sysdata; + int irq_group; + + /* Use default IRQ decoding on PCI BUS0 according slot numbering */ + irq_group = slot & 0x3; + pin = ((pin - 1) + irq_group) & 0x3; + + return priv->irq_map[pin]; +} + +static int grpci1_cfg_r32(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 *val) +{ + u32 *pci_conf, tmp, cfg; + + if (where & 0x3) + return -EINVAL; + + if (bus == 0) { + devfn += (0x8 * 6); /* start at AD16=Device0 */ + } else if (bus == TGT) { + bus = 0; + devfn = 0; /* special case: bridge controller itself */ + } + + /* Select bus */ + cfg = REGLOAD(priv->regs->cfg_stat); + REGSTORE(priv->regs->cfg_stat, (cfg & ~(0xf << 23)) | (bus << 23)); + + /* do read access */ + pci_conf = (u32 *) (priv->pci_conf | (devfn << 8) | (where & 0xfc)); + tmp = LEON3_BYPASS_LOAD_PA(pci_conf); + + /* check if master abort was received */ + if (REGLOAD(priv->regs->cfg_stat) & CFGSTAT_CTO) { + *val = 0xffffffff; + /* Clear Master abort bit in PCI cfg space (is set) */ + tmp = REGLOAD(priv->regs->stat_cmd); + grpci1_cfg_w32(priv, TGT, 0, PCI_COMMAND, tmp); + } else { + /* Bus always little endian (unaffected by byte-swapping) */ + *val = swab32(tmp); + } + + return 0; +} + +static int grpci1_cfg_r16(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 *val) +{ + u32 v; + int ret; + + if (where & 0x1) + return -EINVAL; + ret = grpci1_cfg_r32(priv, bus, devfn, where & ~0x3, &v); + *val = 0xffff & (v >> (8 * (where & 0x3))); + return ret; +} + +static int grpci1_cfg_r8(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 *val) +{ + u32 v; + int ret; + + ret = grpci1_cfg_r32(priv, bus, devfn, where & ~0x3, &v); + *val = 0xff & (v >> (8 * (where & 3))); + + return ret; +} + +static int grpci1_cfg_w32(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val) +{ + unsigned int *pci_conf; + u32 cfg; + + if (where & 0x3) + return -EINVAL; + + if (bus == 0) { + devfn += (0x8 * 6); /* start at AD16=Device0 */ + } else if (bus == TGT) { + bus = 0; + devfn = 0; /* special case: bridge controller itself */ + } + + /* Select bus */ + cfg = REGLOAD(priv->regs->cfg_stat); + REGSTORE(priv->regs->cfg_stat, (cfg & ~(0xf << 23)) | (bus << 23)); + + pci_conf = (unsigned int *) (priv->pci_conf | + (devfn << 8) | (where & 0xfc)); + LEON3_BYPASS_STORE_PA(pci_conf, swab32(val)); + + return 0; +} + +static int grpci1_cfg_w16(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val) +{ + int ret; + u32 v; + + if (where & 0x1) + return -EINVAL; + ret = grpci1_cfg_r32(priv, bus, devfn, where&~3, &v); + if (ret) + return ret; + v = (v & ~(0xffff << (8 * (where & 0x3)))) | + ((0xffff & val) << (8 * (where & 0x3))); + return grpci1_cfg_w32(priv, bus, devfn, where & ~0x3, v); +} + +static int grpci1_cfg_w8(struct grpci1_priv *priv, unsigned int bus, + unsigned int devfn, int where, u32 val) +{ + int ret; + u32 v; + + ret = grpci1_cfg_r32(priv, bus, devfn, where & ~0x3, &v); + if (ret != 0) + return ret; + v = (v & ~(0xff << (8 * (where & 0x3)))) | + ((0xff & val) << (8 * (where & 0x3))); + return grpci1_cfg_w32(priv, bus, devfn, where & ~0x3, v); +} + +/* Read from Configuration Space. When entering here the PCI layer has taken + * the pci_lock spinlock and IRQ is off. + */ +static int grpci1_read_config(struct pci_bus *bus, unsigned int devfn, + int where, int size, u32 *val) +{ + struct grpci1_priv *priv = grpci1priv; + unsigned int busno = bus->number; + int ret; + + if (PCI_SLOT(devfn) > 15 || busno > 15) { + *val = ~0; + return 0; + } + + switch (size) { + case 1: + ret = grpci1_cfg_r8(priv, busno, devfn, where, val); + break; + case 2: + ret = grpci1_cfg_r16(priv, busno, devfn, where, val); + break; + case 4: + ret = grpci1_cfg_r32(priv, busno, devfn, where, val); + break; + default: + ret = -EINVAL; + break; + } + +#ifdef GRPCI1_DEBUG_CFGACCESS + printk(KERN_INFO + "grpci1_read_config: [%02x:%02x:%x] ofs=%d val=%x size=%d\n", + busno, PCI_SLOT(devfn), PCI_FUNC(devfn), where, *val, size); +#endif + + return ret; +} + +/* Write to Configuration Space. When entering here the PCI layer has taken + * the pci_lock spinlock and IRQ is off. + */ +static int grpci1_write_config(struct pci_bus *bus, unsigned int devfn, + int where, int size, u32 val) +{ + struct grpci1_priv *priv = grpci1priv; + unsigned int busno = bus->number; + + if (PCI_SLOT(devfn) > 15 || busno > 15) + return 0; + +#ifdef GRPCI1_DEBUG_CFGACCESS + printk(KERN_INFO + "grpci1_write_config: [%02x:%02x:%x] ofs=%d size=%d val=%x\n", + busno, PCI_SLOT(devfn), PCI_FUNC(devfn), where, size, val); +#endif + + switch (size) { + default: + return -EINVAL; + case 1: + return grpci1_cfg_w8(priv, busno, devfn, where, val); + case 2: + return grpci1_cfg_w16(priv, busno, devfn, where, val); + case 4: + return grpci1_cfg_w32(priv, busno, devfn, where, val); + } +} + +static struct pci_ops grpci1_ops = { + .read = grpci1_read_config, + .write = grpci1_write_config, +}; + +/* GENIRQ IRQ chip implementation for grpci1 irqmode=0..2. In configuration + * 3 where all PCI Interrupts has a separate IRQ on the system IRQ controller + * this is not needed and the standard IRQ controller can be used. + */ + +static void grpci1_mask_irq(struct irq_data *data) +{ + u32 irqidx; + struct grpci1_priv *priv = grpci1priv; + + irqidx = (u32)data->chip_data - 1; + if (irqidx > 3) /* only mask PCI interrupts here */ + return; + irqidx += IRQ_MASK_BIT; + + REGSTORE(priv->regs->irq, REGLOAD(priv->regs->irq) & ~(1 << irqidx)); +} + +static void grpci1_unmask_irq(struct irq_data *data) +{ + u32 irqidx; + struct grpci1_priv *priv = grpci1priv; + + irqidx = (u32)data->chip_data - 1; + if (irqidx > 3) /* only unmask PCI interrupts here */ + return; + irqidx += IRQ_MASK_BIT; + + REGSTORE(priv->regs->irq, REGLOAD(priv->regs->irq) | (1 << irqidx)); +} + +static unsigned int grpci1_startup_irq(struct irq_data *data) +{ + grpci1_unmask_irq(data); + return 0; +} + +static void grpci1_shutdown_irq(struct irq_data *data) +{ + grpci1_mask_irq(data); +} + +static struct irq_chip grpci1_irq = { + .name = "grpci1", + .irq_startup = grpci1_startup_irq, + .irq_shutdown = grpci1_shutdown_irq, + .irq_mask = grpci1_mask_irq, + .irq_unmask = grpci1_unmask_irq, +}; + +/* Handle one or multiple IRQs from the PCI core */ +static void grpci1_pci_flow_irq(unsigned int irq, struct irq_desc *desc) +{ + struct grpci1_priv *priv = grpci1priv; + int i, ack = 0; + unsigned int irqreg; + + irqreg = REGLOAD(priv->regs->irq); + irqreg = (irqreg >> IRQ_MASK_BIT) & irqreg; + + /* Error Interrupt? */ + if (irqreg & IRQ_ALL_ERRORS) { + generic_handle_irq(priv->irq_err); + ack = 1; + } + + /* PCI Interrupt? */ + if (irqreg & IRQ_INTX) { + /* Call respective PCI Interrupt handler */ + for (i = 0; i < 4; i++) { + if (irqreg & (1 << i)) + generic_handle_irq(priv->irq_map[i]); + } + ack = 1; + } + + /* + * Call "first level" IRQ chip end-of-irq handler. It will ACK LEON IRQ + * Controller, this must be done after IRQ sources have been handled to + * avoid double IRQ generation + */ + if (ack) + desc->irq_data.chip->irq_eoi(&desc->irq_data); +} + +/* Create a virtual IRQ */ +static unsigned int grpci1_build_device_irq(unsigned int irq) +{ + unsigned int virq = 0, pil; + + pil = 1 << 8; + virq = irq_alloc(irq, pil); + if (virq == 0) + goto out; + + irq_set_chip_and_handler_name(virq, &grpci1_irq, handle_simple_irq, + "pcilvl"); + irq_set_chip_data(virq, (void *)irq); + +out: + return virq; +} + +/* + * Initialize mappings AMBA<->PCI, clear IRQ state, setup PCI interface + * + * Target BARs: + * BAR0: unused in this implementation + * BAR1: peripheral DMA to host's memory (size at least 256MByte) + * BAR2..BAR5: not implemented in hardware + */ +static void grpci1_hw_init(struct grpci1_priv *priv) +{ + u32 ahbadr, bar_sz, data, pciadr; + struct grpci1_regs __iomem *regs = priv->regs; + + /* set 1:1 mapping between AHB -> PCI memory space */ + REGSTORE(regs->cfg_stat, priv->pci_area & 0xf0000000); + + /* map PCI accesses to target BAR1 to Linux kernel memory 1:1 */ + ahbadr = 0xf0000000 & (u32)__pa(PAGE_ALIGN((unsigned long) &_end)); + REGSTORE(regs->page1, ahbadr); + + /* translate I/O accesses to 0, I/O Space always @ PCI low 64Kbytes */ + REGSTORE(regs->iomap, REGLOAD(regs->iomap) & 0x0000ffff); + + /* disable and clear pending interrupts */ + REGSTORE(regs->irq, 0); + + /* Setup BAR0 outside access range so that it does not conflict with + * peripheral DMA. There is no need to set up the PAGE0 register. + */ + grpci1_cfg_w32(priv, TGT, 0, PCI_BASE_ADDRESS_0, 0xffffffff); + grpci1_cfg_r32(priv, TGT, 0, PCI_BASE_ADDRESS_0, &bar_sz); + bar_sz = ~bar_sz + 1; + pciadr = priv->pci_area - bar_sz; + grpci1_cfg_w32(priv, TGT, 0, PCI_BASE_ADDRESS_0, pciadr); + + /* + * Setup the Host's PCI Target BAR1 for other peripherals to access, + * and do DMA to the host's memory. + */ + grpci1_cfg_w32(priv, TGT, 0, PCI_BASE_ADDRESS_1, ahbadr); + + /* + * Setup Latency Timer and cache line size. Default cache line + * size will result in poor performance (256 word fetches), 0xff + * will set it according to the max size of the PCI FIFO. + */ + grpci1_cfg_w8(priv, TGT, 0, PCI_CACHE_LINE_SIZE, 0xff); + grpci1_cfg_w8(priv, TGT, 0, PCI_LATENCY_TIMER, 0x40); + + /* set as bus master, enable pci memory responses, clear status bits */ + grpci1_cfg_r32(priv, TGT, 0, PCI_COMMAND, &data); + data |= (PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER); + grpci1_cfg_w32(priv, TGT, 0, PCI_COMMAND, data); +} + +static irqreturn_t grpci1_jump_interrupt(int irq, void *arg) +{ + struct grpci1_priv *priv = arg; + dev_err(priv->dev, "Jump IRQ happened\n"); + return IRQ_NONE; +} + +/* Handle GRPCI1 Error Interrupt */ +static irqreturn_t grpci1_err_interrupt(int irq, void *arg) +{ + struct grpci1_priv *priv = arg; + u32 status; + + grpci1_cfg_r16(priv, TGT, 0, PCI_STATUS, &status); + status &= priv->pci_err_mask; + + if (status == 0) + return IRQ_NONE; + + if (status & PCI_STATUS_PARITY) + dev_err(priv->dev, "Data Parity Error\n"); + + if (status & PCI_STATUS_SIG_TARGET_ABORT) + dev_err(priv->dev, "Signalled Target Abort\n"); + + if (status & PCI_STATUS_REC_TARGET_ABORT) + dev_err(priv->dev, "Received Target Abort\n"); + + if (status & PCI_STATUS_REC_MASTER_ABORT) + dev_err(priv->dev, "Received Master Abort\n"); + + if (status & PCI_STATUS_SIG_SYSTEM_ERROR) + dev_err(priv->dev, "Signalled System Error\n"); + + if (status & PCI_STATUS_DETECTED_PARITY) + dev_err(priv->dev, "Parity Error\n"); + + /* Clear handled INT TYPE IRQs */ + grpci1_cfg_w16(priv, TGT, 0, PCI_STATUS, status); + + return IRQ_HANDLED; +} + +static int grpci1_of_probe(struct platform_device *ofdev) +{ + struct grpci1_regs __iomem *regs; + struct grpci1_priv *priv; + int err, len; + const int *tmp; + u32 cfg, size, err_mask; + struct resource *res; + + if (grpci1priv) { + dev_err(&ofdev->dev, "only one GRPCI1 supported\n"); + return -ENODEV; + } + + if (ofdev->num_resources < 3) { + dev_err(&ofdev->dev, "not enough APB/AHB resources\n"); + return -EIO; + } + + priv = devm_kzalloc(&ofdev->dev, sizeof(*priv), GFP_KERNEL); + if (!priv) { + dev_err(&ofdev->dev, "memory allocation failed\n"); + return -ENOMEM; + } + platform_set_drvdata(ofdev, priv); + priv->dev = &ofdev->dev; + + /* find device register base address */ + res = platform_get_resource(ofdev, IORESOURCE_MEM, 0); + regs = devm_ioremap_resource(&ofdev->dev, res); + if (IS_ERR(regs)) + return PTR_ERR(regs); + + /* + * check that we're in Host Slot and that we can act as a Host Bridge + * and not only as target/peripheral. + */ + cfg = REGLOAD(regs->cfg_stat); + if ((cfg & CFGSTAT_HOST) == 0) { + dev_err(&ofdev->dev, "not in host system slot\n"); + return -EIO; + } + + /* check that BAR1 support 256 MByte so that we can map kernel space */ + REGSTORE(regs->page1, 0xffffffff); + size = ~REGLOAD(regs->page1) + 1; + if (size < 0x10000000) { + dev_err(&ofdev->dev, "BAR1 must be at least 256MByte\n"); + return -EIO; + } + + /* hardware must support little-endian PCI (byte-twisting) */ + if ((REGLOAD(regs->page0) & PAGE0_BTEN) == 0) { + dev_err(&ofdev->dev, "byte-twisting is required\n"); + return -EIO; + } + + priv->regs = regs; + priv->irq = irq_of_parse_and_map(ofdev->dev.of_node, 0); + dev_info(&ofdev->dev, "host found at 0x%p, irq%d\n", regs, priv->irq); + + /* Find PCI Memory, I/O and Configuration Space Windows */ + priv->pci_area = ofdev->resource[1].start; + priv->pci_area_end = ofdev->resource[1].end+1; + priv->pci_io = ofdev->resource[2].start; + priv->pci_conf = ofdev->resource[2].start + 0x10000; + priv->pci_conf_end = priv->pci_conf + 0x10000; + priv->pci_io_va = (unsigned long)ioremap(priv->pci_io, 0x10000); + if (!priv->pci_io_va) { + dev_err(&ofdev->dev, "unable to map PCI I/O area\n"); + return -EIO; + } + + printk(KERN_INFO + "GRPCI1: MEMORY SPACE [0x%08lx - 0x%08lx]\n" + " I/O SPACE [0x%08lx - 0x%08lx]\n" + " CONFIG SPACE [0x%08lx - 0x%08lx]\n", + priv->pci_area, priv->pci_area_end-1, + priv->pci_io, priv->pci_conf-1, + priv->pci_conf, priv->pci_conf_end-1); + + /* + * I/O Space resources in I/O Window mapped into Virtual Adr Space + * We never use low 4KB because some devices seem have problems using + * address 0. + */ + priv->info.io_space.name = "GRPCI1 PCI I/O Space"; + priv->info.io_space.start = priv->pci_io_va + 0x1000; + priv->info.io_space.end = priv->pci_io_va + 0x10000 - 1; + priv->info.io_space.flags = IORESOURCE_IO; + + /* + * grpci1 has no prefetchable memory, map everything as + * non-prefetchable memory + */ + priv->info.mem_space.name = "GRPCI1 PCI MEM Space"; + priv->info.mem_space.start = priv->pci_area; + priv->info.mem_space.end = priv->pci_area_end - 1; + priv->info.mem_space.flags = IORESOURCE_MEM; + + if (request_resource(&iomem_resource, &priv->info.mem_space) < 0) { + dev_err(&ofdev->dev, "unable to request PCI memory area\n"); + err = -ENOMEM; + goto err1; + } + + if (request_resource(&ioport_resource, &priv->info.io_space) < 0) { + dev_err(&ofdev->dev, "unable to request PCI I/O area\n"); + err = -ENOMEM; + goto err2; + } + + /* setup maximum supported PCI buses */ + priv->info.busn.name = "GRPCI1 busn"; + priv->info.busn.start = 0; + priv->info.busn.end = 15; + + grpci1priv = priv; + + /* Initialize hardware */ + grpci1_hw_init(priv); + + /* + * Get PCI Interrupt to System IRQ mapping and setup IRQ handling + * Error IRQ. All PCI and PCI-Error interrupts are shared using the + * same system IRQ. + */ + leon_update_virq_handling(priv->irq, grpci1_pci_flow_irq, "pcilvl", 0); + + priv->irq_map[0] = grpci1_build_device_irq(1); + priv->irq_map[1] = grpci1_build_device_irq(2); + priv->irq_map[2] = grpci1_build_device_irq(3); + priv->irq_map[3] = grpci1_build_device_irq(4); + priv->irq_err = grpci1_build_device_irq(5); + + printk(KERN_INFO " PCI INTA..D#: IRQ%d, IRQ%d, IRQ%d, IRQ%d\n", + priv->irq_map[0], priv->irq_map[1], priv->irq_map[2], + priv->irq_map[3]); + + /* Enable IRQs on LEON IRQ controller */ + err = devm_request_irq(&ofdev->dev, priv->irq, grpci1_jump_interrupt, 0, + "GRPCI1_JUMP", priv); + if (err) { + dev_err(&ofdev->dev, "ERR IRQ request failed: %d\n", err); + goto err3; + } + + /* Setup IRQ handler for access errors */ + err = devm_request_irq(&ofdev->dev, priv->irq_err, + grpci1_err_interrupt, IRQF_SHARED, "GRPCI1_ERR", + priv); + if (err) { + dev_err(&ofdev->dev, "ERR VIRQ request failed: %d\n", err); + goto err3; + } + + tmp = of_get_property(ofdev->dev.of_node, "all_pci_errors", &len); + if (tmp && (len == 4)) { + priv->pci_err_mask = ALL_PCI_ERRORS; + err_mask = IRQ_ALL_ERRORS << IRQ_MASK_BIT; + } else { + priv->pci_err_mask = DEF_PCI_ERRORS; + err_mask = IRQ_DEF_ERRORS << IRQ_MASK_BIT; + } + + /* + * Enable Error Interrupts. PCI interrupts are unmasked once request_irq + * is called by the PCI Device drivers + */ + REGSTORE(regs->irq, err_mask); + + /* Init common layer and scan buses */ + priv->info.ops = &grpci1_ops; + priv->info.map_irq = grpci1_map_irq; + leon_pci_init(ofdev, &priv->info); + + return 0; + +err3: + release_resource(&priv->info.io_space); +err2: + release_resource(&priv->info.mem_space); +err1: + iounmap((void __iomem *)priv->pci_io_va); + grpci1priv = NULL; + return err; +} + +static struct of_device_id grpci1_of_match[] = { + { + .name = "GAISLER_PCIFBRG", + }, + { + .name = "01_014", + }, + {}, +}; + +static struct platform_driver grpci1_of_driver = { + .driver = { + .name = "grpci1", + .owner = THIS_MODULE, + .of_match_table = grpci1_of_match, + }, + .probe = grpci1_of_probe, +}; + +static int __init grpci1_init(void) +{ + return platform_driver_register(&grpci1_of_driver); +} + +subsys_initcall(grpci1_init); diff --git a/arch/sparc/kernel/leon_pci_grpci2.c b/arch/sparc/kernel/leon_pci_grpci2.c index b1bc38853a3..e433a4d69fe 100644 --- a/arch/sparc/kernel/leon_pci_grpci2.c +++ b/arch/sparc/kernel/leon_pci_grpci2.c @@ -8,6 +8,7 @@ #include <linux/of_device.h> #include <linux/kernel.h> #include <linux/pci.h> +#include <linux/slab.h> #include <linux/delay.h> #include <linux/export.h> #include <asm/io.h> @@ -186,9 +187,11 @@ struct grpci2_cap_first { #define CAP9_IOMAP_OFS 0x20 #define CAP9_BARSIZE_OFS 0x24 +#define TGT 256 + struct grpci2_priv { struct leon_pci_info info; /* must be on top of this structure */ - struct grpci2_regs *regs; + struct grpci2_regs __iomem *regs; char irq; char irq_mode; /* IRQ Mode from CAPSTS REG */ char bt_enabled; @@ -212,10 +215,10 @@ struct grpci2_priv { struct grpci2_barcfg tgtbars[6]; }; -DEFINE_SPINLOCK(grpci2_dev_lock); -struct grpci2_priv *grpci2priv; +static DEFINE_SPINLOCK(grpci2_dev_lock); +static struct grpci2_priv *grpci2priv; -int grpci2_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) +static int grpci2_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) { struct grpci2_priv *priv = dev->bus->sysdata; int irq_group; @@ -237,8 +240,12 @@ static int grpci2_cfg_r32(struct grpci2_priv *priv, unsigned int bus, if (where & 0x3) return -EINVAL; - if (bus == 0 && PCI_SLOT(devfn) != 0) - devfn += (0x8 * 6); + if (bus == 0) { + devfn += (0x8 * 6); /* start at AD16=Device0 */ + } else if (bus == TGT) { + bus = 0; + devfn = 0; /* special case: bridge controller itself */ + } /* Select bus */ spin_lock_irqsave(&grpci2_dev_lock, flags); @@ -263,7 +270,7 @@ static int grpci2_cfg_r32(struct grpci2_priv *priv, unsigned int bus, *val = 0xffffffff; } else { /* Bus always little endian (unaffected by byte-swapping) */ - *val = flip_dword(tmp); + *val = swab32(tmp); } return 0; @@ -303,8 +310,12 @@ static int grpci2_cfg_w32(struct grpci2_priv *priv, unsigned int bus, if (where & 0x3) return -EINVAL; - if (bus == 0 && PCI_SLOT(devfn) != 0) - devfn += (0x8 * 6); + if (bus == 0) { + devfn += (0x8 * 6); /* start at AD16=Device0 */ + } else if (bus == TGT) { + bus = 0; + devfn = 0; /* special case: bridge controller itself */ + } /* Select bus */ spin_lock_irqsave(&grpci2_dev_lock, flags); @@ -317,7 +328,7 @@ static int grpci2_cfg_w32(struct grpci2_priv *priv, unsigned int bus, pci_conf = (unsigned int *) (priv->pci_conf | (devfn << 8) | (where & 0xfc)); - LEON3_BYPASS_STORE_PA(pci_conf, flip_dword(val)); + LEON3_BYPASS_STORE_PA(pci_conf, swab32(val)); /* Wait until GRPCI2 signals that CFG access is done, it should be * done instantaneously unless a DMA operation is ongoing... @@ -368,7 +379,7 @@ static int grpci2_read_config(struct pci_bus *bus, unsigned int devfn, unsigned int busno = bus->number; int ret; - if (PCI_SLOT(devfn) > 15 || (PCI_SLOT(devfn) == 0 && busno == 0)) { + if (PCI_SLOT(devfn) > 15 || busno > 255) { *val = ~0; return 0; } @@ -406,7 +417,7 @@ static int grpci2_write_config(struct pci_bus *bus, unsigned int devfn, struct grpci2_priv *priv = grpci2priv; unsigned int busno = bus->number; - if (PCI_SLOT(devfn) > 15 || (PCI_SLOT(devfn) == 0 && busno == 0)) + if (PCI_SLOT(devfn) > 15 || busno > 255) return 0; #ifdef GRPCI2_DEBUG_CFGACCESS @@ -550,10 +561,10 @@ out: return virq; } -void grpci2_hw_init(struct grpci2_priv *priv) +static void grpci2_hw_init(struct grpci2_priv *priv) { u32 ahbadr, pciadr, bar_sz, capptr, io_map, data; - struct grpci2_regs *regs = priv->regs; + struct grpci2_regs __iomem *regs = priv->regs; int i; struct grpci2_barcfg *barcfg = priv->tgtbars; @@ -578,15 +589,15 @@ void grpci2_hw_init(struct grpci2_priv *priv) REGSTORE(regs->ahbmst_map[i], priv->pci_area); /* Get the GRPCI2 Host PCI ID */ - grpci2_cfg_r32(priv, 0, 0, PCI_VENDOR_ID, &priv->pciid); + grpci2_cfg_r32(priv, TGT, 0, PCI_VENDOR_ID, &priv->pciid); /* Get address to first (always defined) capability structure */ - grpci2_cfg_r8(priv, 0, 0, PCI_CAPABILITY_LIST, &capptr); + grpci2_cfg_r8(priv, TGT, 0, PCI_CAPABILITY_LIST, &capptr); /* Enable/Disable Byte twisting */ - grpci2_cfg_r32(priv, 0, 0, capptr+CAP9_IOMAP_OFS, &io_map); + grpci2_cfg_r32(priv, TGT, 0, capptr+CAP9_IOMAP_OFS, &io_map); io_map = (io_map & ~0x1) | (priv->bt_enabled ? 1 : 0); - grpci2_cfg_w32(priv, 0, 0, capptr+CAP9_IOMAP_OFS, io_map); + grpci2_cfg_w32(priv, TGT, 0, capptr+CAP9_IOMAP_OFS, io_map); /* Setup the Host's PCI Target BARs for other peripherals to access, * and do DMA to the host's memory. The target BARs can be sized and @@ -617,17 +628,18 @@ void grpci2_hw_init(struct grpci2_priv *priv) pciadr = 0; } } - grpci2_cfg_w32(priv, 0, 0, capptr+CAP9_BARSIZE_OFS+i*4, bar_sz); - grpci2_cfg_w32(priv, 0, 0, PCI_BASE_ADDRESS_0+i*4, pciadr); - grpci2_cfg_w32(priv, 0, 0, capptr+CAP9_BAR_OFS+i*4, ahbadr); + grpci2_cfg_w32(priv, TGT, 0, capptr+CAP9_BARSIZE_OFS+i*4, + bar_sz); + grpci2_cfg_w32(priv, TGT, 0, PCI_BASE_ADDRESS_0+i*4, pciadr); + grpci2_cfg_w32(priv, TGT, 0, capptr+CAP9_BAR_OFS+i*4, ahbadr); printk(KERN_INFO " TGT BAR[%d]: 0x%08x (PCI)-> 0x%08x\n", i, pciadr, ahbadr); } /* set as bus master and enable pci memory responses */ - grpci2_cfg_r32(priv, 0, 0, PCI_COMMAND, &data); + grpci2_cfg_r32(priv, TGT, 0, PCI_COMMAND, &data); data |= (PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER); - grpci2_cfg_w32(priv, 0, 0, PCI_COMMAND, data); + grpci2_cfg_w32(priv, TGT, 0, PCI_COMMAND, data); /* Enable Error respone (CPU-TRAP) on illegal memory access. */ REGSTORE(regs->ctrl, CTRL_ER | CTRL_PE); @@ -643,7 +655,7 @@ static irqreturn_t grpci2_jump_interrupt(int irq, void *arg) static irqreturn_t grpci2_err_interrupt(int irq, void *arg) { struct grpci2_priv *priv = arg; - struct grpci2_regs *regs = priv->regs; + struct grpci2_regs __iomem *regs = priv->regs; unsigned int status; status = REGLOAD(regs->sts_cap); @@ -668,9 +680,9 @@ static irqreturn_t grpci2_err_interrupt(int irq, void *arg) return IRQ_HANDLED; } -static int __devinit grpci2_of_probe(struct platform_device *ofdev) +static int grpci2_of_probe(struct platform_device *ofdev) { - struct grpci2_regs *regs; + struct grpci2_regs __iomem *regs; struct grpci2_priv *priv; int err, i, len; const int *tmp; @@ -788,6 +800,11 @@ static int __devinit grpci2_of_probe(struct platform_device *ofdev) if (request_resource(&ioport_resource, &priv->info.io_space) < 0) goto err4; + /* setup maximum supported PCI buses */ + priv->info.busn.name = "GRPCI2 busn"; + priv->info.busn.start = 0; + priv->info.busn.end = 255; + grpci2_hw_init(priv); /* @@ -861,7 +878,7 @@ err4: release_resource(&priv->info.mem_space); err3: err = -ENOMEM; - iounmap((void *)priv->pci_io_va); + iounmap((void __iomem *)priv->pci_io_va); err2: kfree(priv); err1: diff --git a/arch/sparc/kernel/leon_pmc.c b/arch/sparc/kernel/leon_pmc.c index 4e174321097..ddcf950282e 100644 --- a/arch/sparc/kernel/leon_pmc.c +++ b/arch/sparc/kernel/leon_pmc.c @@ -9,16 +9,17 @@ #include <asm/leon_amba.h> #include <asm/cpu_type.h> #include <asm/leon.h> +#include <asm/processor.h> /* List of Systems that need fixup instructions around power-down instruction */ -unsigned int pmc_leon_fixup_ids[] = { +static unsigned int pmc_leon_fixup_ids[] = { AEROFLEX_UT699, GAISLER_GR712RC, LEON4_NEXTREME1, 0 }; -int pmc_leon_need_fixup(void) +static int pmc_leon_need_fixup(void) { unsigned int systemid = amba_system_id >> 16; unsigned int *id; @@ -37,7 +38,7 @@ int pmc_leon_need_fixup(void) * CPU idle callback function for systems that need some extra handling * See .../arch/sparc/kernel/process.c */ -void pmc_leon_idle_fixup(void) +static void pmc_leon_idle_fixup(void) { /* Prepare an address to a non-cachable region. APB is always * none-cachable. One instruction is executed after the Sleep @@ -46,8 +47,12 @@ void pmc_leon_idle_fixup(void) * MMU does not get a TLB miss here by using the MMU BYPASS ASI. */ register unsigned int address = (unsigned int)leon3_irqctrl_regs; + + /* Interrupts need to be enabled to not hang the CPU */ + local_irq_enable(); + __asm__ __volatile__ ( - "mov %%g0, %%asr19\n" + "wr %%g0, %%asr19\n" "lda [%0] %1, %%g0\n" : : "r"(address), "i"(ASI_LEON_BYPASS)); @@ -57,10 +62,13 @@ void pmc_leon_idle_fixup(void) * CPU idle callback function * See .../arch/sparc/kernel/process.c */ -void pmc_leon_idle(void) +static void pmc_leon_idle(void) { + /* Interrupts need to be enabled to not hang the CPU */ + local_irq_enable(); + /* For systems without power-down, this will be no-op */ - __asm__ __volatile__ ("mov %g0, %asr19\n\t"); + __asm__ __volatile__ ("wr %g0, %asr19\n\t"); } /* Install LEON Power Down function */ @@ -69,9 +77,9 @@ static int __init leon_pmc_install(void) if (sparc_cpu_model == sparc_leon) { /* Assign power management IDLE handler */ if (pmc_leon_need_fixup()) - pm_idle = pmc_leon_idle_fixup; + sparc_idle = pmc_leon_idle_fixup; else - pm_idle = pmc_leon_idle; + sparc_idle = pmc_leon_idle; printk(KERN_INFO "leon: power management initialized\n"); } diff --git a/arch/sparc/kernel/leon_smp.c b/arch/sparc/kernel/leon_smp.c index 0f3fb6d9c8e..018ef11f57d 100644 --- a/arch/sparc/kernel/leon_smp.c +++ b/arch/sparc/kernel/leon_smp.c @@ -54,7 +54,7 @@ extern ctxd_t *srmmu_ctx_table_phys; static int smp_processors_ready; extern volatile unsigned long cpu_callin_map[NR_CPUS]; extern cpumask_t smp_commenced_mask; -void __cpuinit leon_configure_cache_smp(void); +void leon_configure_cache_smp(void); static void leon_ipi_init(void); /* IRQ number of LEON IPIs */ @@ -69,31 +69,19 @@ static inline unsigned long do_swap(volatile unsigned long *ptr, return val; } -void __cpuinit leon_callin(void) +void leon_cpu_pre_starting(void *arg) { - int cpuid = hard_smp_processor_id(); - - local_ops->cache_all(); - local_ops->tlb_all(); leon_configure_cache_smp(); +} - notify_cpu_starting(cpuid); - - /* Get our local ticker going. */ - register_percpu_ce(cpuid); - - calibrate_delay(); - smp_store_cpu_info(cpuid); - - local_ops->cache_all(); - local_ops->tlb_all(); +void leon_cpu_pre_online(void *arg) +{ + int cpuid = hard_smp_processor_id(); - /* - * Unblock the master CPU _only_ when the scheduler state - * of all secondary CPUs will be up-to-date, so after - * the SMP initialization the master will be just allowed - * to call the scheduler code. - * Allow master to continue. + /* Allow master to continue. The master will then give us the + * go-ahead by setting the smp_commenced_mask and will wait without + * timeouts until our setup is completed fully (signified by + * our bit being set in the cpu_online_mask). */ do_swap(&cpu_callin_map[cpuid], 1); @@ -110,9 +98,6 @@ void __cpuinit leon_callin(void) while (!cpumask_test_cpu(cpuid, &smp_commenced_mask)) mb(); - - local_irq_enable(); - set_cpu_online(cpuid, true); } /* @@ -121,7 +106,7 @@ void __cpuinit leon_callin(void) extern struct linux_prom_registers smp_penguin_ctable; -void __cpuinit leon_configure_cache_smp(void) +void leon_configure_cache_smp(void) { unsigned long cfg = sparc_leon3_get_dcachecfg(); int me = smp_processor_id(); @@ -145,7 +130,7 @@ void __cpuinit leon_configure_cache_smp(void) local_ops->tlb_all(); } -void leon_smp_setbroadcast(unsigned int mask) +static void leon_smp_setbroadcast(unsigned int mask) { int broadcast = ((LEON3_BYPASS_LOAD_PA(&(leon3_irqctrl_regs->mpstatus)) >> @@ -163,13 +148,6 @@ void leon_smp_setbroadcast(unsigned int mask) LEON_BYPASS_STORE_PA(&(leon3_irqctrl_regs->mpbroadcast), mask); } -unsigned int leon_smp_getbroadcast(void) -{ - unsigned int mask; - mask = LEON_BYPASS_LOAD_PA(&(leon3_irqctrl_regs->mpbroadcast)); - return mask; -} - int leon_smp_nrcpus(void) { int nrcpu = @@ -201,7 +179,7 @@ void __init leon_boot_cpus(void) } -int __cpuinit leon_boot_one_cpu(int i, struct task_struct *idle) +int leon_boot_one_cpu(int i, struct task_struct *idle) { int timeout; @@ -268,35 +246,19 @@ void __init leon_smp_done(void) /* Free unneeded trap tables */ if (!cpu_present(1)) { - ClearPageReserved(virt_to_page(&trapbase_cpu1)); - init_page_count(virt_to_page(&trapbase_cpu1)); - free_page((unsigned long)&trapbase_cpu1); - totalram_pages++; - num_physpages++; + free_reserved_page(virt_to_page(&trapbase_cpu1)); } if (!cpu_present(2)) { - ClearPageReserved(virt_to_page(&trapbase_cpu2)); - init_page_count(virt_to_page(&trapbase_cpu2)); - free_page((unsigned long)&trapbase_cpu2); - totalram_pages++; - num_physpages++; + free_reserved_page(virt_to_page(&trapbase_cpu2)); } if (!cpu_present(3)) { - ClearPageReserved(virt_to_page(&trapbase_cpu3)); - init_page_count(virt_to_page(&trapbase_cpu3)); - free_page((unsigned long)&trapbase_cpu3); - totalram_pages++; - num_physpages++; + free_reserved_page(virt_to_page(&trapbase_cpu3)); } /* Ok, they are spinning and ready to go. */ smp_processors_ready = 1; } -void leon_irq_rotate(int cpu) -{ -} - struct leon_ipi_work { int single; int msk; diff --git a/arch/sparc/kernel/mdesc.c b/arch/sparc/kernel/mdesc.c index 6dc79628058..a1a4400d402 100644 --- a/arch/sparc/kernel/mdesc.c +++ b/arch/sparc/kernel/mdesc.c @@ -571,9 +571,7 @@ static void __init report_platform_properties(void) mdesc_release(hp); } -static void __cpuinit fill_in_one_cache(cpuinfo_sparc *c, - struct mdesc_handle *hp, - u64 mp) +static void fill_in_one_cache(cpuinfo_sparc *c, struct mdesc_handle *hp, u64 mp) { const u64 *level = mdesc_get_property(hp, mp, "level", NULL); const u64 *size = mdesc_get_property(hp, mp, "size", NULL); @@ -616,7 +614,7 @@ static void __cpuinit fill_in_one_cache(cpuinfo_sparc *c, } } -static void __cpuinit mark_core_ids(struct mdesc_handle *hp, u64 mp, int core_id) +static void mark_core_ids(struct mdesc_handle *hp, u64 mp, int core_id) { u64 a; @@ -649,7 +647,7 @@ static void __cpuinit mark_core_ids(struct mdesc_handle *hp, u64 mp, int core_id } } -static void __cpuinit set_core_ids(struct mdesc_handle *hp) +static void set_core_ids(struct mdesc_handle *hp) { int idx; u64 mp; @@ -674,7 +672,7 @@ static void __cpuinit set_core_ids(struct mdesc_handle *hp) } } -static void __cpuinit mark_proc_ids(struct mdesc_handle *hp, u64 mp, int proc_id) +static void mark_proc_ids(struct mdesc_handle *hp, u64 mp, int proc_id) { u64 a; @@ -693,7 +691,7 @@ static void __cpuinit mark_proc_ids(struct mdesc_handle *hp, u64 mp, int proc_id } } -static void __cpuinit __set_proc_ids(struct mdesc_handle *hp, const char *exec_unit_name) +static void __set_proc_ids(struct mdesc_handle *hp, const char *exec_unit_name) { int idx; u64 mp; @@ -714,14 +712,14 @@ static void __cpuinit __set_proc_ids(struct mdesc_handle *hp, const char *exec_u } } -static void __cpuinit set_proc_ids(struct mdesc_handle *hp) +static void set_proc_ids(struct mdesc_handle *hp) { __set_proc_ids(hp, "exec_unit"); __set_proc_ids(hp, "exec-unit"); } -static void __cpuinit get_one_mondo_bits(const u64 *p, unsigned int *mask, - unsigned long def, unsigned long max) +static void get_one_mondo_bits(const u64 *p, unsigned int *mask, + unsigned long def, unsigned long max) { u64 val; @@ -742,8 +740,8 @@ use_default: *mask = ((1U << def) * 64U) - 1U; } -static void __cpuinit get_mondo_data(struct mdesc_handle *hp, u64 mp, - struct trap_per_cpu *tb) +static void get_mondo_data(struct mdesc_handle *hp, u64 mp, + struct trap_per_cpu *tb) { static int printed; const u64 *val; @@ -769,7 +767,7 @@ static void __cpuinit get_mondo_data(struct mdesc_handle *hp, u64 mp, } } -static void * __cpuinit mdesc_iterate_over_cpus(void *(*func)(struct mdesc_handle *, u64, int, void *), void *arg, cpumask_t *mask) +static void *mdesc_iterate_over_cpus(void *(*func)(struct mdesc_handle *, u64, int, void *), void *arg, cpumask_t *mask) { struct mdesc_handle *hp = mdesc_grab(); void *ret = NULL; @@ -799,7 +797,8 @@ out: return ret; } -static void * __cpuinit record_one_cpu(struct mdesc_handle *hp, u64 mp, int cpuid, void *arg) +static void *record_one_cpu(struct mdesc_handle *hp, u64 mp, int cpuid, + void *arg) { ncpus_probed++; #ifdef CONFIG_SMP @@ -808,7 +807,7 @@ static void * __cpuinit record_one_cpu(struct mdesc_handle *hp, u64 mp, int cpui return NULL; } -void __cpuinit mdesc_populate_present_mask(cpumask_t *mask) +void mdesc_populate_present_mask(cpumask_t *mask) { if (tlb_type != hypervisor) return; @@ -817,7 +816,32 @@ void __cpuinit mdesc_populate_present_mask(cpumask_t *mask) mdesc_iterate_over_cpus(record_one_cpu, NULL, mask); } -static void * __cpuinit fill_in_one_cpu(struct mdesc_handle *hp, u64 mp, int cpuid, void *arg) +static void * __init check_one_pgsz(struct mdesc_handle *hp, u64 mp, int cpuid, void *arg) +{ + const u64 *pgsz_prop = mdesc_get_property(hp, mp, "mmu-page-size-list", NULL); + unsigned long *pgsz_mask = arg; + u64 val; + + val = (HV_PGSZ_MASK_8K | HV_PGSZ_MASK_64K | + HV_PGSZ_MASK_512K | HV_PGSZ_MASK_4MB); + if (pgsz_prop) + val = *pgsz_prop; + + if (!*pgsz_mask) + *pgsz_mask = val; + else + *pgsz_mask &= val; + return NULL; +} + +void __init mdesc_get_page_sizes(cpumask_t *mask, unsigned long *pgsz_mask) +{ + *pgsz_mask = 0; + mdesc_iterate_over_cpus(check_one_pgsz, pgsz_mask, mask); +} + +static void *fill_in_one_cpu(struct mdesc_handle *hp, u64 mp, int cpuid, + void *arg) { const u64 *cfreq = mdesc_get_property(hp, mp, "clock-frequency", NULL); struct trap_per_cpu *tb; @@ -866,16 +890,12 @@ static void * __cpuinit fill_in_one_cpu(struct mdesc_handle *hp, u64 mp, int cpu return NULL; } -void __cpuinit mdesc_fill_in_cpu_data(cpumask_t *mask) +void mdesc_fill_in_cpu_data(cpumask_t *mask) { struct mdesc_handle *hp; mdesc_iterate_over_cpus(fill_in_one_cpu, NULL, mask); -#ifdef CONFIG_SMP - sparc64_multi_core = 1; -#endif - hp = mdesc_grab(); set_core_ids(hp); diff --git a/arch/sparc/kernel/module.c b/arch/sparc/kernel/module.c index 15e0a169397..97655e0fd24 100644 --- a/arch/sparc/kernel/module.c +++ b/arch/sparc/kernel/module.c @@ -29,7 +29,7 @@ static void *module_map(unsigned long size) if (PAGE_ALIGN(size) > MODULES_LEN) return NULL; return __vmalloc_node_range(size, 1, MODULES_VADDR, MODULES_END, - GFP_KERNEL, PAGE_KERNEL, -1, + GFP_KERNEL, PAGE_KERNEL, NUMA_NO_NODE, __builtin_return_address(0)); } #else @@ -43,14 +43,8 @@ void *module_alloc(unsigned long size) { void *ret; - /* We handle the zero case fine, unlike vmalloc */ - if (size == 0) - return NULL; - ret = module_map(size); - if (!ret) - ret = ERR_PTR(-ENOMEM); - else + if (ret) memset(ret, 0, size); return ret; @@ -116,6 +110,10 @@ int apply_relocate_add(Elf_Shdr *sechdrs, v = sym->st_value + rel[i].r_addend; switch (ELF_R_TYPE(rel[i].r_info) & 0xff) { + case R_SPARC_DISP32: + v -= (Elf_Addr) location; + *loc32 = v; + break; #ifdef CONFIG_SPARC64 case R_SPARC_64: location[0] = v >> 56; @@ -128,11 +126,6 @@ int apply_relocate_add(Elf_Shdr *sechdrs, location[7] = v >> 0; break; - case R_SPARC_DISP32: - v -= (Elf_Addr) location; - *loc32 = v; - break; - case R_SPARC_WDISP19: v -= (Elf_Addr) location; *loc32 = (*loc32 & ~0x7ffff) | diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index eb1c1f010a4..33709455691 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -22,7 +22,6 @@ #include <asm/perf_event.h> #include <asm/ptrace.h> #include <asm/pcr.h> -#include <asm/perfctr.h> #include "kstack.h" @@ -69,27 +68,16 @@ EXPORT_SYMBOL(touch_nmi_watchdog); static void die_nmi(const char *str, struct pt_regs *regs, int do_panic) { + int this_cpu = smp_processor_id(); + if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP) return; - console_verbose(); - bust_spinlocks(1); - - printk(KERN_EMERG "%s", str); - printk(" on CPU%d, ip %08lx, registers:\n", - smp_processor_id(), regs->tpc); - show_regs(regs); - dump_stack(); - - bust_spinlocks(0); - if (do_panic || panic_on_oops) - panic("Non maskable interrupt"); - - nmi_exit(); - local_irq_enable(); - do_exit(SIGBUS); + panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu); + else + WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu); } notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) @@ -109,7 +97,7 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP) touched = 1; else - pcr_ops->write(PCR_PIC_PRIV); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); sum = local_cpu_data().irq0_irqs; if (__get_cpu_var(nmi_touch)) { @@ -126,8 +114,8 @@ notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) __this_cpu_write(alert_counter, 0); } if (__get_cpu_var(wd_enabled)) { - write_pic(picl_value(nmi_hz)); - pcr_ops->write(pcr_enable); + pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz)); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable); } restore_hardirq_stack(orig_sp); @@ -166,7 +154,7 @@ static void report_broken_nmi(int cpu, int *prev_nmi_count) void stop_nmi_watchdog(void *unused) { - pcr_ops->write(PCR_PIC_PRIV); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); __get_cpu_var(wd_enabled) = 0; atomic_dec(&nmi_active); } @@ -223,10 +211,10 @@ void start_nmi_watchdog(void *unused) __get_cpu_var(wd_enabled) = 1; atomic_inc(&nmi_active); - pcr_ops->write(PCR_PIC_PRIV); - write_pic(picl_value(nmi_hz)); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); + pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz)); - pcr_ops->write(pcr_enable); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable); } static void nmi_adjust_hz_one(void *unused) @@ -234,10 +222,10 @@ static void nmi_adjust_hz_one(void *unused) if (!__get_cpu_var(wd_enabled)) return; - pcr_ops->write(PCR_PIC_PRIV); - write_pic(picl_value(nmi_hz)); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_disable); + pcr_ops->write_pic(0, pcr_ops->nmi_picl_value(nmi_hz)); - pcr_ops->write(pcr_enable); + pcr_ops->write_pcr(0, pcr_ops->pcr_nmi_enable); } void nmi_adjust_hz(unsigned int new_hz) diff --git a/arch/sparc/kernel/of_device_64.c b/arch/sparc/kernel/of_device_64.c index 7a3be6f6737..7bbdc26d951 100644 --- a/arch/sparc/kernel/of_device_64.c +++ b/arch/sparc/kernel/of_device_64.c @@ -580,7 +580,7 @@ static unsigned int __init build_one_device_irq(struct platform_device *op, printk("%s: Apply [%s:%x] imap --> [%s:%x]\n", op->dev.of_node->full_name, pp->full_name, this_orig_irq, - (iret ? iret->full_name : "NULL"), irq); + of_node_full_name(iret), irq); if (!iret) break; diff --git a/arch/sparc/kernel/of_device_common.c b/arch/sparc/kernel/of_device_common.c index de199bf0cb0..de0ee3971f0 100644 --- a/arch/sparc/kernel/of_device_common.c +++ b/arch/sparc/kernel/of_device_common.c @@ -1,13 +1,14 @@ #include <linux/string.h> #include <linux/kernel.h> #include <linux/of.h> -#include <linux/init.h> #include <linux/export.h> #include <linux/mod_devicetable.h> #include <linux/errno.h> #include <linux/irq.h> -#include <linux/of_device.h> #include <linux/of_platform.h> +#include <linux/of_address.h> +#include <linux/of_device.h> +#include <linux/of_irq.h> #include "of_device_common.h" diff --git a/arch/sparc/kernel/pci.c b/arch/sparc/kernel/pci.c index fdaf2181167..539babf00bb 100644 --- a/arch/sparc/kernel/pci.c +++ b/arch/sparc/kernel/pci.c @@ -28,6 +28,7 @@ #include <asm/apb.h> #include "pci_impl.h" +#include "kernel.h" /* List of all PCI controllers found in the system. */ struct pci_pbm_info *pci_pbm_root = NULL; @@ -254,7 +255,7 @@ static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, const char *type; u32 class; - dev = alloc_pci_dev(); + dev = pci_alloc_dev(bus); if (!dev) return NULL; @@ -281,7 +282,6 @@ static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, printk(" create device, devfn: %x, type: %s\n", devfn, type); - dev->bus = bus; dev->sysdata = node; dev->dev.parent = bus->bridge; dev->dev.bus = &pci_bus_type; @@ -327,7 +327,7 @@ static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, if ((dev->class >> 8) == PCI_CLASS_STORAGE_IDE) pci_set_master(dev); - dev->current_state = 4; /* unknown power state */ + dev->current_state = PCI_UNKNOWN; /* unknown power state */ dev->error_state = pci_channel_io_normal; dev->dma_mask = 0xffffffff; @@ -356,7 +356,7 @@ static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm, return dev; } -static void __devinit apb_calc_first_last(u8 map, u32 *first_p, u32 *last_p) +static void apb_calc_first_last(u8 map, u32 *first_p, u32 *last_p) { u32 idx, first, last; @@ -375,99 +375,12 @@ static void __devinit apb_calc_first_last(u8 map, u32 *first_p, u32 *last_p) *last_p = last; } -/* For PCI bus devices which lack a 'ranges' property we interrogate - * the config space values to set the resources, just like the generic - * Linux PCI probing code does. - */ -static void __devinit pci_cfg_fake_ranges(struct pci_dev *dev, - struct pci_bus *bus, - struct pci_pbm_info *pbm) -{ - struct pci_bus_region region; - struct resource *res, res2; - u8 io_base_lo, io_limit_lo; - u16 mem_base_lo, mem_limit_lo; - unsigned long base, limit; - - pci_read_config_byte(dev, PCI_IO_BASE, &io_base_lo); - pci_read_config_byte(dev, PCI_IO_LIMIT, &io_limit_lo); - base = (io_base_lo & PCI_IO_RANGE_MASK) << 8; - limit = (io_limit_lo & PCI_IO_RANGE_MASK) << 8; - - if ((io_base_lo & PCI_IO_RANGE_TYPE_MASK) == PCI_IO_RANGE_TYPE_32) { - u16 io_base_hi, io_limit_hi; - - pci_read_config_word(dev, PCI_IO_BASE_UPPER16, &io_base_hi); - pci_read_config_word(dev, PCI_IO_LIMIT_UPPER16, &io_limit_hi); - base |= (io_base_hi << 16); - limit |= (io_limit_hi << 16); - } - - res = bus->resource[0]; - if (base <= limit) { - res->flags = (io_base_lo & PCI_IO_RANGE_TYPE_MASK) | IORESOURCE_IO; - res2.flags = res->flags; - region.start = base; - region.end = limit + 0xfff; - pcibios_bus_to_resource(dev, &res2, ®ion); - if (!res->start) - res->start = res2.start; - if (!res->end) - res->end = res2.end; - } - - pci_read_config_word(dev, PCI_MEMORY_BASE, &mem_base_lo); - pci_read_config_word(dev, PCI_MEMORY_LIMIT, &mem_limit_lo); - base = (mem_base_lo & PCI_MEMORY_RANGE_MASK) << 16; - limit = (mem_limit_lo & PCI_MEMORY_RANGE_MASK) << 16; - - res = bus->resource[1]; - if (base <= limit) { - res->flags = ((mem_base_lo & PCI_MEMORY_RANGE_TYPE_MASK) | - IORESOURCE_MEM); - region.start = base; - region.end = limit + 0xfffff; - pcibios_bus_to_resource(dev, res, ®ion); - } - - pci_read_config_word(dev, PCI_PREF_MEMORY_BASE, &mem_base_lo); - pci_read_config_word(dev, PCI_PREF_MEMORY_LIMIT, &mem_limit_lo); - base = (mem_base_lo & PCI_PREF_RANGE_MASK) << 16; - limit = (mem_limit_lo & PCI_PREF_RANGE_MASK) << 16; - - if ((mem_base_lo & PCI_PREF_RANGE_TYPE_MASK) == PCI_PREF_RANGE_TYPE_64) { - u32 mem_base_hi, mem_limit_hi; - - pci_read_config_dword(dev, PCI_PREF_BASE_UPPER32, &mem_base_hi); - pci_read_config_dword(dev, PCI_PREF_LIMIT_UPPER32, &mem_limit_hi); - - /* - * Some bridges set the base > limit by default, and some - * (broken) BIOSes do not initialize them. If we find - * this, just assume they are not being used. - */ - if (mem_base_hi <= mem_limit_hi) { - base |= ((long) mem_base_hi) << 32; - limit |= ((long) mem_limit_hi) << 32; - } - } - - res = bus->resource[2]; - if (base <= limit) { - res->flags = ((mem_base_lo & PCI_MEMORY_RANGE_TYPE_MASK) | - IORESOURCE_MEM | IORESOURCE_PREFETCH); - region.start = base; - region.end = limit + 0xfffff; - pcibios_bus_to_resource(dev, res, ®ion); - } -} - /* Cook up fake bus resources for SUNW,simba PCI bridges which lack * a proper 'ranges' property. */ -static void __devinit apb_fake_ranges(struct pci_dev *dev, - struct pci_bus *bus, - struct pci_pbm_info *pbm) +static void apb_fake_ranges(struct pci_dev *dev, + struct pci_bus *bus, + struct pci_pbm_info *pbm) { struct pci_bus_region region; struct resource *res; @@ -480,26 +393,26 @@ static void __devinit apb_fake_ranges(struct pci_dev *dev, res->flags = IORESOURCE_IO; region.start = (first << 21); region.end = (last << 21) + ((1 << 21) - 1); - pcibios_bus_to_resource(dev, res, ®ion); + pcibios_bus_to_resource(dev->bus, res, ®ion); pci_read_config_byte(dev, APB_MEM_ADDRESS_MAP, &map); apb_calc_first_last(map, &first, &last); res = bus->resource[1]; res->flags = IORESOURCE_MEM; - region.start = (first << 21); - region.end = (last << 21) + ((1 << 21) - 1); - pcibios_bus_to_resource(dev, res, ®ion); + region.start = (first << 29); + region.end = (last << 29) + ((1 << 29) - 1); + pcibios_bus_to_resource(dev->bus, res, ®ion); } -static void __devinit pci_of_scan_bus(struct pci_pbm_info *pbm, - struct device_node *node, - struct pci_bus *bus); +static void pci_of_scan_bus(struct pci_pbm_info *pbm, + struct device_node *node, + struct pci_bus *bus); #define GET_64BIT(prop, i) ((((u64) (prop)[(i)]) << 32) | (prop)[(i)+1]) -static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm, - struct device_node *node, - struct pci_dev *dev) +static void of_scan_pci_bridge(struct pci_pbm_info *pbm, + struct device_node *node, + struct pci_dev *dev) { struct pci_bus *bus; const u32 *busrange, *ranges; @@ -535,7 +448,7 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm, } bus->primary = dev->bus->number; - bus->subordinate = busrange[1]; + pci_bus_insert_busn_res(bus, busrange[0], busrange[1]); bus->bridge_ctl = 0; /* parse ranges property, or cook one up by hand for Simba */ @@ -550,7 +463,7 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm, apb_fake_ranges(dev, bus, pbm); goto after_ranges; } else if (ranges == NULL) { - pci_cfg_fake_ranges(dev, bus, pbm); + pci_read_bridge_bases(bus); goto after_ranges; } i = 1; @@ -579,7 +492,7 @@ static void __devinit of_scan_pci_bridge(struct pci_pbm_info *pbm, res->flags = flags; region.start = GET_64BIT(ranges, 1); region.end = region.start + size - 1; - pcibios_bus_to_resource(dev, res, ®ion); + pcibios_bus_to_resource(dev->bus, res, ®ion); } after_ranges: sprintf(bus->name, "PCI Bus %04x:%02x", pci_domain_nr(bus), @@ -590,9 +503,9 @@ after_ranges: pci_of_scan_bus(pbm, node, bus); } -static void __devinit pci_of_scan_bus(struct pci_pbm_info *pbm, - struct device_node *node, - struct pci_bus *bus) +static void pci_of_scan_bus(struct pci_pbm_info *pbm, + struct device_node *node, + struct pci_bus *bus) { struct device_node *child; const u32 *reg; @@ -631,8 +544,7 @@ static void __devinit pci_of_scan_bus(struct pci_pbm_info *pbm, printk("PCI: dev header type: %x\n", dev->hdr_type); - if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE || - dev->hdr_type == PCI_HEADER_TYPE_CARDBUS) + if (pci_is_bridge(dev)) of_scan_pci_bridge(pbm, child, dev); } } @@ -651,7 +563,7 @@ show_pciobppath_attr(struct device * dev, struct device_attribute * attr, char * static DEVICE_ATTR(obppath, S_IRUSR | S_IRGRP | S_IROTH, show_pciobppath_attr, NULL); -static void __devinit pci_bus_register_of_sysfs(struct pci_bus *bus) +static void pci_bus_register_of_sysfs(struct pci_bus *bus) { struct pci_dev *dev; struct pci_bus *child_bus; @@ -672,8 +584,8 @@ static void __devinit pci_bus_register_of_sysfs(struct pci_bus *bus) pci_bus_register_of_sysfs(child_bus); } -struct pci_bus * __devinit pci_scan_one_pbm(struct pci_pbm_info *pbm, - struct device *parent) +struct pci_bus *pci_scan_one_pbm(struct pci_pbm_info *pbm, + struct device *parent) { LIST_HEAD(resources); struct device_node *node = pbm->op->dev.of_node; @@ -685,6 +597,10 @@ struct pci_bus * __devinit pci_scan_one_pbm(struct pci_pbm_info *pbm, pbm->io_space.start); pci_add_resource_offset(&resources, &pbm->mem_space, pbm->mem_space.start); + pbm->busn.start = pbm->pci_first_busno; + pbm->busn.end = pbm->pci_last_busno; + pbm->busn.flags = IORESOURCE_BUS; + pci_add_resource(&resources, &pbm->busn); bus = pci_create_root_bus(parent, pbm->pci_first_busno, pbm->pci_ops, pbm, &resources); if (!bus) { @@ -693,8 +609,6 @@ struct pci_bus * __devinit pci_scan_one_pbm(struct pci_pbm_info *pbm, pci_free_resource_list(&resources); return NULL; } - bus->secondary = pbm->pci_first_busno; - bus->subordinate = pbm->pci_last_busno; pci_of_scan_bus(pbm, node, bus); pci_bus_add_devices(bus); @@ -703,11 +617,7 @@ struct pci_bus * __devinit pci_scan_one_pbm(struct pci_pbm_info *pbm, return bus; } -void __devinit pcibios_fixup_bus(struct pci_bus *pbus) -{ -} - -void pcibios_update_irq(struct pci_dev *pdev, int irq) +void pcibios_fixup_bus(struct pci_bus *pbus) { } @@ -747,11 +657,6 @@ int pcibios_enable_device(struct pci_dev *dev, int mask) return 0; } -char * __devinit pcibios_setup(char *str) -{ - return str; -} - /* Platform support for /proc/bus/pci/X/Y mmap()s. */ /* If the user uses a host-bridge as the PCI device, he may use @@ -867,15 +772,6 @@ static int __pci_mmap_make_offset(struct pci_dev *pdev, return 0; } -/* Set vm_flags of VMA, as appropriate for this architecture, for a pci device - * mapping. - */ -static void __pci_mmap_set_flags(struct pci_dev *dev, struct vm_area_struct *vma, - enum pci_mmap_state mmap_state) -{ - vma->vm_flags |= (VM_IO | VM_RESERVED); -} - /* Set vm_page_prot of VMA, as appropriate for this architecture, for a pci * device mapping. */ @@ -903,7 +799,6 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, if (ret < 0) return ret; - __pci_mmap_set_flags(dev, vma, mmap_state); __pci_mmap_set_pgprot(dev, vma, mmap_state); vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); @@ -1043,8 +938,7 @@ static int __init pcibios_init(void) subsys_initcall(pcibios_init); #ifdef CONFIG_SYSFS -static void __devinit pci_bus_slot_names(struct device_node *node, - struct pci_bus *bus) +static void pci_bus_slot_names(struct device_node *node, struct pci_bus *bus) { const struct pci_slot_names { u32 slot_mask; @@ -1111,6 +1005,5 @@ static int __init of_pci_slot_init(void) return 0; } - -module_init(of_pci_slot_init); +device_initcall(of_pci_slot_init); #endif diff --git a/arch/sparc/kernel/pci_common.c b/arch/sparc/kernel/pci_common.c index a6895987fb7..944a06536ec 100644 --- a/arch/sparc/kernel/pci_common.c +++ b/arch/sparc/kernel/pci_common.c @@ -5,7 +5,6 @@ #include <linux/string.h> #include <linux/slab.h> -#include <linux/init.h> #include <linux/pci.h> #include <linux/device.h> #include <linux/of_device.h> diff --git a/arch/sparc/kernel/pci_fire.c b/arch/sparc/kernel/pci_fire.c index 188f935276f..e60fc6a67e9 100644 --- a/arch/sparc/kernel/pci_fire.c +++ b/arch/sparc/kernel/pci_fire.c @@ -408,8 +408,8 @@ static void pci_fire_hw_init(struct pci_pbm_info *pbm) upa_writeq(~(u64)0, pbm->pbm_regs + FIRE_PEC_IENAB); } -static int __devinit pci_fire_pbm_init(struct pci_pbm_info *pbm, - struct platform_device *op, u32 portid) +static int pci_fire_pbm_init(struct pci_pbm_info *pbm, + struct platform_device *op, u32 portid) { const struct linux_prom64_registers *regs; struct device_node *dp = op->dev.of_node; @@ -454,7 +454,7 @@ static int __devinit pci_fire_pbm_init(struct pci_pbm_info *pbm, return 0; } -static int __devinit fire_probe(struct platform_device *op) +static int fire_probe(struct platform_device *op) { struct device_node *dp = op->dev.of_node; struct pci_pbm_info *pbm; diff --git a/arch/sparc/kernel/pci_impl.h b/arch/sparc/kernel/pci_impl.h index 6beb60df31d..75803c780af 100644 --- a/arch/sparc/kernel/pci_impl.h +++ b/arch/sparc/kernel/pci_impl.h @@ -48,8 +48,8 @@ struct sparc64_msiq_ops { unsigned long devino); }; -extern void sparc64_pbm_msi_init(struct pci_pbm_info *pbm, - const struct sparc64_msiq_ops *ops); +void sparc64_pbm_msi_init(struct pci_pbm_info *pbm, + const struct sparc64_msiq_ops *ops); struct sparc64_msiq_cookie { struct pci_pbm_info *pbm; @@ -88,7 +88,7 @@ struct pci_pbm_info { int chip_revision; /* Name used for top-level resources. */ - char *name; + const char *name; /* OBP specific information. */ struct platform_device *op; @@ -97,6 +97,7 @@ struct pci_pbm_info { /* PBM I/O and Memory space resources. */ struct resource io_space; struct resource mem_space; + struct resource busn; /* Base of PCI Config space, can be per-PBM or shared. */ unsigned long config_space; @@ -157,23 +158,23 @@ extern struct pci_pbm_info *pci_pbm_root; extern int pci_num_pbms; /* PCI bus scanning and fixup support. */ -extern void pci_get_pbm_props(struct pci_pbm_info *pbm); -extern struct pci_bus *pci_scan_one_pbm(struct pci_pbm_info *pbm, - struct device *parent); -extern void pci_determine_mem_io_space(struct pci_pbm_info *pbm); +void pci_get_pbm_props(struct pci_pbm_info *pbm); +struct pci_bus *pci_scan_one_pbm(struct pci_pbm_info *pbm, + struct device *parent); +void pci_determine_mem_io_space(struct pci_pbm_info *pbm); /* Error reporting support. */ -extern void pci_scan_for_target_abort(struct pci_pbm_info *, struct pci_bus *); -extern void pci_scan_for_master_abort(struct pci_pbm_info *, struct pci_bus *); -extern void pci_scan_for_parity_error(struct pci_pbm_info *, struct pci_bus *); +void pci_scan_for_target_abort(struct pci_pbm_info *, struct pci_bus *); +void pci_scan_for_master_abort(struct pci_pbm_info *, struct pci_bus *); +void pci_scan_for_parity_error(struct pci_pbm_info *, struct pci_bus *); /* Configuration space access. */ -extern void pci_config_read8(u8 *addr, u8 *ret); -extern void pci_config_read16(u16 *addr, u16 *ret); -extern void pci_config_read32(u32 *addr, u32 *ret); -extern void pci_config_write8(u8 *addr, u8 val); -extern void pci_config_write16(u16 *addr, u16 val); -extern void pci_config_write32(u32 *addr, u32 val); +void pci_config_read8(u8 *addr, u8 *ret); +void pci_config_read16(u16 *addr, u16 *ret); +void pci_config_read32(u32 *addr, u32 *ret); +void pci_config_write8(u8 *addr, u8 val); +void pci_config_write16(u16 *addr, u16 val); +void pci_config_write32(u32 *addr, u32 val); extern struct pci_ops sun4u_pci_ops; extern struct pci_ops sun4v_pci_ops; diff --git a/arch/sparc/kernel/pci_psycho.c b/arch/sparc/kernel/pci_psycho.c index f4d29e15ce7..c647634ead2 100644 --- a/arch/sparc/kernel/pci_psycho.c +++ b/arch/sparc/kernel/pci_psycho.c @@ -366,8 +366,8 @@ static void pbm_config_busmastering(struct pci_pbm_info *pbm) pci_config_write8(addr, 64); } -static void __devinit psycho_scan_bus(struct pci_pbm_info *pbm, - struct device *parent) +static void psycho_scan_bus(struct pci_pbm_info *pbm, + struct device *parent) { pbm_config_busmastering(pbm); pbm->is_66mhz_capable = 0; @@ -483,15 +483,15 @@ static void psycho_pbm_strbuf_init(struct pci_pbm_info *pbm, #define PSYCHO_MEMSPACE_B 0x180000000UL #define PSYCHO_MEMSPACE_SIZE 0x07fffffffUL -static void __devinit psycho_pbm_init(struct pci_pbm_info *pbm, - struct platform_device *op, int is_pbm_a) +static void psycho_pbm_init(struct pci_pbm_info *pbm, + struct platform_device *op, int is_pbm_a) { psycho_pbm_init_common(pbm, op, "PSYCHO", PBM_CHIP_TYPE_PSYCHO); psycho_pbm_strbuf_init(pbm, is_pbm_a); psycho_scan_bus(pbm, &op->dev); } -static struct pci_pbm_info * __devinit psycho_find_sibling(u32 upa_portid) +static struct pci_pbm_info *psycho_find_sibling(u32 upa_portid) { struct pci_pbm_info *pbm; @@ -504,7 +504,7 @@ static struct pci_pbm_info * __devinit psycho_find_sibling(u32 upa_portid) #define PSYCHO_CONFIGSPACE 0x001000000UL -static int __devinit psycho_probe(struct platform_device *op) +static int psycho_probe(struct platform_device *op) { const struct linux_prom64_registers *pr_regs; struct device_node *dp = op->dev.of_node; diff --git a/arch/sparc/kernel/pci_sabre.c b/arch/sparc/kernel/pci_sabre.c index 3efaa4644d6..6f00d27e8da 100644 --- a/arch/sparc/kernel/pci_sabre.c +++ b/arch/sparc/kernel/pci_sabre.c @@ -403,8 +403,7 @@ static void apb_init(struct pci_bus *sabre_bus) } } -static void __devinit sabre_scan_bus(struct pci_pbm_info *pbm, - struct device *parent) +static void sabre_scan_bus(struct pci_pbm_info *pbm, struct device *parent) { static int once; @@ -443,8 +442,8 @@ static void __devinit sabre_scan_bus(struct pci_pbm_info *pbm, sabre_register_error_handlers(pbm); } -static void __devinit sabre_pbm_init(struct pci_pbm_info *pbm, - struct platform_device *op) +static void sabre_pbm_init(struct pci_pbm_info *pbm, + struct platform_device *op) { psycho_pbm_init_common(pbm, op, "SABRE", PBM_CHIP_TYPE_SABRE); pbm->pci_afsr = pbm->controller_regs + SABRE_PIOAFSR; @@ -454,7 +453,7 @@ static void __devinit sabre_pbm_init(struct pci_pbm_info *pbm, } static const struct of_device_id sabre_match[]; -static int __devinit sabre_probe(struct platform_device *op) +static int sabre_probe(struct platform_device *op) { const struct of_device_id *match; const struct linux_prom64_registers *pr_regs; diff --git a/arch/sparc/kernel/pci_schizo.c b/arch/sparc/kernel/pci_schizo.c index 13d4aa20b5a..8f76f23dac3 100644 --- a/arch/sparc/kernel/pci_schizo.c +++ b/arch/sparc/kernel/pci_schizo.c @@ -1064,8 +1064,7 @@ static void pbm_config_busmastering(struct pci_pbm_info *pbm) pci_config_write8(addr, 64); } -static void __devinit schizo_scan_bus(struct pci_pbm_info *pbm, - struct device *parent) +static void schizo_scan_bus(struct pci_pbm_info *pbm, struct device *parent) { pbm_config_busmastering(pbm); pbm->is_66mhz_capable = @@ -1307,9 +1306,9 @@ static void schizo_pbm_hw_init(struct pci_pbm_info *pbm) } } -static int __devinit schizo_pbm_init(struct pci_pbm_info *pbm, - struct platform_device *op, u32 portid, - int chip_type) +static int schizo_pbm_init(struct pci_pbm_info *pbm, + struct platform_device *op, u32 portid, + int chip_type) { const struct linux_prom64_registers *regs; struct device_node *dp = op->dev.of_node; @@ -1400,8 +1399,7 @@ static inline int portid_compare(u32 x, u32 y, int chip_type) return (x == y); } -static struct pci_pbm_info * __devinit schizo_find_sibling(u32 portid, - int chip_type) +static struct pci_pbm_info *schizo_find_sibling(u32 portid, int chip_type) { struct pci_pbm_info *pbm; @@ -1412,7 +1410,7 @@ static struct pci_pbm_info * __devinit schizo_find_sibling(u32 portid, return NULL; } -static int __devinit __schizo_init(struct platform_device *op, unsigned long chip_type) +static int __schizo_init(struct platform_device *op, unsigned long chip_type) { struct device_node *dp = op->dev.of_node; struct pci_pbm_info *pbm; @@ -1460,7 +1458,7 @@ out_err: } static const struct of_device_id schizo_match[]; -static int __devinit schizo_probe(struct platform_device *op) +static int schizo_probe(struct platform_device *op) { const struct of_device_id *match; diff --git a/arch/sparc/kernel/pci_sun4v.c b/arch/sparc/kernel/pci_sun4v.c index 7661e84a05a..d07f6b29aed 100644 --- a/arch/sparc/kernel/pci_sun4v.c +++ b/arch/sparc/kernel/pci_sun4v.c @@ -536,8 +536,7 @@ static struct dma_map_ops sun4v_dma_ops = { .unmap_sg = dma_4v_unmap_sg, }; -static void __devinit pci_sun4v_scan_bus(struct pci_pbm_info *pbm, - struct device *parent) +static void pci_sun4v_scan_bus(struct pci_pbm_info *pbm, struct device *parent) { struct property *prop; struct device_node *dp; @@ -550,8 +549,8 @@ static void __devinit pci_sun4v_scan_bus(struct pci_pbm_info *pbm, /* XXX register error interrupt handlers XXX */ } -static unsigned long __devinit probe_existing_entries(struct pci_pbm_info *pbm, - struct iommu *iommu) +static unsigned long probe_existing_entries(struct pci_pbm_info *pbm, + struct iommu *iommu) { struct iommu_arena *arena = &iommu->arena; unsigned long i, cnt = 0; @@ -578,7 +577,7 @@ static unsigned long __devinit probe_existing_entries(struct pci_pbm_info *pbm, return cnt; } -static int __devinit pci_sun4v_iommu_init(struct pci_pbm_info *pbm) +static int pci_sun4v_iommu_init(struct pci_pbm_info *pbm) { static const u32 vdma_default[] = { 0x80000000, 0x80000000 }; struct iommu *iommu = pbm->iommu; @@ -594,7 +593,7 @@ static int __devinit pci_sun4v_iommu_init(struct pci_pbm_info *pbm) printk(KERN_ERR PFX "Strange virtual-dma[%08x:%08x].\n", vdma[0], vdma[1]); return -EINVAL; - }; + } dma_mask = (roundup_pow_of_two(vdma[1]) - 1UL); num_tsb_entries = vdma[1] / IO_PAGE_SIZE; @@ -879,8 +878,8 @@ static void pci_sun4v_msi_init(struct pci_pbm_info *pbm) } #endif /* !(CONFIG_PCI_MSI) */ -static int __devinit pci_sun4v_pbm_init(struct pci_pbm_info *pbm, - struct platform_device *op, u32 devhandle) +static int pci_sun4v_pbm_init(struct pci_pbm_info *pbm, + struct platform_device *op, u32 devhandle) { struct device_node *dp = op->dev.of_node; int err; @@ -919,7 +918,7 @@ static int __devinit pci_sun4v_pbm_init(struct pci_pbm_info *pbm, return 0; } -static int __devinit pci_sun4v_probe(struct platform_device *op) +static int pci_sun4v_probe(struct platform_device *op) { const struct linux_prom64_registers *regs; static int hvapi_negotiated = 0; diff --git a/arch/sparc/kernel/pci_sun4v.h b/arch/sparc/kernel/pci_sun4v.h index 8e9fc3a5b4f..5642212390b 100644 --- a/arch/sparc/kernel/pci_sun4v.h +++ b/arch/sparc/kernel/pci_sun4v.h @@ -6,87 +6,87 @@ #ifndef _PCI_SUN4V_H #define _PCI_SUN4V_H -extern long pci_sun4v_iommu_map(unsigned long devhandle, - unsigned long tsbid, - unsigned long num_ttes, - unsigned long io_attributes, - unsigned long io_page_list_pa); -extern unsigned long pci_sun4v_iommu_demap(unsigned long devhandle, - unsigned long tsbid, - unsigned long num_ttes); -extern unsigned long pci_sun4v_iommu_getmap(unsigned long devhandle, - unsigned long tsbid, - unsigned long *io_attributes, - unsigned long *real_address); -extern unsigned long pci_sun4v_config_get(unsigned long devhandle, - unsigned long pci_device, - unsigned long config_offset, - unsigned long size); -extern int pci_sun4v_config_put(unsigned long devhandle, - unsigned long pci_device, - unsigned long config_offset, - unsigned long size, - unsigned long data); +long pci_sun4v_iommu_map(unsigned long devhandle, + unsigned long tsbid, + unsigned long num_ttes, + unsigned long io_attributes, + unsigned long io_page_list_pa); +unsigned long pci_sun4v_iommu_demap(unsigned long devhandle, + unsigned long tsbid, + unsigned long num_ttes); +unsigned long pci_sun4v_iommu_getmap(unsigned long devhandle, + unsigned long tsbid, + unsigned long *io_attributes, + unsigned long *real_address); +unsigned long pci_sun4v_config_get(unsigned long devhandle, + unsigned long pci_device, + unsigned long config_offset, + unsigned long size); +int pci_sun4v_config_put(unsigned long devhandle, + unsigned long pci_device, + unsigned long config_offset, + unsigned long size, + unsigned long data); -extern unsigned long pci_sun4v_msiq_conf(unsigned long devhandle, +unsigned long pci_sun4v_msiq_conf(unsigned long devhandle, unsigned long msiqid, unsigned long msiq_paddr, unsigned long num_entries); -extern unsigned long pci_sun4v_msiq_info(unsigned long devhandle, - unsigned long msiqid, - unsigned long *msiq_paddr, - unsigned long *num_entries); -extern unsigned long pci_sun4v_msiq_getvalid(unsigned long devhandle, - unsigned long msiqid, - unsigned long *valid); -extern unsigned long pci_sun4v_msiq_setvalid(unsigned long devhandle, - unsigned long msiqid, - unsigned long valid); -extern unsigned long pci_sun4v_msiq_getstate(unsigned long devhandle, - unsigned long msiqid, - unsigned long *state); -extern unsigned long pci_sun4v_msiq_setstate(unsigned long devhandle, - unsigned long msiqid, - unsigned long state); -extern unsigned long pci_sun4v_msiq_gethead(unsigned long devhandle, - unsigned long msiqid, - unsigned long *head); -extern unsigned long pci_sun4v_msiq_sethead(unsigned long devhandle, - unsigned long msiqid, - unsigned long head); -extern unsigned long pci_sun4v_msiq_gettail(unsigned long devhandle, - unsigned long msiqid, - unsigned long *head); -extern unsigned long pci_sun4v_msi_getvalid(unsigned long devhandle, - unsigned long msinum, - unsigned long *valid); -extern unsigned long pci_sun4v_msi_setvalid(unsigned long devhandle, - unsigned long msinum, - unsigned long valid); -extern unsigned long pci_sun4v_msi_getmsiq(unsigned long devhandle, - unsigned long msinum, - unsigned long *msiq); -extern unsigned long pci_sun4v_msi_setmsiq(unsigned long devhandle, - unsigned long msinum, - unsigned long msiq, - unsigned long msitype); -extern unsigned long pci_sun4v_msi_getstate(unsigned long devhandle, - unsigned long msinum, - unsigned long *state); -extern unsigned long pci_sun4v_msi_setstate(unsigned long devhandle, - unsigned long msinum, - unsigned long state); -extern unsigned long pci_sun4v_msg_getmsiq(unsigned long devhandle, - unsigned long msinum, - unsigned long *msiq); -extern unsigned long pci_sun4v_msg_setmsiq(unsigned long devhandle, - unsigned long msinum, - unsigned long msiq); -extern unsigned long pci_sun4v_msg_getvalid(unsigned long devhandle, - unsigned long msinum, - unsigned long *valid); -extern unsigned long pci_sun4v_msg_setvalid(unsigned long devhandle, - unsigned long msinum, - unsigned long valid); +unsigned long pci_sun4v_msiq_info(unsigned long devhandle, + unsigned long msiqid, + unsigned long *msiq_paddr, + unsigned long *num_entries); +unsigned long pci_sun4v_msiq_getvalid(unsigned long devhandle, + unsigned long msiqid, + unsigned long *valid); +unsigned long pci_sun4v_msiq_setvalid(unsigned long devhandle, + unsigned long msiqid, + unsigned long valid); +unsigned long pci_sun4v_msiq_getstate(unsigned long devhandle, + unsigned long msiqid, + unsigned long *state); +unsigned long pci_sun4v_msiq_setstate(unsigned long devhandle, + unsigned long msiqid, + unsigned long state); +unsigned long pci_sun4v_msiq_gethead(unsigned long devhandle, + unsigned long msiqid, + unsigned long *head); +unsigned long pci_sun4v_msiq_sethead(unsigned long devhandle, + unsigned long msiqid, + unsigned long head); +unsigned long pci_sun4v_msiq_gettail(unsigned long devhandle, + unsigned long msiqid, + unsigned long *head); +unsigned long pci_sun4v_msi_getvalid(unsigned long devhandle, + unsigned long msinum, + unsigned long *valid); +unsigned long pci_sun4v_msi_setvalid(unsigned long devhandle, + unsigned long msinum, + unsigned long valid); +unsigned long pci_sun4v_msi_getmsiq(unsigned long devhandle, + unsigned long msinum, + unsigned long *msiq); +unsigned long pci_sun4v_msi_setmsiq(unsigned long devhandle, + unsigned long msinum, + unsigned long msiq, + unsigned long msitype); +unsigned long pci_sun4v_msi_getstate(unsigned long devhandle, + unsigned long msinum, + unsigned long *state); +unsigned long pci_sun4v_msi_setstate(unsigned long devhandle, + unsigned long msinum, + unsigned long state); +unsigned long pci_sun4v_msg_getmsiq(unsigned long devhandle, + unsigned long msinum, + unsigned long *msiq); +unsigned long pci_sun4v_msg_setmsiq(unsigned long devhandle, + unsigned long msinum, + unsigned long msiq); +unsigned long pci_sun4v_msg_getvalid(unsigned long devhandle, + unsigned long msinum, + unsigned long *valid); +unsigned long pci_sun4v_msg_setvalid(unsigned long devhandle, + unsigned long msinum, + unsigned long valid); #endif /* !(_PCI_SUN4V_H) */ diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c index ded3f6090c3..6cc78c213c0 100644 --- a/arch/sparc/kernel/pcic.c +++ b/arch/sparc/kernel/pcic.c @@ -36,6 +36,7 @@ #include <asm/uaccess.h> #include <asm/irq_regs.h> +#include "kernel.h" #include "irq.h" /* @@ -162,8 +163,8 @@ static int pcic0_up; static struct linux_pcic pcic0; void __iomem *pcic_regs; -volatile int pcic_speculative; -volatile int pcic_trapped; +static volatile int pcic_speculative; +static volatile int pcic_trapped; /* forward */ unsigned int pcic_build_device_irq(struct platform_device *op, @@ -329,7 +330,7 @@ int __init pcic_probe(void) pcic->pcic_res_cfg_addr.name = "pcic_cfg_addr"; if ((pcic->pcic_config_space_addr = - ioremap(regs[2].phys_addr, regs[2].reg_size * 2)) == 0) { + ioremap(regs[2].phys_addr, regs[2].reg_size * 2)) == NULL) { prom_printf("PCIC: Error, cannot map " "PCI Configuration Space Address.\n"); prom_halt(); @@ -341,7 +342,7 @@ int __init pcic_probe(void) */ pcic->pcic_res_cfg_data.name = "pcic_cfg_data"; if ((pcic->pcic_config_space_data = - ioremap(regs[3].phys_addr, regs[3].reg_size * 2)) == 0) { + ioremap(regs[3].phys_addr, regs[3].reg_size * 2)) == NULL) { prom_printf("PCIC: Error, cannot map " "PCI Configuration Space Data.\n"); prom_halt(); @@ -353,7 +354,6 @@ int __init pcic_probe(void) strcpy(pbm->prom_name, namebuf); { - extern volatile int t_nmi[4]; extern int pcic_nmi_trap_patch[4]; t_nmi[0] = pcic_nmi_trap_patch[0]; @@ -439,8 +439,7 @@ int pcic_present(void) return pcic0_up; } -static int __devinit pdev_to_pnode(struct linux_pbm_info *pbm, - struct pci_dev *pdev) +static int pdev_to_pnode(struct linux_pbm_info *pbm, struct pci_dev *pdev) { struct linux_prom_pci_registers regs[PROMREG_MAX]; int err; @@ -537,7 +536,7 @@ pcic_fill_irq(struct linux_pcic *pcic, struct pci_dev *dev, int node) prom_getstring(node, "name", namebuf, sizeof(namebuf)); } - if ((p = pcic->pcic_imap) == 0) { + if ((p = pcic->pcic_imap) == NULL) { dev->irq = 0; return; } @@ -595,7 +594,7 @@ pcic_fill_irq(struct linux_pcic *pcic, struct pci_dev *dev, int node) /* * Normally called from {do_}pci_scan_bus... */ -void __devinit pcibios_fixup_bus(struct pci_bus *bus) +void pcibios_fixup_bus(struct pci_bus *bus) { struct pci_dev *dev; int i, has_io, has_mem; @@ -671,30 +670,6 @@ void __devinit pcibios_fixup_bus(struct pci_bus *bus) } } -/* - * pcic_pin_to_irq() is exported to bus probing code - */ -unsigned int -pcic_pin_to_irq(unsigned int pin, const char *name) -{ - struct linux_pcic *pcic = &pcic0; - unsigned int irq; - unsigned int ivec; - - if (pin < 4) { - ivec = readw(pcic->pcic_regs+PCI_INT_SELECT_LO); - irq = ivec >> (pin << 2) & 0xF; - } else if (pin < 8) { - ivec = readw(pcic->pcic_regs+PCI_INT_SELECT_HI); - irq = ivec >> ((pin-4) << 2) & 0xF; - } else { /* Corrupted map */ - printk("PCIC: BAD PIN %d FOR %s\n", pin, name); - for (;;) {} /* XXX Cannot panic properly in case of PROLL */ - } -/* P3 */ /* printk("PCIC: dev %s pin %d ivec 0x%x irq %x\n", name, pin, ivec, irq); */ - return irq; -} - /* Makes compiler happy */ static volatile int pcic_timer_dummy; @@ -767,14 +742,6 @@ static void watchdog_reset() { } #endif -/* - * Other archs parse arguments here. - */ -char * __devinit pcibios_setup(char *str) -{ - return str; -} - resource_size_t pcibios_align_resource(void *data, const struct resource *res, resource_size_t size, resource_size_t align) { @@ -792,7 +759,7 @@ int pcibios_enable_device(struct pci_dev *pdev, int mask) void pcic_nmi(unsigned int pend, struct pt_regs *regs) { - pend = flip_dword(pend); + pend = swab32(pend); if (!pcic_speculative || (pend & PCI_SYS_INT_PENDING_PIO) == 0) { /* @@ -884,87 +851,4 @@ void __init sun4m_pci_init_IRQ(void) sparc_config.load_profile_irq = pcic_load_profile_irq; } -int pcibios_assign_resource(struct pci_dev *pdev, int resource) -{ - return -ENXIO; -} - -/* - * This probably belongs here rather than ioport.c because - * we do not want this crud linked into SBus kernels. - * Also, think for a moment about likes of floppy.c that - * include architecture specific parts. They may want to redefine ins/outs. - * - * We do not use horrible macros here because we want to - * advance pointer by sizeof(size). - */ -void outsb(unsigned long addr, const void *src, unsigned long count) -{ - while (count) { - count -= 1; - outb(*(const char *)src, addr); - src += 1; - /* addr += 1; */ - } -} -EXPORT_SYMBOL(outsb); - -void outsw(unsigned long addr, const void *src, unsigned long count) -{ - while (count) { - count -= 2; - outw(*(const short *)src, addr); - src += 2; - /* addr += 2; */ - } -} -EXPORT_SYMBOL(outsw); - -void outsl(unsigned long addr, const void *src, unsigned long count) -{ - while (count) { - count -= 4; - outl(*(const long *)src, addr); - src += 4; - /* addr += 4; */ - } -} -EXPORT_SYMBOL(outsl); - -void insb(unsigned long addr, void *dst, unsigned long count) -{ - while (count) { - count -= 1; - *(unsigned char *)dst = inb(addr); - dst += 1; - /* addr += 1; */ - } -} -EXPORT_SYMBOL(insb); - -void insw(unsigned long addr, void *dst, unsigned long count) -{ - while (count) { - count -= 2; - *(unsigned short *)dst = inw(addr); - dst += 2; - /* addr += 2; */ - } -} -EXPORT_SYMBOL(insw); - -void insl(unsigned long addr, void *dst, unsigned long count) -{ - while (count) { - count -= 4; - /* - * XXX I am sure we are in for an unaligned trap here. - */ - *(unsigned long *)dst = inl(addr); - dst += 4; - /* addr += 4; */ - } -} -EXPORT_SYMBOL(insl); - subsys_initcall(pcic_init); diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c index 0ce0dd2332a..269af58497a 100644 --- a/arch/sparc/kernel/pcr.c +++ b/arch/sparc/kernel/pcr.c @@ -13,23 +13,14 @@ #include <asm/pil.h> #include <asm/pcr.h> #include <asm/nmi.h> +#include <asm/asi.h> #include <asm/spitfire.h> -#include <asm/perfctr.h> /* This code is shared between various users of the performance * counters. Users will be oprofile, pseudo-NMI watchdog, and the * perf_event support layer. */ -#define PCR_SUN4U_ENABLE (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE) -#define PCR_N2_ENABLE (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE | \ - PCR_N2_TOE_OV1 | \ - (2 << PCR_N2_SL1_SHIFT) | \ - (0xff << PCR_N2_MASK1_SHIFT)) - -u64 pcr_enable; -unsigned int picl_shift; - /* Performance counter interrupts run unmasked at PIL level 15. * Therefore we can't do things like wakeups and other work * that expects IRQ disabling to be adhered to in locking etc. @@ -60,39 +51,144 @@ void arch_irq_work_raise(void) const struct pcr_ops *pcr_ops; EXPORT_SYMBOL_GPL(pcr_ops); -static u64 direct_pcr_read(void) +static u64 direct_pcr_read(unsigned long reg_num) { u64 val; - read_pcr(val); + WARN_ON_ONCE(reg_num != 0); + __asm__ __volatile__("rd %%pcr, %0" : "=r" (val)); return val; } -static void direct_pcr_write(u64 val) +static void direct_pcr_write(unsigned long reg_num, u64 val) +{ + WARN_ON_ONCE(reg_num != 0); + __asm__ __volatile__("wr %0, 0x0, %%pcr" : : "r" (val)); +} + +static u64 direct_pic_read(unsigned long reg_num) { - write_pcr(val); + u64 val; + + WARN_ON_ONCE(reg_num != 0); + __asm__ __volatile__("rd %%pic, %0" : "=r" (val)); + return val; +} + +static void direct_pic_write(unsigned long reg_num, u64 val) +{ + WARN_ON_ONCE(reg_num != 0); + + /* Blackbird errata workaround. See commentary in + * arch/sparc64/kernel/smp.c:smp_percpu_timer_interrupt() + * for more information. + */ + __asm__ __volatile__("ba,pt %%xcc, 99f\n\t" + " nop\n\t" + ".align 64\n" + "99:wr %0, 0x0, %%pic\n\t" + "rd %%pic, %%g0" : : "r" (val)); +} + +static u64 direct_picl_value(unsigned int nmi_hz) +{ + u32 delta = local_cpu_data().clock_tick / nmi_hz; + + return ((u64)((0 - delta) & 0xffffffff)) << 32; } static const struct pcr_ops direct_pcr_ops = { - .read = direct_pcr_read, - .write = direct_pcr_write, + .read_pcr = direct_pcr_read, + .write_pcr = direct_pcr_write, + .read_pic = direct_pic_read, + .write_pic = direct_pic_write, + .nmi_picl_value = direct_picl_value, + .pcr_nmi_enable = (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE), + .pcr_nmi_disable = PCR_PIC_PRIV, }; -static void n2_pcr_write(u64 val) +static void n2_pcr_write(unsigned long reg_num, u64 val) { unsigned long ret; + WARN_ON_ONCE(reg_num != 0); if (val & PCR_N2_HTRACE) { ret = sun4v_niagara2_setperf(HV_N2_PERF_SPARC_CTL, val); if (ret != HV_EOK) - write_pcr(val); + direct_pcr_write(reg_num, val); } else - write_pcr(val); + direct_pcr_write(reg_num, val); +} + +static u64 n2_picl_value(unsigned int nmi_hz) +{ + u32 delta = local_cpu_data().clock_tick / (nmi_hz << 2); + + return ((u64)((0 - delta) & 0xffffffff)) << 32; } static const struct pcr_ops n2_pcr_ops = { - .read = direct_pcr_read, - .write = n2_pcr_write, + .read_pcr = direct_pcr_read, + .write_pcr = n2_pcr_write, + .read_pic = direct_pic_read, + .write_pic = direct_pic_write, + .nmi_picl_value = n2_picl_value, + .pcr_nmi_enable = (PCR_PIC_PRIV | PCR_STRACE | PCR_UTRACE | + PCR_N2_TOE_OV1 | + (2 << PCR_N2_SL1_SHIFT) | + (0xff << PCR_N2_MASK1_SHIFT)), + .pcr_nmi_disable = PCR_PIC_PRIV, +}; + +static u64 n4_pcr_read(unsigned long reg_num) +{ + unsigned long val; + + (void) sun4v_vt_get_perfreg(reg_num, &val); + + return val; +} + +static void n4_pcr_write(unsigned long reg_num, u64 val) +{ + (void) sun4v_vt_set_perfreg(reg_num, val); +} + +static u64 n4_pic_read(unsigned long reg_num) +{ + unsigned long val; + + __asm__ __volatile__("ldxa [%1] %2, %0" + : "=r" (val) + : "r" (reg_num * 0x8UL), "i" (ASI_PIC)); + + return val; +} + +static void n4_pic_write(unsigned long reg_num, u64 val) +{ + __asm__ __volatile__("stxa %0, [%1] %2" + : /* no outputs */ + : "r" (val), "r" (reg_num * 0x8UL), "i" (ASI_PIC)); +} + +static u64 n4_picl_value(unsigned int nmi_hz) +{ + u32 delta = local_cpu_data().clock_tick / (nmi_hz << 2); + + return ((u64)((0 - delta) & 0xffffffff)); +} + +static const struct pcr_ops n4_pcr_ops = { + .read_pcr = n4_pcr_read, + .write_pcr = n4_pcr_write, + .read_pic = n4_pic_read, + .write_pic = n4_pic_write, + .nmi_picl_value = n4_picl_value, + .pcr_nmi_enable = (PCR_N4_PICNPT | PCR_N4_STRACE | + PCR_N4_UTRACE | PCR_N4_TOE | + (26 << PCR_N4_SL_SHIFT)), + .pcr_nmi_disable = PCR_N4_PICNPT, }; static unsigned long perf_hsvc_group; @@ -115,6 +211,10 @@ static int __init register_perf_hsvc(void) perf_hsvc_group = HV_GRP_KT_CPU; break; + case SUN4V_CHIP_NIAGARA4: + perf_hsvc_group = HV_GRP_VT_CPU; + break; + default: return -ENODEV; } @@ -139,6 +239,29 @@ static void __init unregister_perf_hsvc(void) sun4v_hvapi_unregister(perf_hsvc_group); } +static int __init setup_sun4v_pcr_ops(void) +{ + int ret = 0; + + switch (sun4v_chip_type) { + case SUN4V_CHIP_NIAGARA1: + case SUN4V_CHIP_NIAGARA2: + case SUN4V_CHIP_NIAGARA3: + pcr_ops = &n2_pcr_ops; + break; + + case SUN4V_CHIP_NIAGARA4: + pcr_ops = &n4_pcr_ops; + break; + + default: + ret = -ENODEV; + break; + } + + return ret; +} + int __init pcr_arch_init(void) { int err = register_perf_hsvc(); @@ -148,15 +271,14 @@ int __init pcr_arch_init(void) switch (tlb_type) { case hypervisor: - pcr_ops = &n2_pcr_ops; - pcr_enable = PCR_N2_ENABLE; - picl_shift = 2; + err = setup_sun4v_pcr_ops(); + if (err) + goto out_unregister; break; case cheetah: case cheetah_plus: pcr_ops = &direct_pcr_ops; - pcr_enable = PCR_SUN4U_ENABLE; break; case spitfire: diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 5713957dcb8..8efd33753ad 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -25,36 +25,48 @@ #include <linux/atomic.h> #include <asm/nmi.h> #include <asm/pcr.h> -#include <asm/perfctr.h> #include <asm/cacheflush.h> #include "kernel.h" #include "kstack.h" -/* Sparc64 chips have two performance counters, 32-bits each, with - * overflow interrupts generated on transition from 0xffffffff to 0. - * The counters are accessed in one go using a 64-bit register. +/* Two classes of sparc64 chips currently exist. All of which have + * 32-bit counters which can generate overflow interrupts on the + * transition from 0xffffffff to 0. * - * Both counters are controlled using a single control register. The - * only way to stop all sampling is to clear all of the context (user, - * supervisor, hypervisor) sampling enable bits. But these bits apply - * to both counters, thus the two counters can't be enabled/disabled - * individually. + * All chips upto and including SPARC-T3 have two performance + * counters. The two 32-bit counters are accessed in one go using a + * single 64-bit register. * - * The control register has two event fields, one for each of the two - * counters. It's thus nearly impossible to have one counter going - * while keeping the other one stopped. Therefore it is possible to - * get overflow interrupts for counters not currently "in use" and - * that condition must be checked in the overflow interrupt handler. + * On these older chips both counters are controlled using a single + * control register. The only way to stop all sampling is to clear + * all of the context (user, supervisor, hypervisor) sampling enable + * bits. But these bits apply to both counters, thus the two counters + * can't be enabled/disabled individually. + * + * Furthermore, the control register on these older chips have two + * event fields, one for each of the two counters. It's thus nearly + * impossible to have one counter going while keeping the other one + * stopped. Therefore it is possible to get overflow interrupts for + * counters not currently "in use" and that condition must be checked + * in the overflow interrupt handler. * * So we use a hack, in that we program inactive counters with the * "sw_count0" and "sw_count1" events. These count how many times * the instruction "sethi %hi(0xfc000), %g0" is executed. It's an * unusual way to encode a NOP and therefore will not trigger in * normal code. + * + * Starting with SPARC-T4 we have one control register per counter. + * And the counters are stored in individual registers. The registers + * for the counters are 64-bit but only a 32-bit counter is + * implemented. The event selections on SPARC-T4 lack any + * restrictions, therefore we can elide all of the complicated + * conflict resolution code we have for SPARC-T3 and earlier chips. */ -#define MAX_HWEVENTS 2 +#define MAX_HWEVENTS 4 +#define MAX_PCRS 4 #define MAX_PERIOD ((1UL << 32) - 1) #define PIC_UPPER_INDEX 0 @@ -90,19 +102,21 @@ struct cpu_hw_events { */ int current_idx[MAX_HWEVENTS]; - /* Software copy of %pcr register on this cpu. */ - u64 pcr; + /* Software copy of %pcr register(s) on this cpu. */ + u64 pcr[MAX_HWEVENTS]; /* Enabled/disable state. */ int enabled; unsigned int group_flag; }; -DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { .enabled = 1, }; +static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = { .enabled = 1, }; /* An event map describes the characteristics of a performance * counter event. In particular it gives the encoding as well as * a mask telling which counters the event can be measured on. + * + * The mask is unused on SPARC-T4 and later. */ struct perf_event_map { u16 encoding; @@ -142,15 +156,53 @@ struct sparc_pmu { const struct perf_event_map *(*event_map)(int); const cache_map_t *cache_map; int max_events; + u32 (*read_pmc)(int); + void (*write_pmc)(int, u64); int upper_shift; int lower_shift; int event_mask; + int user_bit; + int priv_bit; int hv_bit; int irq_bit; int upper_nop; int lower_nop; + unsigned int flags; +#define SPARC_PMU_ALL_EXCLUDES_SAME 0x00000001 +#define SPARC_PMU_HAS_CONFLICTS 0x00000002 + int max_hw_events; + int num_pcrs; + int num_pic_regs; }; +static u32 sparc_default_read_pmc(int idx) +{ + u64 val; + + val = pcr_ops->read_pic(0); + if (idx == PIC_UPPER_INDEX) + val >>= 32; + + return val & 0xffffffff; +} + +static void sparc_default_write_pmc(int idx, u64 val) +{ + u64 shift, mask, pic; + + shift = 0; + if (idx == PIC_UPPER_INDEX) + shift = 32; + + mask = ((u64) 0xffffffff) << shift; + val <<= shift; + + pic = pcr_ops->read_pic(0); + pic &= ~mask; + pic |= val; + pcr_ops->write_pic(0, pic); +} + static const struct perf_event_map ultra3_perfmon_event_map[] = { [PERF_COUNT_HW_CPU_CYCLES] = { 0x0000, PIC_UPPER | PIC_LOWER }, [PERF_COUNT_HW_INSTRUCTIONS] = { 0x0001, PIC_UPPER | PIC_LOWER }, @@ -268,11 +320,20 @@ static const struct sparc_pmu ultra3_pmu = { .event_map = ultra3_event_map, .cache_map = &ultra3_cache_map, .max_events = ARRAY_SIZE(ultra3_perfmon_event_map), + .read_pmc = sparc_default_read_pmc, + .write_pmc = sparc_default_write_pmc, .upper_shift = 11, .lower_shift = 4, .event_mask = 0x3f, + .user_bit = PCR_UTRACE, + .priv_bit = PCR_STRACE, .upper_nop = 0x1c, .lower_nop = 0x14, + .flags = (SPARC_PMU_ALL_EXCLUDES_SAME | + SPARC_PMU_HAS_CONFLICTS), + .max_hw_events = 2, + .num_pcrs = 1, + .num_pic_regs = 1, }; /* Niagara1 is very limited. The upper PIC is hard-locked to count @@ -397,11 +458,20 @@ static const struct sparc_pmu niagara1_pmu = { .event_map = niagara1_event_map, .cache_map = &niagara1_cache_map, .max_events = ARRAY_SIZE(niagara1_perfmon_event_map), + .read_pmc = sparc_default_read_pmc, + .write_pmc = sparc_default_write_pmc, .upper_shift = 0, .lower_shift = 4, .event_mask = 0x7, + .user_bit = PCR_UTRACE, + .priv_bit = PCR_STRACE, .upper_nop = 0x0, .lower_nop = 0x0, + .flags = (SPARC_PMU_ALL_EXCLUDES_SAME | + SPARC_PMU_HAS_CONFLICTS), + .max_hw_events = 2, + .num_pcrs = 1, + .num_pic_regs = 1, }; static const struct perf_event_map niagara2_perfmon_event_map[] = { @@ -523,13 +593,203 @@ static const struct sparc_pmu niagara2_pmu = { .event_map = niagara2_event_map, .cache_map = &niagara2_cache_map, .max_events = ARRAY_SIZE(niagara2_perfmon_event_map), + .read_pmc = sparc_default_read_pmc, + .write_pmc = sparc_default_write_pmc, .upper_shift = 19, .lower_shift = 6, .event_mask = 0xfff, - .hv_bit = 0x8, + .user_bit = PCR_UTRACE, + .priv_bit = PCR_STRACE, + .hv_bit = PCR_N2_HTRACE, .irq_bit = 0x30, .upper_nop = 0x220, .lower_nop = 0x220, + .flags = (SPARC_PMU_ALL_EXCLUDES_SAME | + SPARC_PMU_HAS_CONFLICTS), + .max_hw_events = 2, + .num_pcrs = 1, + .num_pic_regs = 1, +}; + +static const struct perf_event_map niagara4_perfmon_event_map[] = { + [PERF_COUNT_HW_CPU_CYCLES] = { (26 << 6) }, + [PERF_COUNT_HW_INSTRUCTIONS] = { (3 << 6) | 0x3f }, + [PERF_COUNT_HW_CACHE_REFERENCES] = { (3 << 6) | 0x04 }, + [PERF_COUNT_HW_CACHE_MISSES] = { (16 << 6) | 0x07 }, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = { (4 << 6) | 0x01 }, + [PERF_COUNT_HW_BRANCH_MISSES] = { (25 << 6) | 0x0f }, +}; + +static const struct perf_event_map *niagara4_event_map(int event_id) +{ + return &niagara4_perfmon_event_map[event_id]; +} + +static const cache_map_t niagara4_cache_map = { +[C(L1D)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x04 }, + [C(RESULT_MISS)] = { (16 << 6) | 0x07 }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x08 }, + [C(RESULT_MISS)] = { (16 << 6) | 0x07 }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(L1I)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x3f }, + [C(RESULT_MISS)] = { (11 << 6) | 0x03 }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_NONSENSE }, + [ C(RESULT_MISS) ] = { CACHE_OP_NONSENSE }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(LL)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x04 }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [C(OP_WRITE)] = { + [C(RESULT_ACCESS)] = { (3 << 6) | 0x08 }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [C(OP_PREFETCH)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(DTLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { (17 << 6) | 0x3f }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(ITLB)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { (6 << 6) | 0x3f }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(BPU)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS)] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +[C(NODE)] = { + [C(OP_READ)] = { + [C(RESULT_ACCESS)] = { CACHE_OP_UNSUPPORTED }, + [C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_WRITE) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, + [ C(OP_PREFETCH) ] = { + [ C(RESULT_ACCESS) ] = { CACHE_OP_UNSUPPORTED }, + [ C(RESULT_MISS) ] = { CACHE_OP_UNSUPPORTED }, + }, +}, +}; + +static u32 sparc_vt_read_pmc(int idx) +{ + u64 val = pcr_ops->read_pic(idx); + + return val & 0xffffffff; +} + +static void sparc_vt_write_pmc(int idx, u64 val) +{ + u64 pcr; + + /* There seems to be an internal latch on the overflow event + * on SPARC-T4 that prevents it from triggering unless you + * update the PIC exactly as we do here. The requirement + * seems to be that you have to turn off event counting in the + * PCR around the PIC update. + * + * For example, after the following sequence: + * + * 1) set PIC to -1 + * 2) enable event counting and overflow reporting in PCR + * 3) overflow triggers, softint 15 handler invoked + * 4) clear OV bit in PCR + * 5) write PIC to -1 + * + * a subsequent overflow event will not trigger. This + * sequence works on SPARC-T3 and previous chips. + */ + pcr = pcr_ops->read_pcr(idx); + pcr_ops->write_pcr(idx, PCR_N4_PICNPT); + + pcr_ops->write_pic(idx, val & 0xffffffff); + + pcr_ops->write_pcr(idx, pcr); +} + +static const struct sparc_pmu niagara4_pmu = { + .event_map = niagara4_event_map, + .cache_map = &niagara4_cache_map, + .max_events = ARRAY_SIZE(niagara4_perfmon_event_map), + .read_pmc = sparc_vt_read_pmc, + .write_pmc = sparc_vt_write_pmc, + .upper_shift = 5, + .lower_shift = 5, + .event_mask = 0x7ff, + .user_bit = PCR_N4_UTRACE, + .priv_bit = PCR_N4_STRACE, + + /* We explicitly don't support hypervisor tracing. The T4 + * generates the overflow event for precise events via a trap + * which will not be generated (ie. it's completely lost) if + * we happen to be in the hypervisor when the event triggers. + * Essentially, the overflow event reporting is completely + * unusable when you have hypervisor mode tracing enabled. + */ + .hv_bit = 0, + + .irq_bit = PCR_N4_TOE, + .upper_nop = 0, + .lower_nop = 0, + .flags = 0, + .max_hw_events = 4, + .num_pcrs = 4, + .num_pic_regs = 4, }; static const struct sparc_pmu *sparc_pmu __read_mostly; @@ -557,56 +817,38 @@ static u64 nop_for_index(int idx) static inline void sparc_pmu_enable_event(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc, int idx) { - u64 val, mask = mask_for_index(idx); + u64 enc, val, mask = mask_for_index(idx); + int pcr_index = 0; + + if (sparc_pmu->num_pcrs > 1) + pcr_index = idx; - val = cpuc->pcr; + enc = perf_event_get_enc(cpuc->events[idx]); + + val = cpuc->pcr[pcr_index]; val &= ~mask; - val |= hwc->config; - cpuc->pcr = val; + val |= event_encoding(enc, idx); + cpuc->pcr[pcr_index] = val; - pcr_ops->write(cpuc->pcr); + pcr_ops->write_pcr(pcr_index, cpuc->pcr[pcr_index]); } static inline void sparc_pmu_disable_event(struct cpu_hw_events *cpuc, struct hw_perf_event *hwc, int idx) { u64 mask = mask_for_index(idx); u64 nop = nop_for_index(idx); + int pcr_index = 0; u64 val; - val = cpuc->pcr; + if (sparc_pmu->num_pcrs > 1) + pcr_index = idx; + + val = cpuc->pcr[pcr_index]; val &= ~mask; val |= nop; - cpuc->pcr = val; - - pcr_ops->write(cpuc->pcr); -} - -static u32 read_pmc(int idx) -{ - u64 val; - - read_pic(val); - if (idx == PIC_UPPER_INDEX) - val >>= 32; + cpuc->pcr[pcr_index] = val; - return val & 0xffffffff; -} - -static void write_pmc(int idx, u64 val) -{ - u64 shift, mask, pic; - - shift = 0; - if (idx == PIC_UPPER_INDEX) - shift = 32; - - mask = ((u64) 0xffffffff) << shift; - val <<= shift; - - read_pic(pic); - pic &= ~mask; - pic |= val; - write_pic(pic); + pcr_ops->write_pcr(pcr_index, cpuc->pcr[pcr_index]); } static u64 sparc_perf_event_update(struct perf_event *event, @@ -618,7 +860,7 @@ static u64 sparc_perf_event_update(struct perf_event *event, again: prev_raw_count = local64_read(&hwc->prev_count); - new_raw_count = read_pmc(idx); + new_raw_count = sparc_pmu->read_pmc(idx); if (local64_cmpxchg(&hwc->prev_count, prev_raw_count, new_raw_count) != prev_raw_count) @@ -658,25 +900,17 @@ static int sparc_perf_event_set_period(struct perf_event *event, local64_set(&hwc->prev_count, (u64)-left); - write_pmc(idx, (u64)(-left) & 0xffffffff); + sparc_pmu->write_pmc(idx, (u64)(-left) & 0xffffffff); perf_event_update_userpage(event); return ret; } -/* If performance event entries have been added, move existing - * events around (if necessary) and then assign new entries to - * counters. - */ -static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr) +static void read_in_all_counters(struct cpu_hw_events *cpuc) { int i; - if (!cpuc->n_added) - goto out; - - /* Read in the counters which are moving. */ for (i = 0; i < cpuc->n_events; i++) { struct perf_event *cp = cpuc->event[i]; @@ -687,6 +921,20 @@ static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr) cpuc->current_idx[i] = PIC_NO_INDEX; } } +} + +/* On this PMU all PICs are programmed using a single PCR. Calculate + * the combined control register value. + * + * For such chips we require that all of the events have the same + * configuration, so just fetch the settings from the first entry. + */ +static void calculate_single_pcr(struct cpu_hw_events *cpuc) +{ + int i; + + if (!cpuc->n_added) + goto out; /* Assign to counters all unassigned events. */ for (i = 0; i < cpuc->n_events; i++) { @@ -702,20 +950,71 @@ static u64 maybe_change_configuration(struct cpu_hw_events *cpuc, u64 pcr) cpuc->current_idx[i] = idx; enc = perf_event_get_enc(cpuc->events[i]); - pcr &= ~mask_for_index(idx); + cpuc->pcr[0] &= ~mask_for_index(idx); if (hwc->state & PERF_HES_STOPPED) - pcr |= nop_for_index(idx); + cpuc->pcr[0] |= nop_for_index(idx); else - pcr |= event_encoding(enc, idx); + cpuc->pcr[0] |= event_encoding(enc, idx); } out: - return pcr; + cpuc->pcr[0] |= cpuc->event[0]->hw.config_base; +} + +/* On this PMU each PIC has it's own PCR control register. */ +static void calculate_multiple_pcrs(struct cpu_hw_events *cpuc) +{ + int i; + + if (!cpuc->n_added) + goto out; + + for (i = 0; i < cpuc->n_events; i++) { + struct perf_event *cp = cpuc->event[i]; + struct hw_perf_event *hwc = &cp->hw; + int idx = hwc->idx; + u64 enc; + + if (cpuc->current_idx[i] != PIC_NO_INDEX) + continue; + + sparc_perf_event_set_period(cp, hwc, idx); + cpuc->current_idx[i] = idx; + + enc = perf_event_get_enc(cpuc->events[i]); + cpuc->pcr[idx] &= ~mask_for_index(idx); + if (hwc->state & PERF_HES_STOPPED) + cpuc->pcr[idx] |= nop_for_index(idx); + else + cpuc->pcr[idx] |= event_encoding(enc, idx); + } +out: + for (i = 0; i < cpuc->n_events; i++) { + struct perf_event *cp = cpuc->event[i]; + int idx = cp->hw.idx; + + cpuc->pcr[idx] |= cp->hw.config_base; + } +} + +/* If performance event entries have been added, move existing events + * around (if necessary) and then assign new entries to counters. + */ +static void update_pcrs_for_enable(struct cpu_hw_events *cpuc) +{ + if (cpuc->n_added) + read_in_all_counters(cpuc); + + if (sparc_pmu->num_pcrs == 1) { + calculate_single_pcr(cpuc); + } else { + calculate_multiple_pcrs(cpuc); + } } static void sparc_pmu_enable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 pcr; + int i; if (cpuc->enabled) return; @@ -723,26 +1022,17 @@ static void sparc_pmu_enable(struct pmu *pmu) cpuc->enabled = 1; barrier(); - pcr = cpuc->pcr; - if (!cpuc->n_events) { - pcr = 0; - } else { - pcr = maybe_change_configuration(cpuc, pcr); - - /* We require that all of the events have the same - * configuration, so just fetch the settings from the - * first entry. - */ - cpuc->pcr = pcr | cpuc->event[0]->hw.config_base; - } + if (cpuc->n_events) + update_pcrs_for_enable(cpuc); - pcr_ops->write(cpuc->pcr); + for (i = 0; i < sparc_pmu->num_pcrs; i++) + pcr_ops->write_pcr(i, cpuc->pcr[i]); } static void sparc_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); - u64 val; + int i; if (!cpuc->enabled) return; @@ -750,12 +1040,14 @@ static void sparc_pmu_disable(struct pmu *pmu) cpuc->enabled = 0; cpuc->n_added = 0; - val = cpuc->pcr; - val &= ~(PCR_UTRACE | PCR_STRACE | - sparc_pmu->hv_bit | sparc_pmu->irq_bit); - cpuc->pcr = val; + for (i = 0; i < sparc_pmu->num_pcrs; i++) { + u64 val = cpuc->pcr[i]; - pcr_ops->write(cpuc->pcr); + val &= ~(sparc_pmu->user_bit | sparc_pmu->priv_bit | + sparc_pmu->hv_bit | sparc_pmu->irq_bit); + cpuc->pcr[i] = val; + pcr_ops->write_pcr(i, cpuc->pcr[i]); + } } static int active_event_index(struct cpu_hw_events *cpuc, @@ -854,12 +1146,14 @@ static DEFINE_MUTEX(pmc_grab_mutex); static void perf_stop_nmi_watchdog(void *unused) { struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events); + int i; stop_nmi_watchdog(NULL); - cpuc->pcr = pcr_ops->read(); + for (i = 0; i < sparc_pmu->num_pcrs; i++) + cpuc->pcr[i] = pcr_ops->read_pcr(i); } -void perf_event_grab_pmc(void) +static void perf_event_grab_pmc(void) { if (atomic_inc_not_zero(&active_events)) return; @@ -875,7 +1169,7 @@ void perf_event_grab_pmc(void) mutex_unlock(&pmc_grab_mutex); } -void perf_event_release_pmc(void) +static void perf_event_release_pmc(void) { if (atomic_dec_and_mutex_lock(&active_events, &pmc_grab_mutex)) { if (atomic_read(&nmi_active) == 0) @@ -942,9 +1236,17 @@ static int sparc_check_constraints(struct perf_event **evts, if (!n_ev) return 0; - if (n_ev > MAX_HWEVENTS) + if (n_ev > sparc_pmu->max_hw_events) return -1; + if (!(sparc_pmu->flags & SPARC_PMU_HAS_CONFLICTS)) { + int i; + + for (i = 0; i < n_ev; i++) + evts[i]->hw.idx = i; + return 0; + } + msk0 = perf_event_get_msk(events[0]); if (n_ev == 1) { if (msk0 & PIC_LOWER) @@ -1000,6 +1302,9 @@ static int check_excludes(struct perf_event **evts, int n_prev, int n_new) struct perf_event *event; int i, n, first; + if (!(sparc_pmu->flags & SPARC_PMU_ALL_EXCLUDES_SAME)) + return 0; + n = n_prev + n_new; if (n <= 1) return 0; @@ -1059,7 +1364,7 @@ static int sparc_pmu_add(struct perf_event *event, int ef_flags) perf_pmu_disable(event->pmu); n0 = cpuc->n_events; - if (n0 >= MAX_HWEVENTS) + if (n0 >= sparc_pmu->max_hw_events) goto out; cpuc->event[n0] = event; @@ -1146,16 +1451,16 @@ static int sparc_pmu_event_init(struct perf_event *event) /* We save the enable bits in the config_base. */ hwc->config_base = sparc_pmu->irq_bit; if (!attr->exclude_user) - hwc->config_base |= PCR_UTRACE; + hwc->config_base |= sparc_pmu->user_bit; if (!attr->exclude_kernel) - hwc->config_base |= PCR_STRACE; + hwc->config_base |= sparc_pmu->priv_bit; if (!attr->exclude_hv) hwc->config_base |= sparc_pmu->hv_bit; n = 0; if (event->group_leader != event) { n = collect_events(event->group_leader, - MAX_HWEVENTS - 1, + sparc_pmu->max_hw_events - 1, evts, events, current_idx_dmy); if (n < 0) return -EINVAL; @@ -1254,8 +1559,7 @@ static struct pmu pmu = { void perf_event_print_debug(void) { unsigned long flags; - u64 pcr, pic; - int cpu; + int cpu, i; if (!sparc_pmu) return; @@ -1264,12 +1568,13 @@ void perf_event_print_debug(void) cpu = smp_processor_id(); - pcr = pcr_ops->read(); - read_pic(pic); - pr_info("\n"); - pr_info("CPU#%d: PCR[%016llx] PIC[%016llx]\n", - cpu, pcr, pic); + for (i = 0; i < sparc_pmu->num_pcrs; i++) + pr_info("CPU#%d: PCR%d[%016llx]\n", + cpu, i, pcr_ops->read_pcr(i)); + for (i = 0; i < sparc_pmu->num_pic_regs; i++) + pr_info("CPU#%d: PIC%d[%016llx]\n", + cpu, i, pcr_ops->read_pic(i)); local_irq_restore(flags); } @@ -1305,8 +1610,9 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, * Do this before we peek at the counters to determine * overflow so we don't lose any events. */ - if (sparc_pmu->irq_bit) - pcr_ops->write(cpuc->pcr); + if (sparc_pmu->irq_bit && + sparc_pmu->num_pcrs == 1) + pcr_ops->write_pcr(0, cpuc->pcr[0]); for (i = 0; i < cpuc->n_events; i++) { struct perf_event *event = cpuc->event[i]; @@ -1314,6 +1620,10 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, struct hw_perf_event *hwc; u64 val; + if (sparc_pmu->irq_bit && + sparc_pmu->num_pcrs > 1) + pcr_ops->write_pcr(idx, cpuc->pcr[idx]); + hwc = &event->hw; val = sparc_perf_event_update(event, hwc, idx); if (val & (1ULL << 31)) @@ -1352,10 +1662,14 @@ static bool __init supported_pmu(void) sparc_pmu = &niagara2_pmu; return true; } + if (!strcmp(sparc_pmu_type, "niagara4")) { + sparc_pmu = &niagara4_pmu; + return true; + } return false; } -int __init init_hw_perf_events(void) +static int __init init_hw_perf_events(void) { pr_info("Performance events: "); @@ -1426,14 +1740,13 @@ static void perf_callchain_user_64(struct perf_callchain_entry *entry, { unsigned long ufp; - perf_callchain_store(entry, regs->tpc); - ufp = regs->u_regs[UREG_I6] + STACK_BIAS; do { - struct sparc_stackf *usf, sf; + struct sparc_stackf __user *usf; + struct sparc_stackf sf; unsigned long pc; - usf = (struct sparc_stackf *) ufp; + usf = (struct sparc_stackf __user *)ufp; if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) break; @@ -1448,19 +1761,29 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry, { unsigned long ufp; - perf_callchain_store(entry, regs->tpc); - ufp = regs->u_regs[UREG_I6] & 0xffffffffUL; do { - struct sparc_stackf32 *usf, sf; unsigned long pc; - usf = (struct sparc_stackf32 *) ufp; - if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) - break; + if (thread32_stack_is_64bit(ufp)) { + struct sparc_stackf __user *usf; + struct sparc_stackf sf; - pc = sf.callers_pc; - ufp = (unsigned long)sf.fp; + ufp += STACK_BIAS; + usf = (struct sparc_stackf __user *)ufp; + if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) + break; + pc = sf.callers_pc & 0xffffffff; + ufp = ((unsigned long) sf.fp) & 0xffffffff; + } else { + struct sparc_stackf32 __user *usf; + struct sparc_stackf32 sf; + usf = (struct sparc_stackf32 __user *)ufp; + if (__copy_from_user_inatomic(&sf, usf, sizeof(sf))) + break; + pc = sf.callers_pc; + ufp = (unsigned long)sf.fp; + } perf_callchain_store(entry, pc); } while (entry->nr < PERF_MAX_STACK_DEPTH); } @@ -1468,6 +1791,11 @@ static void perf_callchain_user_32(struct perf_callchain_entry *entry, void perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs) { + perf_callchain_store(entry, regs->tpc); + + if (!current->mm) + return; + flushw_user(); if (test_thread_flag(TIF_32BIT)) perf_callchain_user_32(entry, regs); diff --git a/arch/sparc/kernel/pmc.c b/arch/sparc/kernel/pmc.c index 0e3202239ff..8b7297faca7 100644 --- a/arch/sparc/kernel/pmc.c +++ b/arch/sparc/kernel/pmc.c @@ -17,6 +17,7 @@ #include <asm/oplib.h> #include <asm/uaccess.h> #include <asm/auxio.h> +#include <asm/processor.h> /* Debug * @@ -52,7 +53,7 @@ static void pmc_swift_idle(void) #endif } -static int __devinit pmc_probe(struct platform_device *op) +static int pmc_probe(struct platform_device *op) { regs = of_ioremap(&op->resource[0], 0, resource_size(&op->resource[0]), PMC_OBPNAME); @@ -63,7 +64,7 @@ static int __devinit pmc_probe(struct platform_device *op) #ifndef PMC_NO_IDLE /* Assign power management IDLE handler */ - pm_idle = pmc_swift_idle; + sparc_idle = pmc_swift_idle; #endif printk(KERN_INFO "%s: power management initialized\n", PMC_DEVNAME); diff --git a/arch/sparc/kernel/power.c b/arch/sparc/kernel/power.c index 0d39075063b..4cb23c41553 100644 --- a/arch/sparc/kernel/power.c +++ b/arch/sparc/kernel/power.c @@ -23,7 +23,7 @@ static irqreturn_t power_handler(int irq, void *dev_id) return IRQ_HANDLED; } -static int __devinit has_button_interrupt(unsigned int irq, struct device_node *dp) +static int has_button_interrupt(unsigned int irq, struct device_node *dp) { if (irq == 0xffffffff) return 0; @@ -33,7 +33,7 @@ static int __devinit has_button_interrupt(unsigned int irq, struct device_node * return 1; } -static int __devinit power_probe(struct platform_device *op) +static int power_probe(struct platform_device *op) { struct resource *res = &op->resource[0]; unsigned int irq = op->archdata.irqs[0]; diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c index cb36e82dcd5..50e7b626afe 100644 --- a/arch/sparc/kernel/process_32.c +++ b/arch/sparc/kernel/process_32.c @@ -10,6 +10,7 @@ #include <stdarg.h> +#include <linux/elfcore.h> #include <linux/errno.h> #include <linux/module.h> #include <linux/sched.h> @@ -22,8 +23,8 @@ #include <linux/reboot.h> #include <linux/delay.h> #include <linux/pm.h> -#include <linux/init.h> #include <linux/slab.h> +#include <linux/cpu.h> #include <asm/auxio.h> #include <asm/oplib.h> @@ -39,12 +40,13 @@ #include <asm/unistd.h> #include <asm/setup.h> +#include "kernel.h" + /* * Power management idle function * Set in pm platform drivers (apc.c and pmc.c) */ -void (*pm_idle)(void); -EXPORT_SYMBOL(pm_idle); +void (*sparc_idle)(void); /* * Power-off handler instantiation for pm.h compliance @@ -65,23 +67,12 @@ extern void fpsave(unsigned long *, unsigned long *, void *, unsigned long *); struct task_struct *last_task_used_math = NULL; struct thread_info *current_set[NR_CPUS]; -/* - * the idle loop on a Sparc... ;) - */ -void cpu_idle(void) +/* Idle loop support. */ +void arch_cpu_idle(void) { - set_thread_flag(TIF_POLLING_NRFLAG); - - /* endless idle loop with no priority at all */ - for (;;) { - while (!need_resched()) { - if (pm_idle) - (*pm_idle)(); - else - cpu_relax(); - } - schedule_preempt_disabled(); - } + if (sparc_idle) + (*sparc_idle)(); + local_irq_enable(); } /* XXX cli/sti -> local_irq_xxx here, check this works once SMP is fixed. */ @@ -115,8 +106,12 @@ void machine_restart(char * cmd) void machine_power_off(void) { if (auxio_power_register && - (strcmp(of_console_device->type, "serial") || scons_pwroff)) - *auxio_power_register |= AUXIO_POWER_OFF; + (strcmp(of_console_device->type, "serial") || scons_pwroff)) { + u8 power_register = sbus_readb(auxio_power_register); + power_register |= AUXIO_POWER_OFF; + sbus_writeb(power_register, auxio_power_register); + } + machine_halt(); } @@ -124,6 +119,8 @@ void show_regs(struct pt_regs *r) { struct reg_window32 *rw = (struct reg_window32 *) r->u_regs[14]; + show_regs_print_info(KERN_DEFAULT); + printk("PSR: %08lx PC: %08lx NPC: %08lx Y: %08lx %s\n", r->psr, r->pc, r->npc, r->y, print_tainted()); printk("PC: <%pS>\n", (void *) r->pc); @@ -154,11 +151,13 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp) struct reg_window32 *rw; int count = 0; - if (tsk != NULL) - task_base = (unsigned long) task_stack_page(tsk); - else - task_base = (unsigned long) current_thread_info(); + if (!tsk) + tsk = current; + if (tsk == current && !_ksp) + __asm__ __volatile__("mov %%fp, %0" : "=r" (_ksp)); + + task_base = (unsigned long) task_stack_page(tsk); fp = (unsigned long) _ksp; do { /* Bogus frame pointer? */ @@ -174,17 +173,6 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp) printk("\n"); } -void dump_stack(void) -{ - unsigned long *ksp; - - __asm__ __volatile__("mov %%fp, %0" - : "=r" (ksp)); - show_stack(current, ksp); -} - -EXPORT_SYMBOL(dump_stack); - /* * Note: sparc64 has a pretty intricated thread_saved_pc, check it out. */ @@ -286,8 +274,7 @@ asmlinkage int sparc_do_fork(unsigned long clone_flags, parent_tid_ptr = regs->u_regs[UREG_I2]; child_tid_ptr = regs->u_regs[UREG_I4]; - ret = do_fork(clone_flags, stack_start, - regs, stack_size, + ret = do_fork(clone_flags, stack_start, stack_size, (int __user *) parent_tid_ptr, (int __user *) child_tid_ptr); @@ -316,13 +303,13 @@ asmlinkage int sparc_do_fork(unsigned long clone_flags, * XXX See comment above sys_vfork in sparc64. todo. */ extern void ret_from_fork(void); +extern void ret_from_kernel_thread(void); int copy_thread(unsigned long clone_flags, unsigned long sp, - unsigned long unused, - struct task_struct *p, struct pt_regs *regs) + unsigned long arg, struct task_struct *p) { struct thread_info *ti = task_thread_info(p); - struct pt_regs *childregs; + struct pt_regs *childregs, *regs = current_pt_regs(); char *new_stack; #ifndef CONFIG_SMP @@ -333,22 +320,16 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, put_psr(get_psr() | PSR_EF); fpsave(&p->thread.float_regs[0], &p->thread.fsr, &p->thread.fpqueue[0], &p->thread.fpqdepth); -#ifdef CONFIG_SMP - clear_thread_flag(TIF_USEDFPU); -#endif } /* - * p->thread_info new_stack childregs - * ! ! ! {if(PSR_PS) } - * V V (stk.fr.) V (pt_regs) { (stk.fr.) } - * +----- - - - - - ------+===========+============={+==========}+ + * p->thread_info new_stack childregs stack bottom + * ! ! ! ! + * V V (stk.fr.) V (pt_regs) V + * +----- - - - - - ------+===========+=============+ */ new_stack = task_stack_page(p) + THREAD_SIZE; - if (regs->psr & PSR_PS) - new_stack -= STACKFRAME_SZ; new_stack -= STACKFRAME_SZ + TRACEREG_SZ; - memcpy(new_stack, (char *)regs - STACKFRAME_SZ, STACKFRAME_SZ + TRACEREG_SZ); childregs = (struct pt_regs *) (new_stack + STACKFRAME_SZ); /* @@ -359,60 +340,64 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, * Thus, kpsr|=PSR_PIL. */ ti->ksp = (unsigned long) new_stack; + p->thread.kregs = childregs; + + if (unlikely(p->flags & PF_KTHREAD)) { + extern int nwindows; + unsigned long psr; + memset(new_stack, 0, STACKFRAME_SZ + TRACEREG_SZ); + p->thread.flags |= SPARC_FLAG_KTHREAD; + p->thread.current_ds = KERNEL_DS; + ti->kpc = (((unsigned long) ret_from_kernel_thread) - 0x8); + childregs->u_regs[UREG_G1] = sp; /* function */ + childregs->u_regs[UREG_G2] = arg; + psr = childregs->psr = get_psr(); + ti->kpsr = psr | PSR_PIL; + ti->kwim = 1 << (((psr & PSR_CWP) + 1) % nwindows); + return 0; + } + memcpy(new_stack, (char *)regs - STACKFRAME_SZ, STACKFRAME_SZ + TRACEREG_SZ); + childregs->u_regs[UREG_FP] = sp; + p->thread.flags &= ~SPARC_FLAG_KTHREAD; + p->thread.current_ds = USER_DS; ti->kpc = (((unsigned long) ret_from_fork) - 0x8); ti->kpsr = current->thread.fork_kpsr | PSR_PIL; ti->kwim = current->thread.fork_kwim; - if(regs->psr & PSR_PS) { - extern struct pt_regs fake_swapper_regs; + if (sp != regs->u_regs[UREG_FP]) { + struct sparc_stackf __user *childstack; + struct sparc_stackf __user *parentstack; - p->thread.kregs = &fake_swapper_regs; - new_stack += STACKFRAME_SZ + TRACEREG_SZ; - childregs->u_regs[UREG_FP] = (unsigned long) new_stack; - p->thread.flags |= SPARC_FLAG_KTHREAD; - p->thread.current_ds = KERNEL_DS; - memcpy(new_stack, (void *)regs->u_regs[UREG_FP], STACKFRAME_SZ); - childregs->u_regs[UREG_G6] = (unsigned long) ti; - } else { - p->thread.kregs = childregs; - childregs->u_regs[UREG_FP] = sp; - p->thread.flags &= ~SPARC_FLAG_KTHREAD; - p->thread.current_ds = USER_DS; - - if (sp != regs->u_regs[UREG_FP]) { - struct sparc_stackf __user *childstack; - struct sparc_stackf __user *parentstack; - - /* - * This is a clone() call with supplied user stack. - * Set some valid stack frames to give to the child. - */ - childstack = (struct sparc_stackf __user *) - (sp & ~0xfUL); - parentstack = (struct sparc_stackf __user *) - regs->u_regs[UREG_FP]; + /* + * This is a clone() call with supplied user stack. + * Set some valid stack frames to give to the child. + */ + childstack = (struct sparc_stackf __user *) + (sp & ~0xfUL); + parentstack = (struct sparc_stackf __user *) + regs->u_regs[UREG_FP]; #if 0 - printk("clone: parent stack:\n"); - show_stackframe(parentstack); + printk("clone: parent stack:\n"); + show_stackframe(parentstack); #endif - childstack = clone_stackframe(childstack, parentstack); - if (!childstack) - return -EFAULT; + childstack = clone_stackframe(childstack, parentstack); + if (!childstack) + return -EFAULT; #if 0 - printk("clone: child stack:\n"); - show_stackframe(childstack); + printk("clone: child stack:\n"); + show_stackframe(childstack); #endif - childregs->u_regs[UREG_FP] = (unsigned long)childstack; - } + childregs->u_regs[UREG_FP] = (unsigned long)childstack; } #ifdef CONFIG_SMP /* FPU must be disabled on SMP. */ childregs->psr &= ~PSR_EF; + clear_tsk_thread_flag(p, TIF_USEDFPU); #endif /* Set the return value for the child. */ @@ -477,69 +462,6 @@ int dump_fpu (struct pt_regs * regs, elf_fpregset_t * fpregs) return 1; } -/* - * sparc_execve() executes a new program after the asm stub has set - * things up for us. This should basically do what I want it to. - */ -asmlinkage int sparc_execve(struct pt_regs *regs) -{ - int error, base = 0; - char *filename; - - /* Check for indirect call. */ - if(regs->u_regs[UREG_G1] == 0) - base = 1; - - filename = getname((char __user *)regs->u_regs[base + UREG_I0]); - error = PTR_ERR(filename); - if(IS_ERR(filename)) - goto out; - error = do_execve(filename, - (const char __user *const __user *) - regs->u_regs[base + UREG_I1], - (const char __user *const __user *) - regs->u_regs[base + UREG_I2], - regs); - putname(filename); -out: - return error; -} - -/* - * This is the mechanism for creating a new kernel thread. - * - * NOTE! Only a kernel-only process(ie the swapper or direct descendants - * who haven't done an "execve()") should use this: it will work within - * a system call from a "real" process, but the process memory space will - * not be freed until both the parent and the child have exited. - */ -pid_t kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) -{ - long retval; - - __asm__ __volatile__("mov %4, %%g2\n\t" /* Set aside fn ptr... */ - "mov %5, %%g3\n\t" /* and arg. */ - "mov %1, %%g1\n\t" - "mov %2, %%o0\n\t" /* Clone flags. */ - "mov 0, %%o1\n\t" /* usp arg == 0 */ - "t 0x10\n\t" /* Linux/Sparc clone(). */ - "cmp %%o1, 0\n\t" - "be 1f\n\t" /* The parent, just return. */ - " nop\n\t" /* Delay slot. */ - "jmpl %%g2, %%o7\n\t" /* Call the function. */ - " mov %%g3, %%o0\n\t" /* Get back the arg in delay. */ - "mov %3, %%g1\n\t" - "t 0x10\n\t" /* Linux/Sparc exit(). */ - /* Notreached by child. */ - "1: mov %%o0, %0\n\t" : - "=r" (retval) : - "i" (__NR_clone), "r" (flags | CLONE_VM | CLONE_UNTRACED), - "i" (__NR_exit), "r" (fn), "r" (arg) : - "g1", "g2", "g3", "o0", "o1", "memory", "cc"); - return retval; -} -EXPORT_SYMBOL(kernel_thread); - unsigned long get_wchan(struct task_struct *task) { unsigned long pc, fp, bias = 0; diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c index aff0c72fac0..027e0998619 100644 --- a/arch/sparc/kernel/process_64.c +++ b/arch/sparc/kernel/process_64.c @@ -27,9 +27,11 @@ #include <linux/tick.h> #include <linux/init.h> #include <linux/cpu.h> +#include <linux/perf_event.h> #include <linux/elfcore.h> #include <linux/sysrq.h> #include <linux/nmi.h> +#include <linux/context_tracking.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -47,23 +49,24 @@ #include <asm/syscalls.h> #include <asm/irq_regs.h> #include <asm/smp.h> +#include <asm/pcr.h> #include "kstack.h" -static void sparc64_yield(int cpu) +/* Idle loop support on sparc64. */ +void arch_cpu_idle(void) { if (tlb_type != hypervisor) { touch_nmi_watchdog(); - return; - } - - clear_thread_flag(TIF_POLLING_NRFLAG); - smp_mb__after_clear_bit(); - - while (!need_resched() && !cpu_is_offline(cpu)) { + local_irq_enable(); + } else { unsigned long pstate; - /* Disable interrupts. */ + local_irq_enable(); + + /* The sun4v sleeping code requires that we have PSTATE.IE cleared over + * the cpu sleep hypervisor call. + */ __asm__ __volatile__( "rdpr %%pstate, %0\n\t" "andn %0, %1, %0\n\t" @@ -71,7 +74,7 @@ static void sparc64_yield(int cpu) : "=&r" (pstate) : "i" (PSTATE_IE)); - if (!need_resched() && !cpu_is_offline(cpu)) + if (!need_resched() && !cpu_is_offline(smp_processor_id())) sun4v_cpu_yield(); /* Re-enable interrupts. */ @@ -82,36 +85,15 @@ static void sparc64_yield(int cpu) : "=&r" (pstate) : "i" (PSTATE_IE)); } - - set_thread_flag(TIF_POLLING_NRFLAG); } -/* The idle loop on sparc64. */ -void cpu_idle(void) -{ - int cpu = smp_processor_id(); - - set_thread_flag(TIF_POLLING_NRFLAG); - - while(1) { - tick_nohz_idle_enter(); - rcu_idle_enter(); - - while (!need_resched() && !cpu_is_offline(cpu)) - sparc64_yield(cpu); - - rcu_idle_exit(); - tick_nohz_idle_exit(); - #ifdef CONFIG_HOTPLUG_CPU - if (cpu_is_offline(cpu)) { - sched_preempt_enable_no_resched(); - cpu_play_dead(); - } -#endif - schedule_preempt_disabled(); - } +void arch_cpu_idle_dead(void) +{ + sched_preempt_enable_no_resched(); + cpu_play_dead(); } +#endif #ifdef CONFIG_COMPAT static void show_regwindow32(struct pt_regs *regs) @@ -184,6 +166,8 @@ static void show_regwindow(struct pt_regs *regs) void show_regs(struct pt_regs *regs) { + show_regs_print_info(KERN_DEFAULT); + printk("TSTATE: %016lx TPC: %016lx TNPC: %016lx Y: %08x %s\n", regs->tstate, regs->tpc, regs->tnpc, regs->y, print_tainted()); printk("TPC: <%pS>\n", (void *) regs->tpc); @@ -204,18 +188,22 @@ void show_regs(struct pt_regs *regs) show_stack(current, (unsigned long *) regs->u_regs[UREG_FP]); } -struct global_reg_snapshot global_reg_snapshot[NR_CPUS]; -static DEFINE_SPINLOCK(global_reg_snapshot_lock); +union global_cpu_snapshot global_cpu_snapshot[NR_CPUS]; +static DEFINE_SPINLOCK(global_cpu_snapshot_lock); static void __global_reg_self(struct thread_info *tp, struct pt_regs *regs, int this_cpu) { + struct global_reg_snapshot *rp; + flushw_all(); - global_reg_snapshot[this_cpu].tstate = regs->tstate; - global_reg_snapshot[this_cpu].tpc = regs->tpc; - global_reg_snapshot[this_cpu].tnpc = regs->tnpc; - global_reg_snapshot[this_cpu].o7 = regs->u_regs[UREG_I7]; + rp = &global_cpu_snapshot[this_cpu].reg; + + rp->tstate = regs->tstate; + rp->tpc = regs->tpc; + rp->tnpc = regs->tnpc; + rp->o7 = regs->u_regs[UREG_I7]; if (regs->tstate & TSTATE_PRIV) { struct reg_window *rw; @@ -223,17 +211,17 @@ static void __global_reg_self(struct thread_info *tp, struct pt_regs *regs, rw = (struct reg_window *) (regs->u_regs[UREG_FP] + STACK_BIAS); if (kstack_valid(tp, (unsigned long) rw)) { - global_reg_snapshot[this_cpu].i7 = rw->ins[7]; + rp->i7 = rw->ins[7]; rw = (struct reg_window *) (rw->ins[6] + STACK_BIAS); if (kstack_valid(tp, (unsigned long) rw)) - global_reg_snapshot[this_cpu].rpc = rw->ins[7]; + rp->rpc = rw->ins[7]; } } else { - global_reg_snapshot[this_cpu].i7 = 0; - global_reg_snapshot[this_cpu].rpc = 0; + rp->i7 = 0; + rp->rpc = 0; } - global_reg_snapshot[this_cpu].thread = tp; + rp->thread = tp; } /* In order to avoid hangs we do not try to synchronize with the @@ -251,7 +239,7 @@ static void __global_reg_poll(struct global_reg_snapshot *gp) } } -void arch_trigger_all_cpu_backtrace(void) +void arch_trigger_all_cpu_backtrace(bool include_self) { struct thread_info *tp = current_thread_info(); struct pt_regs *regs = get_irq_regs(); @@ -261,18 +249,24 @@ void arch_trigger_all_cpu_backtrace(void) if (!regs) regs = tp->kregs; - spin_lock_irqsave(&global_reg_snapshot_lock, flags); - - memset(global_reg_snapshot, 0, sizeof(global_reg_snapshot)); + spin_lock_irqsave(&global_cpu_snapshot_lock, flags); this_cpu = raw_smp_processor_id(); - __global_reg_self(tp, regs, this_cpu); + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); + + if (include_self) + __global_reg_self(tp, regs, this_cpu); smp_fetch_global_regs(); for_each_online_cpu(cpu) { - struct global_reg_snapshot *gp = &global_reg_snapshot[cpu]; + struct global_reg_snapshot *gp; + + if (!include_self && cpu == this_cpu) + continue; + + gp = &global_cpu_snapshot[cpu].reg; __global_reg_poll(gp); @@ -295,30 +289,104 @@ void arch_trigger_all_cpu_backtrace(void) } } - memset(global_reg_snapshot, 0, sizeof(global_reg_snapshot)); + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); - spin_unlock_irqrestore(&global_reg_snapshot_lock, flags); + spin_unlock_irqrestore(&global_cpu_snapshot_lock, flags); } #ifdef CONFIG_MAGIC_SYSRQ static void sysrq_handle_globreg(int key) { - arch_trigger_all_cpu_backtrace(); + arch_trigger_all_cpu_backtrace(true); } static struct sysrq_key_op sparc_globalreg_op = { .handler = sysrq_handle_globreg, - .help_msg = "Globalregs", + .help_msg = "global-regs(y)", .action_msg = "Show Global CPU Regs", }; -static int __init sparc_globreg_init(void) +static void __global_pmu_self(int this_cpu) +{ + struct global_pmu_snapshot *pp; + int i, num; + + pp = &global_cpu_snapshot[this_cpu].pmu; + + num = 1; + if (tlb_type == hypervisor && + sun4v_chip_type >= SUN4V_CHIP_NIAGARA4) + num = 4; + + for (i = 0; i < num; i++) { + pp->pcr[i] = pcr_ops->read_pcr(i); + pp->pic[i] = pcr_ops->read_pic(i); + } +} + +static void __global_pmu_poll(struct global_pmu_snapshot *pp) +{ + int limit = 0; + + while (!pp->pcr[0] && ++limit < 100) { + barrier(); + udelay(1); + } +} + +static void pmu_snapshot_all_cpus(void) +{ + unsigned long flags; + int this_cpu, cpu; + + spin_lock_irqsave(&global_cpu_snapshot_lock, flags); + + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); + + this_cpu = raw_smp_processor_id(); + + __global_pmu_self(this_cpu); + + smp_fetch_global_pmu(); + + for_each_online_cpu(cpu) { + struct global_pmu_snapshot *pp = &global_cpu_snapshot[cpu].pmu; + + __global_pmu_poll(pp); + + printk("%c CPU[%3d]: PCR[%08lx:%08lx:%08lx:%08lx] PIC[%08lx:%08lx:%08lx:%08lx]\n", + (cpu == this_cpu ? '*' : ' '), cpu, + pp->pcr[0], pp->pcr[1], pp->pcr[2], pp->pcr[3], + pp->pic[0], pp->pic[1], pp->pic[2], pp->pic[3]); + } + + memset(global_cpu_snapshot, 0, sizeof(global_cpu_snapshot)); + + spin_unlock_irqrestore(&global_cpu_snapshot_lock, flags); +} + +static void sysrq_handle_globpmu(int key) +{ + pmu_snapshot_all_cpus(); +} + +static struct sysrq_key_op sparc_globalpmu_op = { + .handler = sysrq_handle_globpmu, + .help_msg = "global-pmu(x)", + .action_msg = "Show Global PMU Regs", +}; + +static int __init sparc_sysrq_init(void) { - return register_sysrq_key('y', &sparc_globalreg_op); + int ret = register_sysrq_key('y', &sparc_globalreg_op); + + if (!ret) + ret = register_sysrq_key('x', &sparc_globalpmu_op); + return ret; } -core_initcall(sparc_globreg_init); +core_initcall(sparc_sysrq_init); #endif @@ -372,13 +440,16 @@ void flush_thread(void) /* It's a bit more tricky when 64-bit tasks are involved... */ static unsigned long clone_stackframe(unsigned long csp, unsigned long psp) { + bool stack_64bit = test_thread_64bit_stack(psp); unsigned long fp, distance, rval; - if (!(test_thread_flag(TIF_32BIT))) { + if (stack_64bit) { csp += STACK_BIAS; psp += STACK_BIAS; __get_user(fp, &(((struct reg_window __user *)psp)->ins[6])); fp += STACK_BIAS; + if (test_thread_flag(TIF_32BIT)) + fp &= 0xffffffff; } else __get_user(fp, &(((struct reg_window32 __user *)psp)->ins[6])); @@ -392,7 +463,7 @@ static unsigned long clone_stackframe(unsigned long csp, unsigned long psp) rval = (csp - distance); if (copy_in_user((void __user *) rval, (void __user *) psp, distance)) rval = 0; - else if (test_thread_flag(TIF_32BIT)) { + else if (!stack_64bit) { if (put_user(((u32)csp), &(((struct reg_window32 __user *)rval)->ins[6]))) rval = 0; @@ -427,18 +498,18 @@ void synchronize_user_stack(void) flush_user_windows(); if ((window = get_thread_wsaved()) != 0) { - int winsize = sizeof(struct reg_window); - int bias = 0; - - if (test_thread_flag(TIF_32BIT)) - winsize = sizeof(struct reg_window32); - else - bias = STACK_BIAS; - window -= 1; do { - unsigned long sp = (t->rwbuf_stkptrs[window] + bias); struct reg_window *rwin = &t->reg_window[window]; + int winsize = sizeof(struct reg_window); + unsigned long sp; + + sp = t->rwbuf_stkptrs[window]; + + if (test_thread_64bit_stack(sp)) + sp += STACK_BIAS; + else + winsize = sizeof(struct reg_window32); if (!copy_to_user((char __user *)sp, rwin, winsize)) { shift_window_buffer(window, get_thread_wsaved() - 1, t); @@ -464,13 +535,6 @@ void fault_in_user_windows(void) { struct thread_info *t = current_thread_info(); unsigned long window; - int winsize = sizeof(struct reg_window); - int bias = 0; - - if (test_thread_flag(TIF_32BIT)) - winsize = sizeof(struct reg_window32); - else - bias = STACK_BIAS; flush_user_windows(); window = get_thread_wsaved(); @@ -478,8 +542,16 @@ void fault_in_user_windows(void) if (likely(window != 0)) { window -= 1; do { - unsigned long sp = (t->rwbuf_stkptrs[window] + bias); struct reg_window *rwin = &t->reg_window[window]; + int winsize = sizeof(struct reg_window); + unsigned long sp; + + sp = t->rwbuf_stkptrs[window]; + + if (test_thread_64bit_stack(sp)) + sp += STACK_BIAS; + else + winsize = sizeof(struct reg_window32); if (unlikely(sp & 0x7UL)) stack_unaligned(sp); @@ -494,6 +566,7 @@ void fault_in_user_windows(void) barf: set_thread_wsaved(window + 1); + user_exit(); do_exit(SIGILL); } @@ -517,8 +590,7 @@ asmlinkage long sparc_do_fork(unsigned long clone_flags, child_tid_ptr = (int __user *) regs->u_regs[UREG_I4]; } - ret = do_fork(clone_flags, stack_start, - regs, stack_size, + ret = do_fork(clone_flags, stack_start, stack_size, parent_tid_ptr, child_tid_ptr); /* If we get an error and potentially restart the system @@ -538,64 +610,55 @@ asmlinkage long sparc_do_fork(unsigned long clone_flags, * Child --> %o0 == parents pid, %o1 == 1 */ int copy_thread(unsigned long clone_flags, unsigned long sp, - unsigned long unused, - struct task_struct *p, struct pt_regs *regs) + unsigned long arg, struct task_struct *p) { struct thread_info *t = task_thread_info(p); + struct pt_regs *regs = current_pt_regs(); struct sparc_stackf *parent_sf; unsigned long child_stack_sz; char *child_trap_frame; - int kernel_thread; - - kernel_thread = (regs->tstate & TSTATE_PRIV) ? 1 : 0; - parent_sf = ((struct sparc_stackf *) regs) - 1; /* Calculate offset to stack_frame & pt_regs */ - child_stack_sz = ((STACKFRAME_SZ + TRACEREG_SZ) + - (kernel_thread ? STACKFRAME_SZ : 0)); + child_stack_sz = (STACKFRAME_SZ + TRACEREG_SZ); child_trap_frame = (task_stack_page(p) + (THREAD_SIZE - child_stack_sz)); - memcpy(child_trap_frame, parent_sf, child_stack_sz); - t->flags = (t->flags & ~((0xffUL << TI_FLAG_CWP_SHIFT) | - (0xffUL << TI_FLAG_CURRENT_DS_SHIFT))) | - (((regs->tstate + 1) & TSTATE_CWP) << TI_FLAG_CWP_SHIFT); t->new_child = 1; t->ksp = ((unsigned long) child_trap_frame) - STACK_BIAS; t->kregs = (struct pt_regs *) (child_trap_frame + sizeof(struct sparc_stackf)); t->fpsaved[0] = 0; - if (kernel_thread) { - struct sparc_stackf *child_sf = (struct sparc_stackf *) - (child_trap_frame + (STACKFRAME_SZ + TRACEREG_SZ)); - - /* Zero terminate the stack backtrace. */ - child_sf->fp = NULL; - t->kregs->u_regs[UREG_FP] = - ((unsigned long) child_sf) - STACK_BIAS; + if (unlikely(p->flags & PF_KTHREAD)) { + memset(child_trap_frame, 0, child_stack_sz); + __thread_flag_byte_ptr(t)[TI_FLAG_BYTE_CWP] = + (current_pt_regs()->tstate + 1) & TSTATE_CWP; + t->current_ds = ASI_P; + t->kregs->u_regs[UREG_G1] = sp; /* function */ + t->kregs->u_regs[UREG_G2] = arg; + return 0; + } - t->flags |= ((long)ASI_P << TI_FLAG_CURRENT_DS_SHIFT); - t->kregs->u_regs[UREG_G6] = (unsigned long) t; - t->kregs->u_regs[UREG_G4] = (unsigned long) t->task; - } else { - if (t->flags & _TIF_32BIT) { - sp &= 0x00000000ffffffffUL; - regs->u_regs[UREG_FP] &= 0x00000000ffffffffUL; - } - t->kregs->u_regs[UREG_FP] = sp; - t->flags |= ((long)ASI_AIUS << TI_FLAG_CURRENT_DS_SHIFT); - if (sp != regs->u_regs[UREG_FP]) { - unsigned long csp; - - csp = clone_stackframe(sp, regs->u_regs[UREG_FP]); - if (!csp) - return -EFAULT; - t->kregs->u_regs[UREG_FP] = csp; - } - if (t->utraps) - t->utraps[0]++; + parent_sf = ((struct sparc_stackf *) regs) - 1; + memcpy(child_trap_frame, parent_sf, child_stack_sz); + if (t->flags & _TIF_32BIT) { + sp &= 0x00000000ffffffffUL; + regs->u_regs[UREG_FP] &= 0x00000000ffffffffUL; + } + t->kregs->u_regs[UREG_FP] = sp; + __thread_flag_byte_ptr(t)[TI_FLAG_BYTE_CWP] = + (regs->tstate + 1) & TSTATE_CWP; + t->current_ds = ASI_AIUS; + if (sp != regs->u_regs[UREG_FP]) { + unsigned long csp; + + csp = clone_stackframe(sp, regs->u_regs[UREG_FP]); + if (!csp) + return -EFAULT; + t->kregs->u_regs[UREG_FP] = csp; } + if (t->utraps) + t->utraps[0]++; /* Set the return value for the child. */ t->kregs->u_regs[UREG_I0] = current->pid; @@ -610,45 +673,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, return 0; } -/* - * This is the mechanism for creating a new kernel thread. - * - * NOTE! Only a kernel-only process(ie the swapper or direct descendants - * who haven't done an "execve()") should use this: it will work within - * a system call from a "real" process, but the process memory space will - * not be freed until both the parent and the child have exited. - */ -pid_t kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) -{ - long retval; - - /* If the parent runs before fn(arg) is called by the child, - * the input registers of this function can be clobbered. - * So we stash 'fn' and 'arg' into global registers which - * will not be modified by the parent. - */ - __asm__ __volatile__("mov %4, %%g2\n\t" /* Save FN into global */ - "mov %5, %%g3\n\t" /* Save ARG into global */ - "mov %1, %%g1\n\t" /* Clone syscall nr. */ - "mov %2, %%o0\n\t" /* Clone flags. */ - "mov 0, %%o1\n\t" /* usp arg == 0 */ - "t 0x6d\n\t" /* Linux/Sparc clone(). */ - "brz,a,pn %%o1, 1f\n\t" /* Parent, just return. */ - " mov %%o0, %0\n\t" - "jmpl %%g2, %%o7\n\t" /* Call the function. */ - " mov %%g3, %%o0\n\t" /* Set arg in delay. */ - "mov %3, %%g1\n\t" - "t 0x6d\n\t" /* Linux/Sparc exit(). */ - /* Notreached by child. */ - "1:" : - "=r" (retval) : - "i" (__NR_clone), "r" (flags | CLONE_VM | CLONE_UNTRACED), - "i" (__NR_exit), "r" (fn), "r" (arg) : - "g1", "g2", "g3", "o0", "o1", "memory", "cc"); - return retval; -} -EXPORT_SYMBOL(kernel_thread); - typedef struct { union { unsigned int pr_regs[32]; @@ -715,41 +739,6 @@ int dump_fpu (struct pt_regs * regs, elf_fpregset_t * fpregs) } EXPORT_SYMBOL(dump_fpu); -/* - * sparc_execve() executes a new program after the asm stub has set - * things up for us. This should basically do what I want it to. - */ -asmlinkage int sparc_execve(struct pt_regs *regs) -{ - int error, base = 0; - char *filename; - - /* User register window flush is done by entry.S */ - - /* Check for indirect call. */ - if (regs->u_regs[UREG_G1] == 0) - base = 1; - - filename = getname((char __user *)regs->u_regs[base + UREG_I0]); - error = PTR_ERR(filename); - if (IS_ERR(filename)) - goto out; - error = do_execve(filename, - (const char __user *const __user *) - regs->u_regs[base + UREG_I1], - (const char __user *const __user *) - regs->u_regs[base + UREG_I2], regs); - putname(filename); - if (!error) { - fprs_write(0); - current_thread_info()->xfsr[0] = 0; - current_thread_info()->fpsaved[0] = 0; - regs->tstate &= ~TSTATE_PEF; - } -out: - return error; -} - unsigned long get_wchan(struct task_struct *task) { unsigned long pc, fp, bias = 0; diff --git a/arch/sparc/kernel/prom.h b/arch/sparc/kernel/prom.h index cf5fe1c0b02..890281b12b2 100644 --- a/arch/sparc/kernel/prom.h +++ b/arch/sparc/kernel/prom.h @@ -4,7 +4,7 @@ #include <linux/spinlock.h> #include <asm/prom.h> -extern void of_console_init(void); +void of_console_init(void); extern unsigned int prom_early_allocated; diff --git a/arch/sparc/kernel/prom_64.c b/arch/sparc/kernel/prom_64.c index 340c5b976d2..20cc5d80a47 100644 --- a/arch/sparc/kernel/prom_64.c +++ b/arch/sparc/kernel/prom_64.c @@ -15,11 +15,12 @@ * 2 of the License, or (at your option) any later version. */ +#include <linux/memblock.h> #include <linux/kernel.h> -#include <linux/types.h> #include <linux/string.h> +#include <linux/types.h> +#include <linux/cpu.h> #include <linux/mm.h> -#include <linux/memblock.h> #include <linux/of.h> #include <asm/prom.h> @@ -37,7 +38,7 @@ void * __init prom_early_alloc(unsigned long size) void *ret; if (!paddr) { - prom_printf("prom_early_alloc(%lu) failed\n"); + prom_printf("prom_early_alloc(%lu) failed\n", size); prom_halt(); } @@ -373,6 +374,59 @@ static const char *get_mid_prop(void) return (tlb_type == spitfire ? "upa-portid" : "portid"); } +bool arch_find_n_match_cpu_physical_id(struct device_node *cpun, + int cpu, unsigned int *thread) +{ + const char *mid_prop = get_mid_prop(); + int this_cpu_id; + + /* On hypervisor based platforms we interrogate the 'reg' + * property. On everything else we look for a 'upa-portis', + * 'portid', or 'cpuid' property. + */ + + if (tlb_type == hypervisor) { + struct property *prop = of_find_property(cpun, "reg", NULL); + u32 *regs; + + if (!prop) { + pr_warn("CPU node missing reg property\n"); + return false; + } + regs = prop->value; + this_cpu_id = regs[0] & 0x0fffffff; + } else { + this_cpu_id = of_getintprop_default(cpun, mid_prop, -1); + + if (this_cpu_id < 0) { + mid_prop = "cpuid"; + this_cpu_id = of_getintprop_default(cpun, mid_prop, -1); + } + if (this_cpu_id < 0) { + pr_warn("CPU node missing cpu ID property\n"); + return false; + } + } + if (this_cpu_id == cpu) { + if (thread) { + int proc_id = cpu_data(cpu).proc_id; + + /* On sparc64, the cpu thread information is obtained + * either from OBP or the machine description. We've + * actually probed this information already long before + * this interface gets called so instead of interrogating + * both the OF node and the MDESC again, just use what + * we discovered already. + */ + if (proc_id < 0) + proc_id = 0; + *thread = proc_id; + } + return true; + } + return false; +} + static void *of_iterate_over_cpus(void *(*func)(struct device_node *, int, int), int arg) { struct device_node *dp; @@ -502,9 +556,6 @@ static void *fill_in_one_cpu(struct device_node *dp, int cpuid, int arg) cpu_data(cpuid).core_id = portid + 1; cpu_data(cpuid).proc_id = portid; -#ifdef CONFIG_SMP - sparc64_multi_core = 1; -#endif } else { cpu_data(cpuid).dcache_size = of_getintprop_default(dp, "dcache-size", 16 * 1024); diff --git a/arch/sparc/kernel/prom_common.c b/arch/sparc/kernel/prom_common.c index 1303021748c..79cc0d1a477 100644 --- a/arch/sparc/kernel/prom_common.c +++ b/arch/sparc/kernel/prom_common.c @@ -54,6 +54,7 @@ EXPORT_SYMBOL(of_set_property_mutex); int of_set_property(struct device_node *dp, const char *name, void *val, int len) { struct property **prevp; + unsigned long flags; void *new_val; int err; @@ -64,7 +65,7 @@ int of_set_property(struct device_node *dp, const char *name, void *val, int len err = -ENODEV; mutex_lock(&of_set_property_mutex); - write_lock(&devtree_lock); + raw_spin_lock_irqsave(&devtree_lock, flags); prevp = &dp->properties; while (*prevp) { struct property *prop = *prevp; @@ -91,7 +92,7 @@ int of_set_property(struct device_node *dp, const char *name, void *val, int len } prevp = &(*prevp)->next; } - write_unlock(&devtree_lock); + raw_spin_unlock_irqrestore(&devtree_lock, flags); mutex_unlock(&of_set_property_mutex); /* XXX Upate procfs if necessary... */ diff --git a/arch/sparc/kernel/psycho_common.h b/arch/sparc/kernel/psycho_common.h index 590b4ed8ab5..05a6e30a928 100644 --- a/arch/sparc/kernel/psycho_common.h +++ b/arch/sparc/kernel/psycho_common.h @@ -30,19 +30,19 @@ enum psycho_error_type { UE_ERR, CE_ERR, PCI_ERR }; -extern void psycho_check_iommu_error(struct pci_pbm_info *pbm, - unsigned long afsr, - unsigned long afar, - enum psycho_error_type type); +void psycho_check_iommu_error(struct pci_pbm_info *pbm, + unsigned long afsr, + unsigned long afar, + enum psycho_error_type type); -extern irqreturn_t psycho_pcierr_intr(int irq, void *dev_id); +irqreturn_t psycho_pcierr_intr(int irq, void *dev_id); -extern int psycho_iommu_init(struct pci_pbm_info *pbm, int tsbsize, - u32 dvma_offset, u32 dma_mask, - unsigned long write_complete_offset); +int psycho_iommu_init(struct pci_pbm_info *pbm, int tsbsize, + u32 dvma_offset, u32 dma_mask, + unsigned long write_complete_offset); -extern void psycho_pbm_init_common(struct pci_pbm_info *pbm, - struct platform_device *op, - const char *chip_name, int chip_type); +void psycho_pbm_init_common(struct pci_pbm_info *pbm, + struct platform_device *op, + const char *chip_name, int chip_type); #endif /* _PSYCHO_COMMON_H */ diff --git a/arch/sparc/kernel/ptrace_32.c b/arch/sparc/kernel/ptrace_32.c index 896ba7c5cd8..a331fdc11a2 100644 --- a/arch/sparc/kernel/ptrace_32.c +++ b/arch/sparc/kernel/ptrace_32.c @@ -26,6 +26,8 @@ #include <asm/uaccess.h> #include <asm/cacheflush.h> +#include "kernel.h" + /* #define ALLOW_INIT_TRACING */ /* diff --git a/arch/sparc/kernel/ptrace_64.c b/arch/sparc/kernel/ptrace_64.c index 484dabac704..c13c9f25d83 100644 --- a/arch/sparc/kernel/ptrace_64.c +++ b/arch/sparc/kernel/ptrace_64.c @@ -14,6 +14,7 @@ #include <linux/sched.h> #include <linux/mm.h> #include <linux/errno.h> +#include <linux/export.h> #include <linux/ptrace.h> #include <linux/user.h> #include <linux/smp.h> @@ -26,6 +27,7 @@ #include <trace/syscall.h> #include <linux/compat.h> #include <linux/elf.h> +#include <linux/context_tracking.h> #include <asm/asi.h> #include <asm/pgtable.h> @@ -116,6 +118,7 @@ void flush_ptrace_access(struct vm_area_struct *vma, struct page *page, preempt_enable(); } +EXPORT_SYMBOL_GPL(flush_ptrace_access); static int get_from_target(struct task_struct *target, unsigned long uaddr, void *kbuf, int len) @@ -151,7 +154,7 @@ static int regwindow64_get(struct task_struct *target, { unsigned long rw_addr = regs->u_regs[UREG_I6]; - if (test_tsk_thread_flag(current, TIF_32BIT)) { + if (!test_thread_64bit_stack(rw_addr)) { struct reg_window32 win32; int i; @@ -176,7 +179,7 @@ static int regwindow64_set(struct task_struct *target, { unsigned long rw_addr = regs->u_regs[UREG_I6]; - if (test_tsk_thread_flag(current, TIF_32BIT)) { + if (!test_thread_64bit_stack(rw_addr)) { struct reg_window32 win32; int i; @@ -1064,6 +1067,9 @@ asmlinkage int syscall_trace_enter(struct pt_regs *regs) /* do the secure computing check first */ secure_computing_strict(regs->u_regs[UREG_G1]); + if (test_thread_flag(TIF_NOHZ)) + user_exit(); + if (test_thread_flag(TIF_SYSCALL_TRACE)) ret = tracehook_report_syscall_entry(regs); @@ -1084,11 +1090,17 @@ asmlinkage int syscall_trace_enter(struct pt_regs *regs) asmlinkage void syscall_trace_leave(struct pt_regs *regs) { + if (test_thread_flag(TIF_NOHZ)) + user_exit(); + audit_syscall_exit(regs); if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) - trace_sys_exit(regs, regs->u_regs[UREG_G1]); + trace_sys_exit(regs, regs->u_regs[UREG_I0]); if (test_thread_flag(TIF_SYSCALL_TRACE)) tracehook_report_syscall_exit(regs, 0); + + if (test_thread_flag(TIF_NOHZ)) + user_enter(); } diff --git a/arch/sparc/kernel/rtrap_64.S b/arch/sparc/kernel/rtrap_64.S index afa2a9e3d0a..39f0c662f4c 100644 --- a/arch/sparc/kernel/rtrap_64.S +++ b/arch/sparc/kernel/rtrap_64.S @@ -18,10 +18,16 @@ #define RTRAP_PSTATE_IRQOFF (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV) #define RTRAP_PSTATE_AG_IRQOFF (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV|PSTATE_AG) +#ifdef CONFIG_CONTEXT_TRACKING +# define SCHEDULE_USER schedule_user +#else +# define SCHEDULE_USER schedule +#endif + .text .align 32 __handle_preemption: - call schedule + call SCHEDULE_USER wrpr %g0, RTRAP_PSTATE, %pstate ba,pt %xcc, __handle_preemption_continue wrpr %g0, RTRAP_PSTATE_IRQOFF, %pstate @@ -306,12 +312,10 @@ to_kernel: nop cmp %l4, 0 bne,pn %xcc, kern_fpucheck - sethi %hi(PREEMPT_ACTIVE), %l6 - stw %l6, [%g6 + TI_PRE_COUNT] - call schedule + nop + call preempt_schedule_irq nop ba,pt %xcc, rtrap - stw %g0, [%g6 + TI_PRE_COUNT] #endif kern_fpucheck: ldub [%g6 + TI_FPDEPTH], %l5 brz,pt %l5, rt_continue diff --git a/arch/sparc/kernel/sbus.c b/arch/sparc/kernel/sbus.c index 1271b3a27d4..be5bdf93c76 100644 --- a/arch/sparc/kernel/sbus.c +++ b/arch/sparc/kernel/sbus.c @@ -554,10 +554,8 @@ static void __init sbus_iommu_init(struct platform_device *op) regs = pr->phys_addr; iommu = kzalloc(sizeof(*iommu), GFP_ATOMIC); - if (!iommu) - goto fatal_memory_error; strbuf = kzalloc(sizeof(*strbuf), GFP_ATOMIC); - if (!strbuf) + if (!iommu || !strbuf) goto fatal_memory_error; op->dev.archdata.iommu = iommu; @@ -656,6 +654,8 @@ static void __init sbus_iommu_init(struct platform_device *op) return; fatal_memory_error: + kfree(iommu); + kfree(strbuf); prom_printf("sbus_iommu_init: Fatal memory allocation error.\n"); } diff --git a/arch/sparc/kernel/setup_32.c b/arch/sparc/kernel/setup_32.c index efe3e64bba3..baef495c06b 100644 --- a/arch/sparc/kernel/setup_32.c +++ b/arch/sparc/kernel/setup_32.c @@ -267,7 +267,7 @@ static __init void leon_patch(void) } struct tt_entry *sparc_ttable; -struct pt_regs fake_swapper_regs; +static struct pt_regs fake_swapper_regs; /* Called from head_32.S - before we have setup anything * in the kernel. Be very careful with what you do here. @@ -304,7 +304,7 @@ void __init setup_arch(char **cmdline_p) /* Initialize PROM console and command line. */ *cmdline_p = prom_getbootargs(); - strcpy(boot_command_line, *cmdline_p); + strlcpy(boot_command_line, *cmdline_p, COMMAND_LINE_SIZE); parse_early_param(); boot_flags_init(*cmdline_p); @@ -365,13 +365,12 @@ void __init setup_arch(char **cmdline_p) prom_setsync(prom_sync_me); - if((boot_flags&BOOTME_DEBUG) && (linux_dbvec!=0) && + if((boot_flags & BOOTME_DEBUG) && (linux_dbvec != NULL) && ((*(short *)linux_dbvec) != -1)) { printk("Booted under KADB. Syncing trap table.\n"); (*(linux_dbvec->teach_debugger))(); } - init_mm.context = (unsigned long) NO_CONTEXT; init_task.thread.kregs = &fake_swapper_regs; /* Run-time patch instructions to match the cpu model */ diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c index 1414d16712b..3fdb455e331 100644 --- a/arch/sparc/kernel/setup_64.c +++ b/arch/sparc/kernel/setup_64.c @@ -115,7 +115,7 @@ static void __init process_switch(char c) break; } cheetah_pcache_forced_on = 1; - add_taint(TAINT_MACHINE_CHECK); + add_taint(TAINT_MACHINE_CHECK, LOCKDEP_NOW_UNRELIABLE); cheetah_enable_pcache(); break; @@ -316,6 +316,25 @@ static void __init popc_patch(void) } } +static void __init pause_patch(void) +{ + struct pause_patch_entry *p; + + p = &__pause_3insn_patch; + while (p < &__pause_3insn_patch_end) { + unsigned long i, addr = p->addr; + + for (i = 0; i < 3; i++) { + *(unsigned int *) (addr + (i * 4)) = p->insns[i]; + wmb(); + __asm__ __volatile__("flush %0" + : : "r" (addr + (i * 4))); + } + + p++; + } +} + #ifdef CONFIG_SMP void __init boot_cpu_id_too_large(int cpu) { @@ -340,7 +359,12 @@ static const char *hwcaps[] = { */ "mul32", "div32", "fsmuld", "v8plus", "popc", "vis", "vis2", "ASIBlkInit", "fmaf", "vis3", "hpc", "random", "trans", "fjfmau", - "ima", "cspare", + "ima", "cspare", "pause", "cbcond", +}; + +static const char *crypto_hwcaps[] = { + "aes", "des", "kasumi", "camellia", "md5", "sha1", "sha256", + "sha512", "mpmul", "montmul", "montsqr", "crc32c", }; void cpucap_info(struct seq_file *m) @@ -357,27 +381,61 @@ void cpucap_info(struct seq_file *m) printed++; } } + if (caps & HWCAP_SPARC_CRYPTO) { + unsigned long cfr; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + for (i = 0; i < ARRAY_SIZE(crypto_hwcaps); i++) { + unsigned long bit = 1UL << i; + if (cfr & bit) { + seq_printf(m, "%s%s", + printed ? "," : "", crypto_hwcaps[i]); + printed++; + } + } + } seq_putc(m, '\n'); } +static void __init report_one_hwcap(int *printed, const char *name) +{ + if ((*printed) == 0) + printk(KERN_INFO "CPU CAPS: ["); + printk(KERN_CONT "%s%s", + (*printed) ? "," : "", name); + if (++(*printed) == 8) { + printk(KERN_CONT "]\n"); + *printed = 0; + } +} + +static void __init report_crypto_hwcaps(int *printed) +{ + unsigned long cfr; + int i; + + __asm__ __volatile__("rd %%asr26, %0" : "=r" (cfr)); + + for (i = 0; i < ARRAY_SIZE(crypto_hwcaps); i++) { + unsigned long bit = 1UL << i; + if (cfr & bit) + report_one_hwcap(printed, crypto_hwcaps[i]); + } +} + static void __init report_hwcaps(unsigned long caps) { int i, printed = 0; - printk(KERN_INFO "CPU CAPS: ["); for (i = 0; i < ARRAY_SIZE(hwcaps); i++) { unsigned long bit = 1UL << i; - if (caps & bit) { - printk(KERN_CONT "%s%s", - printed ? "," : "", hwcaps[i]); - if (++printed == 8) { - printk(KERN_CONT "]\n"); - printk(KERN_INFO "CPU CAPS: ["); - printed = 0; - } - } + if (caps & bit) + report_one_hwcap(&printed, hwcaps[i]); } - printk(KERN_CONT "]\n"); + if (caps & HWCAP_SPARC_CRYPTO) + report_crypto_hwcaps(&printed); + if (printed != 0) + printk(KERN_CONT "]\n"); } static unsigned long __init mdesc_cpu_hwcap_list(void) @@ -411,6 +469,10 @@ static unsigned long __init mdesc_cpu_hwcap_list(void) break; } } + for (i = 0; i < ARRAY_SIZE(crypto_hwcaps); i++) { + if (!strcmp(prop, crypto_hwcaps[i])) + caps |= HWCAP_SPARC_CRYPTO; + } plen = strlen(prop) + 1; prop += plen; @@ -437,12 +499,14 @@ static void __init init_sparc64_elf_hwcap(void) sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || sun4v_chip_type == SUN4V_CHIP_NIAGARA3 || sun4v_chip_type == SUN4V_CHIP_NIAGARA4 || - sun4v_chip_type == SUN4V_CHIP_NIAGARA5) + sun4v_chip_type == SUN4V_CHIP_NIAGARA5 || + sun4v_chip_type == SUN4V_CHIP_SPARC64X) cap |= HWCAP_SPARC_BLKINIT; if (sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || sun4v_chip_type == SUN4V_CHIP_NIAGARA3 || sun4v_chip_type == SUN4V_CHIP_NIAGARA4 || - sun4v_chip_type == SUN4V_CHIP_NIAGARA5) + sun4v_chip_type == SUN4V_CHIP_NIAGARA5 || + sun4v_chip_type == SUN4V_CHIP_SPARC64X) cap |= HWCAP_SPARC_N2; } @@ -468,13 +532,15 @@ static void __init init_sparc64_elf_hwcap(void) if (sun4v_chip_type == SUN4V_CHIP_NIAGARA2 || sun4v_chip_type == SUN4V_CHIP_NIAGARA3 || sun4v_chip_type == SUN4V_CHIP_NIAGARA4 || - sun4v_chip_type == SUN4V_CHIP_NIAGARA5) + sun4v_chip_type == SUN4V_CHIP_NIAGARA5 || + sun4v_chip_type == SUN4V_CHIP_SPARC64X) cap |= (AV_SPARC_VIS | AV_SPARC_VIS2 | AV_SPARC_ASI_BLK_INIT | AV_SPARC_POPC); if (sun4v_chip_type == SUN4V_CHIP_NIAGARA3 || sun4v_chip_type == SUN4V_CHIP_NIAGARA4 || - sun4v_chip_type == SUN4V_CHIP_NIAGARA5) + sun4v_chip_type == SUN4V_CHIP_NIAGARA5 || + sun4v_chip_type == SUN4V_CHIP_SPARC64X) cap |= (AV_SPARC_VIS3 | AV_SPARC_HPC | AV_SPARC_FMAF); } @@ -485,13 +551,15 @@ static void __init init_sparc64_elf_hwcap(void) if (sparc64_elf_hwcap & AV_SPARC_POPC) popc_patch(); + if (sparc64_elf_hwcap & AV_SPARC_PAUSE) + pause_patch(); } void __init setup_arch(char **cmdline_p) { /* Initialize PROM console and command line. */ *cmdline_p = prom_getbootargs(); - strcpy(boot_command_line, *cmdline_p); + strlcpy(boot_command_line, *cmdline_p, COMMAND_LINE_SIZE); parse_early_param(); boot_flags_init(*cmdline_p); diff --git a/arch/sparc/kernel/signal32.c b/arch/sparc/kernel/signal32.c index a53e0a5fd3a..62deba7be1a 100644 --- a/arch/sparc/kernel/signal32.c +++ b/arch/sparc/kernel/signal32.c @@ -31,6 +31,7 @@ #include <asm/switch_to.h> #include "sigutil.h" +#include "kernel.h" /* This magic should be in g_upper[0] for all upper parts * to be valid. @@ -54,58 +55,6 @@ struct signal_frame32 { /* __siginfo_rwin_t * */u32 rwin_save; } __attribute__((aligned(8))); -typedef struct compat_siginfo{ - int si_signo; - int si_errno; - int si_code; - - union { - int _pad[SI_PAD_SIZE32]; - - /* kill() */ - struct { - compat_pid_t _pid; /* sender's pid */ - unsigned int _uid; /* sender's uid */ - } _kill; - - /* POSIX.1b timers */ - struct { - compat_timer_t _tid; /* timer id */ - int _overrun; /* overrun count */ - compat_sigval_t _sigval; /* same as below */ - int _sys_private; /* not to be passed to user */ - } _timer; - - /* POSIX.1b signals */ - struct { - compat_pid_t _pid; /* sender's pid */ - unsigned int _uid; /* sender's uid */ - compat_sigval_t _sigval; - } _rt; - - /* SIGCHLD */ - struct { - compat_pid_t _pid; /* which child */ - unsigned int _uid; /* sender's uid */ - int _status; /* exit code */ - compat_clock_t _utime; - compat_clock_t _stime; - } _sigchld; - - /* SIGILL, SIGFPE, SIGSEGV, SIGBUS, SIGEMT */ - struct { - u32 _addr; /* faulting insn/memory ref. */ - int _trapno; - } _sigfault; - - /* SIGPOLL */ - struct { - int _band; /* POLL_IN, POLL_OUT, POLL_MSG */ - int _fd; - } _sigpoll; - } _sifields; -}compat_siginfo_t; - struct rt_signal_frame32 { struct sparc_stackf32 ss; compat_siginfo_t info; @@ -113,14 +62,14 @@ struct rt_signal_frame32 { compat_sigset_t mask; /* __siginfo_fpu_t * */ u32 fpu_save; unsigned int insns[2]; - stack_t32 stack; + compat_stack_t stack; unsigned int extra_size; /* Should be sizeof(siginfo_extra_v8plus_t) */ /* Only valid if (regs.psr & (PSR_VERS|PSR_IMPL)) == PSR_V8PLUS */ siginfo_extra_v8plus_t v8plus; /* __siginfo_rwin_t * */u32 rwin_save; } __attribute__((aligned(8))); -int copy_siginfo_to_user32(compat_siginfo_t __user *to, siginfo_t *from) +int copy_siginfo_to_user32(compat_siginfo_t __user *to, const siginfo_t *from) { int err; @@ -197,7 +146,7 @@ void do_sigreturn32(struct pt_regs *regs) unsigned int psr; unsigned pc, npc; sigset_t set; - unsigned seta[_COMPAT_NSIG_WORDS]; + compat_sigset_t seta; int err, i; /* Always make any pending restarted system calls return -EINTR */ @@ -261,17 +210,13 @@ void do_sigreturn32(struct pt_regs *regs) if (restore_rwin_state(compat_ptr(rwin_save))) goto segv; } - err |= __get_user(seta[0], &sf->info.si_mask); - err |= copy_from_user(seta+1, &sf->extramask, + err |= __get_user(seta.sig[0], &sf->info.si_mask); + err |= copy_from_user(&seta.sig[1], &sf->extramask, (_COMPAT_NSIG_WORDS - 1) * sizeof(unsigned int)); if (err) goto segv; - switch (_NSIG_WORDS) { - case 4: set.sig[3] = seta[6] + (((long)seta[7]) << 32); - case 3: set.sig[2] = seta[4] + (((long)seta[5]) << 32); - case 2: set.sig[1] = seta[2] + (((long)seta[3]) << 32); - case 1: set.sig[0] = seta[0] + (((long)seta[1]) << 32); - } + + set.sig[0] = seta.sig[0] + (((long)seta.sig[1]) << 32); set_current_blocked(&set); return; @@ -282,13 +227,11 @@ segv: asmlinkage void do_rt_sigreturn32(struct pt_regs *regs) { struct rt_signal_frame32 __user *sf; - unsigned int psr, pc, npc, u_ss_sp; + unsigned int psr, pc, npc; compat_uptr_t fpu_save; compat_uptr_t rwin_save; - mm_segment_t old_fs; sigset_t set; compat_sigset_t seta; - stack_t st; int err, i; /* Always make any pending restarted system calls return -EINTR */ @@ -347,32 +290,17 @@ asmlinkage void do_rt_sigreturn32(struct pt_regs *regs) if (!err && fpu_save) err |= restore_fpu_state(regs, compat_ptr(fpu_save)); err |= copy_from_user(&seta, &sf->mask, sizeof(compat_sigset_t)); - err |= __get_user(u_ss_sp, &sf->stack.ss_sp); - st.ss_sp = compat_ptr(u_ss_sp); - err |= __get_user(st.ss_flags, &sf->stack.ss_flags); - err |= __get_user(st.ss_size, &sf->stack.ss_size); + err |= compat_restore_altstack(&sf->stack); if (err) goto segv; - /* It is more difficult to avoid calling this function than to - call it and ignore errors. */ - old_fs = get_fs(); - set_fs(KERNEL_DS); - do_sigaltstack((stack_t __user *) &st, NULL, (unsigned long)sf); - set_fs(old_fs); - err |= __get_user(rwin_save, &sf->rwin_save); if (!err && rwin_save) { if (restore_rwin_state(compat_ptr(rwin_save))) goto segv; } - switch (_NSIG_WORDS) { - case 4: set.sig[3] = seta.sig[6] + (((long)seta.sig[7]) << 32); - case 3: set.sig[2] = seta.sig[4] + (((long)seta.sig[5]) << 32); - case 2: set.sig[1] = seta.sig[2] + (((long)seta.sig[3]) << 32); - case 1: set.sig[0] = seta.sig[0] + (((long)seta.sig[1]) << 32); - } + set.sig[0] = seta.sig[0] + (((long)seta.sig[1]) << 32); set_current_blocked(&set); return; segv: @@ -387,7 +315,7 @@ static int invalid_frame_pointer(void __user *fp, int fplen) return 0; } -static void __user *get_sigframe(struct sigaction *sa, struct pt_regs *regs, unsigned long framesize) +static void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs, unsigned long framesize) { unsigned long sp; @@ -402,12 +330,7 @@ static void __user *get_sigframe(struct sigaction *sa, struct pt_regs *regs, uns return (void __user *) -1L; /* This is the X/Open sanctioned signal stack switching. */ - if (sa->sa_flags & SA_ONSTACK) { - if (sas_ss_flags(sp) == 0) - sp = current->sas_ss_sp + current->sas_ss_size; - } - - sp -= framesize; + sp = sigsp(sp, ksig) - framesize; /* Always align the stack frame. This handles two cases. First, * sigaltstack need not be mindful of platform specific stack @@ -478,15 +401,15 @@ out_irqs_on: } -static int setup_frame32(struct k_sigaction *ka, struct pt_regs *regs, - int signo, sigset_t *oldset) +static int setup_frame32(struct ksignal *ksig, struct pt_regs *regs, + sigset_t *oldset) { struct signal_frame32 __user *sf; int i, err, wsaved; void __user *tail; int sigframe_size; u32 psr; - unsigned int seta[_COMPAT_NSIG_WORDS]; + compat_sigset_t seta; /* 1. Make sure everything is clean */ synchronize_user_stack(); @@ -501,10 +424,12 @@ static int setup_frame32(struct k_sigaction *ka, struct pt_regs *regs, sigframe_size += sizeof(__siginfo_rwin_t); sf = (struct signal_frame32 __user *) - get_sigframe(&ka->sa, regs, sigframe_size); + get_sigframe(ksig, regs, sigframe_size); - if (invalid_frame_pointer(sf, sigframe_size)) - goto sigill; + if (invalid_frame_pointer(sf, sigframe_size)) { + do_exit(SIGILL); + return -EINVAL; + } tail = (sf + 1); @@ -548,18 +473,14 @@ static int setup_frame32(struct k_sigaction *ka, struct pt_regs *regs, err |= __put_user(0, &sf->rwin_save); } - switch (_NSIG_WORDS) { - case 4: seta[7] = (oldset->sig[3] >> 32); - seta[6] = oldset->sig[3]; - case 3: seta[5] = (oldset->sig[2] >> 32); - seta[4] = oldset->sig[2]; - case 2: seta[3] = (oldset->sig[1] >> 32); - seta[2] = oldset->sig[1]; - case 1: seta[1] = (oldset->sig[0] >> 32); - seta[0] = oldset->sig[0]; - } - err |= __put_user(seta[0], &sf->info.si_mask); - err |= __copy_to_user(sf->extramask, seta + 1, + /* If these change we need to know - assignments to seta relies on these sizes */ + BUILD_BUG_ON(_NSIG_WORDS != 1); + BUILD_BUG_ON(_COMPAT_NSIG_WORDS != 2); + seta.sig[1] = (oldset->sig[0] >> 32); + seta.sig[0] = oldset->sig[0]; + + err |= __put_user(seta.sig[0], &sf->info.si_mask); + err |= __copy_to_user(sf->extramask, &seta.sig[1], (_COMPAT_NSIG_WORDS - 1) * sizeof(unsigned int)); if (!wsaved) { @@ -578,16 +499,16 @@ static int setup_frame32(struct k_sigaction *ka, struct pt_regs *regs, err |= __put_user(rp->ins[7], &sf->ss.callers_pc); } if (err) - goto sigsegv; + return err; /* 3. signal handler back-trampoline and parameters */ regs->u_regs[UREG_FP] = (unsigned long) sf; - regs->u_regs[UREG_I0] = signo; + regs->u_regs[UREG_I0] = ksig->sig; regs->u_regs[UREG_I1] = (unsigned long) &sf->info; regs->u_regs[UREG_I2] = (unsigned long) &sf->info; /* 4. signal handler */ - regs->tpc = (unsigned long) ka->sa.sa_handler; + regs->tpc = (unsigned long) ksig->ka.sa.sa_handler; regs->tnpc = (regs->tpc + 4); if (test_thread_flag(TIF_32BIT)) { regs->tpc &= 0xffffffff; @@ -595,8 +516,8 @@ static int setup_frame32(struct k_sigaction *ka, struct pt_regs *regs, } /* 5. return to kernel instructions */ - if (ka->ka_restorer) { - regs->u_regs[UREG_I7] = (unsigned long)ka->ka_restorer; + if (ksig->ka.ka_restorer) { + regs->u_regs[UREG_I7] = (unsigned long)ksig->ka.ka_restorer; } else { unsigned long address = ((unsigned long)&(sf->insns[0])); @@ -605,23 +526,14 @@ static int setup_frame32(struct k_sigaction *ka, struct pt_regs *regs, err = __put_user(0x821020d8, &sf->insns[0]); /*mov __NR_sigreturn, %g1*/ err |= __put_user(0x91d02010, &sf->insns[1]); /*t 0x10*/ if (err) - goto sigsegv; + return err; flush_signal_insns(address); } return 0; - -sigill: - do_exit(SIGILL); - return -EINVAL; - -sigsegv: - force_sigsegv(signo, current); - return -EFAULT; } -static int setup_rt_frame32(struct k_sigaction *ka, struct pt_regs *regs, - unsigned long signr, sigset_t *oldset, - siginfo_t *info) +static int setup_rt_frame32(struct ksignal *ksig, struct pt_regs *regs, + sigset_t *oldset) { struct rt_signal_frame32 __user *sf; int i, err, wsaved; @@ -643,10 +555,12 @@ static int setup_rt_frame32(struct k_sigaction *ka, struct pt_regs *regs, sigframe_size += sizeof(__siginfo_rwin_t); sf = (struct rt_signal_frame32 __user *) - get_sigframe(&ka->sa, regs, sigframe_size); + get_sigframe(ksig, regs, sigframe_size); - if (invalid_frame_pointer(sf, sigframe_size)) - goto sigill; + if (invalid_frame_pointer(sf, sigframe_size)) { + do_exit(SIGILL); + return -EINVAL; + } tail = (sf + 1); @@ -691,23 +605,13 @@ static int setup_rt_frame32(struct k_sigaction *ka, struct pt_regs *regs, } /* Update the siginfo structure. */ - err |= copy_siginfo_to_user32(&sf->info, info); + err |= copy_siginfo_to_user32(&sf->info, &ksig->info); /* Setup sigaltstack */ - err |= __put_user(current->sas_ss_sp, &sf->stack.ss_sp); - err |= __put_user(sas_ss_flags(regs->u_regs[UREG_FP]), &sf->stack.ss_flags); - err |= __put_user(current->sas_ss_size, &sf->stack.ss_size); - - switch (_NSIG_WORDS) { - case 4: seta.sig[7] = (oldset->sig[3] >> 32); - seta.sig[6] = oldset->sig[3]; - case 3: seta.sig[5] = (oldset->sig[2] >> 32); - seta.sig[4] = oldset->sig[2]; - case 2: seta.sig[3] = (oldset->sig[1] >> 32); - seta.sig[2] = oldset->sig[1]; - case 1: seta.sig[1] = (oldset->sig[0] >> 32); - seta.sig[0] = oldset->sig[0]; - } + err |= __compat_save_altstack(&sf->stack, regs->u_regs[UREG_FP]); + + seta.sig[1] = (oldset->sig[0] >> 32); + seta.sig[0] = oldset->sig[0]; err |= __copy_to_user(&sf->mask, &seta, sizeof(compat_sigset_t)); if (!wsaved) { @@ -726,16 +630,16 @@ static int setup_rt_frame32(struct k_sigaction *ka, struct pt_regs *regs, err |= __put_user(rp->ins[7], &sf->ss.callers_pc); } if (err) - goto sigsegv; + return err; /* 3. signal handler back-trampoline and parameters */ regs->u_regs[UREG_FP] = (unsigned long) sf; - regs->u_regs[UREG_I0] = signr; + regs->u_regs[UREG_I0] = ksig->sig; regs->u_regs[UREG_I1] = (unsigned long) &sf->info; regs->u_regs[UREG_I2] = (unsigned long) &sf->regs; /* 4. signal handler */ - regs->tpc = (unsigned long) ka->sa.sa_handler; + regs->tpc = (unsigned long) ksig->ka.sa.sa_handler; regs->tnpc = (regs->tpc + 4); if (test_thread_flag(TIF_32BIT)) { regs->tpc &= 0xffffffff; @@ -743,8 +647,8 @@ static int setup_rt_frame32(struct k_sigaction *ka, struct pt_regs *regs, } /* 5. return to kernel instructions */ - if (ka->ka_restorer) - regs->u_regs[UREG_I7] = (unsigned long)ka->ka_restorer; + if (ksig->ka.ka_restorer) + regs->u_regs[UREG_I7] = (unsigned long)ksig->ka.ka_restorer; else { unsigned long address = ((unsigned long)&(sf->insns[0])); @@ -756,36 +660,25 @@ static int setup_rt_frame32(struct k_sigaction *ka, struct pt_regs *regs, /* t 0x10 */ err |= __put_user(0x91d02010, &sf->insns[1]); if (err) - goto sigsegv; + return err; flush_signal_insns(address); } return 0; - -sigill: - do_exit(SIGILL); - return -EINVAL; - -sigsegv: - force_sigsegv(signr, current); - return -EFAULT; } -static inline void handle_signal32(unsigned long signr, struct k_sigaction *ka, - siginfo_t *info, - sigset_t *oldset, struct pt_regs *regs) +static inline void handle_signal32(struct ksignal *ksig, + struct pt_regs *regs) { + sigset_t *oldset = sigmask_to_save(); int err; - if (ka->sa.sa_flags & SA_SIGINFO) - err = setup_rt_frame32(ka, regs, signr, oldset, info); + if (ksig->ka.sa.sa_flags & SA_SIGINFO) + err = setup_rt_frame32(ksig, regs, oldset); else - err = setup_frame32(ka, regs, signr, oldset); - - if (err) - return; + err = setup_frame32(ksig, regs, oldset); - signal_delivered(signr, info, ka, regs, 0); + signal_setup_done(err, ksig, 0); } static inline void syscall_restart32(unsigned long orig_i0, struct pt_regs *regs, @@ -813,52 +706,43 @@ static inline void syscall_restart32(unsigned long orig_i0, struct pt_regs *regs * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -void do_signal32(sigset_t *oldset, struct pt_regs * regs) +void do_signal32(struct pt_regs * regs) { - struct k_sigaction ka; - unsigned long orig_i0; - int restart_syscall; - siginfo_t info; - int signr; - - signr = get_signal_to_deliver(&info, &ka, regs, NULL); + struct ksignal ksig; + unsigned long orig_i0 = 0; + int restart_syscall = 0; + bool has_handler = get_signal(&ksig); - restart_syscall = 0; - orig_i0 = 0; if (pt_regs_is_syscall(regs) && (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) { restart_syscall = 1; orig_i0 = regs->u_regs[UREG_G6]; } - if (signr > 0) { + if (has_handler) { if (restart_syscall) - syscall_restart32(orig_i0, regs, &ka.sa); - handle_signal32(signr, &ka, &info, oldset, regs); - return; - } - if (restart_syscall && - (regs->u_regs[UREG_I0] == ERESTARTNOHAND || - regs->u_regs[UREG_I0] == ERESTARTSYS || - regs->u_regs[UREG_I0] == ERESTARTNOINTR)) { - /* replay the system call when we are done */ - regs->u_regs[UREG_I0] = orig_i0; - regs->tpc -= 4; - regs->tnpc -= 4; - pt_regs_clear_syscall(regs); - } - if (restart_syscall && - regs->u_regs[UREG_I0] == ERESTART_RESTARTBLOCK) { - regs->u_regs[UREG_G1] = __NR_restart_syscall; - regs->tpc -= 4; - regs->tnpc -= 4; - pt_regs_clear_syscall(regs); + syscall_restart32(orig_i0, regs, &ksig.ka.sa); + handle_signal32(&ksig, regs); + } else { + if (restart_syscall) { + switch (regs->u_regs[UREG_I0]) { + case ERESTARTNOHAND: + case ERESTARTSYS: + case ERESTARTNOINTR: + /* replay the system call when we are done */ + regs->u_regs[UREG_I0] = orig_i0; + regs->tpc -= 4; + regs->tnpc -= 4; + pt_regs_clear_syscall(regs); + case ERESTART_RESTARTBLOCK: + regs->u_regs[UREG_G1] = __NR_restart_syscall; + regs->tpc -= 4; + regs->tnpc -= 4; + pt_regs_clear_syscall(regs); + } + } + restore_saved_sigmask(); } - - /* If there's no signal to deliver, we just put the saved sigmask - * back - */ - restore_saved_sigmask(); } struct sigstack32 { @@ -908,29 +792,3 @@ asmlinkage int do_sys32_sigstack(u32 u_ssptr, u32 u_ossptr, unsigned long sp) out: return ret; } - -asmlinkage long do_sys32_sigaltstack(u32 ussa, u32 uossa, unsigned long sp) -{ - stack_t uss, uoss; - u32 u_ss_sp = 0; - int ret; - mm_segment_t old_fs; - stack_t32 __user *uss32 = compat_ptr(ussa); - stack_t32 __user *uoss32 = compat_ptr(uossa); - - if (ussa && (get_user(u_ss_sp, &uss32->ss_sp) || - __get_user(uss.ss_flags, &uss32->ss_flags) || - __get_user(uss.ss_size, &uss32->ss_size))) - return -EFAULT; - uss.ss_sp = compat_ptr(u_ss_sp); - old_fs = get_fs(); - set_fs(KERNEL_DS); - ret = do_sigaltstack(ussa ? (stack_t __user *) &uss : NULL, - uossa ? (stack_t __user *) &uoss : NULL, sp); - set_fs(old_fs); - if (!ret && uossa && (put_user(ptr_to_compat(uoss.ss_sp), &uoss32->ss_sp) || - __put_user(uoss.ss_flags, &uoss32->ss_flags) || - __put_user(uoss.ss_size, &uoss32->ss_size))) - return -EFAULT; - return ret; -} diff --git a/arch/sparc/kernel/signal_32.c b/arch/sparc/kernel/signal_32.c index 68f9c8650af..9ee72fc8e0e 100644 --- a/arch/sparc/kernel/signal_32.c +++ b/arch/sparc/kernel/signal_32.c @@ -28,6 +28,7 @@ #include <asm/switch_to.h> #include "sigutil.h" +#include "kernel.h" extern void fpsave(unsigned long *fpregs, unsigned long *fsr, void *fpqueue, unsigned long *fpqdepth); @@ -59,18 +60,6 @@ struct rt_signal_frame { #define SF_ALIGNEDSZ (((sizeof(struct signal_frame) + 7) & (~7))) #define RT_ALIGNEDSZ (((sizeof(struct rt_signal_frame) + 7) & (~7))) -static int _sigpause_common(old_sigset_t set) -{ - sigset_t blocked; - siginitset(&blocked, set); - return sigsuspend(&blocked); -} - -asmlinkage int sys_sigsuspend(old_sigset_t set) -{ - return _sigpause_common(set); -} - asmlinkage void do_sigreturn(struct pt_regs *regs) { struct signal_frame __user *sf; @@ -141,9 +130,7 @@ asmlinkage void do_rt_sigreturn(struct pt_regs *regs) unsigned int psr, pc, npc; __siginfo_fpu_t __user *fpu_save; __siginfo_rwin_t __user *rwin_save; - mm_segment_t old_fs; sigset_t set; - stack_t st; int err; synchronize_user_stack(); @@ -171,8 +158,7 @@ asmlinkage void do_rt_sigreturn(struct pt_regs *regs) if (!err && fpu_save) err |= restore_fpu_state(regs, fpu_save); err |= __copy_from_user(&set, &sf->mask, sizeof(sigset_t)); - - err |= __copy_from_user(&st, &sf->stack, sizeof(stack_t)); + err |= restore_altstack(&sf->stack); if (err) goto segv; @@ -180,14 +166,6 @@ asmlinkage void do_rt_sigreturn(struct pt_regs *regs) regs->pc = pc; regs->npc = npc; - /* It is more difficult to avoid calling this function than to - * call it and ignore errors. - */ - old_fs = get_fs(); - set_fs(KERNEL_DS); - do_sigaltstack((const stack_t __user *) &st, NULL, (unsigned long)sf); - set_fs(old_fs); - err |= __get_user(rwin_save, &sf->rwin_save); if (!err && rwin_save) { if (restore_rwin_state(rwin_save)) @@ -209,7 +187,7 @@ static inline int invalid_frame_pointer(void __user *fp, int fplen) return 0; } -static inline void __user *get_sigframe(struct sigaction *sa, struct pt_regs *regs, unsigned long framesize) +static inline void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs, unsigned long framesize) { unsigned long sp = regs->u_regs[UREG_FP]; @@ -221,12 +199,7 @@ static inline void __user *get_sigframe(struct sigaction *sa, struct pt_regs *re return (void __user *) -1L; /* This is the X/Open sanctioned signal stack switching. */ - if (sa->sa_flags & SA_ONSTACK) { - if (sas_ss_flags(sp) == 0) - sp = current->sas_ss_sp + current->sas_ss_size; - } - - sp -= framesize; + sp = sigsp(sp, ksig) - framesize; /* Always align the stack frame. This handles two cases. First, * sigaltstack need not be mindful of platform specific stack @@ -239,8 +212,8 @@ static inline void __user *get_sigframe(struct sigaction *sa, struct pt_regs *re return (void __user *) sp; } -static int setup_frame(struct k_sigaction *ka, struct pt_regs *regs, - int signo, sigset_t *oldset) +static int setup_frame(struct ksignal *ksig, struct pt_regs *regs, + sigset_t *oldset) { struct signal_frame __user *sf; int sigframe_size, err, wsaved; @@ -258,10 +231,12 @@ static int setup_frame(struct k_sigaction *ka, struct pt_regs *regs, sigframe_size += sizeof(__siginfo_rwin_t); sf = (struct signal_frame __user *) - get_sigframe(&ka->sa, regs, sigframe_size); + get_sigframe(ksig, regs, sigframe_size); - if (invalid_frame_pointer(sf, sigframe_size)) - goto sigill_and_return; + if (invalid_frame_pointer(sf, sigframe_size)) { + do_exit(SIGILL); + return -EINVAL; + } tail = sf + 1; @@ -300,21 +275,21 @@ static int setup_frame(struct k_sigaction *ka, struct pt_regs *regs, err |= __copy_to_user(sf, rp, sizeof(struct reg_window32)); } if (err) - goto sigsegv; + return err; /* 3. signal handler back-trampoline and parameters */ regs->u_regs[UREG_FP] = (unsigned long) sf; - regs->u_regs[UREG_I0] = signo; + regs->u_regs[UREG_I0] = ksig->sig; regs->u_regs[UREG_I1] = (unsigned long) &sf->info; regs->u_regs[UREG_I2] = (unsigned long) &sf->info; /* 4. signal handler */ - regs->pc = (unsigned long) ka->sa.sa_handler; + regs->pc = (unsigned long) ksig->ka.sa.sa_handler; regs->npc = (regs->pc + 4); /* 5. return to kernel instructions */ - if (ka->ka_restorer) - regs->u_regs[UREG_I7] = (unsigned long)ka->ka_restorer; + if (ksig->ka.ka_restorer) + regs->u_regs[UREG_I7] = (unsigned long)ksig->ka.ka_restorer; else { regs->u_regs[UREG_I7] = (unsigned long)(&(sf->insns[0]) - 2); @@ -324,24 +299,16 @@ static int setup_frame(struct k_sigaction *ka, struct pt_regs *regs, /* t 0x10 */ err |= __put_user(0x91d02010, &sf->insns[1]); if (err) - goto sigsegv; + return err; /* Flush instruction space. */ flush_sig_insns(current->mm, (unsigned long) &(sf->insns[0])); } return 0; - -sigill_and_return: - do_exit(SIGILL); - return -EINVAL; - -sigsegv: - force_sigsegv(signo, current); - return -EFAULT; } -static int setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, - int signo, sigset_t *oldset, siginfo_t *info) +static int setup_rt_frame(struct ksignal *ksig, struct pt_regs *regs, + sigset_t *oldset) { struct rt_signal_frame __user *sf; int sigframe_size, wsaved; @@ -357,9 +324,11 @@ static int setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, if (wsaved) sigframe_size += sizeof(__siginfo_rwin_t); sf = (struct rt_signal_frame __user *) - get_sigframe(&ka->sa, regs, sigframe_size); - if (invalid_frame_pointer(sf, sigframe_size)) - goto sigill; + get_sigframe(ksig, regs, sigframe_size); + if (invalid_frame_pointer(sf, sigframe_size)) { + do_exit(SIGILL); + return -EINVAL; + } tail = sf + 1; err = __put_user(regs->pc, &sf->regs.pc); @@ -373,7 +342,7 @@ static int setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, err |= __put_user(0, &sf->extra_size); if (psr & PSR_EF) { - __siginfo_fpu_t *fp = tail; + __siginfo_fpu_t __user *fp = tail; tail += sizeof(*fp); err |= save_fpu_state(regs, fp); err |= __put_user(fp, &sf->fpu_save); @@ -381,7 +350,7 @@ static int setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, err |= __put_user(0, &sf->fpu_save); } if (wsaved) { - __siginfo_rwin_t *rwp = tail; + __siginfo_rwin_t __user *rwp = tail; tail += sizeof(*rwp); err |= save_rwin_state(wsaved, rwp); err |= __put_user(rwp, &sf->rwin_save); @@ -391,9 +360,7 @@ static int setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, err |= __copy_to_user(&sf->mask, &oldset->sig[0], sizeof(sigset_t)); /* Setup sigaltstack */ - err |= __put_user(current->sas_ss_sp, &sf->stack.ss_sp); - err |= __put_user(sas_ss_flags(regs->u_regs[UREG_FP]), &sf->stack.ss_flags); - err |= __put_user(current->sas_ss_size, &sf->stack.ss_size); + err |= __save_altstack(&sf->stack, regs->u_regs[UREG_FP]); if (!wsaved) { err |= __copy_to_user(sf, (char *) regs->u_regs[UREG_FP], @@ -405,21 +372,21 @@ static int setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, err |= __copy_to_user(sf, rp, sizeof(struct reg_window32)); } - err |= copy_siginfo_to_user(&sf->info, info); + err |= copy_siginfo_to_user(&sf->info, &ksig->info); if (err) - goto sigsegv; + return err; regs->u_regs[UREG_FP] = (unsigned long) sf; - regs->u_regs[UREG_I0] = signo; + regs->u_regs[UREG_I0] = ksig->sig; regs->u_regs[UREG_I1] = (unsigned long) &sf->info; regs->u_regs[UREG_I2] = (unsigned long) &sf->regs; - regs->pc = (unsigned long) ka->sa.sa_handler; + regs->pc = (unsigned long) ksig->ka.sa.sa_handler; regs->npc = (regs->pc + 4); - if (ka->ka_restorer) - regs->u_regs[UREG_I7] = (unsigned long)ka->ka_restorer; + if (ksig->ka.ka_restorer) + regs->u_regs[UREG_I7] = (unsigned long)ksig->ka.ka_restorer; else { regs->u_regs[UREG_I7] = (unsigned long)(&(sf->insns[0]) - 2); @@ -429,38 +396,25 @@ static int setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, /* t 0x10 */ err |= __put_user(0x91d02010, &sf->insns[1]); if (err) - goto sigsegv; + return err; /* Flush instruction space. */ flush_sig_insns(current->mm, (unsigned long) &(sf->insns[0])); } return 0; - -sigill: - do_exit(SIGILL); - return -EINVAL; - -sigsegv: - force_sigsegv(signo, current); - return -EFAULT; } static inline void -handle_signal(unsigned long signr, struct k_sigaction *ka, - siginfo_t *info, struct pt_regs *regs) +handle_signal(struct ksignal *ksig, struct pt_regs *regs) { sigset_t *oldset = sigmask_to_save(); int err; - if (ka->sa.sa_flags & SA_SIGINFO) - err = setup_rt_frame(ka, regs, signr, oldset, info); + if (ksig->ka.sa.sa_flags & SA_SIGINFO) + err = setup_rt_frame(ksig, regs, oldset); else - err = setup_frame(ka, regs, signr, oldset); - - if (err) - return; - - signal_delivered(signr, info, ka, regs, 0); + err = setup_frame(ksig, regs, oldset); + signal_setup_done(err, ksig, 0); } static inline void syscall_restart(unsigned long orig_i0, struct pt_regs *regs, @@ -490,10 +444,9 @@ static inline void syscall_restart(unsigned long orig_i0, struct pt_regs *regs, */ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) { - struct k_sigaction ka; + struct ksignal ksig; int restart_syscall; - siginfo_t info; - int signr; + bool has_handler; /* It's a lot of work and synchronization to add a new ptrace * register for GDB to save and restore in order to get @@ -516,7 +469,7 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) if (pt_regs_is_syscall(regs) && (regs->psr & PSR_C)) regs->u_regs[UREG_G6] = orig_i0; - signr = get_signal_to_deliver(&info, &ka, regs, NULL); + has_handler = get_signal(&ksig); /* If the debugger messes with the program counter, it clears * the software "in syscall" bit, directing us to not perform @@ -528,35 +481,30 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) orig_i0 = regs->u_regs[UREG_G6]; } - - if (signr > 0) { + if (has_handler) { if (restart_syscall) - syscall_restart(orig_i0, regs, &ka.sa); - handle_signal(signr, &ka, &info, regs); - return; - } - if (restart_syscall && - (regs->u_regs[UREG_I0] == ERESTARTNOHAND || - regs->u_regs[UREG_I0] == ERESTARTSYS || - regs->u_regs[UREG_I0] == ERESTARTNOINTR)) { - /* replay the system call when we are done */ - regs->u_regs[UREG_I0] = orig_i0; - regs->pc -= 4; - regs->npc -= 4; - pt_regs_clear_syscall(regs); - } - if (restart_syscall && - regs->u_regs[UREG_I0] == ERESTART_RESTARTBLOCK) { - regs->u_regs[UREG_G1] = __NR_restart_syscall; - regs->pc -= 4; - regs->npc -= 4; - pt_regs_clear_syscall(regs); + syscall_restart(orig_i0, regs, &ksig.ka.sa); + handle_signal(&ksig, regs); + } else { + if (restart_syscall) { + switch (regs->u_regs[UREG_I0]) { + case ERESTARTNOHAND: + case ERESTARTSYS: + case ERESTARTNOINTR: + /* replay the system call when we are done */ + regs->u_regs[UREG_I0] = orig_i0; + regs->pc -= 4; + regs->npc -= 4; + pt_regs_clear_syscall(regs); + case ERESTART_RESTARTBLOCK: + regs->u_regs[UREG_G1] = __NR_restart_syscall; + regs->pc -= 4; + regs->npc -= 4; + pt_regs_clear_syscall(regs); + } + } + restore_saved_sigmask(); } - - /* if there's no signal to deliver, we just put the saved sigmask - * back - */ - restore_saved_sigmask(); } void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0, @@ -570,9 +518,9 @@ void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0, } } -asmlinkage int -do_sys_sigstack(struct sigstack __user *ssptr, struct sigstack __user *ossptr, - unsigned long sp) +asmlinkage int do_sys_sigstack(struct sigstack __user *ssptr, + struct sigstack __user *ossptr, + unsigned long sp) { int ret = -EFAULT; diff --git a/arch/sparc/kernel/signal_64.c b/arch/sparc/kernel/signal_64.c index 867de2f8189..1a699986803 100644 --- a/arch/sparc/kernel/signal_64.c +++ b/arch/sparc/kernel/signal_64.c @@ -23,6 +23,7 @@ #include <linux/tty.h> #include <linux/binfmts.h> #include <linux/bitops.h> +#include <linux/context_tracking.h> #include <asm/uaccess.h> #include <asm/ptrace.h> @@ -34,15 +35,17 @@ #include <asm/switch_to.h> #include <asm/cacheflush.h> -#include "entry.h" -#include "systbls.h" #include "sigutil.h" +#include "systbls.h" +#include "kernel.h" +#include "entry.h" /* {set, get}context() needed for 64-bit SparcLinux userland. */ asmlinkage void sparc64_set_context(struct pt_regs *regs) { struct ucontext __user *ucp = (struct ucontext __user *) regs->u_regs[UREG_I0]; + enum ctx_state prev_state = exception_enter(); mc_gregset_t __user *grp; unsigned long pc, npc, tstate; unsigned long fp, i7; @@ -129,16 +132,19 @@ asmlinkage void sparc64_set_context(struct pt_regs *regs) } if (err) goto do_sigsegv; - +out: + exception_exit(prev_state); return; do_sigsegv: force_sig(SIGSEGV, current); + goto out; } asmlinkage void sparc64_get_context(struct pt_regs *regs) { struct ucontext __user *ucp = (struct ucontext __user *) regs->u_regs[UREG_I0]; + enum ctx_state prev_state = exception_enter(); mc_gregset_t __user *grp; mcontext_t __user *mcp; unsigned long fp, i7; @@ -220,10 +226,12 @@ asmlinkage void sparc64_get_context(struct pt_regs *regs) } if (err) goto do_sigsegv; - +out: + exception_exit(prev_state); return; do_sigsegv: force_sig(SIGSEGV, current); + goto out; } struct rt_signal_frame { @@ -236,23 +244,6 @@ struct rt_signal_frame { __siginfo_rwin_t *rwin_save; }; -static long _sigpause_common(old_sigset_t set) -{ - sigset_t blocked; - siginitset(&blocked, set); - return sigsuspend(&blocked); -} - -asmlinkage long sys_sigpause(unsigned int set) -{ - return _sigpause_common(set); -} - -asmlinkage long sys_sigsuspend(old_sigset_t set) -{ - return _sigpause_common(set); -} - void do_rt_sigreturn(struct pt_regs *regs) { struct rt_signal_frame __user *sf; @@ -295,8 +286,7 @@ void do_rt_sigreturn(struct pt_regs *regs) err |= restore_fpu_state(regs, fpu_save); err |= __copy_from_user(&set, &sf->mask, sizeof(sigset_t)); - err |= do_sigaltstack(&sf->stack, NULL, (unsigned long)sf); - + err |= restore_altstack(&sf->stack); if (err) goto segv; @@ -326,7 +316,7 @@ static int invalid_frame_pointer(void __user *fp) return 0; } -static inline void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, unsigned long framesize) +static inline void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs, unsigned long framesize) { unsigned long sp = regs->u_regs[UREG_FP] + STACK_BIAS; @@ -338,12 +328,7 @@ static inline void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs * return (void __user *) -1L; /* This is the X/Open sanctioned signal stack switching. */ - if (ka->sa.sa_flags & SA_ONSTACK) { - if (sas_ss_flags(sp) == 0) - sp = current->sas_ss_sp + current->sas_ss_size; - } - - sp -= framesize; + sp = sigsp(sp, ksig) - framesize; /* Always align the stack frame. This handles two cases. First, * sigaltstack need not be mindful of platform specific stack @@ -357,8 +342,7 @@ static inline void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs * } static inline int -setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, - int signo, sigset_t *oldset, siginfo_t *info) +setup_rt_frame(struct ksignal *ksig, struct pt_regs *regs) { struct rt_signal_frame __user *sf; int wsaved, err, sf_size; @@ -376,10 +360,12 @@ setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, if (wsaved) sf_size += sizeof(__siginfo_rwin_t); sf = (struct rt_signal_frame __user *) - get_sigframe(ka, regs, sf_size); + get_sigframe(ksig, regs, sf_size); - if (invalid_frame_pointer (sf)) - goto sigill; + if (invalid_frame_pointer (sf)) { + do_exit(SIGILL); /* won't return, actually */ + return -EINVAL; + } tail = (sf + 1); @@ -405,11 +391,9 @@ setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, } /* Setup sigaltstack */ - err |= __put_user(current->sas_ss_sp, &sf->stack.ss_sp); - err |= __put_user(sas_ss_flags(regs->u_regs[UREG_FP]), &sf->stack.ss_flags); - err |= __put_user(current->sas_ss_size, &sf->stack.ss_size); + err |= __save_altstack(&sf->stack, regs->u_regs[UREG_FP]); - err |= copy_to_user(&sf->mask, oldset, sizeof(sigset_t)); + err |= copy_to_user(&sf->mask, sigmask_to_save(), sizeof(sigset_t)); if (!wsaved) { err |= copy_in_user((u64 __user *)sf, @@ -422,18 +406,18 @@ setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, rp = ¤t_thread_info()->reg_window[wsaved - 1]; err |= copy_to_user(sf, rp, sizeof(struct reg_window)); } - if (info) - err |= copy_siginfo_to_user(&sf->info, info); + if (ksig->ka.sa.sa_flags & SA_SIGINFO) + err |= copy_siginfo_to_user(&sf->info, &ksig->info); else { - err |= __put_user(signo, &sf->info.si_signo); + err |= __put_user(ksig->sig, &sf->info.si_signo); err |= __put_user(SI_NOINFO, &sf->info.si_code); } if (err) - goto sigsegv; + return err; /* 3. signal handler back-trampoline and parameters */ regs->u_regs[UREG_FP] = ((unsigned long) sf) - STACK_BIAS; - regs->u_regs[UREG_I0] = signo; + regs->u_regs[UREG_I0] = ksig->sig; regs->u_regs[UREG_I1] = (unsigned long) &sf->info; /* The sigcontext is passed in this way because of how it @@ -443,37 +427,15 @@ setup_rt_frame(struct k_sigaction *ka, struct pt_regs *regs, regs->u_regs[UREG_I2] = (unsigned long) &sf->info; /* 5. signal handler */ - regs->tpc = (unsigned long) ka->sa.sa_handler; + regs->tpc = (unsigned long) ksig->ka.sa.sa_handler; regs->tnpc = (regs->tpc + 4); if (test_thread_flag(TIF_32BIT)) { regs->tpc &= 0xffffffff; regs->tnpc &= 0xffffffff; } /* 4. return to kernel instructions */ - regs->u_regs[UREG_I7] = (unsigned long)ka->ka_restorer; + regs->u_regs[UREG_I7] = (unsigned long)ksig->ka.ka_restorer; return 0; - -sigill: - do_exit(SIGILL); - return -EINVAL; - -sigsegv: - force_sigsegv(signo, current); - return -EFAULT; -} - -static inline void handle_signal(unsigned long signr, struct k_sigaction *ka, - siginfo_t *info, - sigset_t *oldset, struct pt_regs *regs) -{ - int err; - - err = setup_rt_frame(ka, regs, signr, oldset, - (ka->sa.sa_flags & SA_SIGINFO) ? info : NULL); - if (err) - return; - - signal_delivered(signr, info, ka, regs, 0); } static inline void syscall_restart(unsigned long orig_i0, struct pt_regs *regs, @@ -503,11 +465,9 @@ static inline void syscall_restart(unsigned long orig_i0, struct pt_regs *regs, */ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) { - struct k_sigaction ka; + struct ksignal ksig; int restart_syscall; - sigset_t *oldset = sigmask_to_save(); - siginfo_t info; - int signr; + bool has_handler; /* It's a lot of work and synchronization to add a new ptrace * register for GDB to save and restore in order to get @@ -533,13 +493,12 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) #ifdef CONFIG_COMPAT if (test_thread_flag(TIF_32BIT)) { - extern void do_signal32(sigset_t *, struct pt_regs *); - do_signal32(oldset, regs); + do_signal32(regs); return; } #endif - signr = get_signal_to_deliver(&info, &ka, regs, NULL); + has_handler = get_signal(&ksig); restart_syscall = 0; if (pt_regs_is_syscall(regs) && @@ -548,43 +507,41 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) orig_i0 = regs->u_regs[UREG_G6]; } - if (signr > 0) { + if (has_handler) { if (restart_syscall) - syscall_restart(orig_i0, regs, &ka.sa); - handle_signal(signr, &ka, &info, oldset, regs); - return; - } - if (restart_syscall && - (regs->u_regs[UREG_I0] == ERESTARTNOHAND || - regs->u_regs[UREG_I0] == ERESTARTSYS || - regs->u_regs[UREG_I0] == ERESTARTNOINTR)) { - /* replay the system call when we are done */ - regs->u_regs[UREG_I0] = orig_i0; - regs->tpc -= 4; - regs->tnpc -= 4; - pt_regs_clear_syscall(regs); - } - if (restart_syscall && - regs->u_regs[UREG_I0] == ERESTART_RESTARTBLOCK) { - regs->u_regs[UREG_G1] = __NR_restart_syscall; - regs->tpc -= 4; - regs->tnpc -= 4; - pt_regs_clear_syscall(regs); + syscall_restart(orig_i0, regs, &ksig.ka.sa); + signal_setup_done(setup_rt_frame(&ksig, regs), &ksig, 0); + } else { + if (restart_syscall) { + switch (regs->u_regs[UREG_I0]) { + case ERESTARTNOHAND: + case ERESTARTSYS: + case ERESTARTNOINTR: + /* replay the system call when we are done */ + regs->u_regs[UREG_I0] = orig_i0; + regs->tpc -= 4; + regs->tnpc -= 4; + pt_regs_clear_syscall(regs); + case ERESTART_RESTARTBLOCK: + regs->u_regs[UREG_G1] = __NR_restart_syscall; + regs->tpc -= 4; + regs->tnpc -= 4; + pt_regs_clear_syscall(regs); + } + } + restore_saved_sigmask(); } - - /* If there's no signal to deliver, we just put the saved sigmask - * back - */ - restore_saved_sigmask(); } void do_notify_resume(struct pt_regs *regs, unsigned long orig_i0, unsigned long thread_info_flags) { + user_exit(); if (thread_info_flags & _TIF_SIGPENDING) do_signal(regs, orig_i0); if (thread_info_flags & _TIF_NOTIFY_RESUME) { clear_thread_flag(TIF_NOTIFY_RESUME); tracehook_notify_resume(regs); } + user_enter(); } diff --git a/arch/sparc/kernel/smp_32.c b/arch/sparc/kernel/smp_32.c index 79db45e5134..7958242d63c 100644 --- a/arch/sparc/kernel/smp_32.c +++ b/arch/sparc/kernel/smp_32.c @@ -20,6 +20,8 @@ #include <linux/seq_file.h> #include <linux/cache.h> #include <linux/delay.h> +#include <linux/profile.h> +#include <linux/cpu.h> #include <asm/ptrace.h> #include <linux/atomic.h> @@ -32,11 +34,13 @@ #include <asm/cacheflush.h> #include <asm/tlbflush.h> #include <asm/cpudata.h> +#include <asm/timer.h> #include <asm/leon.h> +#include "kernel.h" #include "irq.h" -volatile unsigned long cpu_callin_map[NR_CPUS] __cpuinitdata = {0,}; +volatile unsigned long cpu_callin_map[NR_CPUS] = {0,}; cpumask_t smp_commenced_mask = CPU_MASK_NONE; @@ -50,7 +54,7 @@ const struct sparc32_ipi_ops *sparc32_ipi_ops; * instruction which is much better... */ -void __cpuinit smp_store_cpu_info(int id) +void smp_store_cpu_info(int id) { int cpu_node; int mid; @@ -72,8 +76,6 @@ void __cpuinit smp_store_cpu_info(int id) void __init smp_cpus_done(unsigned int max_cpus) { - extern void smp4m_smp_done(void); - extern void smp4d_smp_done(void); unsigned long bogosum = 0; int cpu, num = 0; @@ -117,7 +119,7 @@ void cpu_panic(void) panic("SMP bolixed\n"); } -struct linux_prom_registers smp_penguin_ctable __cpuinitdata = { 0 }; +struct linux_prom_registers smp_penguin_ctable = { 0 }; void smp_send_reschedule(int cpu) { @@ -180,8 +182,6 @@ int setup_profiling_timer(unsigned int multiplier) void __init smp_prepare_cpus(unsigned int max_cpus) { - extern void __init smp4m_boot_cpus(void); - extern void __init smp4d_boot_cpus(void); int i, cpuid, extra; printk("Entering SMP Mode...\n"); @@ -256,10 +256,8 @@ void __init smp_prepare_boot_cpu(void) set_cpu_possible(cpuid, true); } -int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *tidle) +int __cpu_up(unsigned int cpu, struct task_struct *tidle) { - extern int __cpuinit smp4m_boot_one_cpu(int, struct task_struct *); - extern int __cpuinit smp4d_boot_one_cpu(int, struct task_struct *); int ret=0; switch(sparc_cpu_model) { @@ -294,6 +292,89 @@ int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *tidle) return ret; } +static void arch_cpu_pre_starting(void *arg) +{ + local_ops->cache_all(); + local_ops->tlb_all(); + + switch(sparc_cpu_model) { + case sun4m: + sun4m_cpu_pre_starting(arg); + break; + case sun4d: + sun4d_cpu_pre_starting(arg); + break; + case sparc_leon: + leon_cpu_pre_starting(arg); + break; + default: + BUG(); + } +} + +static void arch_cpu_pre_online(void *arg) +{ + unsigned int cpuid = hard_smp_processor_id(); + + register_percpu_ce(cpuid); + + calibrate_delay(); + smp_store_cpu_info(cpuid); + + local_ops->cache_all(); + local_ops->tlb_all(); + + switch(sparc_cpu_model) { + case sun4m: + sun4m_cpu_pre_online(arg); + break; + case sun4d: + sun4d_cpu_pre_online(arg); + break; + case sparc_leon: + leon_cpu_pre_online(arg); + break; + default: + BUG(); + } +} + +static void sparc_start_secondary(void *arg) +{ + unsigned int cpu; + + /* + * SMP booting is extremely fragile in some architectures. So run + * the cpu initialization code first before anything else. + */ + arch_cpu_pre_starting(arg); + + preempt_disable(); + cpu = smp_processor_id(); + + /* Invoke the CPU_STARTING notifier callbacks */ + notify_cpu_starting(cpu); + + arch_cpu_pre_online(arg); + + /* Set the CPU in the cpu_online_mask */ + set_cpu_online(cpu, true); + + /* Enable local interrupts now */ + local_irq_enable(); + + wmb(); + cpu_startup_entry(CPUHP_ONLINE); + + /* We should never reach here! */ + BUG(); +} + +void smp_callin(void) +{ + sparc_start_secondary(NULL); +} + void smp_bogo(struct seq_file *m) { int i; diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c index f591598d92f..41aa2478f3c 100644 --- a/arch/sparc/kernel/smp_64.c +++ b/arch/sparc/kernel/smp_64.c @@ -25,6 +25,7 @@ #include <linux/ftrace.h> #include <linux/cpu.h> #include <linux/slab.h> +#include <linux/kgdb.h> #include <asm/head.h> #include <asm/ptrace.h> @@ -35,6 +36,7 @@ #include <asm/hvtramp.h> #include <asm/io.h> #include <asm/timer.h> +#include <asm/setup.h> #include <asm/irq.h> #include <asm/irq_regs.h> @@ -52,8 +54,7 @@ #include <asm/pcr.h> #include "cpumap.h" - -int sparc64_multi_core __read_mostly; +#include "kernel.h" DEFINE_PER_CPU(cpumask_t, cpu_sibling_map) = CPU_MASK_NONE; cpumask_t cpu_core_map[NR_CPUS] __read_mostly = @@ -87,7 +88,7 @@ extern void setup_sparc64_timer(void); static volatile unsigned long callin_flag = 0; -void __cpuinit smp_callin(void) +void smp_callin(void) { int cpuid = hard_smp_processor_id(); @@ -103,8 +104,6 @@ void __cpuinit smp_callin(void) if (cheetah_pcache_forced_on) cheetah_enable_pcache(); - local_irq_enable(); - callin_flag = 1; __asm__ __volatile__("membar #Sync\n\t" "flush %%g6" : : : "memory"); @@ -124,12 +123,14 @@ void __cpuinit smp_callin(void) while (!cpumask_test_cpu(cpuid, &smp_commenced_mask)) rmb(); - ipi_call_lock_irq(); set_cpu_online(cpuid, true); - ipi_call_unlock_irq(); /* idle thread is expected to have preempt disabled */ preempt_disable(); + + local_irq_enable(); + + cpu_startup_entry(CPUHP_ONLINE); } void cpu_panic(void) @@ -151,7 +152,7 @@ void cpu_panic(void) #define NUM_ROUNDS 64 /* magic value */ #define NUM_ITERS 5 /* likewise */ -static DEFINE_SPINLOCK(itc_sync_lock); +static DEFINE_RAW_SPINLOCK(itc_sync_lock); static unsigned long go[SLAVE + 1]; #define DEBUG_TICK_SYNC 0 @@ -259,7 +260,7 @@ static void smp_synchronize_one_tick(int cpu) go[MASTER] = 0; membar_safe("#StoreLoad"); - spin_lock_irqsave(&itc_sync_lock, flags); + raw_spin_lock_irqsave(&itc_sync_lock, flags); { for (i = 0; i < NUM_ROUNDS*NUM_ITERS; i++) { while (!go[MASTER]) @@ -270,19 +271,12 @@ static void smp_synchronize_one_tick(int cpu) membar_safe("#StoreLoad"); } } - spin_unlock_irqrestore(&itc_sync_lock, flags); + raw_spin_unlock_irqrestore(&itc_sync_lock, flags); } #if defined(CONFIG_SUN_LDOMS) && defined(CONFIG_HOTPLUG_CPU) -/* XXX Put this in some common place. XXX */ -static unsigned long kimage_addr_to_ra(void *p) -{ - unsigned long val = (unsigned long) p; - - return kern_base + (val - KERNBASE); -} - -static void __cpuinit ldom_startcpu_cpuid(unsigned int cpu, unsigned long thread_reg, void **descrp) +static void ldom_startcpu_cpuid(unsigned int cpu, unsigned long thread_reg, + void **descrp) { extern unsigned long sparc64_ttable_tl0; extern unsigned long kern_locked_tte_data; @@ -343,7 +337,7 @@ extern unsigned long sparc64_cpu_startup; */ static struct thread_info *cpu_new_thread = NULL; -static int __cpuinit smp_boot_one_cpu(unsigned int cpu, struct task_struct *idle) +static int smp_boot_one_cpu(unsigned int cpu, struct task_struct *idle) { unsigned long entry = (unsigned long)(&sparc64_cpu_startup); @@ -852,9 +846,11 @@ void smp_tsb_sync(struct mm_struct *mm) } extern unsigned long xcall_flush_tlb_mm; -extern unsigned long xcall_flush_tlb_pending; +extern unsigned long xcall_flush_tlb_page; extern unsigned long xcall_flush_tlb_kernel_range; extern unsigned long xcall_fetch_glob_regs; +extern unsigned long xcall_fetch_glob_pmu; +extern unsigned long xcall_fetch_glob_pmu_n4; extern unsigned long xcall_receive_signal; extern unsigned long xcall_new_mmu_context_version; #ifdef CONFIG_KGDB @@ -866,11 +862,6 @@ extern unsigned long xcall_flush_dcache_page_cheetah; #endif extern unsigned long xcall_flush_dcache_page_spitfire; -#ifdef CONFIG_DEBUG_DCFLUSH -extern atomic_t dcpage_flushes; -extern atomic_t dcpage_flushes_xcall; -#endif - static inline void __local_flush_dcache_page(struct page *page) { #ifdef DCACHE_ALIASING_POSSIBLE @@ -1003,6 +994,15 @@ void smp_fetch_global_regs(void) smp_cross_call(&xcall_fetch_glob_regs, 0, 0, 0); } +void smp_fetch_global_pmu(void) +{ + if (tlb_type == hypervisor && + sun4v_chip_type >= SUN4V_CHIP_NIAGARA4) + smp_cross_call(&xcall_fetch_glob_pmu_n4, 0, 0, 0); + else + smp_cross_call(&xcall_fetch_glob_pmu, 0, 0, 0); +} + /* We know that the window frames of the user have been flushed * to the stack before we get here because all callers of us * are flush_tlb_*() routines, and these run after flush_cache_*() @@ -1066,23 +1066,56 @@ local_flush_and_out: put_cpu(); } +struct tlb_pending_info { + unsigned long ctx; + unsigned long nr; + unsigned long *vaddrs; +}; + +static void tlb_pending_func(void *info) +{ + struct tlb_pending_info *t = info; + + __flush_tlb_pending(t->ctx, t->nr, t->vaddrs); +} + void smp_flush_tlb_pending(struct mm_struct *mm, unsigned long nr, unsigned long *vaddrs) { u32 ctx = CTX_HWBITS(mm->context); + struct tlb_pending_info info; int cpu = get_cpu(); + info.ctx = ctx; + info.nr = nr; + info.vaddrs = vaddrs; + if (mm == current->mm && atomic_read(&mm->mm_users) == 1) cpumask_copy(mm_cpumask(mm), cpumask_of(cpu)); else - smp_cross_call_masked(&xcall_flush_tlb_pending, - ctx, nr, (unsigned long) vaddrs, - mm_cpumask(mm)); + smp_call_function_many(mm_cpumask(mm), tlb_pending_func, + &info, 1); __flush_tlb_pending(ctx, nr, vaddrs); put_cpu(); } +void smp_flush_tlb_page(struct mm_struct *mm, unsigned long vaddr) +{ + unsigned long context = CTX_HWBITS(mm->context); + int cpu = get_cpu(); + + if (mm == current->mm && atomic_read(&mm->mm_users) == 1) + cpumask_copy(mm_cpumask(mm), cpumask_of(cpu)); + else + smp_cross_call_masked(&xcall_flush_tlb_page, + context, vaddr, 0, + mm_cpumask(mm)); + __flush_tlb_page(context, vaddr); + + put_cpu(); +} + void smp_flush_tlb_kernel_range(unsigned long start, unsigned long end) { start &= PAGE_MASK; @@ -1172,7 +1205,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus) { } -void __devinit smp_prepare_boot_cpu(void) +void smp_prepare_boot_cpu(void) { } @@ -1186,7 +1219,7 @@ void __init smp_setup_processor_id(void) xcall_deliver_impl = hypervisor_xcall_deliver; } -void __devinit smp_fill_in_sib_core_maps(void) +void smp_fill_in_sib_core_maps(void) { unsigned int i; @@ -1223,7 +1256,7 @@ void __devinit smp_fill_in_sib_core_maps(void) } } -int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *tidle) +int __cpu_up(unsigned int cpu, struct task_struct *tidle) { int ret = smp_boot_one_cpu(cpu, tidle); @@ -1308,9 +1341,7 @@ int __cpu_disable(void) mdelay(1); local_irq_disable(); - ipi_call_lock(); set_cpu_online(cpu, false); - ipi_call_unlock(); cpu_map_rebuild(); @@ -1357,8 +1388,13 @@ void __init smp_cpus_done(unsigned int max_cpus) void smp_send_reschedule(int cpu) { - xcall_deliver((u64) &xcall_receive_signal, 0, 0, - cpumask_of(cpu)); + if (cpu == smp_processor_id()) { + WARN_ON_ONCE(preemptible()); + set_softint(1 << PIL_SMP_RECEIVE_SIGNAL); + } else { + xcall_deliver((u64) &xcall_receive_signal, + 0, 0, cpumask_of(cpu)); + } } void __irq_entry smp_receive_signal_client(int irq, struct pt_regs *regs) diff --git a/arch/sparc/kernel/sparc_ksyms_32.c b/arch/sparc/kernel/sparc_ksyms_32.c index e521c54560f..bf4ccb10a78 100644 --- a/arch/sparc/kernel/sparc_ksyms_32.c +++ b/arch/sparc/kernel/sparc_ksyms_32.c @@ -6,7 +6,6 @@ */ #include <linux/module.h> -#include <linux/init.h> #include <asm/pgtable.h> #include <asm/uaccess.h> diff --git a/arch/sparc/kernel/sparc_ksyms_64.c b/arch/sparc/kernel/sparc_ksyms_64.c index 9f5e24ddcc7..a92d5d2c46a 100644 --- a/arch/sparc/kernel/sparc_ksyms_64.c +++ b/arch/sparc/kernel/sparc_ksyms_64.c @@ -7,7 +7,6 @@ #include <linux/export.h> #include <linux/pci.h> -#include <linux/init.h> #include <linux/bitops.h> #include <asm/cpudata.h> diff --git a/arch/sparc/kernel/sun4d_irq.c b/arch/sparc/kernel/sun4d_irq.c index e490ac9327c..a1bb2675b28 100644 --- a/arch/sparc/kernel/sun4d_irq.c +++ b/arch/sparc/kernel/sun4d_irq.c @@ -6,6 +6,7 @@ */ #include <linux/kernel_stat.h> +#include <linux/slab.h> #include <linux/seq_file.h> #include <asm/timer.h> @@ -142,7 +143,7 @@ static void sun4d_sbus_handler_irq(int sbusl) } } -void sun4d_handler_irq(int pil, struct pt_regs *regs) +void sun4d_handler_irq(unsigned int pil, struct pt_regs *regs) { struct pt_regs *old_regs; /* SBUS IRQ level (1 - 7) */ @@ -235,7 +236,7 @@ static void sun4d_shutdown_irq(struct irq_data *data) irq_unlink(data->irq); } -struct irq_chip sun4d_irq = { +static struct irq_chip sun4d_irq = { .name = "sun4d", .irq_startup = sun4d_startup_irq, .irq_shutdown = sun4d_shutdown_irq, @@ -284,9 +285,9 @@ static void __init sun4d_load_profile_irqs(void) } } -unsigned int _sun4d_build_device_irq(unsigned int real_irq, - unsigned int pil, - unsigned int board) +static unsigned int _sun4d_build_device_irq(unsigned int real_irq, + unsigned int pil, + unsigned int board) { struct sun4d_handler_data *handler_data; unsigned int irq; @@ -319,8 +320,8 @@ err_out: -unsigned int sun4d_build_device_irq(struct platform_device *op, - unsigned int real_irq) +static unsigned int sun4d_build_device_irq(struct platform_device *op, + unsigned int real_irq) { struct device_node *dp = op->dev.of_node; struct device_node *board_parent, *bus = dp->parent; @@ -382,7 +383,8 @@ err_out: return irq; } -unsigned int sun4d_build_timer_irq(unsigned int board, unsigned int real_irq) +static unsigned int sun4d_build_timer_irq(unsigned int board, + unsigned int real_irq) { return _sun4d_build_device_irq(real_irq, real_irq, board); } diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c index ddaea31de58..d5c319553fd 100644 --- a/arch/sparc/kernel/sun4d_smp.c +++ b/arch/sparc/kernel/sun4d_smp.c @@ -50,10 +50,9 @@ static inline void show_leds(int cpuid) "i" (ASI_M_CTL)); } -void __cpuinit smp4d_callin(void) +void sun4d_cpu_pre_starting(void *arg) { int cpuid = hard_smp_processor_id(); - unsigned long flags; /* Show we are alive */ cpu_leds[cpuid] = 0x6; @@ -61,26 +60,20 @@ void __cpuinit smp4d_callin(void) /* Enable level15 interrupt, disable level14 interrupt for now */ cc_set_imsk((cc_get_imsk() & ~0x8000) | 0x4000); +} - local_ops->cache_all(); - local_ops->tlb_all(); +void sun4d_cpu_pre_online(void *arg) +{ + unsigned long flags; + int cpuid; - notify_cpu_starting(cpuid); - /* - * Unblock the master CPU _only_ when the scheduler state + cpuid = hard_smp_processor_id(); + + /* Unblock the master CPU _only_ when the scheduler state * of all secondary CPUs will be up-to-date, so after * the SMP initialization the master will be just allowed * to call the scheduler code. */ - /* Get our local ticker going. */ - register_percpu_ce(cpuid); - - calibrate_delay(); - smp_store_cpu_info(cpuid); - local_ops->cache_all(); - local_ops->tlb_all(); - - /* Allow master to continue. */ sun4d_swap((unsigned long *)&cpu_callin_map[cpuid], 1); local_ops->cache_all(); local_ops->tlb_all(); @@ -106,16 +99,12 @@ void __cpuinit smp4d_callin(void) local_ops->cache_all(); local_ops->tlb_all(); - local_irq_enable(); /* We don't allow PIL 14 yet */ - while (!cpumask_test_cpu(cpuid, &smp_commenced_mask)) barrier(); spin_lock_irqsave(&sun4d_imsk_lock, flags); cc_set_imsk(cc_get_imsk() & ~0x4000); /* Allow PIL 14 as well */ spin_unlock_irqrestore(&sun4d_imsk_lock, flags); - set_cpu_online(cpuid, true); - } /* @@ -129,7 +118,7 @@ void __init smp4d_boot_cpus(void) local_ops->cache_all(); } -int __cpuinit smp4d_boot_one_cpu(int i, struct task_struct *idle) +int smp4d_boot_one_cpu(int i, struct task_struct *idle) { unsigned long *entry = &sun4d_cpu_startup; int timeout; diff --git a/arch/sparc/kernel/sun4m_irq.c b/arch/sparc/kernel/sun4m_irq.c index c5ade9d27a1..8bb3b3fddea 100644 --- a/arch/sparc/kernel/sun4m_irq.c +++ b/arch/sparc/kernel/sun4m_irq.c @@ -9,6 +9,8 @@ * Copyright (C) 1996 Dave Redman (djhr@tadpole.co.uk) */ +#include <linux/slab.h> + #include <asm/timer.h> #include <asm/traps.h> #include <asm/pgalloc.h> diff --git a/arch/sparc/kernel/sun4m_smp.c b/arch/sparc/kernel/sun4m_smp.c index 128af730428..d3408e72d20 100644 --- a/arch/sparc/kernel/sun4m_smp.c +++ b/arch/sparc/kernel/sun4m_smp.c @@ -34,30 +34,19 @@ swap_ulong(volatile unsigned long *ptr, unsigned long val) return val; } -void __cpuinit smp4m_callin(void) +void sun4m_cpu_pre_starting(void *arg) { - int cpuid = hard_smp_processor_id(); - - local_ops->cache_all(); - local_ops->tlb_all(); - - notify_cpu_starting(cpuid); - - register_percpu_ce(cpuid); - - calibrate_delay(); - smp_store_cpu_info(cpuid); +} - local_ops->cache_all(); - local_ops->tlb_all(); +void sun4m_cpu_pre_online(void *arg) +{ + int cpuid = hard_smp_processor_id(); - /* - * Unblock the master CPU _only_ when the scheduler state - * of all secondary CPUs will be up-to-date, so after - * the SMP initialization the master will be just allowed - * to call the scheduler code. + /* Allow master to continue. The master will then give us the + * go-ahead by setting the smp_commenced_mask and will wait without + * timeouts until our setup is completed fully (signified by + * our bit being set in the cpu_online_mask). */ - /* Allow master to continue. */ swap_ulong(&cpu_callin_map[cpuid], 1); /* XXX: What's up with all the flushes? */ @@ -75,10 +64,6 @@ void __cpuinit smp4m_callin(void) while (!cpumask_test_cpu(cpuid, &smp_commenced_mask)) mb(); - - local_irq_enable(); - - set_cpu_online(cpuid, true); } /* @@ -90,7 +75,7 @@ void __init smp4m_boot_cpus(void) local_ops->cache_all(); } -int __cpuinit smp4m_boot_one_cpu(int i, struct task_struct *idle) +int smp4m_boot_one_cpu(int i, struct task_struct *idle) { unsigned long *entry = &sun4m_cpu_startup; int timeout; diff --git a/arch/sparc/kernel/sun4v_tlb_miss.S b/arch/sparc/kernel/sun4v_tlb_miss.S index e1fbf8c7578..e0c09bf8561 100644 --- a/arch/sparc/kernel/sun4v_tlb_miss.S +++ b/arch/sparc/kernel/sun4v_tlb_miss.S @@ -176,13 +176,13 @@ sun4v_tsb_miss_common: sub %g2, TRAP_PER_CPU_FAULT_INFO, %g2 -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) mov SCRATCHPAD_UTSBREG2, %g5 ldxa [%g5] ASI_SCRATCHPAD, %g5 cmp %g5, -1 be,pt %xcc, 80f nop - COMPUTE_TSB_PTR(%g5, %g4, HPAGE_SHIFT, %g2, %g7) + COMPUTE_TSB_PTR(%g5, %g4, REAL_HPAGE_SHIFT, %g2, %g7) /* That clobbered %g2, reload it. */ ldxa [%g0] ASI_SCRATCHPAD, %g2 diff --git a/arch/sparc/kernel/sys32.S b/arch/sparc/kernel/sys32.S index d97f3eb72e0..f834224208e 100644 --- a/arch/sparc/kernel/sys32.S +++ b/arch/sparc/kernel/sys32.S @@ -36,108 +36,19 @@ STUB: sra REG1, 0, REG1; \ jmpl %g1 + %lo(SYSCALL), %g0; \ sra REG3, 0, REG3 -#define SIGN4(STUB,SYSCALL,REG1,REG2,REG3,REG4) \ - .align 32; \ - .globl STUB; \ -STUB: sra REG1, 0, REG1; \ - sethi %hi(SYSCALL), %g1; \ - sra REG2, 0, REG2; \ - sra REG3, 0, REG3; \ - jmpl %g1 + %lo(SYSCALL), %g0; \ - sra REG4, 0, REG4 - -SIGN1(sys32_exit, sparc_exit, %o0) -SIGN1(sys32_exit_group, sys_exit_group, %o0) -SIGN1(sys32_wait4, compat_sys_wait4, %o2) -SIGN1(sys32_creat, sys_creat, %o1) -SIGN1(sys32_mknod, sys_mknod, %o1) -SIGN1(sys32_umount, sys_umount, %o1) -SIGN1(sys32_signal, sys_signal, %o0) -SIGN1(sys32_access, sys_access, %o1) -SIGN1(sys32_msync, sys_msync, %o2) -SIGN2(sys32_reboot, sys_reboot, %o0, %o1) -SIGN1(sys32_setitimer, compat_sys_setitimer, %o0) -SIGN1(sys32_getitimer, compat_sys_getitimer, %o0) -SIGN1(sys32_sethostname, sys_sethostname, %o1) -SIGN1(sys32_swapon, sys_swapon, %o1) -SIGN1(sys32_sigaction, compat_sys_sigaction, %o0) -SIGN1(sys32_rt_sigaction, compat_sys_rt_sigaction, %o0) -SIGN1(sys32_sigprocmask, compat_sys_sigprocmask, %o0) -SIGN1(sys32_rt_sigprocmask, compat_sys_rt_sigprocmask, %o0) -SIGN2(sys32_rt_sigqueueinfo, compat_sys_rt_sigqueueinfo, %o0, %o1) -SIGN1(sys32_getrusage, compat_sys_getrusage, %o0) -SIGN1(sys32_setxattr, sys_setxattr, %o4) -SIGN1(sys32_lsetxattr, sys_lsetxattr, %o4) -SIGN1(sys32_fsetxattr, sys_fsetxattr, %o4) -SIGN1(sys32_fgetxattr, sys_fgetxattr, %o0) -SIGN1(sys32_flistxattr, sys_flistxattr, %o0) -SIGN1(sys32_fremovexattr, sys_fremovexattr, %o0) -SIGN2(sys32_tkill, sys_tkill, %o0, %o1) -SIGN1(sys32_epoll_create, sys_epoll_create, %o0) -SIGN3(sys32_epoll_ctl, sys_epoll_ctl, %o0, %o1, %o2) -SIGN3(sys32_epoll_wait, sys_epoll_wait, %o0, %o2, %o3) SIGN1(sys32_readahead, compat_sys_readahead, %o0) SIGN2(sys32_fadvise64, compat_sys_fadvise64, %o0, %o4) SIGN2(sys32_fadvise64_64, compat_sys_fadvise64_64, %o0, %o5) -SIGN2(sys32_bdflush, sys_bdflush, %o0, %o1) -SIGN1(sys32_mlockall, sys_mlockall, %o0) SIGN1(sys32_clock_nanosleep, compat_sys_clock_nanosleep, %o1) SIGN1(sys32_timer_settime, compat_sys_timer_settime, %o1) SIGN1(sys32_io_submit, compat_sys_io_submit, %o1) SIGN1(sys32_mq_open, compat_sys_mq_open, %o1) SIGN1(sys32_select, compat_sys_select, %o0) -SIGN1(sys32_mkdir, sys_mkdir, %o1) -SIGN3(sys32_futex, compat_sys_futex, %o1, %o2, %o5) -SIGN1(sys32_sysfs, compat_sys_sysfs, %o0) -SIGN2(sys32_sendfile, compat_sys_sendfile, %o0, %o1) -SIGN2(sys32_sendfile64, compat_sys_sendfile64, %o0, %o1) -SIGN1(sys32_prctl, sys_prctl, %o0) -SIGN1(sys32_sched_rr_get_interval, compat_sys_sched_rr_get_interval, %o0) -SIGN2(sys32_waitpid, sys_waitpid, %o0, %o2) -SIGN1(sys32_getgroups, sys_getgroups, %o0) -SIGN1(sys32_getpgid, sys_getpgid, %o0) -SIGN2(sys32_getpriority, sys_getpriority, %o0, %o1) -SIGN1(sys32_getsid, sys_getsid, %o0) -SIGN2(sys32_kill, sys_kill, %o0, %o1) -SIGN1(sys32_nice, sys_nice, %o0) -SIGN1(sys32_lseek, sys_lseek, %o1) -SIGN2(sys32_open, sparc32_open, %o1, %o2) -SIGN1(sys32_readlink, sys_readlink, %o2) -SIGN1(sys32_sched_get_priority_max, sys_sched_get_priority_max, %o0) -SIGN1(sys32_sched_get_priority_min, sys_sched_get_priority_min, %o0) -SIGN1(sys32_sched_getparam, sys_sched_getparam, %o0) -SIGN1(sys32_sched_getscheduler, sys_sched_getscheduler, %o0) -SIGN1(sys32_sched_setparam, sys_sched_setparam, %o0) -SIGN2(sys32_sched_setscheduler, sys_sched_setscheduler, %o0, %o1) -SIGN1(sys32_getdomainname, sys_getdomainname, %o1) -SIGN1(sys32_setdomainname, sys_setdomainname, %o1) -SIGN1(sys32_setgroups, sys_setgroups, %o0) -SIGN2(sys32_setpgid, sys_setpgid, %o0, %o1) -SIGN3(sys32_setpriority, sys_setpriority, %o0, %o1, %o2) -SIGN1(sys32_ssetmask, sys_ssetmask, %o0) -SIGN2(sys32_syslog, sys_syslog, %o0, %o2) -SIGN1(sys32_umask, sys_umask, %o0) -SIGN3(sys32_tgkill, sys_tgkill, %o0, %o1, %o2) -SIGN1(sys32_sendto, sys_sendto, %o0) +SIGN1(sys32_futex, compat_sys_futex, %o1) SIGN1(sys32_recvfrom, compat_sys_recvfrom, %o0) -SIGN3(sys32_socket, sys_socket, %o0, %o1, %o2) -SIGN2(sys32_connect, sys_connect, %o0, %o2) -SIGN2(sys32_bind, sys_bind, %o0, %o2) -SIGN2(sys32_listen, sys_listen, %o0, %o1) SIGN1(sys32_recvmsg, compat_sys_recvmsg, %o0) SIGN1(sys32_sendmsg, compat_sys_sendmsg, %o0) -SIGN2(sys32_shutdown, sys_shutdown, %o0, %o1) -SIGN3(sys32_socketpair, sys_socketpair, %o0, %o1, %o2) -SIGN1(sys32_getpeername, sys_getpeername, %o0) -SIGN1(sys32_getsockname, sys_getsockname, %o0) -SIGN2(sys32_ioprio_get, sys_ioprio_get, %o0, %o1) -SIGN3(sys32_ioprio_set, sys_ioprio_set, %o0, %o1, %o2) -SIGN2(sys32_splice, sys_splice, %o0, %o2) -SIGN2(sys32_sync_file_range, compat_sync_file_range, %o0, %o5) -SIGN2(sys32_tee, sys_tee, %o0, %o1) -SIGN1(sys32_vmsplice, compat_sys_vmsplice, %o0) -SIGN1(sys32_truncate, sys_truncate, %o1) -SIGN1(sys32_ftruncate, sys_ftruncate, %o1) +SIGN2(sys32_renameat2, sys_renameat2, %o0, %o2) .globl sys32_mmap2 sys32_mmap2: @@ -329,15 +240,6 @@ do_sys_accept4: /* sys_accept4(int, struct sockaddr *, int *, int) */ nop nop - .globl sys32_fanotify_mark -sys32_fanotify_mark: - sethi %hi(sys_fanotify_mark), %g1 - sllx %o2, 32, %o2 - or %o2, %o3, %o2 - mov %o4, %o3 - jmpl %g1 + %lo(sys_fanotify_mark), %g0 - mov %o5, %o4 - .section __ex_table,"a" .align 4 .word 1b, __retl_efault, 2b, __retl_efault diff --git a/arch/sparc/kernel/sys_sparc32.c b/arch/sparc/kernel/sys_sparc32.c index f7392336961..022c30c72eb 100644 --- a/arch/sparc/kernel/sys_sparc32.c +++ b/arch/sparc/kernel/sys_sparc32.c @@ -49,70 +49,7 @@ #include <asm/mmu_context.h> #include <asm/compat_signal.h> -#ifdef CONFIG_SYSVIPC -asmlinkage long compat_sys_ipc(u32 call, u32 first, u32 second, u32 third, compat_uptr_t ptr, u32 fifth) -{ - int version; - - version = call >> 16; /* hack for backward compatibility */ - call &= 0xffff; - - switch (call) { - case SEMTIMEDOP: - if (fifth) - /* sign extend semid */ - return compat_sys_semtimedop((int)first, - compat_ptr(ptr), second, - compat_ptr(fifth)); - /* else fall through for normal semop() */ - case SEMOP: - /* struct sembuf is the same on 32 and 64bit :)) */ - /* sign extend semid */ - return sys_semtimedop((int)first, compat_ptr(ptr), second, - NULL); - case SEMGET: - /* sign extend key, nsems */ - return sys_semget((int)first, (int)second, third); - case SEMCTL: - /* sign extend semid, semnum */ - return compat_sys_semctl((int)first, (int)second, third, - compat_ptr(ptr)); - - case MSGSND: - /* sign extend msqid */ - return compat_sys_msgsnd((int)first, (int)second, third, - compat_ptr(ptr)); - case MSGRCV: - /* sign extend msqid, msgtyp */ - return compat_sys_msgrcv((int)first, second, (int)fifth, - third, version, compat_ptr(ptr)); - case MSGGET: - /* sign extend key */ - return sys_msgget((int)first, second); - case MSGCTL: - /* sign extend msqid */ - return compat_sys_msgctl((int)first, second, compat_ptr(ptr)); - - case SHMAT: - /* sign extend shmid */ - return compat_sys_shmat((int)first, second, third, version, - compat_ptr(ptr)); - case SHMDT: - return sys_shmdt(compat_ptr(ptr)); - case SHMGET: - /* sign extend key_t */ - return sys_shmget((int)first, second, third); - case SHMCTL: - /* sign extend shmid */ - return compat_sys_shmctl((int)first, second, compat_ptr(ptr)); - - default: - return -ENOSYS; - } - - return -ENOSYS; -} -#endif +#include "systbls.h" asmlinkage long sys32_truncate64(const char __user * path, unsigned long high, unsigned long low) { @@ -206,147 +143,19 @@ asmlinkage long compat_sys_fstatat64(unsigned int dfd, return cp_compat_stat64(&stat, statbuf); } -asmlinkage long compat_sys_sysfs(int option, u32 arg1, u32 arg2) -{ - return sys_sysfs(option, arg1, arg2); -} - -asmlinkage long compat_sys_sched_rr_get_interval(compat_pid_t pid, struct compat_timespec __user *interval) -{ - struct timespec t; - int ret; - mm_segment_t old_fs = get_fs (); - - set_fs (KERNEL_DS); - ret = sys_sched_rr_get_interval(pid, (struct timespec __user *) &t); - set_fs (old_fs); - if (put_compat_timespec(&t, interval)) - return -EFAULT; - return ret; -} - -asmlinkage long compat_sys_rt_sigprocmask(int how, - compat_sigset_t __user *set, - compat_sigset_t __user *oset, - compat_size_t sigsetsize) +COMPAT_SYSCALL_DEFINE3(sparc_sigaction, int, sig, + struct compat_old_sigaction __user *,act, + struct compat_old_sigaction __user *,oact) { - sigset_t s; - compat_sigset_t s32; - int ret; - mm_segment_t old_fs = get_fs(); - - if (set) { - if (copy_from_user (&s32, set, sizeof(compat_sigset_t))) - return -EFAULT; - switch (_NSIG_WORDS) { - case 4: s.sig[3] = s32.sig[6] | (((long)s32.sig[7]) << 32); - case 3: s.sig[2] = s32.sig[4] | (((long)s32.sig[5]) << 32); - case 2: s.sig[1] = s32.sig[2] | (((long)s32.sig[3]) << 32); - case 1: s.sig[0] = s32.sig[0] | (((long)s32.sig[1]) << 32); - } - } - set_fs (KERNEL_DS); - ret = sys_rt_sigprocmask(how, - set ? (sigset_t __user *) &s : NULL, - oset ? (sigset_t __user *) &s : NULL, - sigsetsize); - set_fs (old_fs); - if (ret) return ret; - if (oset) { - switch (_NSIG_WORDS) { - case 4: s32.sig[7] = (s.sig[3] >> 32); s32.sig[6] = s.sig[3]; - case 3: s32.sig[5] = (s.sig[2] >> 32); s32.sig[4] = s.sig[2]; - case 2: s32.sig[3] = (s.sig[1] >> 32); s32.sig[2] = s.sig[1]; - case 1: s32.sig[1] = (s.sig[0] >> 32); s32.sig[0] = s.sig[0]; - } - if (copy_to_user (oset, &s32, sizeof(compat_sigset_t))) - return -EFAULT; - } - return 0; -} - -asmlinkage long sys32_rt_sigpending(compat_sigset_t __user *set, - compat_size_t sigsetsize) -{ - sigset_t s; - compat_sigset_t s32; - int ret; - mm_segment_t old_fs = get_fs(); - - set_fs (KERNEL_DS); - ret = sys_rt_sigpending((sigset_t __user *) &s, sigsetsize); - set_fs (old_fs); - if (!ret) { - switch (_NSIG_WORDS) { - case 4: s32.sig[7] = (s.sig[3] >> 32); s32.sig[6] = s.sig[3]; - case 3: s32.sig[5] = (s.sig[2] >> 32); s32.sig[4] = s.sig[2]; - case 2: s32.sig[3] = (s.sig[1] >> 32); s32.sig[2] = s.sig[1]; - case 1: s32.sig[1] = (s.sig[0] >> 32); s32.sig[0] = s.sig[0]; - } - if (copy_to_user (set, &s32, sizeof(compat_sigset_t))) - return -EFAULT; - } - return ret; -} - -asmlinkage long compat_sys_rt_sigqueueinfo(int pid, int sig, - struct compat_siginfo __user *uinfo) -{ - siginfo_t info; - int ret; - mm_segment_t old_fs = get_fs(); - - if (copy_siginfo_from_user32(&info, uinfo)) - return -EFAULT; - - set_fs (KERNEL_DS); - ret = sys_rt_sigqueueinfo(pid, sig, (siginfo_t __user *) &info); - set_fs (old_fs); - return ret; -} - -asmlinkage long compat_sys_sigaction(int sig, struct old_sigaction32 __user *act, - struct old_sigaction32 __user *oact) -{ - struct k_sigaction new_ka, old_ka; - int ret; - WARN_ON_ONCE(sig >= 0); - sig = -sig; - - if (act) { - compat_old_sigset_t mask; - u32 u_handler, u_restorer; - - ret = get_user(u_handler, &act->sa_handler); - new_ka.sa.sa_handler = compat_ptr(u_handler); - ret |= __get_user(u_restorer, &act->sa_restorer); - new_ka.sa.sa_restorer = compat_ptr(u_restorer); - ret |= __get_user(new_ka.sa.sa_flags, &act->sa_flags); - ret |= __get_user(mask, &act->sa_mask); - if (ret) - return ret; - new_ka.ka_restorer = NULL; - siginitset(&new_ka.sa.sa_mask, mask); - } - - ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); - - if (!ret && oact) { - ret = put_user(ptr_to_compat(old_ka.sa.sa_handler), &oact->sa_handler); - ret |= __put_user(ptr_to_compat(old_ka.sa.sa_restorer), &oact->sa_restorer); - ret |= __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - ret |= __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); - } - - return ret; + return compat_sys_sigaction(-sig, act, oact); } -asmlinkage long compat_sys_rt_sigaction(int sig, - struct sigaction32 __user *act, - struct sigaction32 __user *oact, - void __user *restorer, - compat_size_t sigsetsize) +COMPAT_SYSCALL_DEFINE5(rt_sigaction, int, sig, + struct compat_sigaction __user *,act, + struct compat_sigaction __user *,oact, + void __user *,restorer, + compat_size_t,sigsetsize) { struct k_sigaction new_ka, old_ka; int ret; @@ -362,15 +171,10 @@ asmlinkage long compat_sys_rt_sigaction(int sig, new_ka.ka_restorer = restorer; ret = get_user(u_handler, &act->sa_handler); new_ka.sa.sa_handler = compat_ptr(u_handler); - ret |= __copy_from_user(&set32, &act->sa_mask, sizeof(compat_sigset_t)); - switch (_NSIG_WORDS) { - case 4: new_ka.sa.sa_mask.sig[3] = set32.sig[6] | (((long)set32.sig[7]) << 32); - case 3: new_ka.sa.sa_mask.sig[2] = set32.sig[4] | (((long)set32.sig[5]) << 32); - case 2: new_ka.sa.sa_mask.sig[1] = set32.sig[2] | (((long)set32.sig[3]) << 32); - case 1: new_ka.sa.sa_mask.sig[0] = set32.sig[0] | (((long)set32.sig[1]) << 32); - } - ret |= __get_user(new_ka.sa.sa_flags, &act->sa_flags); - ret |= __get_user(u_restorer, &act->sa_restorer); + ret |= copy_from_user(&set32, &act->sa_mask, sizeof(compat_sigset_t)); + sigset_from_compat(&new_ka.sa.sa_mask, &set32); + ret |= get_user(new_ka.sa.sa_flags, &act->sa_flags); + ret |= get_user(u_restorer, &act->sa_restorer); new_ka.sa.sa_restorer = compat_ptr(u_restorer); if (ret) return -EFAULT; @@ -379,16 +183,11 @@ asmlinkage long compat_sys_rt_sigaction(int sig, ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); if (!ret && oact) { - switch (_NSIG_WORDS) { - case 4: set32.sig[7] = (old_ka.sa.sa_mask.sig[3] >> 32); set32.sig[6] = old_ka.sa.sa_mask.sig[3]; - case 3: set32.sig[5] = (old_ka.sa.sa_mask.sig[2] >> 32); set32.sig[4] = old_ka.sa.sa_mask.sig[2]; - case 2: set32.sig[3] = (old_ka.sa.sa_mask.sig[1] >> 32); set32.sig[2] = old_ka.sa.sa_mask.sig[1]; - case 1: set32.sig[1] = (old_ka.sa.sa_mask.sig[0] >> 32); set32.sig[0] = old_ka.sa.sa_mask.sig[0]; - } + sigset_to_compat(&set32, &old_ka.sa.sa_mask); ret = put_user(ptr_to_compat(old_ka.sa.sa_handler), &oact->sa_handler); - ret |= __copy_to_user(&oact->sa_mask, &set32, sizeof(compat_sigset_t)); - ret |= __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - ret |= __put_user(ptr_to_compat(old_ka.sa.sa_restorer), &oact->sa_restorer); + ret |= copy_to_user(&oact->sa_mask, &set32, sizeof(compat_sigset_t)); + ret |= put_user(old_ka.sa.sa_flags, &oact->sa_flags); + ret |= put_user(ptr_to_compat(old_ka.sa.sa_restorer), &oact->sa_restorer); if (ret) ret = -EFAULT; } @@ -396,71 +195,6 @@ asmlinkage long compat_sys_rt_sigaction(int sig, return ret; } -/* - * sparc32_execve() executes a new program after the asm stub has set - * things up for us. This should basically do what I want it to. - */ -asmlinkage long sparc32_execve(struct pt_regs *regs) -{ - int error, base = 0; - char *filename; - - /* User register window flush is done by entry.S */ - - /* Check for indirect call. */ - if ((u32)regs->u_regs[UREG_G1] == 0) - base = 1; - - filename = getname(compat_ptr(regs->u_regs[base + UREG_I0])); - error = PTR_ERR(filename); - if (IS_ERR(filename)) - goto out; - - error = compat_do_execve(filename, - compat_ptr(regs->u_regs[base + UREG_I1]), - compat_ptr(regs->u_regs[base + UREG_I2]), regs); - - putname(filename); - - if (!error) { - fprs_write(0); - current_thread_info()->xfsr[0] = 0; - current_thread_info()->fpsaved[0] = 0; - regs->tstate &= ~TSTATE_PEF; - } -out: - return error; -} - -#ifdef CONFIG_MODULES - -asmlinkage long sys32_init_module(void __user *umod, u32 len, - const char __user *uargs) -{ - return sys_init_module(umod, len, uargs); -} - -asmlinkage long sys32_delete_module(const char __user *name_user, - unsigned int flags) -{ - return sys_delete_module(name_user, flags); -} - -#else /* CONFIG_MODULES */ - -asmlinkage long sys32_init_module(const char __user *name_user, - struct module __user *mod_user) -{ - return -ENOSYS; -} - -asmlinkage long sys32_delete_module(const char __user *name_user) -{ - return -ENOSYS; -} - -#endif /* CONFIG_MODULES */ - asmlinkage compat_ssize_t sys32_pread64(unsigned int fd, char __user *ubuf, compat_size_t count, @@ -506,71 +240,7 @@ long compat_sys_fadvise64_64(int fd, advice); } -asmlinkage long compat_sys_sendfile(int out_fd, int in_fd, - compat_off_t __user *offset, - compat_size_t count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - off_t of; - - if (offset && get_user(of, offset)) - return -EFAULT; - - set_fs(KERNEL_DS); - ret = sys_sendfile(out_fd, in_fd, - offset ? (off_t __user *) &of : NULL, - count); - set_fs(old_fs); - - if (offset && put_user(of, offset)) - return -EFAULT; - - return ret; -} - -asmlinkage long compat_sys_sendfile64(int out_fd, int in_fd, - compat_loff_t __user *offset, - compat_size_t count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - loff_t lof; - - if (offset && get_user(lof, offset)) - return -EFAULT; - - set_fs(KERNEL_DS); - ret = sys_sendfile64(out_fd, in_fd, - offset ? (loff_t __user *) &lof : NULL, - count); - set_fs(old_fs); - - if (offset && put_user(lof, offset)) - return -EFAULT; - - return ret; -} - -/* This is just a version for 32-bit applications which does - * not force O_LARGEFILE on. - */ - -asmlinkage long sparc32_open(const char __user *filename, - int flags, int mode) -{ - return do_sys_open(AT_FDCWD, filename, flags, mode); -} - -long sys32_lookup_dcookie(unsigned long cookie_high, - unsigned long cookie_low, - char __user *buf, size_t len) -{ - return sys_lookup_dcookie((cookie_high << 32) | cookie_low, - buf, len); -} - -long compat_sync_file_range(int fd, unsigned long off_high, unsigned long off_low, unsigned long nb_high, unsigned long nb_low, int flags) +long sys32_sync_file_range(unsigned int fd, unsigned long off_high, unsigned long off_low, unsigned long nb_high, unsigned long nb_low, unsigned int flags) { return sys_sync_file_range(fd, (off_high << 32) | off_low, diff --git a/arch/sparc/kernel/sys_sparc_32.c b/arch/sparc/kernel/sys_sparc_32.c index 0c9b31b22e0..646988d4c1a 100644 --- a/arch/sparc/kernel/sys_sparc_32.c +++ b/arch/sparc/kernel/sys_sparc_32.c @@ -24,6 +24,8 @@ #include <asm/uaccess.h> #include <asm/unistd.h> +#include "systbls.h" + /* #define DEBUG_UNIMP_SYSCALL */ /* XXX Make this per-binary type, this way we can detect the type of @@ -34,11 +36,9 @@ asmlinkage unsigned long sys_getpagesize(void) return PAGE_SIZE; /* Possibly older binaries want 8192 on sun4's? */ } -#define COLOUR_ALIGN(addr) (((addr)+SHMLBA-1)&~(SHMLBA-1)) - unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { - struct vm_area_struct * vmm; + struct vm_unmapped_area_info info; if (flags & MAP_FIXED) { /* We do not accept a shared mapping if it would violate @@ -56,28 +56,21 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi if (!addr) addr = TASK_UNMAPPED_BASE; - if (flags & MAP_SHARED) - addr = COLOUR_ALIGN(addr); - else - addr = PAGE_ALIGN(addr); - - for (vmm = find_vma(current->mm, addr); ; vmm = vmm->vm_next) { - /* At this point: (!vmm || addr < vmm->vm_end). */ - if (TASK_SIZE - PAGE_SIZE - len < addr) - return -ENOMEM; - if (!vmm || addr + len <= vmm->vm_start) - return addr; - addr = vmm->vm_end; - if (flags & MAP_SHARED) - addr = COLOUR_ALIGN(addr); - } + info.flags = 0; + info.length = len; + info.low_limit = addr; + info.high_limit = TASK_SIZE; + info.align_mask = (flags & MAP_SHARED) ? + (PAGE_MASK & (SHMLBA - 1)) : 0; + info.align_offset = pgoff << PAGE_SHIFT; + return vm_unmapped_area(&info); } /* * sys_pipe() is the normal C calling standard for creating * a pipe. It's not the way unix traditionally does this, though. */ -asmlinkage int sparc_pipe(struct pt_regs *regs) +asmlinkage long sparc_pipe(struct pt_regs *regs) { int fd[2]; int error; @@ -102,7 +95,7 @@ int sparc_mmap_check(unsigned long addr, unsigned long len) /* Linux version of mmap */ -asmlinkage unsigned long sys_mmap2(unsigned long addr, unsigned long len, +asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) { @@ -112,7 +105,7 @@ asmlinkage unsigned long sys_mmap2(unsigned long addr, unsigned long len, pgoff >> (PAGE_SHIFT - 12)); } -asmlinkage unsigned long sys_mmap(unsigned long addr, unsigned long len, +asmlinkage long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long off) { @@ -169,49 +162,19 @@ sparc_breakpoint (struct pt_regs *regs) #endif } -asmlinkage int -sparc_sigaction (int sig, const struct old_sigaction __user *act, - struct old_sigaction __user *oact) +SYSCALL_DEFINE3(sparc_sigaction, int, sig, + struct old_sigaction __user *,act, + struct old_sigaction __user *,oact) { - struct k_sigaction new_ka, old_ka; - int ret; - WARN_ON_ONCE(sig >= 0); - sig = -sig; - - if (act) { - unsigned long mask; - - if (!access_ok(VERIFY_READ, act, sizeof(*act)) || - __get_user(new_ka.sa.sa_handler, &act->sa_handler) || - __get_user(new_ka.sa.sa_restorer, &act->sa_restorer) || - __get_user(new_ka.sa.sa_flags, &act->sa_flags) || - __get_user(mask, &act->sa_mask)) - return -EFAULT; - siginitset(&new_ka.sa.sa_mask, mask); - new_ka.ka_restorer = NULL; - } - - ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); - - if (!ret && oact) { - if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || - __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || - __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer) || - __put_user(old_ka.sa.sa_flags, &oact->sa_flags) || - __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask)) - return -EFAULT; - } - - return ret; + return sys_sigaction(-sig, act, oact); } -asmlinkage long -sys_rt_sigaction(int sig, - const struct sigaction __user *act, - struct sigaction __user *oact, - void __user *restorer, - size_t sigsetsize) +SYSCALL_DEFINE5(rt_sigaction, int, sig, + const struct sigaction __user *, act, + struct sigaction __user *, oact, + void __user *, restorer, + size_t, sigsetsize) { struct k_sigaction new_ka, old_ka; int ret; @@ -236,7 +199,7 @@ sys_rt_sigaction(int sig, return ret; } -asmlinkage int sys_getdomainname(char __user *name, int len) +asmlinkage long sys_getdomainname(char __user *name, int len) { int nlen, err; @@ -258,27 +221,3 @@ out: up_read(&uts_sem); return err; } - -/* - * Do a system call from kernel instead of calling sys_execve so we - * end up with proper pt_regs. - */ -int kernel_execve(const char *filename, - const char *const argv[], - const char *const envp[]) -{ - long __res; - register long __g1 __asm__ ("g1") = __NR_execve; - register long __o0 __asm__ ("o0") = (long)(filename); - register long __o1 __asm__ ("o1") = (long)(argv); - register long __o2 __asm__ ("o2") = (long)(envp); - asm volatile ("t 0x10\n\t" - "bcc 1f\n\t" - "mov %%o0, %0\n\t" - "sub %%g0, %%o0, %0\n\t" - "1:\n\t" - : "=r" (__res), "=&r" (__o0) - : "1" (__o0), "r" (__o1), "r" (__o2), "r" (__g1) - : "cc"); - return __res; -} diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c index 275f74fd6f6..c85403d0496 100644 --- a/arch/sparc/kernel/sys_sparc_64.c +++ b/arch/sparc/kernel/sys_sparc_64.c @@ -24,12 +24,14 @@ #include <linux/personality.h> #include <linux/random.h> #include <linux/export.h> +#include <linux/context_tracking.h> #include <asm/uaccess.h> #include <asm/utrap.h> #include <asm/unistd.h> #include "entry.h" +#include "kernel.h" #include "systbls.h" /* #define DEBUG_UNIMP_SYSCALL */ @@ -39,9 +41,6 @@ asmlinkage unsigned long sys_getpagesize(void) return PAGE_SIZE; } -#define VA_EXCLUDE_START (0x0000080000000000UL - (1UL << 32UL)) -#define VA_EXCLUDE_END (0xfffff80000000000UL + (1UL << 32UL)) - /* Does addr --> addr+len fall within 4GB of the VA-space hole or * overflow past the end of the 64-bit address space? */ @@ -66,23 +65,6 @@ static inline int invalid_64bit_range(unsigned long addr, unsigned long len) return 0; } -/* Does start,end straddle the VA-space hole? */ -static inline int straddles_64bit_va_hole(unsigned long start, unsigned long end) -{ - unsigned long va_exclude_start, va_exclude_end; - - va_exclude_start = VA_EXCLUDE_START; - va_exclude_end = VA_EXCLUDE_END; - - if (likely(start < va_exclude_start && end < va_exclude_start)) - return 0; - - if (likely(start >= va_exclude_end && end >= va_exclude_end)) - return 0; - - return 1; -} - /* These functions differ from the default implementations in * mm/mmap.c in two ways: * @@ -92,7 +74,7 @@ static inline int straddles_64bit_va_hole(unsigned long start, unsigned long end * the spitfire/niagara VA-hole. */ -static inline unsigned long COLOUR_ALIGN(unsigned long addr, +static inline unsigned long COLOR_ALIGN(unsigned long addr, unsigned long pgoff) { unsigned long base = (addr+SHMLBA-1)&~(SHMLBA-1); @@ -101,24 +83,13 @@ static inline unsigned long COLOUR_ALIGN(unsigned long addr, return base + off; } -static inline unsigned long COLOUR_ALIGN_DOWN(unsigned long addr, - unsigned long pgoff) -{ - unsigned long base = addr & ~(SHMLBA-1); - unsigned long off = (pgoff<<PAGE_SHIFT) & (SHMLBA-1); - - if (base + off <= addr) - return base + off; - return base - off; -} - unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { struct mm_struct *mm = current->mm; struct vm_area_struct * vma; unsigned long task_size = TASK_SIZE; - unsigned long start_addr; int do_color_align; + struct vm_unmapped_area_info info; if (flags & MAP_FIXED) { /* We do not accept a shared mapping if it would violate @@ -141,7 +112,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi if (addr) { if (do_color_align) - addr = COLOUR_ALIGN(addr, pgoff); + addr = COLOR_ALIGN(addr, pgoff); else addr = PAGE_ALIGN(addr); @@ -151,50 +122,22 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi return addr; } - if (len > mm->cached_hole_size) { - start_addr = addr = mm->free_area_cache; - } else { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; + info.flags = 0; + info.length = len; + info.low_limit = TASK_UNMAPPED_BASE; + info.high_limit = min(task_size, VA_EXCLUDE_START); + info.align_mask = do_color_align ? (PAGE_MASK & (SHMLBA - 1)) : 0; + info.align_offset = pgoff << PAGE_SHIFT; + addr = vm_unmapped_area(&info); + + if ((addr & ~PAGE_MASK) && task_size > VA_EXCLUDE_END) { + VM_BUG_ON(addr != -ENOMEM); + info.low_limit = VA_EXCLUDE_END; + info.high_limit = task_size; + addr = vm_unmapped_area(&info); } - task_size -= len; - -full_search: - if (do_color_align) - addr = COLOUR_ALIGN(addr, pgoff); - else - addr = PAGE_ALIGN(addr); - - for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { - /* At this point: (!vma || addr < vma->vm_end). */ - if (addr < VA_EXCLUDE_START && - (addr + len) >= VA_EXCLUDE_START) { - addr = VA_EXCLUDE_END; - vma = find_vma(mm, VA_EXCLUDE_END); - } - if (unlikely(task_size < addr)) { - if (start_addr != TASK_UNMAPPED_BASE) { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; - goto full_search; - } - return -ENOMEM; - } - if (likely(!vma || addr + len <= vma->vm_start)) { - /* - * Remember the place where we stopped the search: - */ - mm->free_area_cache = addr + len; - return addr; - } - if (addr + mm->cached_hole_size < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; - - addr = vma->vm_end; - if (do_color_align) - addr = COLOUR_ALIGN(addr, pgoff); - } + return addr; } unsigned long @@ -207,6 +150,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, unsigned long task_size = STACK_TOP32; unsigned long addr = addr0; int do_color_align; + struct vm_unmapped_area_info info; /* This should only ever run for 32-bit processes. */ BUG_ON(!test_thread_flag(TIF_32BIT)); @@ -231,7 +175,7 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, /* requesting a specific address */ if (addr) { if (do_color_align) - addr = COLOUR_ALIGN(addr, pgoff); + addr = COLOR_ALIGN(addr, pgoff); else addr = PAGE_ALIGN(addr); @@ -241,73 +185,27 @@ arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, return addr; } - /* check if free_area_cache is useful for us */ - if (len <= mm->cached_hole_size) { - mm->cached_hole_size = 0; - mm->free_area_cache = mm->mmap_base; - } - - /* either no address requested or can't fit in requested address hole */ - addr = mm->free_area_cache; - if (do_color_align) { - unsigned long base = COLOUR_ALIGN_DOWN(addr-len, pgoff); - - addr = base + len; - } - - /* make sure it can fit in the remaining address space */ - if (likely(addr > len)) { - vma = find_vma(mm, addr-len); - if (!vma || addr <= vma->vm_start) { - /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr-len); - } - } - - if (unlikely(mm->mmap_base < len)) - goto bottomup; - - addr = mm->mmap_base-len; - if (do_color_align) - addr = COLOUR_ALIGN_DOWN(addr, pgoff); - - do { - /* - * Lookup failure means no vma is above this address, - * else if new region fits below vma->vm_start, - * return with success: - */ - vma = find_vma(mm, addr); - if (likely(!vma || addr+len <= vma->vm_start)) { - /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr); - } - - /* remember the largest hole we saw so far */ - if (addr + mm->cached_hole_size < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; - - /* try just below the current vma->vm_start */ - addr = vma->vm_start-len; - if (do_color_align) - addr = COLOUR_ALIGN_DOWN(addr, pgoff); - } while (likely(len < vma->vm_start)); + info.flags = VM_UNMAPPED_AREA_TOPDOWN; + info.length = len; + info.low_limit = PAGE_SIZE; + info.high_limit = mm->mmap_base; + info.align_mask = do_color_align ? (PAGE_MASK & (SHMLBA - 1)) : 0; + info.align_offset = pgoff << PAGE_SHIFT; + addr = vm_unmapped_area(&info); -bottomup: /* * A failed mmap() very likely causes application failure, * so fall back to the bottom-up function here. This scenario * can happen with large stack limits and large mmap() * allocations. */ - mm->cached_hole_size = ~0UL; - mm->free_area_cache = TASK_UNMAPPED_BASE; - addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); - /* - * Restore the topdown base: - */ - mm->free_area_cache = mm->mmap_base; - mm->cached_hole_size = ~0UL; + if (addr & ~PAGE_MASK) { + VM_BUG_ON(addr != -ENOMEM); + info.flags = 0; + info.low_limit = TASK_UNMAPPED_BASE; + info.high_limit = STACK_TOP32; + addr = vm_unmapped_area(&info); + } return addr; } @@ -391,7 +289,6 @@ void arch_pick_mmap_layout(struct mm_struct *mm) sysctl_legacy_va_layout) { mm->mmap_base = TASK_UNMAPPED_BASE + random_factor; mm->get_unmapped_area = arch_get_unmapped_area; - mm->unmap_area = arch_unmap_area; } else { /* We know it's 32-bit */ unsigned long task_size = STACK_TOP32; @@ -403,7 +300,6 @@ void arch_pick_mmap_layout(struct mm_struct *mm) mm->mmap_base = PAGE_ALIGN(task_size - gap - random_factor); mm->get_unmapped_area = arch_get_unmapped_area_topdown; - mm->unmap_area = arch_unmap_area_topdown; } } @@ -454,7 +350,7 @@ SYSCALL_DEFINE6(sparc_ipc, unsigned int, call, int, first, unsigned long, second case SEMCTL: { err = sys_semctl(first, second, (int)third | IPC_64, - (union semun) ptr); + (unsigned long) ptr); goto out; } default: @@ -487,7 +383,7 @@ SYSCALL_DEFINE6(sparc_ipc, unsigned int, call, int, first, unsigned long, second switch (call) { case SHMAT: { ulong raddr; - err = do_shmat(first, ptr, (int)second, &raddr); + err = do_shmat(first, ptr, (int)second, &raddr, SHMLBA); if (!err) { if (put_user(raddr, (ulong __user *) third)) @@ -519,12 +415,12 @@ SYSCALL_DEFINE1(sparc64_personality, unsigned long, personality) { int ret; - if (current->personality == PER_LINUX32 && - personality == PER_LINUX) - personality = PER_LINUX32; + if (personality(current->personality) == PER_LINUX32 && + personality(personality) == PER_LINUX) + personality |= PER_LINUX32; ret = sys_personality(personality); - if (ret == PER_LINUX32) - ret = PER_LINUX; + if (personality(ret) == PER_LINUX32) + ret &= ~PER_LINUX32; return ret; } @@ -571,10 +467,6 @@ SYSCALL_DEFINE2(64_munmap, unsigned long, addr, size_t, len) return vm_munmap(addr, len); } - -extern unsigned long do_mremap(unsigned long addr, - unsigned long old_len, unsigned long new_len, - unsigned long flags, unsigned long new_addr); SYSCALL_DEFINE5(64_mremap, unsigned long, addr, unsigned long, old_len, unsigned long, new_len, unsigned long, flags, @@ -606,6 +498,7 @@ asmlinkage unsigned long c_sys_nis_syscall(struct pt_regs *regs) asmlinkage void sparc_breakpoint(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); siginfo_t info; if (test_thread_flag(TIF_32BIT)) { @@ -624,6 +517,7 @@ asmlinkage void sparc_breakpoint(struct pt_regs *regs) #ifdef DEBUG_SPARC_BREAKPOINT printk ("TRAP: Returning to space: PC=%lx nPC=%lx\n", regs->tpc, regs->tnpc); #endif + exception_exit(prev_state); } extern void check_pending(int signum); @@ -747,24 +641,7 @@ SYSCALL_DEFINE5(rt_sigaction, int, sig, const struct sigaction __user *, act, return ret; } -/* - * Do a system call from kernel instead of calling sys_execve so we - * end up with proper pt_regs. - */ -int kernel_execve(const char *filename, - const char *const argv[], - const char *const envp[]) +asmlinkage long sys_kern_features(void) { - long __res; - register long __g1 __asm__ ("g1") = __NR_execve; - register long __o0 __asm__ ("o0") = (long)(filename); - register long __o1 __asm__ ("o1") = (long)(argv); - register long __o2 __asm__ ("o2") = (long)(envp); - asm volatile ("t 0x6d\n\t" - "sub %%g0, %%o0, %0\n\t" - "movcc %%xcc, %%o0, %0\n\t" - : "=r" (__res), "=&r" (__o0) - : "1" (__o0), "r" (__o1), "r" (__o2), "r" (__g1) - : "cc"); - return __res; + return KERN_FEATURE_MIXED_MODE_STACK; } diff --git a/arch/sparc/kernel/syscalls.S b/arch/sparc/kernel/syscalls.S index 1d7e274f3f2..33a17e7b3cc 100644 --- a/arch/sparc/kernel/syscalls.S +++ b/arch/sparc/kernel/syscalls.S @@ -1,23 +1,19 @@ /* SunOS's execv() call only specifies the argv argument, the * environment settings are the same as the calling processes. */ -sys_execve: - sethi %hi(sparc_execve), %g1 - ba,pt %xcc, execve_merge - or %g1, %lo(sparc_execve), %g1 +sys64_execve: + set sys_execve, %g1 + jmpl %g1, %g0 + flushw #ifdef CONFIG_COMPAT sunos_execv: - stx %g0, [%sp + PTREGS_OFF + PT_V9_I2] + mov %g0, %o2 sys32_execve: - sethi %hi(sparc32_execve), %g1 - or %g1, %lo(sparc32_execve), %g1 -#endif - -execve_merge: - flushw + set compat_sys_execve, %g1 jmpl %g1, %g0 - add %sp, PTREGS_OFF, %o0 + flushw +#endif .align 32 sys_sparc_pipe: @@ -29,16 +25,10 @@ sys_nis_syscall: sys_memory_ordering: ba,pt %xcc, sparc_memory_ordering add %sp, PTREGS_OFF, %o1 -sys_sigaltstack: - ba,pt %xcc, do_sigaltstack - add %i6, STACK_BIAS, %o2 #ifdef CONFIG_COMPAT sys32_sigstack: ba,pt %xcc, do_sys32_sigstack mov %i6, %o2 -sys32_sigaltstack: - ba,pt %xcc, do_sys32_sigaltstack - mov %i6, %o2 #endif .align 32 #ifdef CONFIG_COMPAT @@ -62,7 +52,7 @@ sys32_rt_sigreturn: #endif .align 32 1: ldx [%g6 + TI_FLAGS], %l5 - andcc %l5, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0 + andcc %l5, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0 be,pt %icc, rtrap nop call syscall_trace_leave @@ -108,20 +98,35 @@ sys_clone: ba,pt %xcc, sparc_do_fork add %sp, PTREGS_OFF, %o2 - .globl ret_from_syscall -ret_from_syscall: + .globl ret_from_fork +ret_from_fork: /* Clear current_thread_info()->new_child. */ stb %g0, [%g6 + TI_NEW_CHILD] - ldx [%g6 + TI_FLAGS], %l0 call schedule_tail mov %g7, %o0 + ldx [%sp + PTREGS_OFF + PT_V9_I0], %o0 + brnz,pt %o0, ret_sys_call + ldx [%g6 + TI_FLAGS], %l0 + ldx [%sp + PTREGS_OFF + PT_V9_G1], %l1 + call %l1 + ldx [%sp + PTREGS_OFF + PT_V9_G2], %o0 ba,pt %xcc, ret_sys_call - ldx [%sp + PTREGS_OFF + PT_V9_I0], %o0 + mov 0, %o0 + + .globl sparc_exit_group + .type sparc_exit_group,#function +sparc_exit_group: + sethi %hi(sys_exit_group), %g7 + ba,pt %xcc, 1f + or %g7, %lo(sys_exit_group), %g7 + .size sparc_exit_group,.-sparc_exit_group .globl sparc_exit .type sparc_exit,#function sparc_exit: - rdpr %pstate, %g2 + sethi %hi(sys_exit), %g7 + or %g7, %lo(sys_exit), %g7 +1: rdpr %pstate, %g2 wrpr %g2, PSTATE_IE, %pstate rdpr %otherwin, %g1 rdpr %cansave, %g3 @@ -129,7 +134,7 @@ sparc_exit: wrpr %g3, 0x0, %cansave wrpr %g0, 0x0, %otherwin wrpr %g2, 0x0, %pstate - ba,pt %xcc, sys_exit + jmpl %g7, %g0 stb %g0, [%g6 + TI_WSAVED] .size sparc_exit,.-sparc_exit @@ -147,7 +152,7 @@ linux_syscall_trace32: srl %i4, 0, %o4 srl %i1, 0, %o1 srl %i2, 0, %o2 - ba,pt %xcc, 2f + ba,pt %xcc, 5f srl %i3, 0, %o3 linux_syscall_trace: @@ -177,14 +182,15 @@ linux_sparc_syscall32: srl %i1, 0, %o1 ! IEU0 Group ldx [%g6 + TI_FLAGS], %l0 ! Load - srl %i5, 0, %o5 ! IEU1 + srl %i3, 0, %o3 ! IEU0 srl %i2, 0, %o2 ! IEU0 Group - andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0 + andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0 bne,pn %icc, linux_syscall_trace32 ! CTI mov %i0, %l5 ! IEU1 - call %l7 ! CTI Group brk forced - srl %i3, 0, %o3 ! IEU0 - ba,a,pt %xcc, 3f +5: call %l7 ! CTI Group brk forced + srl %i5, 0, %o5 ! IEU1 + ba,pt %xcc, 3f + sra %o0, 0, %o0 /* Linux native system calls enter here... */ .align 32 @@ -202,7 +208,7 @@ linux_sparc_syscall: mov %i3, %o3 ! IEU1 mov %i4, %o4 ! IEU0 Group - andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %g0 + andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0 bne,pn %icc, linux_syscall_trace ! CTI Group mov %i0, %l5 ! IEU0 2: call %l7 ! CTI Group brk forced @@ -212,24 +218,18 @@ linux_sparc_syscall: 3: stx %o0, [%sp + PTREGS_OFF + PT_V9_I0] ret_sys_call: ldx [%sp + PTREGS_OFF + PT_V9_TSTATE], %g3 - ldx [%sp + PTREGS_OFF + PT_V9_TNPC], %l1 ! pc = npc - sra %o0, 0, %o0 mov %ulo(TSTATE_XCARRY | TSTATE_ICARRY), %g2 sllx %g2, 32, %g2 - /* Check if force_successful_syscall_return() - * was invoked. - */ - ldub [%g6 + TI_SYS_NOERROR], %l2 - brnz,a,pn %l2, 80f - stb %g0, [%g6 + TI_SYS_NOERROR] - cmp %o0, -ERESTART_RESTARTBLOCK bgeu,pn %xcc, 1f - andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %l6 -80: + andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT|_TIF_NOHZ), %g0 + ldx [%sp + PTREGS_OFF + PT_V9_TNPC], %l1 ! pc = npc + +2: /* System call success, clear Carry condition code. */ andn %g3, %g2, %g3 +3: stx %g3, [%sp + PTREGS_OFF + PT_V9_TSTATE] bne,pn %icc, linux_syscall_trace2 add %l1, 0x4, %l2 ! npc = npc+4 @@ -238,20 +238,20 @@ ret_sys_call: stx %l2, [%sp + PTREGS_OFF + PT_V9_TNPC] 1: + /* Check if force_successful_syscall_return() + * was invoked. + */ + ldub [%g6 + TI_SYS_NOERROR], %l2 + brnz,pn %l2, 2b + ldx [%sp + PTREGS_OFF + PT_V9_TNPC], %l1 ! pc = npc /* System call failure, set Carry condition code. * Also, get abs(errno) to return to the process. */ - andcc %l0, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT|_TIF_SYSCALL_TRACEPOINT), %l6 sub %g0, %o0, %o0 - or %g3, %g2, %g3 stx %o0, [%sp + PTREGS_OFF + PT_V9_I0] - stx %g3, [%sp + PTREGS_OFF + PT_V9_TSTATE] - bne,pn %icc, linux_syscall_trace2 - add %l1, 0x4, %l2 ! npc = npc+4 - stx %l1, [%sp + PTREGS_OFF + PT_V9_TPC] + ba,pt %xcc, 3b + or %g3, %g2, %g3 - b,pt %xcc, rtrap - stx %l2, [%sp + PTREGS_OFF + PT_V9_TNPC] linux_syscall_trace2: call syscall_trace_leave add %sp, PTREGS_OFF, %o0 diff --git a/arch/sparc/kernel/sysfs.c b/arch/sparc/kernel/sysfs.c index 654e8aad3bb..7f41d40b7e6 100644 --- a/arch/sparc/kernel/sysfs.c +++ b/arch/sparc/kernel/sysfs.c @@ -151,7 +151,7 @@ static ssize_t store_mmustat_enable(struct device *s, size_t count) { unsigned long val, err; - int ret = sscanf(buf, "%ld", &val); + int ret = sscanf(buf, "%lu", &val); if (ret != 1) return -EINVAL; @@ -246,7 +246,7 @@ static void unregister_cpu_online(unsigned int cpu) } #endif -static int __cpuinit sysfs_cpu_notify(struct notifier_block *self, +static int sysfs_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned int)(long)hcpu; @@ -266,7 +266,7 @@ static int __cpuinit sysfs_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __cpuinitdata sysfs_cpu_nb = { +static struct notifier_block sysfs_cpu_nb = { .notifier_call = sysfs_cpu_notify, }; @@ -300,7 +300,7 @@ static int __init topology_init(void) check_mmu_stats(); - register_cpu_notifier(&sysfs_cpu_nb); + cpu_notifier_register_begin(); for_each_possible_cpu(cpu) { struct cpu *c = &per_cpu(cpu_devices, cpu); @@ -310,6 +310,10 @@ static int __init topology_init(void) register_cpu_online(cpu); } + __register_cpu_notifier(&sysfs_cpu_nb); + + cpu_notifier_register_done(); + return 0; } diff --git a/arch/sparc/kernel/systbls.h b/arch/sparc/kernel/systbls.h index 118759cd734..2dab8236d49 100644 --- a/arch/sparc/kernel/systbls.h +++ b/arch/sparc/kernel/systbls.h @@ -1,43 +1,103 @@ #ifndef _SYSTBLS_H #define _SYSTBLS_H +#include <linux/signal.h> #include <linux/kernel.h> +#include <linux/compat.h> #include <linux/types.h> + #include <asm/utrap.h> -#include <asm/signal.h> -extern asmlinkage unsigned long sys_getpagesize(void); -extern asmlinkage long sparc_pipe(struct pt_regs *regs); -extern asmlinkage long sys_sparc_ipc(unsigned int call, int first, - unsigned long second, - unsigned long third, - void __user *ptr, long fifth); -extern asmlinkage long sparc64_personality(unsigned long personality); -extern asmlinkage long sys64_munmap(unsigned long addr, size_t len); -extern asmlinkage unsigned long sys64_mremap(unsigned long addr, - unsigned long old_len, - unsigned long new_len, - unsigned long flags, - unsigned long new_addr); -extern asmlinkage unsigned long c_sys_nis_syscall(struct pt_regs *regs); -extern asmlinkage long sys_getdomainname(char __user *name, int len); -extern asmlinkage long sys_utrap_install(utrap_entry_t type, - utrap_handler_t new_p, - utrap_handler_t new_d, - utrap_handler_t __user *old_p, - utrap_handler_t __user *old_d); -extern asmlinkage long sparc_memory_ordering(unsigned long model, - struct pt_regs *regs); -extern asmlinkage long sys_rt_sigaction(int sig, - const struct sigaction __user *act, - struct sigaction __user *oact, - void __user *restorer, - size_t sigsetsize); +asmlinkage unsigned long sys_getpagesize(void); +asmlinkage long sparc_pipe(struct pt_regs *regs); +asmlinkage unsigned long c_sys_nis_syscall(struct pt_regs *regs); +asmlinkage long sys_getdomainname(char __user *name, int len); +void do_rt_sigreturn(struct pt_regs *regs); +asmlinkage long sys_mmap(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long off); +asmlinkage void sparc_breakpoint(struct pt_regs *regs); + +#ifdef CONFIG_SPARC32 +asmlinkage long sys_mmap2(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long pgoff); +long sparc_remap_file_pages(unsigned long start, unsigned long size, + unsigned long prot, unsigned long pgoff, + unsigned long flags); -extern asmlinkage void sparc64_set_context(struct pt_regs *regs); -extern asmlinkage void sparc64_get_context(struct pt_regs *regs); -extern asmlinkage long sys_sigpause(unsigned int set); -extern asmlinkage long sys_sigsuspend(old_sigset_t set); -extern void do_rt_sigreturn(struct pt_regs *regs); +#endif /* CONFIG_SPARC32 */ +#ifdef CONFIG_SPARC64 +asmlinkage long sys_sparc_ipc(unsigned int call, int first, + unsigned long second, + unsigned long third, + void __user *ptr, long fifth); +asmlinkage long sparc64_personality(unsigned long personality); +asmlinkage long sys64_munmap(unsigned long addr, size_t len); +asmlinkage unsigned long sys64_mremap(unsigned long addr, + unsigned long old_len, + unsigned long new_len, + unsigned long flags, + unsigned long new_addr); +asmlinkage long sys_utrap_install(utrap_entry_t type, + utrap_handler_t new_p, + utrap_handler_t new_d, + utrap_handler_t __user *old_p, + utrap_handler_t __user *old_d); +asmlinkage long sparc_memory_ordering(unsigned long model, + struct pt_regs *regs); +asmlinkage void sparc64_set_context(struct pt_regs *regs); +asmlinkage void sparc64_get_context(struct pt_regs *regs); +asmlinkage long sys32_truncate64(const char __user * path, + unsigned long high, + unsigned long low); +asmlinkage long sys32_ftruncate64(unsigned int fd, + unsigned long high, + unsigned long low); +struct compat_stat64; +asmlinkage long compat_sys_stat64(const char __user * filename, + struct compat_stat64 __user *statbuf); +asmlinkage long compat_sys_lstat64(const char __user * filename, + struct compat_stat64 __user *statbuf); +asmlinkage long compat_sys_fstat64(unsigned int fd, + struct compat_stat64 __user * statbuf); +asmlinkage long compat_sys_fstatat64(unsigned int dfd, + const char __user *filename, + struct compat_stat64 __user * statbuf, int flag); +asmlinkage compat_ssize_t sys32_pread64(unsigned int fd, + char __user *ubuf, + compat_size_t count, + unsigned long poshi, + unsigned long poslo); +asmlinkage compat_ssize_t sys32_pwrite64(unsigned int fd, + char __user *ubuf, + compat_size_t count, + unsigned long poshi, + unsigned long poslo); +asmlinkage long compat_sys_readahead(int fd, + unsigned long offhi, + unsigned long offlo, + compat_size_t count); +long compat_sys_fadvise64(int fd, + unsigned long offhi, + unsigned long offlo, + compat_size_t len, int advice); +long compat_sys_fadvise64_64(int fd, + unsigned long offhi, unsigned long offlo, + unsigned long lenhi, unsigned long lenlo, + int advice); +long sys32_sync_file_range(unsigned int fd, + unsigned long off_high, unsigned long off_low, + unsigned long nb_high, unsigned long nb_low, + unsigned int flags); +asmlinkage long compat_sys_fallocate(int fd, int mode, u32 offhi, u32 offlo, + u32 lenhi, u32 lenlo); +asmlinkage long compat_sys_fstat64(unsigned int fd, + struct compat_stat64 __user * statbuf); +asmlinkage long compat_sys_fstatat64(unsigned int dfd, + const char __user *filename, + struct compat_stat64 __user * statbuf, + int flag); +#endif /* CONFIG_SPARC64 */ #endif /* _SYSTBLS_H */ diff --git a/arch/sparc/kernel/systbls_32.S b/arch/sparc/kernel/systbls_32.S index 63402f9e9f5..85fe9b1087c 100644 --- a/arch/sparc/kernel/systbls_32.S +++ b/arch/sparc/kernel/systbls_32.S @@ -55,7 +55,7 @@ sys_call_table: /*180*/ .long sys_flistxattr, sys_removexattr, sys_lremovexattr, sys_sigpending, sys_ni_syscall /*185*/ .long sys_setpgid, sys_fremovexattr, sys_tkill, sys_exit_group, sys_newuname /*190*/ .long sys_init_module, sys_personality, sparc_remap_file_pages, sys_epoll_create, sys_epoll_ctl -/*195*/ .long sys_epoll_wait, sys_ioprio_set, sys_getppid, sparc_sigaction, sys_sgetmask +/*195*/ .long sys_epoll_wait, sys_ioprio_set, sys_getppid, sys_sparc_sigaction, sys_sgetmask /*200*/ .long sys_ssetmask, sys_sigsuspend, sys_newlstat, sys_uselib, sys_old_readdir /*205*/ .long sys_readahead, sys_socketcall, sys_syslog, sys_lookup_dcookie, sys_fadvise64 /*210*/ .long sys_fadvise64_64, sys_tgkill, sys_waitpid, sys_swapoff, sys_sysinfo @@ -85,3 +85,5 @@ sys_call_table: /*325*/ .long sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init /*330*/ .long sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime /*335*/ .long sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev +/*340*/ .long sys_ni_syscall, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr +/*345*/ .long sys_renameat2 diff --git a/arch/sparc/kernel/systbls_64.S b/arch/sparc/kernel/systbls_64.S index 3a58e0d66f5..33ecba2826e 100644 --- a/arch/sparc/kernel/systbls_64.S +++ b/arch/sparc/kernel/systbls_64.S @@ -18,63 +18,63 @@ .globl sys_call_table32 sys_call_table32: -/*0*/ .word sys_restart_syscall, sys32_exit, sys_fork, sys_read, sys_write -/*5*/ .word sys32_open, sys_close, sys32_wait4, sys32_creat, sys_link -/*10*/ .word sys_unlink, sunos_execv, sys_chdir, sys_chown16, sys32_mknod -/*15*/ .word sys_chmod, sys_lchown16, sys_brk, sys_nis_syscall, sys32_lseek +/*0*/ .word sys_restart_syscall, sparc_exit, sys_fork, sys_read, sys_write +/*5*/ .word compat_sys_open, sys_close, compat_sys_wait4, sys_creat, sys_link +/*10*/ .word sys_unlink, sunos_execv, sys_chdir, sys_chown16, sys_mknod +/*15*/ .word sys_chmod, sys_lchown16, sys_brk, sys_nis_syscall, compat_sys_lseek /*20*/ .word sys_getpid, sys_capget, sys_capset, sys_setuid16, sys_getuid16 -/*25*/ .word sys32_vmsplice, compat_sys_ptrace, sys_alarm, sys32_sigaltstack, sys_pause -/*30*/ .word compat_sys_utime, sys_lchown, sys_fchown, sys32_access, sys32_nice - .word sys_chown, sys_sync, sys32_kill, compat_sys_newstat, sys32_sendfile +/*25*/ .word compat_sys_vmsplice, compat_sys_ptrace, sys_alarm, compat_sys_sigaltstack, sys_pause +/*30*/ .word compat_sys_utime, sys_lchown, sys_fchown, sys_access, sys_nice + .word sys_chown, sys_sync, sys_kill, compat_sys_newstat, compat_sys_sendfile /*40*/ .word compat_sys_newlstat, sys_dup, sys_sparc_pipe, compat_sys_times, sys_getuid - .word sys32_umount, sys_setgid16, sys_getgid16, sys32_signal, sys_geteuid16 + .word sys_umount, sys_setgid16, sys_getgid16, sys_signal, sys_geteuid16 /*50*/ .word sys_getegid16, sys_acct, sys_nis_syscall, sys_getgid, compat_sys_ioctl - .word sys32_reboot, sys32_mmap2, sys_symlink, sys32_readlink, sys32_execve -/*60*/ .word sys32_umask, sys_chroot, compat_sys_newfstat, compat_sys_fstat64, sys_getpagesize - .word sys32_msync, sys_vfork, sys32_pread64, sys32_pwrite64, sys_geteuid + .word sys_reboot, sys32_mmap2, sys_symlink, sys_readlink, sys32_execve +/*60*/ .word sys_umask, sys_chroot, compat_sys_newfstat, compat_sys_fstat64, sys_getpagesize + .word sys_msync, sys_vfork, sys32_pread64, sys32_pwrite64, sys_geteuid /*70*/ .word sys_getegid, sys_mmap, sys_setreuid, sys_munmap, sys_mprotect .word sys_madvise, sys_vhangup, sys32_truncate64, sys_mincore, sys_getgroups16 -/*80*/ .word sys_setgroups16, sys_getpgrp, sys32_setgroups, sys32_setitimer, sys32_ftruncate64 - .word sys32_swapon, sys32_getitimer, sys_setuid, sys32_sethostname, sys_setgid +/*80*/ .word sys_setgroups16, sys_getpgrp, sys_setgroups, compat_sys_setitimer, sys32_ftruncate64 + .word sys_swapon, compat_sys_getitimer, sys_setuid, sys_sethostname, sys_setgid /*90*/ .word sys_dup2, sys_setfsuid, compat_sys_fcntl, sys32_select, sys_setfsgid - .word sys_fsync, sys32_setpriority, sys_nis_syscall, sys_nis_syscall, sys_nis_syscall -/*100*/ .word sys32_getpriority, sys32_rt_sigreturn, sys32_rt_sigaction, sys32_rt_sigprocmask, sys32_rt_sigpending - .word compat_sys_rt_sigtimedwait, sys32_rt_sigqueueinfo, compat_sys_rt_sigsuspend, sys_setresuid, sys_getresuid + .word sys_fsync, sys_setpriority, sys_nis_syscall, sys_nis_syscall, sys_nis_syscall +/*100*/ .word sys_getpriority, sys32_rt_sigreturn, compat_sys_rt_sigaction, compat_sys_rt_sigprocmask, compat_sys_rt_sigpending + .word compat_sys_rt_sigtimedwait, compat_sys_rt_sigqueueinfo, compat_sys_rt_sigsuspend, sys_setresuid, sys_getresuid /*110*/ .word sys_setresgid, sys_getresgid, sys_setregid, sys_nis_syscall, sys_nis_syscall - .word sys32_getgroups, compat_sys_gettimeofday, sys32_getrusage, sys_nis_syscall, sys_getcwd + .word sys_getgroups, compat_sys_gettimeofday, compat_sys_getrusage, sys_nis_syscall, sys_getcwd /*120*/ .word compat_sys_readv, compat_sys_writev, compat_sys_settimeofday, sys_fchown16, sys_fchmod - .word sys_nis_syscall, sys_setreuid16, sys_setregid16, sys_rename, sys32_truncate -/*130*/ .word sys32_ftruncate, sys_flock, compat_sys_lstat64, sys_nis_syscall, sys_nis_syscall - .word sys_nis_syscall, sys32_mkdir, sys_rmdir, compat_sys_utimes, compat_sys_stat64 -/*140*/ .word sys32_sendfile64, sys_nis_syscall, sys32_futex, sys_gettid, compat_sys_getrlimit - .word compat_sys_setrlimit, sys_pivot_root, sys32_prctl, sys_pciconfig_read, sys_pciconfig_write + .word sys_nis_syscall, sys_setreuid16, sys_setregid16, sys_rename, compat_sys_truncate +/*130*/ .word compat_sys_ftruncate, sys_flock, compat_sys_lstat64, sys_nis_syscall, sys_nis_syscall + .word sys_nis_syscall, sys_mkdir, sys_rmdir, compat_sys_utimes, compat_sys_stat64 +/*140*/ .word sys_sendfile64, sys_nis_syscall, sys32_futex, sys_gettid, compat_sys_getrlimit + .word compat_sys_setrlimit, sys_pivot_root, sys_prctl, sys_pciconfig_read, sys_pciconfig_write /*150*/ .word sys_nis_syscall, sys_inotify_init, sys_inotify_add_watch, sys_poll, sys_getdents64 .word compat_sys_fcntl64, sys_inotify_rm_watch, compat_sys_statfs, compat_sys_fstatfs, sys_oldumount -/*160*/ .word compat_sys_sched_setaffinity, compat_sys_sched_getaffinity, sys32_getdomainname, sys32_setdomainname, sys_nis_syscall - .word sys_quotactl, sys_set_tid_address, compat_sys_mount, compat_sys_ustat, sys32_setxattr -/*170*/ .word sys32_lsetxattr, sys32_fsetxattr, sys_getxattr, sys_lgetxattr, compat_sys_getdents - .word sys_setsid, sys_fchdir, sys32_fgetxattr, sys_listxattr, sys_llistxattr -/*180*/ .word sys32_flistxattr, sys_removexattr, sys_lremovexattr, compat_sys_sigpending, sys_ni_syscall - .word sys32_setpgid, sys32_fremovexattr, sys32_tkill, sys32_exit_group, sys_newuname -/*190*/ .word sys32_init_module, sys_sparc64_personality, sys_remap_file_pages, sys32_epoll_create, sys32_epoll_ctl - .word sys32_epoll_wait, sys32_ioprio_set, sys_getppid, sys32_sigaction, sys_sgetmask -/*200*/ .word sys32_ssetmask, sys_sigsuspend, compat_sys_newlstat, sys_uselib, compat_sys_old_readdir - .word sys32_readahead, sys32_socketcall, sys32_syslog, sys32_lookup_dcookie, sys32_fadvise64 -/*210*/ .word sys32_fadvise64_64, sys32_tgkill, sys32_waitpid, sys_swapoff, compat_sys_sysinfo - .word compat_sys_ipc, sys32_sigreturn, sys_clone, sys32_ioprio_get, compat_sys_adjtimex -/*220*/ .word sys32_sigprocmask, sys_ni_syscall, sys32_delete_module, sys_ni_syscall, sys32_getpgid - .word sys32_bdflush, sys32_sysfs, sys_nis_syscall, sys_setfsuid16, sys_setfsgid16 -/*230*/ .word sys32_select, compat_sys_time, sys32_splice, compat_sys_stime, compat_sys_statfs64 - .word compat_sys_fstatfs64, sys_llseek, sys_mlock, sys_munlock, sys32_mlockall -/*240*/ .word sys_munlockall, sys32_sched_setparam, sys32_sched_getparam, sys32_sched_setscheduler, sys32_sched_getscheduler - .word sys_sched_yield, sys32_sched_get_priority_max, sys32_sched_get_priority_min, sys32_sched_rr_get_interval, compat_sys_nanosleep -/*250*/ .word sys_mremap, compat_sys_sysctl, sys32_getsid, sys_fdatasync, sys_nis_syscall +/*160*/ .word compat_sys_sched_setaffinity, compat_sys_sched_getaffinity, sys_getdomainname, sys_setdomainname, sys_nis_syscall + .word sys_quotactl, sys_set_tid_address, compat_sys_mount, compat_sys_ustat, sys_setxattr +/*170*/ .word sys_lsetxattr, sys_fsetxattr, sys_getxattr, sys_lgetxattr, compat_sys_getdents + .word sys_setsid, sys_fchdir, sys_fgetxattr, sys_listxattr, sys_llistxattr +/*180*/ .word sys_flistxattr, sys_removexattr, sys_lremovexattr, compat_sys_sigpending, sys_ni_syscall + .word sys_setpgid, sys_fremovexattr, sys_tkill, sparc_exit_group, sys_newuname +/*190*/ .word sys_init_module, sys_sparc64_personality, sys_remap_file_pages, sys_epoll_create, sys_epoll_ctl + .word sys_epoll_wait, sys_ioprio_set, sys_getppid, compat_sys_sparc_sigaction, sys_sgetmask +/*200*/ .word sys_ssetmask, sys_sigsuspend, compat_sys_newlstat, sys_uselib, compat_sys_old_readdir + .word sys32_readahead, sys32_socketcall, sys_syslog, compat_sys_lookup_dcookie, sys32_fadvise64 +/*210*/ .word sys32_fadvise64_64, sys_tgkill, sys_waitpid, sys_swapoff, compat_sys_sysinfo + .word compat_sys_ipc, sys32_sigreturn, sys_clone, sys_ioprio_get, compat_sys_adjtimex +/*220*/ .word compat_sys_sigprocmask, sys_ni_syscall, sys_delete_module, sys_ni_syscall, sys_getpgid + .word sys_bdflush, sys_sysfs, sys_nis_syscall, sys_setfsuid16, sys_setfsgid16 +/*230*/ .word sys32_select, compat_sys_time, sys_splice, compat_sys_stime, compat_sys_statfs64 + .word compat_sys_fstatfs64, sys_llseek, sys_mlock, sys_munlock, sys_mlockall +/*240*/ .word sys_munlockall, sys_sched_setparam, sys_sched_getparam, sys_sched_setscheduler, sys_sched_getscheduler + .word sys_sched_yield, sys_sched_get_priority_max, sys_sched_get_priority_min, compat_sys_sched_rr_get_interval, compat_sys_nanosleep +/*250*/ .word sys_mremap, compat_sys_sysctl, sys_getsid, sys_fdatasync, sys_nis_syscall .word sys32_sync_file_range, compat_sys_clock_settime, compat_sys_clock_gettime, compat_sys_clock_getres, sys32_clock_nanosleep /*260*/ .word compat_sys_sched_getaffinity, compat_sys_sched_setaffinity, sys32_timer_settime, compat_sys_timer_gettime, sys_timer_getoverrun .word sys_timer_delete, compat_sys_timer_create, sys_ni_syscall, compat_sys_io_setup, sys_io_destroy /*270*/ .word sys32_io_submit, sys_io_cancel, compat_sys_io_getevents, sys32_mq_open, sys_mq_unlink .word compat_sys_mq_timedsend, compat_sys_mq_timedreceive, compat_sys_mq_notify, compat_sys_mq_getsetattr, compat_sys_waitid -/*280*/ .word sys32_tee, sys_add_key, sys_request_key, compat_sys_keyctl, compat_sys_openat +/*280*/ .word sys_tee, sys_add_key, sys_request_key, compat_sys_keyctl, compat_sys_openat .word sys_mkdirat, sys_mknodat, sys_fchownat, compat_sys_futimesat, compat_sys_fstatat64 /*290*/ .word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat .word sys_fchmodat, sys_faccessat, compat_sys_pselect6, compat_sys_ppoll, sys_unshare @@ -84,8 +84,10 @@ sys_call_table32: .word compat_sys_timerfd_settime, compat_sys_timerfd_gettime, compat_sys_signalfd4, sys_eventfd2, sys_epoll_create1 /*320*/ .word sys_dup3, sys_pipe2, sys_inotify_init1, sys_accept4, compat_sys_preadv .word compat_sys_pwritev, compat_sys_rt_tgsigqueueinfo, sys_perf_event_open, compat_sys_recvmmsg, sys_fanotify_init -/*330*/ .word sys32_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, compat_sys_open_by_handle_at, compat_sys_clock_adjtime +/*330*/ .word compat_sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, compat_sys_open_by_handle_at, compat_sys_clock_adjtime .word sys_syncfs, compat_sys_sendmmsg, sys_setns, compat_sys_process_vm_readv, compat_sys_process_vm_writev +/*340*/ .word sys_kern_features, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr + .word sys32_renameat2 #endif /* CONFIG_COMPAT */ @@ -106,7 +108,7 @@ sys_call_table: /*40*/ .word sys_newlstat, sys_dup, sys_sparc_pipe, sys_times, sys_nis_syscall .word sys_umount, sys_setgid, sys_getgid, sys_signal, sys_geteuid /*50*/ .word sys_getegid, sys_acct, sys_memory_ordering, sys_nis_syscall, sys_ioctl - .word sys_reboot, sys_nis_syscall, sys_symlink, sys_readlink, sys_execve + .word sys_reboot, sys_nis_syscall, sys_symlink, sys_readlink, sys64_execve /*60*/ .word sys_umask, sys_chroot, sys_newfstat, sys_fstat64, sys_getpagesize .word sys_msync, sys_vfork, sys_pread64, sys_pwrite64, sys_nis_syscall /*70*/ .word sys_nis_syscall, sys_mmap, sys_nis_syscall, sys_64_munmap, sys_mprotect @@ -132,7 +134,7 @@ sys_call_table: /*170*/ .word sys_lsetxattr, sys_fsetxattr, sys_getxattr, sys_lgetxattr, sys_getdents .word sys_setsid, sys_fchdir, sys_fgetxattr, sys_listxattr, sys_llistxattr /*180*/ .word sys_flistxattr, sys_removexattr, sys_lremovexattr, sys_nis_syscall, sys_ni_syscall - .word sys_setpgid, sys_fremovexattr, sys_tkill, sys_exit_group, sys_newuname + .word sys_setpgid, sys_fremovexattr, sys_tkill, sparc_exit_group, sys_newuname /*190*/ .word sys_init_module, sys_sparc64_personality, sys_remap_file_pages, sys_epoll_create, sys_epoll_ctl .word sys_epoll_wait, sys_ioprio_set, sys_getppid, sys_nis_syscall, sys_sgetmask /*200*/ .word sys_ssetmask, sys_nis_syscall, sys_newlstat, sys_uselib, sys_nis_syscall @@ -163,3 +165,5 @@ sys_call_table: .word sys_pwritev, sys_rt_tgsigqueueinfo, sys_perf_event_open, sys_recvmmsg, sys_fanotify_init /*330*/ .word sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime .word sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev +/*340*/ .word sys_kern_features, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr + .word sys_renameat2 diff --git a/arch/sparc/kernel/tadpole.c b/arch/sparc/kernel/tadpole.c deleted file mode 100644 index 9aba8bd5a78..00000000000 --- a/arch/sparc/kernel/tadpole.c +++ /dev/null @@ -1,126 +0,0 @@ -/* tadpole.c: Probing for the tadpole clock stopping h/w at boot time. - * - * Copyright (C) 1996 David Redman (djhr@tadpole.co.uk) - */ - -#include <linux/string.h> -#include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/init.h> - -#include <asm/asi.h> -#include <asm/oplib.h> -#include <asm/io.h> - -#define MACIO_SCSI_CSR_ADDR 0x78400000 -#define MACIO_EN_DMA 0x00000200 -#define CLOCK_INIT_DONE 1 - -static int clk_state; -static volatile unsigned char *clk_ctrl; -void (*cpu_pwr_save)(void); - -static inline unsigned int ldphys(unsigned int addr) -{ - unsigned long data; - - __asm__ __volatile__("\n\tlda [%1] %2, %0\n\t" : - "=r" (data) : - "r" (addr), "i" (ASI_M_BYPASS)); - return data; -} - -static void clk_init(void) -{ - __asm__ __volatile__("mov 0x6c, %%g1\n\t" - "mov 0x4c, %%g2\n\t" - "mov 0xdf, %%g3\n\t" - "stb %%g1, [%0+3]\n\t" - "stb %%g2, [%0+3]\n\t" - "stb %%g3, [%0+3]\n\t" : : - "r" (clk_ctrl) : - "g1", "g2", "g3"); -} - -static void clk_slow(void) -{ - __asm__ __volatile__("mov 0xcc, %%g2\n\t" - "mov 0x4c, %%g3\n\t" - "mov 0xcf, %%g4\n\t" - "mov 0xdf, %%g5\n\t" - "stb %%g2, [%0+3]\n\t" - "stb %%g3, [%0+3]\n\t" - "stb %%g4, [%0+3]\n\t" - "stb %%g5, [%0+3]\n\t" : : - "r" (clk_ctrl) : - "g2", "g3", "g4", "g5"); -} - -/* - * Tadpole is guaranteed to be UP, using local_irq_save. - */ -static void tsu_clockstop(void) -{ - unsigned int mcsr; - unsigned long flags; - - if (!clk_ctrl) - return; - if (!(clk_state & CLOCK_INIT_DONE)) { - local_irq_save(flags); - clk_init(); - clk_state |= CLOCK_INIT_DONE; /* all done */ - local_irq_restore(flags); - return; - } - if (!(clk_ctrl[2] & 1)) - return; /* no speed up yet */ - - local_irq_save(flags); - - /* if SCSI DMA in progress, don't slow clock */ - mcsr = ldphys(MACIO_SCSI_CSR_ADDR); - if ((mcsr&MACIO_EN_DMA) != 0) { - local_irq_restore(flags); - return; - } - /* TODO... the minimum clock setting ought to increase the - * memory refresh interval.. - */ - clk_slow(); - local_irq_restore(flags); -} - -static void swift_clockstop(void) -{ - if (!clk_ctrl) - return; - clk_ctrl[0] = 0; -} - -void __init clock_stop_probe(void) -{ - phandle node, clk_nd; - char name[20]; - - prom_getstring(prom_root_node, "name", name, sizeof(name)); - if (strncmp(name, "Tadpole", 7)) - return; - node = prom_getchild(prom_root_node); - node = prom_searchsiblings(node, "obio"); - node = prom_getchild(node); - clk_nd = prom_searchsiblings(node, "clk-ctrl"); - if (!clk_nd) - return; - printk("Clock Stopping h/w detected... "); - clk_ctrl = (char *) prom_getint(clk_nd, "address"); - clk_state = 0; - if (name[10] == '\0') { - cpu_pwr_save = tsu_clockstop; - printk("enabled (S3)\n"); - } else if ((name[10] == 'X') || (name[10] == 'G')) { - cpu_pwr_save = swift_clockstop; - printk("enabled (%s)\n",name+7); - } else - printk("disabled %s\n",name+7); -} diff --git a/arch/sparc/kernel/time_32.c b/arch/sparc/kernel/time_32.c index 953641549e8..5923d1e4e7c 100644 --- a/arch/sparc/kernel/time_32.c +++ b/arch/sparc/kernel/time_32.c @@ -36,6 +36,7 @@ #include <linux/of_device.h> #include <linux/platform_device.h> +#include <asm/mc146818rtc.h> #include <asm/oplib.h> #include <asm/timex.h> #include <asm/timer.h> @@ -47,6 +48,7 @@ #include <asm/irq_regs.h> #include <asm/setup.h> +#include "kernel.h" #include "irq.h" static __cacheline_aligned_in_smp DEFINE_SEQLOCK(timer_cs_lock); @@ -83,7 +85,7 @@ unsigned long profile_pc(struct pt_regs *regs) EXPORT_SYMBOL(profile_pc); -__volatile__ unsigned int *master_l10_counter; +volatile u32 __iomem *master_l10_counter; int update_persistent_clock(struct timespec now) { @@ -143,9 +145,9 @@ static __init void setup_timer_ce(void) static unsigned int sbus_cycles_offset(void) { - unsigned int val, offset; + u32 val, offset; - val = *master_l10_counter; + val = sbus_readl(master_l10_counter); offset = (val >> TIMER_VALUE_SHIFT) & TIMER_VALUE_MASK; /* Limit hit? */ @@ -278,7 +280,7 @@ static struct platform_device m48t59_rtc = { }, }; -static int __devinit clock_probe(struct platform_device *op) +static int clock_probe(struct platform_device *op) { struct device_node *dp = op->dev.of_node; const char *model = of_get_property(dp, "model", NULL); diff --git a/arch/sparc/kernel/time_64.c b/arch/sparc/kernel/time_64.c index e861072b9c5..3fddf64c7fc 100644 --- a/arch/sparc/kernel/time_64.c +++ b/arch/sparc/kernel/time_64.c @@ -419,7 +419,7 @@ static struct platform_device rtc_cmos_device = { .num_resources = 1, }; -static int __devinit rtc_probe(struct platform_device *op) +static int rtc_probe(struct platform_device *op) { struct resource *r; @@ -477,7 +477,7 @@ static struct platform_device rtc_bq4802_device = { .num_resources = 1, }; -static int __devinit bq4802_probe(struct platform_device *op) +static int bq4802_probe(struct platform_device *op) { printk(KERN_INFO "%s: BQ4802 regs at 0x%llx\n", @@ -534,7 +534,7 @@ static struct platform_device m48t59_rtc = { }, }; -static int __devinit mostek_probe(struct platform_device *op) +static int mostek_probe(struct platform_device *op) { struct device_node *dp = op->dev.of_node; @@ -659,8 +659,7 @@ static int sparc64_cpufreq_notifier(struct notifier_block *nb, unsigned long val ft->clock_tick_ref = cpu_data(cpu).clock_tick; } if ((val == CPUFREQ_PRECHANGE && freq->old < freq->new) || - (val == CPUFREQ_POSTCHANGE && freq->old > freq->new) || - (val == CPUFREQ_RESUMECHANGE)) { + (val == CPUFREQ_POSTCHANGE && freq->old > freq->new)) { cpu_data(cpu).clock_tick = cpufreq_scale(ft->clock_tick_ref, ft->ref_freq, @@ -733,7 +732,7 @@ void __irq_entry timer_interrupt(int irq, struct pt_regs *regs) irq_enter(); local_cpu_data().irq0_irqs++; - kstat_incr_irqs_this_cpu(0, irq_to_desc(0)); + kstat_incr_irq_this_cpu(0); if (unlikely(!evt->event_handler)) { printk(KERN_WARNING @@ -746,7 +745,7 @@ void __irq_entry timer_interrupt(int irq, struct pt_regs *regs) set_irq_regs(old_regs); } -void __devinit setup_sparc64_timer(void) +void setup_sparc64_timer(void) { struct clock_event_device *sevt; unsigned long pstate; @@ -844,7 +843,7 @@ unsigned long long sched_clock(void) >> SPARC64_NSEC_PER_CYC_SHIFT; } -int __devinit read_current_timer(unsigned long *timer_val) +int read_current_timer(unsigned long *timer_val) { *timer_val = tick_ops->get_tick(); return 0; diff --git a/arch/sparc/kernel/trampoline_32.S b/arch/sparc/kernel/trampoline_32.S index af27acab448..3eed99fc698 100644 --- a/arch/sparc/kernel/trampoline_32.S +++ b/arch/sparc/kernel/trampoline_32.S @@ -5,7 +5,6 @@ * Copyright (C) 1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz) */ -#include <linux/init.h> #include <asm/head.h> #include <asm/psr.h> #include <asm/page.h> @@ -18,7 +17,6 @@ .globl sun4m_cpu_startup .globl sun4d_cpu_startup - __CPUINIT .align 4 /* When we start up a cpu for the first time it enters this routine. @@ -79,25 +77,21 @@ cpu3_startup: nop /* Start this processor. */ - call smp4m_callin + call smp_callin nop - b,a smp_do_cpu_idle + b,a smp_panic .text .align 4 -smp_do_cpu_idle: - call cpu_idle - mov 0, %o0 - +smp_panic: call cpu_panic nop /* CPUID in bootbus can be found at PA 0xff0140000 */ #define SUN4D_BOOTBUS_CPUID 0xf0140000 - __CPUINIT .align 4 sun4d_cpu_startup: @@ -144,12 +138,11 @@ sun4d_cpu_startup: nop /* Start this processor. */ - call smp4d_callin + call smp_callin nop - b,a smp_do_cpu_idle + b,a smp_panic - __CPUINIT .align 4 .global leon_smp_cpu_startup, smp_penguin_ctable @@ -201,7 +194,7 @@ leon_smp_cpu_startup: nop /* Start this processor. */ - call leon_callin + call smp_callin nop - b,a smp_do_cpu_idle + b,a smp_panic diff --git a/arch/sparc/kernel/trampoline_64.S b/arch/sparc/kernel/trampoline_64.S index da1b781b5e6..737f8cbc7d5 100644 --- a/arch/sparc/kernel/trampoline_64.S +++ b/arch/sparc/kernel/trampoline_64.S @@ -4,7 +4,6 @@ * Copyright (C) 1997 David S. Miller (davem@caip.rutgers.edu) */ -#include <linux/init.h> #include <asm/head.h> #include <asm/asi.h> @@ -32,13 +31,11 @@ itlb_load: dtlb_load: .asciz "SUNW,dtlb-load" - /* XXX __cpuinit this thing XXX */ #define TRAMP_STACK_SIZE 1024 .align 16 tramp_stack: .skip TRAMP_STACK_SIZE - __CPUINIT .align 8 .globl sparc64_cpu_startup, sparc64_cpu_startup_end sparc64_cpu_startup: @@ -131,7 +128,6 @@ startup_continue: clr %l5 sethi %hi(num_kernel_image_mappings), %l6 lduw [%l6 + %lo(num_kernel_image_mappings)], %l6 - add %l6, 1, %l6 mov 15, %l7 BRANCH_IF_ANY_CHEETAH(g1,g5,2f) @@ -224,7 +220,6 @@ niagara_lock_tlb: clr %l5 sethi %hi(num_kernel_image_mappings), %l6 lduw [%l6 + %lo(num_kernel_image_mappings)], %l6 - add %l6, 1, %l6 1: mov HV_FAST_MMU_MAP_PERM_ADDR, %o5 @@ -407,8 +402,7 @@ after_lock_tlb: call smp_callin nop - call cpu_idle - mov 0, %o0 + call cpu_panic nop 1: b,a,pt %xcc, 1b diff --git a/arch/sparc/kernel/traps_32.c b/arch/sparc/kernel/traps_32.c index a5785ea2a85..6fd386c5232 100644 --- a/arch/sparc/kernel/traps_32.c +++ b/arch/sparc/kernel/traps_32.c @@ -44,7 +44,7 @@ static void instruction_dump(unsigned long *pc) #define __SAVE __asm__ __volatile__("save %sp, -0x40, %sp\n\t") #define __RESTORE __asm__ __volatile__("restore %g0, %g0, %g0\n\t") -void die_if_kernel(char *str, struct pt_regs *regs) +void __noreturn die_if_kernel(char *str, struct pt_regs *regs) { static int die_counter; int count = 0; @@ -58,7 +58,7 @@ void die_if_kernel(char *str, struct pt_regs *regs) printk("%s(%d): %s [#%d]\n", current->comm, task_pid_nr(current), str, ++die_counter); show_regs(regs); - add_taint(TAINT_DIE); + add_taint(TAINT_DIE, LOCKDEP_NOW_UNRELIABLE); __SAVE; __SAVE; __SAVE; __SAVE; __SAVE; __SAVE; __SAVE; __SAVE; @@ -219,8 +219,6 @@ static unsigned long fake_fsr; static unsigned long fake_queue[32] __attribute__ ((aligned (8))); static unsigned long fake_depth; -extern int do_mathemu(struct pt_regs *, struct task_struct *); - void do_fpe_trap(struct pt_regs *regs, unsigned long pc, unsigned long npc, unsigned long psr) { diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index 3b05e669771..fb6640ec855 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -1,6 +1,6 @@ /* arch/sparc64/kernel/traps.c * - * Copyright (C) 1995,1997,2008,2009 David S. Miller (davem@davemloft.net) + * Copyright (C) 1995,1997,2008,2009,2012 David S. Miller (davem@davemloft.net) * Copyright (C) 1997,1999,2000 Jakub Jelinek (jakub@redhat.com) */ @@ -18,7 +18,9 @@ #include <linux/init.h> #include <linux/kdebug.h> #include <linux/ftrace.h> +#include <linux/reboot.h> #include <linux/gfp.h> +#include <linux/context_tracking.h> #include <asm/smp.h> #include <asm/delay.h> @@ -41,8 +43,10 @@ #include <asm/prom.h> #include <asm/memctrl.h> #include <asm/cacheflush.h> +#include <asm/setup.h> #include "entry.h" +#include "kernel.h" #include "kstack.h" /* When an irrecoverable trap occurs at tl > 0, the trap entry @@ -185,11 +189,12 @@ EXPORT_SYMBOL_GPL(unregister_dimm_printer); void spitfire_insn_access_exception(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar) { + enum ctx_state prev_state = exception_enter(); siginfo_t info; if (notify_die(DIE_TRAP, "instruction access exception", regs, 0, 0x8, SIGTRAP) == NOTIFY_STOP) - return; + goto out; if (regs->tstate & TSTATE_PRIV) { printk("spitfire_insn_access_exception: SFSR[%016lx] " @@ -206,6 +211,8 @@ void spitfire_insn_access_exception(struct pt_regs *regs, unsigned long sfsr, un info.si_addr = (void __user *)regs->tpc; info.si_trapno = 0; force_sig_info(SIGSEGV, &info, current); +out: + exception_exit(prev_state); } void spitfire_insn_access_exception_tl1(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar) @@ -259,11 +266,12 @@ void sun4v_insn_access_exception_tl1(struct pt_regs *regs, unsigned long addr, u void spitfire_data_access_exception(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar) { + enum ctx_state prev_state = exception_enter(); siginfo_t info; if (notify_die(DIE_TRAP, "data access exception", regs, 0, 0x30, SIGTRAP) == NOTIFY_STOP) - return; + goto out; if (regs->tstate & TSTATE_PRIV) { /* Test if this comes from uaccess places. */ @@ -279,7 +287,7 @@ void spitfire_data_access_exception(struct pt_regs *regs, unsigned long sfsr, un #endif regs->tpc = entry->fixup; regs->tnpc = regs->tpc + 4; - return; + goto out; } /* Shit... */ printk("spitfire_data_access_exception: SFSR[%016lx] " @@ -293,6 +301,8 @@ void spitfire_data_access_exception(struct pt_regs *regs, unsigned long sfsr, un info.si_addr = (void __user *)sfar; info.si_trapno = 0; force_sig_info(SIGSEGV, &info, current); +out: + exception_exit(prev_state); } void spitfire_data_access_exception_tl1(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar) @@ -850,7 +860,7 @@ void __init cheetah_ecache_flush_init(void) ecache_flush_physbase = find_ecache_flush_span(ecache_flush_size); if (ecache_flush_physbase == ~0UL) { - prom_printf("cheetah_ecache_flush_init: Cannot find %d byte " + prom_printf("cheetah_ecache_flush_init: Cannot find %ld byte " "contiguous physical memory.\n", ecache_flush_size); prom_halt(); @@ -1760,85 +1770,223 @@ void cheetah_plus_parity_error(int type, struct pt_regs *regs) } struct sun4v_error_entry { - u64 err_handle; - u64 err_stick; + /* Unique error handle */ +/*0x00*/u64 err_handle; + + /* %stick value at the time of the error */ +/*0x08*/u64 err_stick; - u32 err_type; +/*0x10*/u8 reserved_1[3]; + + /* Error type */ +/*0x13*/u8 err_type; #define SUN4V_ERR_TYPE_UNDEFINED 0 #define SUN4V_ERR_TYPE_UNCORRECTED_RES 1 #define SUN4V_ERR_TYPE_PRECISE_NONRES 2 #define SUN4V_ERR_TYPE_DEFERRED_NONRES 3 -#define SUN4V_ERR_TYPE_WARNING_RES 4 +#define SUN4V_ERR_TYPE_SHUTDOWN_RQST 4 +#define SUN4V_ERR_TYPE_DUMP_CORE 5 +#define SUN4V_ERR_TYPE_SP_STATE_CHANGE 6 +#define SUN4V_ERR_TYPE_NUM 7 - u32 err_attrs; + /* Error attributes */ +/*0x14*/u32 err_attrs; #define SUN4V_ERR_ATTRS_PROCESSOR 0x00000001 #define SUN4V_ERR_ATTRS_MEMORY 0x00000002 #define SUN4V_ERR_ATTRS_PIO 0x00000004 #define SUN4V_ERR_ATTRS_INT_REGISTERS 0x00000008 #define SUN4V_ERR_ATTRS_FPU_REGISTERS 0x00000010 -#define SUN4V_ERR_ATTRS_USER_MODE 0x01000000 -#define SUN4V_ERR_ATTRS_PRIV_MODE 0x02000000 +#define SUN4V_ERR_ATTRS_SHUTDOWN_RQST 0x00000020 +#define SUN4V_ERR_ATTRS_ASR 0x00000040 +#define SUN4V_ERR_ATTRS_ASI 0x00000080 +#define SUN4V_ERR_ATTRS_PRIV_REG 0x00000100 +#define SUN4V_ERR_ATTRS_SPSTATE_MSK 0x00000600 +#define SUN4V_ERR_ATTRS_SPSTATE_SHFT 9 +#define SUN4V_ERR_ATTRS_MODE_MSK 0x03000000 +#define SUN4V_ERR_ATTRS_MODE_SHFT 24 #define SUN4V_ERR_ATTRS_RES_QUEUE_FULL 0x80000000 - u64 err_raddr; - u32 err_size; - u16 err_cpu; - u16 err_pad; +#define SUN4V_ERR_SPSTATE_FAULTED 0 +#define SUN4V_ERR_SPSTATE_AVAILABLE 1 +#define SUN4V_ERR_SPSTATE_NOT_PRESENT 2 + +#define SUN4V_ERR_MODE_USER 1 +#define SUN4V_ERR_MODE_PRIV 2 + + /* Real address of the memory region or PIO transaction */ +/*0x18*/u64 err_raddr; + + /* Size of the operation triggering the error, in bytes */ +/*0x20*/u32 err_size; + + /* ID of the CPU */ +/*0x24*/u16 err_cpu; + + /* Grace periof for shutdown, in seconds */ +/*0x26*/u16 err_secs; + + /* Value of the %asi register */ +/*0x28*/u8 err_asi; + +/*0x29*/u8 reserved_2; + + /* Value of the ASR register number */ +/*0x2a*/u16 err_asr; +#define SUN4V_ERR_ASR_VALID 0x8000 + +/*0x2c*/u32 reserved_3; +/*0x30*/u64 reserved_4; +/*0x38*/u64 reserved_5; }; static atomic_t sun4v_resum_oflow_cnt = ATOMIC_INIT(0); static atomic_t sun4v_nonresum_oflow_cnt = ATOMIC_INIT(0); -static const char *sun4v_err_type_to_str(u32 type) -{ - switch (type) { - case SUN4V_ERR_TYPE_UNDEFINED: - return "undefined"; - case SUN4V_ERR_TYPE_UNCORRECTED_RES: - return "uncorrected resumable"; - case SUN4V_ERR_TYPE_PRECISE_NONRES: - return "precise nonresumable"; - case SUN4V_ERR_TYPE_DEFERRED_NONRES: - return "deferred nonresumable"; - case SUN4V_ERR_TYPE_WARNING_RES: - return "warning resumable"; - default: - return "unknown"; +static const char *sun4v_err_type_to_str(u8 type) +{ + static const char *types[SUN4V_ERR_TYPE_NUM] = { + "undefined", + "uncorrected resumable", + "precise nonresumable", + "deferred nonresumable", + "shutdown request", + "dump core", + "SP state change", + }; + + if (type < SUN4V_ERR_TYPE_NUM) + return types[type]; + + return "unknown"; +} + +static void sun4v_emit_err_attr_strings(u32 attrs) +{ + static const char *attr_names[] = { + "processor", + "memory", + "PIO", + "int-registers", + "fpu-registers", + "shutdown-request", + "ASR", + "ASI", + "priv-reg", + }; + static const char *sp_states[] = { + "sp-faulted", + "sp-available", + "sp-not-present", + "sp-state-reserved", + }; + static const char *modes[] = { + "mode-reserved0", + "user", + "priv", + "mode-reserved1", + }; + u32 sp_state, mode; + int i; + + for (i = 0; i < ARRAY_SIZE(attr_names); i++) { + if (attrs & (1U << i)) { + const char *s = attr_names[i]; + + pr_cont("%s ", s); + } } + + sp_state = ((attrs & SUN4V_ERR_ATTRS_SPSTATE_MSK) >> + SUN4V_ERR_ATTRS_SPSTATE_SHFT); + pr_cont("%s ", sp_states[sp_state]); + + mode = ((attrs & SUN4V_ERR_ATTRS_MODE_MSK) >> + SUN4V_ERR_ATTRS_MODE_SHFT); + pr_cont("%s ", modes[mode]); + + if (attrs & SUN4V_ERR_ATTRS_RES_QUEUE_FULL) + pr_cont("res-queue-full "); +} + +/* When the report contains a real-address of "-1" it means that the + * hardware did not provide the address. So we compute the effective + * address of the load or store instruction at regs->tpc and report + * that. Usually when this happens it's a PIO and in such a case we + * are using physical addresses with bypass ASIs anyways, so what we + * report here is exactly what we want. + */ +static void sun4v_report_real_raddr(const char *pfx, struct pt_regs *regs) +{ + unsigned int insn; + u64 addr; + + if (!(regs->tstate & TSTATE_PRIV)) + return; + + insn = *(unsigned int *) regs->tpc; + + addr = compute_effective_address(regs, insn, 0); + + printk("%s: insn effective address [0x%016llx]\n", + pfx, addr); } -static void sun4v_log_error(struct pt_regs *regs, struct sun4v_error_entry *ent, int cpu, const char *pfx, atomic_t *ocnt) +static void sun4v_log_error(struct pt_regs *regs, struct sun4v_error_entry *ent, + int cpu, const char *pfx, atomic_t *ocnt) { + u64 *raw_ptr = (u64 *) ent; + u32 attrs; int cnt; printk("%s: Reporting on cpu %d\n", pfx, cpu); - printk("%s: err_handle[%llx] err_stick[%llx] err_type[%08x:%s]\n", - pfx, - ent->err_handle, ent->err_stick, - ent->err_type, - sun4v_err_type_to_str(ent->err_type)); - printk("%s: err_attrs[%08x:%s %s %s %s %s %s %s %s]\n", - pfx, - ent->err_attrs, - ((ent->err_attrs & SUN4V_ERR_ATTRS_PROCESSOR) ? - "processor" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_MEMORY) ? - "memory" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_PIO) ? - "pio" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_INT_REGISTERS) ? - "integer-regs" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_FPU_REGISTERS) ? - "fpu-regs" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_USER_MODE) ? - "user" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_PRIV_MODE) ? - "privileged" : ""), - ((ent->err_attrs & SUN4V_ERR_ATTRS_RES_QUEUE_FULL) ? - "queue-full" : "")); - printk("%s: err_raddr[%016llx] err_size[%u] err_cpu[%u]\n", - pfx, - ent->err_raddr, ent->err_size, ent->err_cpu); + printk("%s: TPC [0x%016lx] <%pS>\n", + pfx, regs->tpc, (void *) regs->tpc); + + printk("%s: RAW [%016llx:%016llx:%016llx:%016llx\n", + pfx, raw_ptr[0], raw_ptr[1], raw_ptr[2], raw_ptr[3]); + printk("%s: %016llx:%016llx:%016llx:%016llx]\n", + pfx, raw_ptr[4], raw_ptr[5], raw_ptr[6], raw_ptr[7]); + + printk("%s: handle [0x%016llx] stick [0x%016llx]\n", + pfx, ent->err_handle, ent->err_stick); + + printk("%s: type [%s]\n", pfx, sun4v_err_type_to_str(ent->err_type)); + + attrs = ent->err_attrs; + printk("%s: attrs [0x%08x] < ", pfx, attrs); + sun4v_emit_err_attr_strings(attrs); + pr_cont(">\n"); + + /* Various fields in the error report are only valid if + * certain attribute bits are set. + */ + if (attrs & (SUN4V_ERR_ATTRS_MEMORY | + SUN4V_ERR_ATTRS_PIO | + SUN4V_ERR_ATTRS_ASI)) { + printk("%s: raddr [0x%016llx]\n", pfx, ent->err_raddr); + + if (ent->err_raddr == ~(u64)0) + sun4v_report_real_raddr(pfx, regs); + } + + if (attrs & (SUN4V_ERR_ATTRS_MEMORY | SUN4V_ERR_ATTRS_ASI)) + printk("%s: size [0x%x]\n", pfx, ent->err_size); + + if (attrs & (SUN4V_ERR_ATTRS_PROCESSOR | + SUN4V_ERR_ATTRS_INT_REGISTERS | + SUN4V_ERR_ATTRS_FPU_REGISTERS | + SUN4V_ERR_ATTRS_PRIV_REG)) + printk("%s: cpu[%u]\n", pfx, ent->err_cpu); + + if (attrs & SUN4V_ERR_ATTRS_ASI) + printk("%s: asi [0x%02x]\n", pfx, ent->err_asi); + + if ((attrs & (SUN4V_ERR_ATTRS_INT_REGISTERS | + SUN4V_ERR_ATTRS_FPU_REGISTERS | + SUN4V_ERR_ATTRS_PRIV_REG)) && + (ent->err_asr & SUN4V_ERR_ASR_VALID) != 0) + printk("%s: reg [0x%04x]\n", + pfx, ent->err_asr & ~SUN4V_ERR_ASR_VALID); show_regs(regs); @@ -1855,6 +2003,7 @@ static void sun4v_log_error(struct pt_regs *regs, struct sun4v_error_entry *ent, */ void sun4v_resum_error(struct pt_regs *regs, unsigned long offset) { + enum ctx_state prev_state = exception_enter(); struct sun4v_error_entry *ent, local_copy; struct trap_per_cpu *tb; unsigned long paddr; @@ -1874,19 +2023,23 @@ void sun4v_resum_error(struct pt_regs *regs, unsigned long offset) put_cpu(); - if (ent->err_type == SUN4V_ERR_TYPE_WARNING_RES) { - /* If err_type is 0x4, it's a powerdown request. Do - * not do the usual resumable error log because that - * makes it look like some abnormal error. + if (local_copy.err_type == SUN4V_ERR_TYPE_SHUTDOWN_RQST) { + /* We should really take the seconds field of + * the error report and use it for the shutdown + * invocation, but for now do the same thing we + * do for a DS shutdown request. */ - printk(KERN_INFO "Power down request...\n"); - kill_cad_pid(SIGINT, 1); - return; + pr_info("Shutdown request, %u seconds...\n", + local_copy.err_secs); + orderly_poweroff(true); + goto out; } sun4v_log_error(regs, &local_copy, cpu, KERN_ERR "RESUMABLE ERROR", &sun4v_resum_oflow_cnt); +out: + exception_exit(prev_state); } /* If we try to printk() we'll probably make matters worse, by trying @@ -2011,7 +2164,7 @@ void hypervisor_tlbop_error_xcall(unsigned long err, unsigned long op) err, op); } -void do_fpe_common(struct pt_regs *regs) +static void do_fpe_common(struct pt_regs *regs) { if (regs->tstate & TSTATE_PRIV) { regs->tpc = regs->tnpc; @@ -2047,23 +2200,26 @@ void do_fpe_common(struct pt_regs *regs) void do_fpieee(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); + if (notify_die(DIE_TRAP, "fpu exception ieee", regs, 0, 0x24, SIGFPE) == NOTIFY_STOP) - return; + goto out; do_fpe_common(regs); +out: + exception_exit(prev_state); } -extern int do_mathemu(struct pt_regs *, struct fpustate *, bool); - void do_fpother(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); struct fpustate *f = FPUSTATE; int ret = 0; if (notify_die(DIE_TRAP, "fpu exception other", regs, 0, 0x25, SIGFPE) == NOTIFY_STOP) - return; + goto out; switch ((current_thread_info()->xfsr[0] & 0x1c000)) { case (2 << 14): /* unfinished_FPop */ @@ -2072,17 +2228,20 @@ void do_fpother(struct pt_regs *regs) break; } if (ret) - return; + goto out; do_fpe_common(regs); +out: + exception_exit(prev_state); } void do_tof(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); siginfo_t info; if (notify_die(DIE_TRAP, "tagged arithmetic overflow", regs, 0, 0x26, SIGEMT) == NOTIFY_STOP) - return; + goto out; if (regs->tstate & TSTATE_PRIV) die_if_kernel("Penguin overflow trap from kernel mode", regs); @@ -2096,15 +2255,18 @@ void do_tof(struct pt_regs *regs) info.si_addr = (void __user *)regs->tpc; info.si_trapno = 0; force_sig_info(SIGEMT, &info, current); +out: + exception_exit(prev_state); } void do_div0(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); siginfo_t info; if (notify_die(DIE_TRAP, "integer division by zero", regs, 0, 0x28, SIGFPE) == NOTIFY_STOP) - return; + goto out; if (regs->tstate & TSTATE_PRIV) die_if_kernel("TL0: Kernel divide by zero.", regs); @@ -2118,6 +2280,8 @@ void do_div0(struct pt_regs *regs) info.si_addr = (void __user *)regs->tpc; info.si_trapno = 0; force_sig_info(SIGFPE, &info, current); +out: + exception_exit(prev_state); } static void instruction_dump(unsigned int *pc) @@ -2209,13 +2373,6 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp) } while (++count < 16); } -void dump_stack(void) -{ - show_stack(current, NULL); -} - -EXPORT_SYMBOL(dump_stack); - static inline struct reg_window *kernel_stack_up(struct reg_window *rw) { unsigned long fp = rw->ins[6]; @@ -2226,7 +2383,7 @@ static inline struct reg_window *kernel_stack_up(struct reg_window *rw) return (struct reg_window *) (fp + STACK_BIAS); } -void die_if_kernel(char *str, struct pt_regs *regs) +void __noreturn die_if_kernel(char *str, struct pt_regs *regs) { static int die_counter; int count = 0; @@ -2242,7 +2399,7 @@ void die_if_kernel(char *str, struct pt_regs *regs) notify_die(DIE_OOPS, str, regs, 0, 255, SIGSEGV); __asm__ __volatile__("flushw"); show_regs(regs); - add_taint(TAINT_DIE); + add_taint(TAINT_DIE, LOCKDEP_NOW_UNRELIABLE); if (regs->tstate & TSTATE_PRIV) { struct thread_info *tp = current_thread_info(); struct reg_window *rw = (struct reg_window *) @@ -2276,11 +2433,9 @@ EXPORT_SYMBOL(die_if_kernel); #define VIS_OPCODE_MASK ((0x3 << 30) | (0x3f << 19)) #define VIS_OPCODE_VAL ((0x2 << 30) | (0x36 << 19)) -extern int handle_popc(u32 insn, struct pt_regs *regs); -extern int handle_ldf_stq(u32 insn, struct pt_regs *regs); - void do_illegal_instruction(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); unsigned long pc = regs->tpc; unsigned long tstate = regs->tstate; u32 insn; @@ -2288,7 +2443,7 @@ void do_illegal_instruction(struct pt_regs *regs) if (notify_die(DIE_TRAP, "illegal instruction", regs, 0, 0x10, SIGILL) == NOTIFY_STOP) - return; + goto out; if (tstate & TSTATE_PRIV) die_if_kernel("Kernel illegal instruction", regs); @@ -2297,14 +2452,14 @@ void do_illegal_instruction(struct pt_regs *regs) if (get_user(insn, (u32 __user *) pc) != -EFAULT) { if ((insn & 0xc1ffc000) == 0x81700000) /* POPC */ { if (handle_popc(insn, regs)) - return; + goto out; } else if ((insn & 0xc1580000) == 0xc1100000) /* LDQ/STQ */ { if (handle_ldf_stq(insn, regs)) - return; + goto out; } else if (tlb_type == hypervisor) { if ((insn & VIS_OPCODE_MASK) == VIS_OPCODE_VAL) { if (!vis_emul(regs, insn)) - return; + goto out; } else { struct fpustate *f = FPUSTATE; @@ -2314,7 +2469,7 @@ void do_illegal_instruction(struct pt_regs *regs) * Trap in the %fsr to unimplemented_FPop. */ if (do_mathemu(regs, f, true)) - return; + goto out; } } } @@ -2324,21 +2479,22 @@ void do_illegal_instruction(struct pt_regs *regs) info.si_addr = (void __user *)pc; info.si_trapno = 0; force_sig_info(SIGILL, &info, current); +out: + exception_exit(prev_state); } -extern void kernel_unaligned_trap(struct pt_regs *regs, unsigned int insn); - void mem_address_unaligned(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr) { + enum ctx_state prev_state = exception_enter(); siginfo_t info; if (notify_die(DIE_TRAP, "memory address unaligned", regs, 0, 0x34, SIGSEGV) == NOTIFY_STOP) - return; + goto out; if (regs->tstate & TSTATE_PRIV) { kernel_unaligned_trap(regs, *((unsigned int *)regs->tpc)); - return; + goto out; } info.si_signo = SIGBUS; info.si_errno = 0; @@ -2346,6 +2502,8 @@ void mem_address_unaligned(struct pt_regs *regs, unsigned long sfar, unsigned lo info.si_addr = (void __user *)sfar; info.si_trapno = 0; force_sig_info(SIGBUS, &info, current); +out: + exception_exit(prev_state); } void sun4v_do_mna(struct pt_regs *regs, unsigned long addr, unsigned long type_ctx) @@ -2370,11 +2528,12 @@ void sun4v_do_mna(struct pt_regs *regs, unsigned long addr, unsigned long type_c void do_privop(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); siginfo_t info; if (notify_die(DIE_TRAP, "privileged operation", regs, 0, 0x11, SIGILL) == NOTIFY_STOP) - return; + goto out; if (test_thread_flag(TIF_32BIT)) { regs->tpc &= 0xffffffff; @@ -2386,6 +2545,8 @@ void do_privop(struct pt_regs *regs) info.si_addr = (void __user *)regs->tpc; info.si_trapno = 0; force_sig_info(SIGILL, &info, current); +out: + exception_exit(prev_state); } void do_privact(struct pt_regs *regs) @@ -2396,99 +2557,116 @@ void do_privact(struct pt_regs *regs) /* Trap level 1 stuff or other traps we should never see... */ void do_cee(struct pt_regs *regs) { + exception_enter(); die_if_kernel("TL0: Cache Error Exception", regs); } void do_cee_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: Cache Error Exception", regs); } void do_dae_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: Data Access Exception", regs); } void do_iae_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: Instruction Access Exception", regs); } void do_div0_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: DIV0 Exception", regs); } void do_fpdis_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: FPU Disabled", regs); } void do_fpieee_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: FPU IEEE Exception", regs); } void do_fpother_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: FPU Other Exception", regs); } void do_ill_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: Illegal Instruction Exception", regs); } void do_irq_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: IRQ Exception", regs); } void do_lddfmna_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: LDDF Exception", regs); } void do_stdfmna_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: STDF Exception", regs); } void do_paw(struct pt_regs *regs) { + exception_enter(); die_if_kernel("TL0: Phys Watchpoint Exception", regs); } void do_paw_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: Phys Watchpoint Exception", regs); } void do_vaw(struct pt_regs *regs) { + exception_enter(); die_if_kernel("TL0: Virt Watchpoint Exception", regs); } void do_vaw_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: Virt Watchpoint Exception", regs); } void do_tof_tl1(struct pt_regs *regs) { + exception_enter(); dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); die_if_kernel("TL1: Tag Overflow Exception", regs); } @@ -2547,8 +2725,8 @@ void __init trap_init(void) TI_PRE_COUNT != offsetof(struct thread_info, preempt_count) || TI_NEW_CHILD != offsetof(struct thread_info, new_child) || - TI_SYS_NOERROR != offsetof(struct thread_info, - syscall_noerror) || + TI_CURRENT_DS != offsetof(struct thread_info, + current_ds) || TI_RESTART_BLOCK != offsetof(struct thread_info, restart_block) || TI_KUNA_REGS != offsetof(struct thread_info, diff --git a/arch/sparc/kernel/tsb.S b/arch/sparc/kernel/tsb.S index db15d123f05..14158d40ba7 100644 --- a/arch/sparc/kernel/tsb.S +++ b/arch/sparc/kernel/tsb.S @@ -49,7 +49,7 @@ tsb_miss_page_table_walk: /* Before committing to a full page table walk, * check the huge page TSB. */ -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) 661: ldx [%g7 + TRAP_PER_CPU_TSB_HUGE], %g5 nop @@ -75,7 +75,7 @@ tsb_miss_page_table_walk: mov 512, %g7 andn %g5, 0x7, %g5 sllx %g7, %g6, %g7 - srlx %g4, HPAGE_SHIFT, %g6 + srlx %g4, REAL_HPAGE_SHIFT, %g6 sub %g7, 1, %g7 and %g6, %g7, %g6 sllx %g6, 4, %g6 @@ -110,12 +110,9 @@ tsb_miss_page_table_walk: tsb_miss_page_table_walk_sun4v_fastpath: USER_PGTABLE_WALK_TL1(%g4, %g7, %g5, %g2, tsb_do_fault) - /* Load and check PTE. */ - ldxa [%g5] ASI_PHYS_USE_EC, %g5 - brgez,pn %g5, tsb_do_fault - nop + /* Valid PTE is now in %g5. */ -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) 661: sethi %uhi(_PAGE_SZALL_4U), %g7 sllx %g7, 32, %g7 .section .sun4v_2insn_patch, "ax" @@ -139,12 +136,43 @@ tsb_miss_page_table_walk_sun4v_fastpath: nop /* It is a huge page, use huge page TSB entry address we - * calculated above. + * calculated above. If the huge page TSB has not been + * allocated, setup a trap stack and call hugetlb_setup() + * to do so, then return from the trap to replay the TLB + * miss. + * + * This is necessary to handle the case of transparent huge + * pages where we don't really have a non-atomic context + * in which to allocate the hugepage TSB hash table. When + * the 'mm' faults in the hugepage for the first time, we + * thus handle it here. This also makes sure that we can + * allocate the TSB hash table on the correct NUMA node. */ TRAP_LOAD_TRAP_BLOCK(%g7, %g2) - ldx [%g7 + TRAP_PER_CPU_TSB_HUGE_TEMP], %g2 - cmp %g2, -1 - movne %xcc, %g2, %g1 + ldx [%g7 + TRAP_PER_CPU_TSB_HUGE_TEMP], %g1 + cmp %g1, -1 + bne,pt %xcc, 60f + nop + +661: rdpr %pstate, %g5 + wrpr %g5, PSTATE_AG | PSTATE_MG, %pstate + .section .sun4v_2insn_patch, "ax" + .word 661b + SET_GL(1) + nop + .previous + + rdpr %tl, %g3 + cmp %g3, 1 + bne,pn %xcc, winfix_trampoline + nop + ba,pt %xcc, etrap + rd %pc, %g7 + call hugetlb_setup + add %sp, PTREGS_OFF, %o0 + ba,pt %xcc, rtrap + nop + 60: #endif diff --git a/arch/sparc/kernel/unaligned_32.c b/arch/sparc/kernel/unaligned_32.c index c0ec8978619..c5c61b3c6b5 100644 --- a/arch/sparc/kernel/unaligned_32.c +++ b/arch/sparc/kernel/unaligned_32.c @@ -16,6 +16,10 @@ #include <linux/smp.h> #include <linux/perf_event.h> +#include <asm/setup.h> + +#include "kernel.h" + enum direction { load, /* ld, ldd, ldh, ldsh */ store, /* st, std, sth, stsh */ diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c index f81d038f734..62098a89bbb 100644 --- a/arch/sparc/kernel/unaligned_64.c +++ b/arch/sparc/kernel/unaligned_64.c @@ -21,8 +21,13 @@ #include <linux/bitops.h> #include <linux/perf_event.h> #include <linux/ratelimit.h> +#include <linux/context_tracking.h> #include <asm/fpumacro.h> #include <asm/cacheflush.h> +#include <asm/setup.h> + +#include "entry.h" +#include "kernel.h" enum direction { load, /* ld, ldd, ldh, ldsh */ @@ -113,21 +118,24 @@ static inline long sign_extend_imm13(long imm) static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) { - unsigned long value; + unsigned long value, fp; if (reg < 16) return (!reg ? 0 : regs->u_regs[reg]); + + fp = regs->u_regs[UREG_FP]; + if (regs->tstate & TSTATE_PRIV) { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); value = win->locals[reg - 16]; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); get_user(value, &win32->locals[reg - 16]); } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); get_user(value, &win->locals[reg - 16]); } return value; @@ -135,19 +143,24 @@ static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) static unsigned long *fetch_reg_addr(unsigned int reg, struct pt_regs *regs) { + unsigned long fp; + if (reg < 16) return ®s->u_regs[reg]; + + fp = regs->u_regs[UREG_FP]; + if (regs->tstate & TSTATE_PRIV) { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); return &win->locals[reg - 16]; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(fp)) { struct reg_window32 *win32; - win32 = (struct reg_window32 *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 *)((unsigned long)((u32)fp)); return (unsigned long *)&win32->locals[reg - 16]; } else { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); return &win->locals[reg - 16]; } } @@ -155,17 +168,23 @@ static unsigned long *fetch_reg_addr(unsigned int reg, struct pt_regs *regs) unsigned long compute_effective_address(struct pt_regs *regs, unsigned int insn, unsigned int rd) { + int from_kernel = (regs->tstate & TSTATE_PRIV) != 0; unsigned int rs1 = (insn >> 14) & 0x1f; unsigned int rs2 = insn & 0x1f; - int from_kernel = (regs->tstate & TSTATE_PRIV) != 0; + unsigned long addr; if (insn & 0x2000) { maybe_flush_windows(rs1, 0, rd, from_kernel); - return (fetch_reg(rs1, regs) + sign_extend_imm13(insn)); + addr = (fetch_reg(rs1, regs) + sign_extend_imm13(insn)); } else { maybe_flush_windows(rs1, rs2, rd, from_kernel); - return (fetch_reg(rs1, regs) + fetch_reg(rs2, regs)); + addr = (fetch_reg(rs1, regs) + fetch_reg(rs2, regs)); } + + if (!from_kernel && test_thread_flag(TIF_32BIT)) + addr &= 0xffffffff; + + return addr; } /* This is just to make gcc think die_if_kernel does return... */ @@ -392,13 +411,15 @@ int handle_popc(u32 insn, struct pt_regs *regs) if (rd) regs->u_regs[rd] = ret; } else { - if (test_thread_flag(TIF_32BIT)) { + unsigned long fp = regs->u_regs[UREG_FP]; + + if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); put_user(ret, &win32->locals[rd - 16]); } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); put_user(ret, &win->locals[rd - 16]); } } @@ -408,9 +429,6 @@ int handle_popc(u32 insn, struct pt_regs *regs) extern void do_fpother(struct pt_regs *regs); extern void do_privact(struct pt_regs *regs); -extern void spitfire_data_access_exception(struct pt_regs *regs, - unsigned long sfsr, - unsigned long sfar); extern void sun4v_data_access_exception(struct pt_regs *regs, unsigned long addr, unsigned long type_ctx); @@ -554,7 +572,7 @@ void handle_ld_nf(u32 insn, struct pt_regs *regs) reg[0] = 0; if ((insn & 0x780000) == 0x180000) reg[1] = 0; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(regs->u_regs[UREG_FP])) { put_user(0, (int __user *) reg); if ((insn & 0x780000) == 0x180000) put_user(0, ((int __user *) reg) + 1); @@ -568,6 +586,7 @@ void handle_ld_nf(u32 insn, struct pt_regs *regs) void handle_lddfmna(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr) { + enum ctx_state prev_state = exception_enter(); unsigned long pc = regs->tpc; unsigned long tstate = regs->tstate; u32 insn; @@ -622,13 +641,16 @@ daex: sun4v_data_access_exception(regs, sfar, sfsr); else spitfire_data_access_exception(regs, sfsr, sfar); - return; + goto out; } advance(regs); +out: + exception_exit(prev_state); } void handle_stdfmna(struct pt_regs *regs, unsigned long sfar, unsigned long sfsr) { + enum ctx_state prev_state = exception_enter(); unsigned long pc = regs->tpc; unsigned long tstate = regs->tstate; u32 insn; @@ -670,7 +692,9 @@ daex: sun4v_data_access_exception(regs, sfar, sfsr); else spitfire_data_access_exception(regs, sfsr, sfar); - return; + goto out; } advance(regs); +out: + exception_exit(prev_state); } diff --git a/arch/sparc/kernel/us2e_cpufreq.c b/arch/sparc/kernel/us2e_cpufreq.c deleted file mode 100644 index 489fc15f319..00000000000 --- a/arch/sparc/kernel/us2e_cpufreq.c +++ /dev/null @@ -1,413 +0,0 @@ -/* us2e_cpufreq.c: UltraSPARC-IIe cpu frequency support - * - * Copyright (C) 2003 David S. Miller (davem@redhat.com) - * - * Many thanks to Dominik Brodowski for fixing up the cpufreq - * infrastructure in order to make this driver easier to implement. - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <linux/smp.h> -#include <linux/cpufreq.h> -#include <linux/threads.h> -#include <linux/slab.h> -#include <linux/delay.h> -#include <linux/init.h> - -#include <asm/asi.h> -#include <asm/timer.h> - -static struct cpufreq_driver *cpufreq_us2e_driver; - -struct us2e_freq_percpu_info { - struct cpufreq_frequency_table table[6]; -}; - -/* Indexed by cpu number. */ -static struct us2e_freq_percpu_info *us2e_freq_table; - -#define HBIRD_MEM_CNTL0_ADDR 0x1fe0000f010UL -#define HBIRD_ESTAR_MODE_ADDR 0x1fe0000f080UL - -/* UltraSPARC-IIe has five dividers: 1, 2, 4, 6, and 8. These are controlled - * in the ESTAR mode control register. - */ -#define ESTAR_MODE_DIV_1 0x0000000000000000UL -#define ESTAR_MODE_DIV_2 0x0000000000000001UL -#define ESTAR_MODE_DIV_4 0x0000000000000003UL -#define ESTAR_MODE_DIV_6 0x0000000000000002UL -#define ESTAR_MODE_DIV_8 0x0000000000000004UL -#define ESTAR_MODE_DIV_MASK 0x0000000000000007UL - -#define MCTRL0_SREFRESH_ENAB 0x0000000000010000UL -#define MCTRL0_REFR_COUNT_MASK 0x0000000000007f00UL -#define MCTRL0_REFR_COUNT_SHIFT 8 -#define MCTRL0_REFR_INTERVAL 7800 -#define MCTRL0_REFR_CLKS_P_CNT 64 - -static unsigned long read_hbreg(unsigned long addr) -{ - unsigned long ret; - - __asm__ __volatile__("ldxa [%1] %2, %0" - : "=&r" (ret) - : "r" (addr), "i" (ASI_PHYS_BYPASS_EC_E)); - return ret; -} - -static void write_hbreg(unsigned long addr, unsigned long val) -{ - __asm__ __volatile__("stxa %0, [%1] %2\n\t" - "membar #Sync" - : /* no outputs */ - : "r" (val), "r" (addr), "i" (ASI_PHYS_BYPASS_EC_E) - : "memory"); - if (addr == HBIRD_ESTAR_MODE_ADDR) { - /* Need to wait 16 clock cycles for the PLL to lock. */ - udelay(1); - } -} - -static void self_refresh_ctl(int enable) -{ - unsigned long mctrl = read_hbreg(HBIRD_MEM_CNTL0_ADDR); - - if (enable) - mctrl |= MCTRL0_SREFRESH_ENAB; - else - mctrl &= ~MCTRL0_SREFRESH_ENAB; - write_hbreg(HBIRD_MEM_CNTL0_ADDR, mctrl); - (void) read_hbreg(HBIRD_MEM_CNTL0_ADDR); -} - -static void frob_mem_refresh(int cpu_slowing_down, - unsigned long clock_tick, - unsigned long old_divisor, unsigned long divisor) -{ - unsigned long old_refr_count, refr_count, mctrl; - - refr_count = (clock_tick * MCTRL0_REFR_INTERVAL); - refr_count /= (MCTRL0_REFR_CLKS_P_CNT * divisor * 1000000000UL); - - mctrl = read_hbreg(HBIRD_MEM_CNTL0_ADDR); - old_refr_count = (mctrl & MCTRL0_REFR_COUNT_MASK) - >> MCTRL0_REFR_COUNT_SHIFT; - - mctrl &= ~MCTRL0_REFR_COUNT_MASK; - mctrl |= refr_count << MCTRL0_REFR_COUNT_SHIFT; - write_hbreg(HBIRD_MEM_CNTL0_ADDR, mctrl); - mctrl = read_hbreg(HBIRD_MEM_CNTL0_ADDR); - - if (cpu_slowing_down && !(mctrl & MCTRL0_SREFRESH_ENAB)) { - unsigned long usecs; - - /* We have to wait for both refresh counts (old - * and new) to go to zero. - */ - usecs = (MCTRL0_REFR_CLKS_P_CNT * - (refr_count + old_refr_count) * - 1000000UL * - old_divisor) / clock_tick; - udelay(usecs + 1UL); - } -} - -static void us2e_transition(unsigned long estar, unsigned long new_bits, - unsigned long clock_tick, - unsigned long old_divisor, unsigned long divisor) -{ - unsigned long flags; - - local_irq_save(flags); - - estar &= ~ESTAR_MODE_DIV_MASK; - - /* This is based upon the state transition diagram in the IIe manual. */ - if (old_divisor == 2 && divisor == 1) { - self_refresh_ctl(0); - write_hbreg(HBIRD_ESTAR_MODE_ADDR, estar | new_bits); - frob_mem_refresh(0, clock_tick, old_divisor, divisor); - } else if (old_divisor == 1 && divisor == 2) { - frob_mem_refresh(1, clock_tick, old_divisor, divisor); - write_hbreg(HBIRD_ESTAR_MODE_ADDR, estar | new_bits); - self_refresh_ctl(1); - } else if (old_divisor == 1 && divisor > 2) { - us2e_transition(estar, ESTAR_MODE_DIV_2, clock_tick, - 1, 2); - us2e_transition(estar, new_bits, clock_tick, - 2, divisor); - } else if (old_divisor > 2 && divisor == 1) { - us2e_transition(estar, ESTAR_MODE_DIV_2, clock_tick, - old_divisor, 2); - us2e_transition(estar, new_bits, clock_tick, - 2, divisor); - } else if (old_divisor < divisor) { - frob_mem_refresh(0, clock_tick, old_divisor, divisor); - write_hbreg(HBIRD_ESTAR_MODE_ADDR, estar | new_bits); - } else if (old_divisor > divisor) { - write_hbreg(HBIRD_ESTAR_MODE_ADDR, estar | new_bits); - frob_mem_refresh(1, clock_tick, old_divisor, divisor); - } else { - BUG(); - } - - local_irq_restore(flags); -} - -static unsigned long index_to_estar_mode(unsigned int index) -{ - switch (index) { - case 0: - return ESTAR_MODE_DIV_1; - - case 1: - return ESTAR_MODE_DIV_2; - - case 2: - return ESTAR_MODE_DIV_4; - - case 3: - return ESTAR_MODE_DIV_6; - - case 4: - return ESTAR_MODE_DIV_8; - - default: - BUG(); - } -} - -static unsigned long index_to_divisor(unsigned int index) -{ - switch (index) { - case 0: - return 1; - - case 1: - return 2; - - case 2: - return 4; - - case 3: - return 6; - - case 4: - return 8; - - default: - BUG(); - } -} - -static unsigned long estar_to_divisor(unsigned long estar) -{ - unsigned long ret; - - switch (estar & ESTAR_MODE_DIV_MASK) { - case ESTAR_MODE_DIV_1: - ret = 1; - break; - case ESTAR_MODE_DIV_2: - ret = 2; - break; - case ESTAR_MODE_DIV_4: - ret = 4; - break; - case ESTAR_MODE_DIV_6: - ret = 6; - break; - case ESTAR_MODE_DIV_8: - ret = 8; - break; - default: - BUG(); - } - - return ret; -} - -static unsigned int us2e_freq_get(unsigned int cpu) -{ - cpumask_t cpus_allowed; - unsigned long clock_tick, estar; - - if (!cpu_online(cpu)) - return 0; - - cpumask_copy(&cpus_allowed, tsk_cpus_allowed(current)); - set_cpus_allowed_ptr(current, cpumask_of(cpu)); - - clock_tick = sparc64_get_clock_tick(cpu) / 1000; - estar = read_hbreg(HBIRD_ESTAR_MODE_ADDR); - - set_cpus_allowed_ptr(current, &cpus_allowed); - - return clock_tick / estar_to_divisor(estar); -} - -static void us2e_set_cpu_divider_index(unsigned int cpu, unsigned int index) -{ - unsigned long new_bits, new_freq; - unsigned long clock_tick, divisor, old_divisor, estar; - cpumask_t cpus_allowed; - struct cpufreq_freqs freqs; - - if (!cpu_online(cpu)) - return; - - cpumask_copy(&cpus_allowed, tsk_cpus_allowed(current)); - set_cpus_allowed_ptr(current, cpumask_of(cpu)); - - new_freq = clock_tick = sparc64_get_clock_tick(cpu) / 1000; - new_bits = index_to_estar_mode(index); - divisor = index_to_divisor(index); - new_freq /= divisor; - - estar = read_hbreg(HBIRD_ESTAR_MODE_ADDR); - - old_divisor = estar_to_divisor(estar); - - freqs.old = clock_tick / old_divisor; - freqs.new = new_freq; - freqs.cpu = cpu; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - if (old_divisor != divisor) - us2e_transition(estar, new_bits, clock_tick * 1000, - old_divisor, divisor); - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - - set_cpus_allowed_ptr(current, &cpus_allowed); -} - -static int us2e_freq_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int new_index = 0; - - if (cpufreq_frequency_table_target(policy, - &us2e_freq_table[policy->cpu].table[0], - target_freq, relation, &new_index)) - return -EINVAL; - - us2e_set_cpu_divider_index(policy->cpu, new_index); - - return 0; -} - -static int us2e_freq_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, - &us2e_freq_table[policy->cpu].table[0]); -} - -static int __init us2e_freq_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int cpu = policy->cpu; - unsigned long clock_tick = sparc64_get_clock_tick(cpu) / 1000; - struct cpufreq_frequency_table *table = - &us2e_freq_table[cpu].table[0]; - - table[0].index = 0; - table[0].frequency = clock_tick / 1; - table[1].index = 1; - table[1].frequency = clock_tick / 2; - table[2].index = 2; - table[2].frequency = clock_tick / 4; - table[2].index = 3; - table[2].frequency = clock_tick / 6; - table[2].index = 4; - table[2].frequency = clock_tick / 8; - table[2].index = 5; - table[3].frequency = CPUFREQ_TABLE_END; - - policy->cpuinfo.transition_latency = 0; - policy->cur = clock_tick; - - return cpufreq_frequency_table_cpuinfo(policy, table); -} - -static int us2e_freq_cpu_exit(struct cpufreq_policy *policy) -{ - if (cpufreq_us2e_driver) - us2e_set_cpu_divider_index(policy->cpu, 0); - - return 0; -} - -static int __init us2e_freq_init(void) -{ - unsigned long manuf, impl, ver; - int ret; - - if (tlb_type != spitfire) - return -ENODEV; - - __asm__("rdpr %%ver, %0" : "=r" (ver)); - manuf = ((ver >> 48) & 0xffff); - impl = ((ver >> 32) & 0xffff); - - if (manuf == 0x17 && impl == 0x13) { - struct cpufreq_driver *driver; - - ret = -ENOMEM; - driver = kzalloc(sizeof(struct cpufreq_driver), GFP_KERNEL); - if (!driver) - goto err_out; - - us2e_freq_table = kzalloc( - (NR_CPUS * sizeof(struct us2e_freq_percpu_info)), - GFP_KERNEL); - if (!us2e_freq_table) - goto err_out; - - driver->init = us2e_freq_cpu_init; - driver->verify = us2e_freq_verify; - driver->target = us2e_freq_target; - driver->get = us2e_freq_get; - driver->exit = us2e_freq_cpu_exit; - driver->owner = THIS_MODULE, - strcpy(driver->name, "UltraSPARC-IIe"); - - cpufreq_us2e_driver = driver; - ret = cpufreq_register_driver(driver); - if (ret) - goto err_out; - - return 0; - -err_out: - if (driver) { - kfree(driver); - cpufreq_us2e_driver = NULL; - } - kfree(us2e_freq_table); - us2e_freq_table = NULL; - return ret; - } - - return -ENODEV; -} - -static void __exit us2e_freq_exit(void) -{ - if (cpufreq_us2e_driver) { - cpufreq_unregister_driver(cpufreq_us2e_driver); - kfree(cpufreq_us2e_driver); - cpufreq_us2e_driver = NULL; - kfree(us2e_freq_table); - us2e_freq_table = NULL; - } -} - -MODULE_AUTHOR("David S. Miller <davem@redhat.com>"); -MODULE_DESCRIPTION("cpufreq driver for UltraSPARC-IIe"); -MODULE_LICENSE("GPL"); - -module_init(us2e_freq_init); -module_exit(us2e_freq_exit); diff --git a/arch/sparc/kernel/us3_cpufreq.c b/arch/sparc/kernel/us3_cpufreq.c deleted file mode 100644 index eb1624b931d..00000000000 --- a/arch/sparc/kernel/us3_cpufreq.c +++ /dev/null @@ -1,274 +0,0 @@ -/* us3_cpufreq.c: UltraSPARC-III cpu frequency support - * - * Copyright (C) 2003 David S. Miller (davem@redhat.com) - * - * Many thanks to Dominik Brodowski for fixing up the cpufreq - * infrastructure in order to make this driver easier to implement. - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/sched.h> -#include <linux/smp.h> -#include <linux/cpufreq.h> -#include <linux/threads.h> -#include <linux/slab.h> -#include <linux/init.h> - -#include <asm/head.h> -#include <asm/timer.h> - -static struct cpufreq_driver *cpufreq_us3_driver; - -struct us3_freq_percpu_info { - struct cpufreq_frequency_table table[4]; -}; - -/* Indexed by cpu number. */ -static struct us3_freq_percpu_info *us3_freq_table; - -/* UltraSPARC-III has three dividers: 1, 2, and 32. These are controlled - * in the Safari config register. - */ -#define SAFARI_CFG_DIV_1 0x0000000000000000UL -#define SAFARI_CFG_DIV_2 0x0000000040000000UL -#define SAFARI_CFG_DIV_32 0x0000000080000000UL -#define SAFARI_CFG_DIV_MASK 0x00000000C0000000UL - -static unsigned long read_safari_cfg(void) -{ - unsigned long ret; - - __asm__ __volatile__("ldxa [%%g0] %1, %0" - : "=&r" (ret) - : "i" (ASI_SAFARI_CONFIG)); - return ret; -} - -static void write_safari_cfg(unsigned long val) -{ - __asm__ __volatile__("stxa %0, [%%g0] %1\n\t" - "membar #Sync" - : /* no outputs */ - : "r" (val), "i" (ASI_SAFARI_CONFIG) - : "memory"); -} - -static unsigned long get_current_freq(unsigned int cpu, unsigned long safari_cfg) -{ - unsigned long clock_tick = sparc64_get_clock_tick(cpu) / 1000; - unsigned long ret; - - switch (safari_cfg & SAFARI_CFG_DIV_MASK) { - case SAFARI_CFG_DIV_1: - ret = clock_tick / 1; - break; - case SAFARI_CFG_DIV_2: - ret = clock_tick / 2; - break; - case SAFARI_CFG_DIV_32: - ret = clock_tick / 32; - break; - default: - BUG(); - } - - return ret; -} - -static unsigned int us3_freq_get(unsigned int cpu) -{ - cpumask_t cpus_allowed; - unsigned long reg; - unsigned int ret; - - if (!cpu_online(cpu)) - return 0; - - cpumask_copy(&cpus_allowed, tsk_cpus_allowed(current)); - set_cpus_allowed_ptr(current, cpumask_of(cpu)); - - reg = read_safari_cfg(); - ret = get_current_freq(cpu, reg); - - set_cpus_allowed_ptr(current, &cpus_allowed); - - return ret; -} - -static void us3_set_cpu_divider_index(unsigned int cpu, unsigned int index) -{ - unsigned long new_bits, new_freq, reg; - cpumask_t cpus_allowed; - struct cpufreq_freqs freqs; - - if (!cpu_online(cpu)) - return; - - cpumask_copy(&cpus_allowed, tsk_cpus_allowed(current)); - set_cpus_allowed_ptr(current, cpumask_of(cpu)); - - new_freq = sparc64_get_clock_tick(cpu) / 1000; - switch (index) { - case 0: - new_bits = SAFARI_CFG_DIV_1; - new_freq /= 1; - break; - case 1: - new_bits = SAFARI_CFG_DIV_2; - new_freq /= 2; - break; - case 2: - new_bits = SAFARI_CFG_DIV_32; - new_freq /= 32; - break; - - default: - BUG(); - } - - reg = read_safari_cfg(); - - freqs.old = get_current_freq(cpu, reg); - freqs.new = new_freq; - freqs.cpu = cpu; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - reg &= ~SAFARI_CFG_DIV_MASK; - reg |= new_bits; - write_safari_cfg(reg); - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - - set_cpus_allowed_ptr(current, &cpus_allowed); -} - -static int us3_freq_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int new_index = 0; - - if (cpufreq_frequency_table_target(policy, - &us3_freq_table[policy->cpu].table[0], - target_freq, - relation, - &new_index)) - return -EINVAL; - - us3_set_cpu_divider_index(policy->cpu, new_index); - - return 0; -} - -static int us3_freq_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, - &us3_freq_table[policy->cpu].table[0]); -} - -static int __init us3_freq_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int cpu = policy->cpu; - unsigned long clock_tick = sparc64_get_clock_tick(cpu) / 1000; - struct cpufreq_frequency_table *table = - &us3_freq_table[cpu].table[0]; - - table[0].index = 0; - table[0].frequency = clock_tick / 1; - table[1].index = 1; - table[1].frequency = clock_tick / 2; - table[2].index = 2; - table[2].frequency = clock_tick / 32; - table[3].index = 0; - table[3].frequency = CPUFREQ_TABLE_END; - - policy->cpuinfo.transition_latency = 0; - policy->cur = clock_tick; - - return cpufreq_frequency_table_cpuinfo(policy, table); -} - -static int us3_freq_cpu_exit(struct cpufreq_policy *policy) -{ - if (cpufreq_us3_driver) - us3_set_cpu_divider_index(policy->cpu, 0); - - return 0; -} - -static int __init us3_freq_init(void) -{ - unsigned long manuf, impl, ver; - int ret; - - if (tlb_type != cheetah && tlb_type != cheetah_plus) - return -ENODEV; - - __asm__("rdpr %%ver, %0" : "=r" (ver)); - manuf = ((ver >> 48) & 0xffff); - impl = ((ver >> 32) & 0xffff); - - if (manuf == CHEETAH_MANUF && - (impl == CHEETAH_IMPL || - impl == CHEETAH_PLUS_IMPL || - impl == JAGUAR_IMPL || - impl == PANTHER_IMPL)) { - struct cpufreq_driver *driver; - - ret = -ENOMEM; - driver = kzalloc(sizeof(struct cpufreq_driver), GFP_KERNEL); - if (!driver) - goto err_out; - - us3_freq_table = kzalloc( - (NR_CPUS * sizeof(struct us3_freq_percpu_info)), - GFP_KERNEL); - if (!us3_freq_table) - goto err_out; - - driver->init = us3_freq_cpu_init; - driver->verify = us3_freq_verify; - driver->target = us3_freq_target; - driver->get = us3_freq_get; - driver->exit = us3_freq_cpu_exit; - driver->owner = THIS_MODULE, - strcpy(driver->name, "UltraSPARC-III"); - - cpufreq_us3_driver = driver; - ret = cpufreq_register_driver(driver); - if (ret) - goto err_out; - - return 0; - -err_out: - if (driver) { - kfree(driver); - cpufreq_us3_driver = NULL; - } - kfree(us3_freq_table); - us3_freq_table = NULL; - return ret; - } - - return -ENODEV; -} - -static void __exit us3_freq_exit(void) -{ - if (cpufreq_us3_driver) { - cpufreq_unregister_driver(cpufreq_us3_driver); - kfree(cpufreq_us3_driver); - cpufreq_us3_driver = NULL; - kfree(us3_freq_table); - us3_freq_table = NULL; - } -} - -MODULE_AUTHOR("David S. Miller <davem@redhat.com>"); -MODULE_DESCRIPTION("cpufreq driver for UltraSPARC-III"); -MODULE_LICENSE("GPL"); - -module_init(us3_freq_init); -module_exit(us3_freq_exit); diff --git a/arch/sparc/kernel/vio.c b/arch/sparc/kernel/vio.c index 3e244f31e56..8647fcc5ca6 100644 --- a/arch/sparc/kernel/vio.c +++ b/arch/sparc/kernel/vio.c @@ -342,6 +342,7 @@ static void vio_remove(struct mdesc_handle *hp, u64 node) printk(KERN_INFO "VIO: Removing device %s\n", dev_name(dev)); device_unregister(dev); + put_device(dev); } } diff --git a/arch/sparc/kernel/visemul.c b/arch/sparc/kernel/visemul.c index 08e074b7eb6..c096c624ac4 100644 --- a/arch/sparc/kernel/visemul.c +++ b/arch/sparc/kernel/visemul.c @@ -149,21 +149,24 @@ static inline void maybe_flush_windows(unsigned int rs1, unsigned int rs2, static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) { - unsigned long value; + unsigned long value, fp; if (reg < 16) return (!reg ? 0 : regs->u_regs[reg]); + + fp = regs->u_regs[UREG_FP]; + if (regs->tstate & TSTATE_PRIV) { struct reg_window *win; - win = (struct reg_window *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window *)(fp + STACK_BIAS); value = win->locals[reg - 16]; - } else if (test_thread_flag(TIF_32BIT)) { + } else if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); get_user(value, &win32->locals[reg - 16]); } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); get_user(value, &win->locals[reg - 16]); } return value; @@ -172,16 +175,18 @@ static unsigned long fetch_reg(unsigned int reg, struct pt_regs *regs) static inline unsigned long __user *__fetch_reg_addr_user(unsigned int reg, struct pt_regs *regs) { + unsigned long fp = regs->u_regs[UREG_FP]; + BUG_ON(reg < 16); BUG_ON(regs->tstate & TSTATE_PRIV); - if (test_thread_flag(TIF_32BIT)) { + if (!test_thread_64bit_stack(fp)) { struct reg_window32 __user *win32; - win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); + win32 = (struct reg_window32 __user *)((unsigned long)((u32)fp)); return (unsigned long __user *)&win32->locals[reg - 16]; } else { struct reg_window __user *win; - win = (struct reg_window __user *)(regs->u_regs[UREG_FP] + STACK_BIAS); + win = (struct reg_window __user *)(fp + STACK_BIAS); return &win->locals[reg - 16]; } } @@ -204,7 +209,7 @@ static void store_reg(struct pt_regs *regs, unsigned long val, unsigned long rd) } else { unsigned long __user *rd_user = __fetch_reg_addr_user(rd, regs); - if (test_thread_flag(TIF_32BIT)) + if (!test_thread_64bit_stack(regs->u_regs[UREG_FP])) __put_user((u32)val, (u32 __user *)rd_user); else __put_user(val, rd_user); diff --git a/arch/sparc/kernel/vmlinux.lds.S b/arch/sparc/kernel/vmlinux.lds.S index 89c2c29f154..932ff90fd76 100644 --- a/arch/sparc/kernel/vmlinux.lds.S +++ b/arch/sparc/kernel/vmlinux.lds.S @@ -122,6 +122,11 @@ SECTIONS *(.swapper_4m_tsb_phys_patch) __swapper_4m_tsb_phys_patch_end = .; } + .page_offset_shift_patch : { + __page_offset_shift_patch = .; + *(.page_offset_shift_patch) + __page_offset_shift_patch_end = .; + } .popc_3insn_patch : { __popc_3insn_patch = .; *(.popc_3insn_patch) @@ -132,6 +137,11 @@ SECTIONS *(.popc_6insn_patch) __popc_6insn_patch_end = .; } + .pause_3insn_patch : { + __pause_3insn_patch = .; + *(.pause_3insn_patch) + __pause_3insn_patch_end = .; + } PERCPU_SECTION(SMP_CACHE_BYTES) . = ALIGN(PAGE_SIZE); diff --git a/arch/sparc/kernel/windows.c b/arch/sparc/kernel/windows.c index 3107381e576..87bab0a3857 100644 --- a/arch/sparc/kernel/windows.c +++ b/arch/sparc/kernel/windows.c @@ -10,8 +10,11 @@ #include <linux/mm.h> #include <linux/smp.h> +#include <asm/cacheflush.h> #include <asm/uaccess.h> +#include "kernel.h" + /* Do save's until all user register windows are out of the cpu. */ void flush_user_windows(void) { diff --git a/arch/sparc/kernel/winfixup.S b/arch/sparc/kernel/winfixup.S index a6b0863c27d..1e67ce95836 100644 --- a/arch/sparc/kernel/winfixup.S +++ b/arch/sparc/kernel/winfixup.S @@ -43,6 +43,8 @@ spill_fixup_mna: spill_fixup_dax: TRAP_LOAD_THREAD_REG(%g6, %g1) ldx [%g6 + TI_FLAGS], %g1 + andcc %sp, 0x1, %g0 + movne %icc, 0, %g1 andcc %g1, _TIF_32BIT, %g0 ldub [%g6 + TI_WSAVED], %g1 sll %g1, 3, %g3 diff --git a/arch/sparc/lib/Makefile b/arch/sparc/lib/Makefile index dff4096f3de..3269b023409 100644 --- a/arch/sparc/lib/Makefile +++ b/arch/sparc/lib/Makefile @@ -32,14 +32,16 @@ lib-$(CONFIG_SPARC64) += NGpatch.o NGpage.o NGbzero.o lib-$(CONFIG_SPARC64) += NG2memcpy.o NG2copy_from_user.o NG2copy_to_user.o lib-$(CONFIG_SPARC64) += NG2patch.o +lib-$(CONFIG_SPARC64) += NG4memcpy.o NG4copy_from_user.o NG4copy_to_user.o +lib-$(CONFIG_SPARC64) += NG4patch.o NG4copy_page.o NG4clear_page.o NG4memset.o + lib-$(CONFIG_SPARC64) += GENmemcpy.o GENcopy_from_user.o GENcopy_to_user.o lib-$(CONFIG_SPARC64) += GENpatch.o GENpage.o GENbzero.o lib-$(CONFIG_SPARC64) += copy_in_user.o user_fixup.o memmove.o lib-$(CONFIG_SPARC64) += mcount.o ipcsum.o xor.o hweight.o ffs.o -obj-y += iomap.o +obj-$(CONFIG_SPARC64) += iomap.o obj-$(CONFIG_SPARC32) += atomic32.o ucmpdi2.o obj-y += ksyms.o obj-$(CONFIG_SPARC64) += PeeCeeI.o -obj-y += usercopy.o diff --git a/arch/sparc/lib/NG2memcpy.S b/arch/sparc/lib/NG2memcpy.S index 0aed75653b5..30eee6e8a81 100644 --- a/arch/sparc/lib/NG2memcpy.S +++ b/arch/sparc/lib/NG2memcpy.S @@ -14,7 +14,7 @@ #define FPRS_FEF 0x04 #ifdef MEMCPY_DEBUG #define VISEntryHalf rd %fprs, %o5; wr %g0, FPRS_FEF, %fprs; \ - clr %g1; clr %g2; clr %g3; subcc %g0, %g0, %g0; + clr %g1; clr %g2; clr %g3; clr %g5; subcc %g0, %g0, %g0; #define VISExitHalf and %o5, FPRS_FEF, %o5; wr %o5, 0x0, %fprs #else #define VISEntryHalf rd %fprs, %o5; wr %g0, FPRS_FEF, %fprs @@ -90,49 +90,49 @@ faligndata %x7, %x8, %f14; #define FREG_MOVE_1(x0) \ - fmovd %x0, %f0; + fsrc2 %x0, %f0; #define FREG_MOVE_2(x0, x1) \ - fmovd %x0, %f0; \ - fmovd %x1, %f2; + fsrc2 %x0, %f0; \ + fsrc2 %x1, %f2; #define FREG_MOVE_3(x0, x1, x2) \ - fmovd %x0, %f0; \ - fmovd %x1, %f2; \ - fmovd %x2, %f4; + fsrc2 %x0, %f0; \ + fsrc2 %x1, %f2; \ + fsrc2 %x2, %f4; #define FREG_MOVE_4(x0, x1, x2, x3) \ - fmovd %x0, %f0; \ - fmovd %x1, %f2; \ - fmovd %x2, %f4; \ - fmovd %x3, %f6; + fsrc2 %x0, %f0; \ + fsrc2 %x1, %f2; \ + fsrc2 %x2, %f4; \ + fsrc2 %x3, %f6; #define FREG_MOVE_5(x0, x1, x2, x3, x4) \ - fmovd %x0, %f0; \ - fmovd %x1, %f2; \ - fmovd %x2, %f4; \ - fmovd %x3, %f6; \ - fmovd %x4, %f8; + fsrc2 %x0, %f0; \ + fsrc2 %x1, %f2; \ + fsrc2 %x2, %f4; \ + fsrc2 %x3, %f6; \ + fsrc2 %x4, %f8; #define FREG_MOVE_6(x0, x1, x2, x3, x4, x5) \ - fmovd %x0, %f0; \ - fmovd %x1, %f2; \ - fmovd %x2, %f4; \ - fmovd %x3, %f6; \ - fmovd %x4, %f8; \ - fmovd %x5, %f10; + fsrc2 %x0, %f0; \ + fsrc2 %x1, %f2; \ + fsrc2 %x2, %f4; \ + fsrc2 %x3, %f6; \ + fsrc2 %x4, %f8; \ + fsrc2 %x5, %f10; #define FREG_MOVE_7(x0, x1, x2, x3, x4, x5, x6) \ - fmovd %x0, %f0; \ - fmovd %x1, %f2; \ - fmovd %x2, %f4; \ - fmovd %x3, %f6; \ - fmovd %x4, %f8; \ - fmovd %x5, %f10; \ - fmovd %x6, %f12; + fsrc2 %x0, %f0; \ + fsrc2 %x1, %f2; \ + fsrc2 %x2, %f4; \ + fsrc2 %x3, %f6; \ + fsrc2 %x4, %f8; \ + fsrc2 %x5, %f10; \ + fsrc2 %x6, %f12; #define FREG_MOVE_8(x0, x1, x2, x3, x4, x5, x6, x7) \ - fmovd %x0, %f0; \ - fmovd %x1, %f2; \ - fmovd %x2, %f4; \ - fmovd %x3, %f6; \ - fmovd %x4, %f8; \ - fmovd %x5, %f10; \ - fmovd %x6, %f12; \ - fmovd %x7, %f14; + fsrc2 %x0, %f0; \ + fsrc2 %x1, %f2; \ + fsrc2 %x2, %f4; \ + fsrc2 %x3, %f6; \ + fsrc2 %x4, %f8; \ + fsrc2 %x5, %f10; \ + fsrc2 %x6, %f12; \ + fsrc2 %x7, %f14; #define FREG_LOAD_1(base, x0) \ EX_LD(LOAD(ldd, base + 0x00, %x0)) #define FREG_LOAD_2(base, x0, x1) \ @@ -182,13 +182,13 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ cmp %g2, 0 tne %xcc, 5 PREAMBLE - mov %o0, GLOBAL_SPARE + mov %o0, %o3 cmp %o2, 0 be,pn %XCC, 85f - or %o0, %o1, %o3 + or %o0, %o1, GLOBAL_SPARE cmp %o2, 16 blu,a,pn %XCC, 80f - or %o3, %o2, %o3 + or GLOBAL_SPARE, %o2, GLOBAL_SPARE /* 2 blocks (128 bytes) is the minimum we can do the block * copy with. We need to ensure that we'll iterate at least @@ -202,7 +202,7 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ */ cmp %o2, (4 * 64) blu,pt %XCC, 75f - andcc %o3, 0x7, %g0 + andcc GLOBAL_SPARE, 0x7, %g0 /* %o0: dst * %o1: src @@ -236,6 +236,7 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ */ VISEntryHalf + membar #Sync alignaddr %o1, %g0, %g0 add %o1, (64 - 1), %o4 @@ -404,13 +405,13 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ * over. If anything is left, we copy it one byte at a time. */ brz,pt %o2, 85f - sub %o0, %o1, %o3 + sub %o0, %o1, GLOBAL_SPARE ba,a,pt %XCC, 90f .align 64 75: /* 16 < len <= 64 */ bne,pn %XCC, 75f - sub %o0, %o1, %o3 + sub %o0, %o1, GLOBAL_SPARE 72: andn %o2, 0xf, %o4 @@ -420,9 +421,9 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ add %o1, 0x08, %o1 EX_LD(LOAD(ldx, %o1, %g1)) sub %o1, 0x08, %o1 - EX_ST(STORE(stx, %o5, %o1 + %o3)) + EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE)) add %o1, 0x8, %o1 - EX_ST(STORE(stx, %g1, %o1 + %o3)) + EX_ST(STORE(stx, %g1, %o1 + GLOBAL_SPARE)) bgu,pt %XCC, 1b add %o1, 0x8, %o1 73: andcc %o2, 0x8, %g0 @@ -430,14 +431,14 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ nop sub %o2, 0x8, %o2 EX_LD(LOAD(ldx, %o1, %o5)) - EX_ST(STORE(stx, %o5, %o1 + %o3)) + EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE)) add %o1, 0x8, %o1 1: andcc %o2, 0x4, %g0 be,pt %XCC, 1f nop sub %o2, 0x4, %o2 EX_LD(LOAD(lduw, %o1, %o5)) - EX_ST(STORE(stw, %o5, %o1 + %o3)) + EX_ST(STORE(stw, %o5, %o1 + GLOBAL_SPARE)) add %o1, 0x4, %o1 1: cmp %o2, 0 be,pt %XCC, 85f @@ -454,11 +455,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ 1: subcc %g1, 1, %g1 EX_LD(LOAD(ldub, %o1, %o5)) - EX_ST(STORE(stb, %o5, %o1 + %o3)) + EX_ST(STORE(stb, %o5, %o1 + GLOBAL_SPARE)) bgu,pt %icc, 1b add %o1, 1, %o1 -2: add %o1, %o3, %o0 +2: add %o1, GLOBAL_SPARE, %o0 andcc %o1, 0x7, %g1 bne,pt %icc, 8f sll %g1, 3, %g1 @@ -468,16 +469,16 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ nop ba,a,pt %xcc, 73b -8: mov 64, %o3 +8: mov 64, GLOBAL_SPARE andn %o1, 0x7, %o1 EX_LD(LOAD(ldx, %o1, %g2)) - sub %o3, %g1, %o3 + sub GLOBAL_SPARE, %g1, GLOBAL_SPARE andn %o2, 0x7, %o4 sllx %g2, %g1, %g2 1: add %o1, 0x8, %o1 EX_LD(LOAD(ldx, %o1, %g3)) subcc %o4, 0x8, %o4 - srlx %g3, %o3, %o5 + srlx %g3, GLOBAL_SPARE, %o5 or %o5, %g2, %o5 EX_ST(STORE(stx, %o5, %o0)) add %o0, 0x8, %o0 @@ -489,32 +490,32 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ be,pn %icc, 85f add %o1, %g1, %o1 ba,pt %xcc, 90f - sub %o0, %o1, %o3 + sub %o0, %o1, GLOBAL_SPARE .align 64 80: /* 0 < len <= 16 */ - andcc %o3, 0x3, %g0 + andcc GLOBAL_SPARE, 0x3, %g0 bne,pn %XCC, 90f - sub %o0, %o1, %o3 + sub %o0, %o1, GLOBAL_SPARE 1: subcc %o2, 4, %o2 EX_LD(LOAD(lduw, %o1, %g1)) - EX_ST(STORE(stw, %g1, %o1 + %o3)) + EX_ST(STORE(stw, %g1, %o1 + GLOBAL_SPARE)) bgu,pt %XCC, 1b add %o1, 4, %o1 85: retl - mov EX_RETVAL(GLOBAL_SPARE), %o0 + mov EX_RETVAL(%o3), %o0 .align 32 90: subcc %o2, 1, %o2 EX_LD(LOAD(ldub, %o1, %g1)) - EX_ST(STORE(stb, %g1, %o1 + %o3)) + EX_ST(STORE(stb, %g1, %o1 + GLOBAL_SPARE)) bgu,pt %XCC, 90b add %o1, 1, %o1 retl - mov EX_RETVAL(GLOBAL_SPARE), %o0 + mov EX_RETVAL(%o3), %o0 .size FUNC_NAME, .-FUNC_NAME diff --git a/arch/sparc/lib/NG4clear_page.S b/arch/sparc/lib/NG4clear_page.S new file mode 100644 index 00000000000..e16c88204a4 --- /dev/null +++ b/arch/sparc/lib/NG4clear_page.S @@ -0,0 +1,29 @@ +/* NG4copy_page.S: Niagara-4 optimized clear page. + * + * Copyright (C) 2012 (davem@davemloft.net) + */ + +#include <asm/asi.h> +#include <asm/page.h> + + .text + + .register %g3, #scratch + + .align 32 + .globl NG4clear_page + .globl NG4clear_user_page +NG4clear_page: /* %o0=dest */ +NG4clear_user_page: /* %o0=dest, %o1=vaddr */ + set PAGE_SIZE, %g7 + mov 0x20, %g3 +1: stxa %g0, [%o0 + %g0] ASI_ST_BLKINIT_MRU_P + subcc %g7, 0x40, %g7 + stxa %g0, [%o0 + %g3] ASI_ST_BLKINIT_MRU_P + bne,pt %xcc, 1b + add %o0, 0x40, %o0 + membar #StoreLoad|#StoreStore + retl + nop + .size NG4clear_page,.-NG4clear_page + .size NG4clear_user_page,.-NG4clear_user_page
\ No newline at end of file diff --git a/arch/sparc/lib/NG4copy_from_user.S b/arch/sparc/lib/NG4copy_from_user.S new file mode 100644 index 00000000000..fd9f903ffa3 --- /dev/null +++ b/arch/sparc/lib/NG4copy_from_user.S @@ -0,0 +1,30 @@ +/* NG4copy_from_user.S: Niagara-4 optimized copy from userspace. + * + * Copyright (C) 2012 David S. Miller (davem@davemloft.net) + */ + +#define EX_LD(x) \ +98: x; \ + .section __ex_table,"a";\ + .align 4; \ + .word 98b, __retl_one_asi;\ + .text; \ + .align 4; + +#ifndef ASI_AIUS +#define ASI_AIUS 0x11 +#endif + +#define FUNC_NAME NG4copy_from_user +#define LOAD(type,addr,dest) type##a [addr] %asi, dest +#define EX_RETVAL(x) 0 + +#ifdef __KERNEL__ +#define PREAMBLE \ + rd %asi, %g1; \ + cmp %g1, ASI_AIUS; \ + bne,pn %icc, ___copy_in_user; \ + nop +#endif + +#include "NG4memcpy.S" diff --git a/arch/sparc/lib/NG4copy_page.S b/arch/sparc/lib/NG4copy_page.S new file mode 100644 index 00000000000..28504e88c53 --- /dev/null +++ b/arch/sparc/lib/NG4copy_page.S @@ -0,0 +1,57 @@ +/* NG4copy_page.S: Niagara-4 optimized copy page. + * + * Copyright (C) 2012 (davem@davemloft.net) + */ + +#include <asm/asi.h> +#include <asm/page.h> + + .text + .align 32 + + .register %g2, #scratch + .register %g3, #scratch + + .globl NG4copy_user_page +NG4copy_user_page: /* %o0=dest, %o1=src, %o2=vaddr */ + prefetch [%o1 + 0x000], #n_reads_strong + prefetch [%o1 + 0x040], #n_reads_strong + prefetch [%o1 + 0x080], #n_reads_strong + prefetch [%o1 + 0x0c0], #n_reads_strong + set PAGE_SIZE, %g7 + prefetch [%o1 + 0x100], #n_reads_strong + prefetch [%o1 + 0x140], #n_reads_strong + prefetch [%o1 + 0x180], #n_reads_strong + prefetch [%o1 + 0x1c0], #n_reads_strong +1: + ldx [%o1 + 0x00], %o2 + subcc %g7, 0x40, %g7 + ldx [%o1 + 0x08], %o3 + ldx [%o1 + 0x10], %o4 + ldx [%o1 + 0x18], %o5 + ldx [%o1 + 0x20], %g1 + stxa %o2, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + ldx [%o1 + 0x28], %g2 + stxa %o3, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + ldx [%o1 + 0x30], %g3 + stxa %o4, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + ldx [%o1 + 0x38], %o2 + add %o1, 0x40, %o1 + stxa %o5, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + stxa %g1, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + stxa %g2, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + stxa %g3, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + stxa %o2, [%o0] ASI_ST_BLKINIT_MRU_P + add %o0, 0x08, %o0 + bne,pt %icc, 1b + prefetch [%o1 + 0x200], #n_reads_strong + retl + membar #StoreLoad | #StoreStore + .size NG4copy_user_page,.-NG4copy_user_page diff --git a/arch/sparc/lib/NG4copy_to_user.S b/arch/sparc/lib/NG4copy_to_user.S new file mode 100644 index 00000000000..9744c4540a8 --- /dev/null +++ b/arch/sparc/lib/NG4copy_to_user.S @@ -0,0 +1,39 @@ +/* NG4copy_to_user.S: Niagara-4 optimized copy to userspace. + * + * Copyright (C) 2012 David S. Miller (davem@davemloft.net) + */ + +#define EX_ST(x) \ +98: x; \ + .section __ex_table,"a";\ + .align 4; \ + .word 98b, __retl_one_asi;\ + .text; \ + .align 4; + +#ifndef ASI_AIUS +#define ASI_AIUS 0x11 +#endif + +#ifndef ASI_BLK_INIT_QUAD_LDD_AIUS +#define ASI_BLK_INIT_QUAD_LDD_AIUS 0x23 +#endif + +#define FUNC_NAME NG4copy_to_user +#define STORE(type,src,addr) type##a src, [addr] %asi +#define STORE_ASI ASI_BLK_INIT_QUAD_LDD_AIUS +#define EX_RETVAL(x) 0 + +#ifdef __KERNEL__ + /* Writing to %asi is _expensive_ so we hardcode it. + * Reading %asi to check for KERNEL_DS is comparatively + * cheap. + */ +#define PREAMBLE \ + rd %asi, %g1; \ + cmp %g1, ASI_AIUS; \ + bne,pn %icc, ___copy_in_user; \ + nop +#endif + +#include "NG4memcpy.S" diff --git a/arch/sparc/lib/NG4memcpy.S b/arch/sparc/lib/NG4memcpy.S new file mode 100644 index 00000000000..9cf2ee01cee --- /dev/null +++ b/arch/sparc/lib/NG4memcpy.S @@ -0,0 +1,360 @@ +/* NG4memcpy.S: Niagara-4 optimized memcpy. + * + * Copyright (C) 2012 David S. Miller (davem@davemloft.net) + */ + +#ifdef __KERNEL__ +#include <asm/visasm.h> +#include <asm/asi.h> +#define GLOBAL_SPARE %g7 +#else +#define ASI_BLK_INIT_QUAD_LDD_P 0xe2 +#define FPRS_FEF 0x04 + +/* On T4 it is very expensive to access ASRs like %fprs and + * %asi, avoiding a read or a write can save ~50 cycles. + */ +#define FPU_ENTER \ + rd %fprs, %o5; \ + andcc %o5, FPRS_FEF, %g0; \ + be,a,pn %icc, 999f; \ + wr %g0, FPRS_FEF, %fprs; \ + 999: + +#ifdef MEMCPY_DEBUG +#define VISEntryHalf FPU_ENTER; \ + clr %g1; clr %g2; clr %g3; clr %g5; subcc %g0, %g0, %g0; +#define VISExitHalf and %o5, FPRS_FEF, %o5; wr %o5, 0x0, %fprs +#else +#define VISEntryHalf FPU_ENTER +#define VISExitHalf and %o5, FPRS_FEF, %o5; wr %o5, 0x0, %fprs +#endif + +#define GLOBAL_SPARE %g5 +#endif + +#ifndef STORE_ASI +#ifndef SIMULATE_NIAGARA_ON_NON_NIAGARA +#define STORE_ASI ASI_BLK_INIT_QUAD_LDD_P +#else +#define STORE_ASI 0x80 /* ASI_P */ +#endif +#endif + +#ifndef EX_LD +#define EX_LD(x) x +#endif + +#ifndef EX_ST +#define EX_ST(x) x +#endif + +#ifndef EX_RETVAL +#define EX_RETVAL(x) x +#endif + +#ifndef LOAD +#define LOAD(type,addr,dest) type [addr], dest +#endif + +#ifndef STORE +#ifndef MEMCPY_DEBUG +#define STORE(type,src,addr) type src, [addr] +#else +#define STORE(type,src,addr) type##a src, [addr] %asi +#endif +#endif + +#ifndef STORE_INIT +#define STORE_INIT(src,addr) stxa src, [addr] STORE_ASI +#endif + +#ifndef FUNC_NAME +#define FUNC_NAME NG4memcpy +#endif +#ifndef PREAMBLE +#define PREAMBLE +#endif + +#ifndef XCC +#define XCC xcc +#endif + + .register %g2,#scratch + .register %g3,#scratch + + .text + .align 64 + + .globl FUNC_NAME + .type FUNC_NAME,#function +FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ +#ifdef MEMCPY_DEBUG + wr %g0, 0x80, %asi +#endif + srlx %o2, 31, %g2 + cmp %g2, 0 + tne %XCC, 5 + PREAMBLE + mov %o0, %o3 + brz,pn %o2, .Lexit + cmp %o2, 3 + ble,pn %icc, .Ltiny + cmp %o2, 19 + ble,pn %icc, .Lsmall + or %o0, %o1, %g2 + cmp %o2, 128 + bl,pn %icc, .Lmedium + nop + +.Llarge:/* len >= 0x80 */ + /* First get dest 8 byte aligned. */ + sub %g0, %o0, %g1 + and %g1, 0x7, %g1 + brz,pt %g1, 51f + sub %o2, %g1, %o2 + +1: EX_LD(LOAD(ldub, %o1 + 0x00, %g2)) + add %o1, 1, %o1 + subcc %g1, 1, %g1 + add %o0, 1, %o0 + bne,pt %icc, 1b + EX_ST(STORE(stb, %g2, %o0 - 0x01)) + +51: LOAD(prefetch, %o1 + 0x040, #n_reads_strong) + LOAD(prefetch, %o1 + 0x080, #n_reads_strong) + LOAD(prefetch, %o1 + 0x0c0, #n_reads_strong) + LOAD(prefetch, %o1 + 0x100, #n_reads_strong) + LOAD(prefetch, %o1 + 0x140, #n_reads_strong) + LOAD(prefetch, %o1 + 0x180, #n_reads_strong) + LOAD(prefetch, %o1 + 0x1c0, #n_reads_strong) + LOAD(prefetch, %o1 + 0x200, #n_reads_strong) + + /* Check if we can use the straight fully aligned + * loop, or we require the alignaddr/faligndata variant. + */ + andcc %o1, 0x7, %o5 + bne,pn %icc, .Llarge_src_unaligned + sub %g0, %o0, %g1 + + /* Legitimize the use of initializing stores by getting dest + * to be 64-byte aligned. + */ + and %g1, 0x3f, %g1 + brz,pt %g1, .Llarge_aligned + sub %o2, %g1, %o2 + +1: EX_LD(LOAD(ldx, %o1 + 0x00, %g2)) + add %o1, 8, %o1 + subcc %g1, 8, %g1 + add %o0, 8, %o0 + bne,pt %icc, 1b + EX_ST(STORE(stx, %g2, %o0 - 0x08)) + +.Llarge_aligned: + /* len >= 0x80 && src 8-byte aligned && dest 8-byte aligned */ + andn %o2, 0x3f, %o4 + sub %o2, %o4, %o2 + +1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1)) + add %o1, 0x40, %o1 + EX_LD(LOAD(ldx, %o1 - 0x38, %g2)) + subcc %o4, 0x40, %o4 + EX_LD(LOAD(ldx, %o1 - 0x30, %g3)) + EX_LD(LOAD(ldx, %o1 - 0x28, GLOBAL_SPARE)) + EX_LD(LOAD(ldx, %o1 - 0x20, %o5)) + EX_ST(STORE_INIT(%g1, %o0)) + add %o0, 0x08, %o0 + EX_ST(STORE_INIT(%g2, %o0)) + add %o0, 0x08, %o0 + EX_LD(LOAD(ldx, %o1 - 0x18, %g2)) + EX_ST(STORE_INIT(%g3, %o0)) + add %o0, 0x08, %o0 + EX_LD(LOAD(ldx, %o1 - 0x10, %g3)) + EX_ST(STORE_INIT(GLOBAL_SPARE, %o0)) + add %o0, 0x08, %o0 + EX_LD(LOAD(ldx, %o1 - 0x08, GLOBAL_SPARE)) + EX_ST(STORE_INIT(%o5, %o0)) + add %o0, 0x08, %o0 + EX_ST(STORE_INIT(%g2, %o0)) + add %o0, 0x08, %o0 + EX_ST(STORE_INIT(%g3, %o0)) + add %o0, 0x08, %o0 + EX_ST(STORE_INIT(GLOBAL_SPARE, %o0)) + add %o0, 0x08, %o0 + bne,pt %icc, 1b + LOAD(prefetch, %o1 + 0x200, #n_reads_strong) + + membar #StoreLoad | #StoreStore + + brz,pn %o2, .Lexit + cmp %o2, 19 + ble,pn %icc, .Lsmall_unaligned + nop + ba,a,pt %icc, .Lmedium_noprefetch + +.Lexit: retl + mov EX_RETVAL(%o3), %o0 + +.Llarge_src_unaligned: + andn %o2, 0x3f, %o4 + sub %o2, %o4, %o2 + VISEntryHalf + alignaddr %o1, %g0, %g1 + add %o1, %o4, %o1 + EX_LD(LOAD(ldd, %g1 + 0x00, %f0)) +1: EX_LD(LOAD(ldd, %g1 + 0x08, %f2)) + subcc %o4, 0x40, %o4 + EX_LD(LOAD(ldd, %g1 + 0x10, %f4)) + EX_LD(LOAD(ldd, %g1 + 0x18, %f6)) + EX_LD(LOAD(ldd, %g1 + 0x20, %f8)) + EX_LD(LOAD(ldd, %g1 + 0x28, %f10)) + EX_LD(LOAD(ldd, %g1 + 0x30, %f12)) + EX_LD(LOAD(ldd, %g1 + 0x38, %f14)) + faligndata %f0, %f2, %f16 + EX_LD(LOAD(ldd, %g1 + 0x40, %f0)) + faligndata %f2, %f4, %f18 + add %g1, 0x40, %g1 + faligndata %f4, %f6, %f20 + faligndata %f6, %f8, %f22 + faligndata %f8, %f10, %f24 + faligndata %f10, %f12, %f26 + faligndata %f12, %f14, %f28 + faligndata %f14, %f0, %f30 + EX_ST(STORE(std, %f16, %o0 + 0x00)) + EX_ST(STORE(std, %f18, %o0 + 0x08)) + EX_ST(STORE(std, %f20, %o0 + 0x10)) + EX_ST(STORE(std, %f22, %o0 + 0x18)) + EX_ST(STORE(std, %f24, %o0 + 0x20)) + EX_ST(STORE(std, %f26, %o0 + 0x28)) + EX_ST(STORE(std, %f28, %o0 + 0x30)) + EX_ST(STORE(std, %f30, %o0 + 0x38)) + add %o0, 0x40, %o0 + bne,pt %icc, 1b + LOAD(prefetch, %g1 + 0x200, #n_reads_strong) + VISExitHalf + + brz,pn %o2, .Lexit + cmp %o2, 19 + ble,pn %icc, .Lsmall_unaligned + nop + ba,a,pt %icc, .Lmedium_unaligned + +.Lmedium: + LOAD(prefetch, %o1 + 0x40, #n_reads_strong) + andcc %g2, 0x7, %g0 + bne,pn %icc, .Lmedium_unaligned + nop +.Lmedium_noprefetch: + andncc %o2, 0x20 - 1, %o5 + be,pn %icc, 2f + sub %o2, %o5, %o2 +1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1)) + EX_LD(LOAD(ldx, %o1 + 0x08, %g2)) + EX_LD(LOAD(ldx, %o1 + 0x10, GLOBAL_SPARE)) + EX_LD(LOAD(ldx, %o1 + 0x18, %o4)) + add %o1, 0x20, %o1 + subcc %o5, 0x20, %o5 + EX_ST(STORE(stx, %g1, %o0 + 0x00)) + EX_ST(STORE(stx, %g2, %o0 + 0x08)) + EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x10)) + EX_ST(STORE(stx, %o4, %o0 + 0x18)) + bne,pt %icc, 1b + add %o0, 0x20, %o0 +2: andcc %o2, 0x18, %o5 + be,pt %icc, 3f + sub %o2, %o5, %o2 +1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1)) + add %o1, 0x08, %o1 + add %o0, 0x08, %o0 + subcc %o5, 0x08, %o5 + bne,pt %icc, 1b + EX_ST(STORE(stx, %g1, %o0 - 0x08)) +3: brz,pt %o2, .Lexit + cmp %o2, 0x04 + bl,pn %icc, .Ltiny + nop + EX_LD(LOAD(lduw, %o1 + 0x00, %g1)) + add %o1, 0x04, %o1 + add %o0, 0x04, %o0 + subcc %o2, 0x04, %o2 + bne,pn %icc, .Ltiny + EX_ST(STORE(stw, %g1, %o0 - 0x04)) + ba,a,pt %icc, .Lexit +.Lmedium_unaligned: + /* First get dest 8 byte aligned. */ + sub %g0, %o0, %g1 + and %g1, 0x7, %g1 + brz,pt %g1, 2f + sub %o2, %g1, %o2 + +1: EX_LD(LOAD(ldub, %o1 + 0x00, %g2)) + add %o1, 1, %o1 + subcc %g1, 1, %g1 + add %o0, 1, %o0 + bne,pt %icc, 1b + EX_ST(STORE(stb, %g2, %o0 - 0x01)) +2: + and %o1, 0x7, %g1 + brz,pn %g1, .Lmedium_noprefetch + sll %g1, 3, %g1 + mov 64, %g2 + sub %g2, %g1, %g2 + andn %o1, 0x7, %o1 + EX_LD(LOAD(ldx, %o1 + 0x00, %o4)) + sllx %o4, %g1, %o4 + andn %o2, 0x08 - 1, %o5 + sub %o2, %o5, %o2 +1: EX_LD(LOAD(ldx, %o1 + 0x08, %g3)) + add %o1, 0x08, %o1 + subcc %o5, 0x08, %o5 + srlx %g3, %g2, GLOBAL_SPARE + or GLOBAL_SPARE, %o4, GLOBAL_SPARE + EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x00)) + add %o0, 0x08, %o0 + bne,pt %icc, 1b + sllx %g3, %g1, %o4 + srl %g1, 3, %g1 + add %o1, %g1, %o1 + brz,pn %o2, .Lexit + nop + ba,pt %icc, .Lsmall_unaligned + +.Ltiny: + EX_LD(LOAD(ldub, %o1 + 0x00, %g1)) + subcc %o2, 1, %o2 + be,pn %icc, .Lexit + EX_ST(STORE(stb, %g1, %o0 + 0x00)) + EX_LD(LOAD(ldub, %o1 + 0x01, %g1)) + subcc %o2, 1, %o2 + be,pn %icc, .Lexit + EX_ST(STORE(stb, %g1, %o0 + 0x01)) + EX_LD(LOAD(ldub, %o1 + 0x02, %g1)) + ba,pt %icc, .Lexit + EX_ST(STORE(stb, %g1, %o0 + 0x02)) + +.Lsmall: + andcc %g2, 0x3, %g0 + bne,pn %icc, .Lsmall_unaligned + andn %o2, 0x4 - 1, %o5 + sub %o2, %o5, %o2 +1: + EX_LD(LOAD(lduw, %o1 + 0x00, %g1)) + add %o1, 0x04, %o1 + subcc %o5, 0x04, %o5 + add %o0, 0x04, %o0 + bne,pt %icc, 1b + EX_ST(STORE(stw, %g1, %o0 - 0x04)) + brz,pt %o2, .Lexit + nop + ba,a,pt %icc, .Ltiny + +.Lsmall_unaligned: +1: EX_LD(LOAD(ldub, %o1 + 0x00, %g1)) + add %o1, 1, %o1 + add %o0, 1, %o0 + subcc %o2, 1, %o2 + bne,pt %icc, 1b + EX_ST(STORE(stb, %g1, %o0 - 0x01)) + ba,a,pt %icc, .Lexit + .size FUNC_NAME, .-FUNC_NAME diff --git a/arch/sparc/lib/NG4memset.S b/arch/sparc/lib/NG4memset.S new file mode 100644 index 00000000000..41da4bdd95c --- /dev/null +++ b/arch/sparc/lib/NG4memset.S @@ -0,0 +1,105 @@ +/* NG4memset.S: Niagara-4 optimized memset/bzero. + * + * Copyright (C) 2012 David S. Miller (davem@davemloft.net) + */ + +#include <asm/asi.h> + + .register %g2, #scratch + .register %g3, #scratch + + .text + .align 32 + .globl NG4memset +NG4memset: + andcc %o1, 0xff, %o4 + be,pt %icc, 1f + mov %o2, %o1 + sllx %o4, 8, %g1 + or %g1, %o4, %o2 + sllx %o2, 16, %g1 + or %g1, %o2, %o2 + sllx %o2, 32, %g1 + ba,pt %icc, 1f + or %g1, %o2, %o4 + .size NG4memset,.-NG4memset + + .align 32 + .globl NG4bzero +NG4bzero: + clr %o4 +1: cmp %o1, 16 + ble %icc, .Ltiny + mov %o0, %o3 + sub %g0, %o0, %g1 + and %g1, 0x7, %g1 + brz,pt %g1, .Laligned8 + sub %o1, %g1, %o1 +1: stb %o4, [%o0 + 0x00] + subcc %g1, 1, %g1 + bne,pt %icc, 1b + add %o0, 1, %o0 +.Laligned8: + cmp %o1, 64 + (64 - 8) + ble .Lmedium + sub %g0, %o0, %g1 + andcc %g1, (64 - 1), %g1 + brz,pn %g1, .Laligned64 + sub %o1, %g1, %o1 +1: stx %o4, [%o0 + 0x00] + subcc %g1, 8, %g1 + bne,pt %icc, 1b + add %o0, 0x8, %o0 +.Laligned64: + andn %o1, 64 - 1, %g1 + sub %o1, %g1, %o1 + brnz,pn %o4, .Lnon_bzero_loop + mov 0x20, %g2 +1: stxa %o4, [%o0 + %g0] ASI_BLK_INIT_QUAD_LDD_P + subcc %g1, 0x40, %g1 + stxa %o4, [%o0 + %g2] ASI_BLK_INIT_QUAD_LDD_P + bne,pt %icc, 1b + add %o0, 0x40, %o0 +.Lpostloop: + cmp %o1, 8 + bl,pn %icc, .Ltiny + membar #StoreStore|#StoreLoad +.Lmedium: + andn %o1, 0x7, %g1 + sub %o1, %g1, %o1 +1: stx %o4, [%o0 + 0x00] + subcc %g1, 0x8, %g1 + bne,pt %icc, 1b + add %o0, 0x08, %o0 + andcc %o1, 0x4, %g1 + be,pt %icc, .Ltiny + sub %o1, %g1, %o1 + stw %o4, [%o0 + 0x00] + add %o0, 0x4, %o0 +.Ltiny: + cmp %o1, 0 + be,pn %icc, .Lexit +1: subcc %o1, 1, %o1 + stb %o4, [%o0 + 0x00] + bne,pt %icc, 1b + add %o0, 1, %o0 +.Lexit: + retl + mov %o3, %o0 +.Lnon_bzero_loop: + mov 0x08, %g3 + mov 0x28, %o5 +1: stxa %o4, [%o0 + %g0] ASI_BLK_INIT_QUAD_LDD_P + subcc %g1, 0x40, %g1 + stxa %o4, [%o0 + %g2] ASI_BLK_INIT_QUAD_LDD_P + stxa %o4, [%o0 + %g3] ASI_BLK_INIT_QUAD_LDD_P + stxa %o4, [%o0 + %o5] ASI_BLK_INIT_QUAD_LDD_P + add %o0, 0x10, %o0 + stxa %o4, [%o0 + %g0] ASI_BLK_INIT_QUAD_LDD_P + stxa %o4, [%o0 + %g2] ASI_BLK_INIT_QUAD_LDD_P + stxa %o4, [%o0 + %g3] ASI_BLK_INIT_QUAD_LDD_P + stxa %o4, [%o0 + %o5] ASI_BLK_INIT_QUAD_LDD_P + bne,pt %icc, 1b + add %o0, 0x30, %o0 + ba,a,pt %icc, .Lpostloop + .size NG4bzero,.-NG4bzero diff --git a/arch/sparc/lib/NG4patch.S b/arch/sparc/lib/NG4patch.S new file mode 100644 index 00000000000..a114cbcf2a4 --- /dev/null +++ b/arch/sparc/lib/NG4patch.S @@ -0,0 +1,54 @@ +/* NG4patch.S: Patch Ultra-I routines with Niagara-4 variant. + * + * Copyright (C) 2012 David S. Miller <davem@davemloft.net> + */ + +#define BRANCH_ALWAYS 0x10680000 +#define NOP 0x01000000 +#define NG_DO_PATCH(OLD, NEW) \ + sethi %hi(NEW), %g1; \ + or %g1, %lo(NEW), %g1; \ + sethi %hi(OLD), %g2; \ + or %g2, %lo(OLD), %g2; \ + sub %g1, %g2, %g1; \ + sethi %hi(BRANCH_ALWAYS), %g3; \ + sll %g1, 11, %g1; \ + srl %g1, 11 + 2, %g1; \ + or %g3, %lo(BRANCH_ALWAYS), %g3; \ + or %g3, %g1, %g3; \ + stw %g3, [%g2]; \ + sethi %hi(NOP), %g3; \ + or %g3, %lo(NOP), %g3; \ + stw %g3, [%g2 + 0x4]; \ + flush %g2; + + .globl niagara4_patch_copyops + .type niagara4_patch_copyops,#function +niagara4_patch_copyops: + NG_DO_PATCH(memcpy, NG4memcpy) + NG_DO_PATCH(___copy_from_user, NG4copy_from_user) + NG_DO_PATCH(___copy_to_user, NG4copy_to_user) + retl + nop + .size niagara4_patch_copyops,.-niagara4_patch_copyops + + .globl niagara4_patch_bzero + .type niagara4_patch_bzero,#function +niagara4_patch_bzero: + NG_DO_PATCH(memset, NG4memset) + NG_DO_PATCH(__bzero, NG4bzero) + NG_DO_PATCH(__clear_user, NGclear_user) + NG_DO_PATCH(tsb_init, NGtsb_init) + retl + nop + .size niagara4_patch_bzero,.-niagara4_patch_bzero + + .globl niagara4_patch_pageops + .type niagara4_patch_pageops,#function +niagara4_patch_pageops: + NG_DO_PATCH(copy_user_page, NG4copy_user_page) + NG_DO_PATCH(_clear_page, NG4clear_page) + NG_DO_PATCH(clear_user_page, NG4clear_user_page) + retl + nop + .size niagara4_patch_pageops,.-niagara4_patch_pageops diff --git a/arch/sparc/lib/NGpage.S b/arch/sparc/lib/NGpage.S index b9e790b9c6b..423d46e2258 100644 --- a/arch/sparc/lib/NGpage.S +++ b/arch/sparc/lib/NGpage.S @@ -59,6 +59,8 @@ NGcopy_user_page: /* %o0=dest, %o1=src, %o2=vaddr */ restore .align 32 + .globl NGclear_page + .globl NGclear_user_page NGclear_page: /* %o0=dest */ NGclear_user_page: /* %o0=dest, %o1=vaddr */ rd %asi, %g3 diff --git a/arch/sparc/lib/U1memcpy.S b/arch/sparc/lib/U1memcpy.S index bafd2fc07ac..b67142b7768 100644 --- a/arch/sparc/lib/U1memcpy.S +++ b/arch/sparc/lib/U1memcpy.S @@ -109,7 +109,7 @@ #define UNEVEN_VISCHUNK_LAST(dest, f0, f1, left) \ subcc %left, 8, %left; \ bl,pn %xcc, 95f; \ - fsrc1 %f0, %f1; + fsrc2 %f0, %f1; #define UNEVEN_VISCHUNK(dest, f0, f1, left) \ UNEVEN_VISCHUNK_LAST(dest, f0, f1, left) \ @@ -201,7 +201,7 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */ andn %o1, (0x40 - 1), %o1 and %g2, 7, %g2 andncc %g3, 0x7, %g3 - fmovd %f0, %f2 + fsrc2 %f0, %f2 sub %g3, 0x8, %g3 sub %o2, %GLOBAL_SPARE, %o2 diff --git a/arch/sparc/lib/atomic_64.S b/arch/sparc/lib/atomic_64.S index 4d502da3de7..85c233d0a34 100644 --- a/arch/sparc/lib/atomic_64.S +++ b/arch/sparc/lib/atomic_64.S @@ -1,6 +1,6 @@ /* atomic.S: These things are too big to do inline. * - * Copyright (C) 1999, 2007 David S. Miller (davem@davemloft.net) + * Copyright (C) 1999, 2007 2012 David S. Miller (davem@davemloft.net) */ #include <linux/linkage.h> @@ -117,3 +117,17 @@ ENTRY(atomic64_sub_ret) /* %o0 = decrement, %o1 = atomic_ptr */ sub %g1, %o0, %o0 2: BACKOFF_SPIN(%o2, %o3, 1b) ENDPROC(atomic64_sub_ret) + +ENTRY(atomic64_dec_if_positive) /* %o0 = atomic_ptr */ + BACKOFF_SETUP(%o2) +1: ldx [%o0], %g1 + brlez,pn %g1, 3f + sub %g1, 1, %g7 + casx [%o0], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, BACKOFF_LABEL(2f, 1b) + nop +3: retl + sub %g1, 1, %o0 +2: BACKOFF_SPIN(%o2, %o3, 1b) +ENDPROC(atomic64_dec_if_positive) diff --git a/arch/sparc/lib/bitext.c b/arch/sparc/lib/bitext.c index 48d00e72ce1..8ec4e9c0251 100644 --- a/arch/sparc/lib/bitext.c +++ b/arch/sparc/lib/bitext.c @@ -119,11 +119,7 @@ void bit_map_clear(struct bit_map *t, int offset, int len) void bit_map_init(struct bit_map *t, unsigned long *map, int size) { - - if ((size & 07) != 0) - BUG(); - memset(map, 0, size>>3); - + bitmap_zero(map, size); memset(t, 0, sizeof *t); spin_lock_init(&t->lock); t->map = map; diff --git a/arch/sparc/lib/clear_page.S b/arch/sparc/lib/clear_page.S index 77e531f6c2a..46272dfc26e 100644 --- a/arch/sparc/lib/clear_page.S +++ b/arch/sparc/lib/clear_page.S @@ -37,10 +37,10 @@ _clear_page: /* %o0=dest */ .globl clear_user_page clear_user_page: /* %o0=dest, %o1=vaddr */ lduw [%g6 + TI_PRE_COUNT], %o2 - sethi %uhi(PAGE_OFFSET), %g2 + sethi %hi(PAGE_OFFSET), %g2 sethi %hi(PAGE_SIZE), %o4 - sllx %g2, 32, %g2 + ldx [%g2 + %lo(PAGE_OFFSET)], %g2 sethi %hi(PAGE_KERNEL_LOCKED), %g3 ldx [%g3 + %lo(PAGE_KERNEL_LOCKED)], %g3 diff --git a/arch/sparc/lib/copy_page.S b/arch/sparc/lib/copy_page.S index b243d3b606b..dd16c61f326 100644 --- a/arch/sparc/lib/copy_page.S +++ b/arch/sparc/lib/copy_page.S @@ -34,10 +34,10 @@ #endif #define TOUCH(reg0, reg1, reg2, reg3, reg4, reg5, reg6, reg7) \ - fmovd %reg0, %f48; fmovd %reg1, %f50; \ - fmovd %reg2, %f52; fmovd %reg3, %f54; \ - fmovd %reg4, %f56; fmovd %reg5, %f58; \ - fmovd %reg6, %f60; fmovd %reg7, %f62; + fsrc2 %reg0, %f48; fsrc2 %reg1, %f50; \ + fsrc2 %reg2, %f52; fsrc2 %reg3, %f54; \ + fsrc2 %reg4, %f56; fsrc2 %reg5, %f58; \ + fsrc2 %reg6, %f60; fsrc2 %reg7, %f62; .text @@ -46,10 +46,10 @@ .type copy_user_page,#function copy_user_page: /* %o0=dest, %o1=src, %o2=vaddr */ lduw [%g6 + TI_PRE_COUNT], %o4 - sethi %uhi(PAGE_OFFSET), %g2 + sethi %hi(PAGE_OFFSET), %g2 sethi %hi(PAGE_SIZE), %o3 - sllx %g2, 32, %g2 + ldx [%g2 + %lo(PAGE_OFFSET)], %g2 sethi %hi(PAGE_KERNEL_LOCKED), %g3 ldx [%g3 + %lo(PAGE_KERNEL_LOCKED)], %g3 @@ -104,60 +104,60 @@ cheetah_copy_page_insn: prefetch [%o1 + 0x140], #one_read ldd [%o1 + 0x010], %f4 prefetch [%o1 + 0x180], #one_read - fmovd %f0, %f16 + fsrc2 %f0, %f16 ldd [%o1 + 0x018], %f6 - fmovd %f2, %f18 + fsrc2 %f2, %f18 ldd [%o1 + 0x020], %f8 - fmovd %f4, %f20 + fsrc2 %f4, %f20 ldd [%o1 + 0x028], %f10 - fmovd %f6, %f22 + fsrc2 %f6, %f22 ldd [%o1 + 0x030], %f12 - fmovd %f8, %f24 + fsrc2 %f8, %f24 ldd [%o1 + 0x038], %f14 - fmovd %f10, %f26 + fsrc2 %f10, %f26 ldd [%o1 + 0x040], %f0 1: ldd [%o1 + 0x048], %f2 - fmovd %f12, %f28 + fsrc2 %f12, %f28 ldd [%o1 + 0x050], %f4 - fmovd %f14, %f30 + fsrc2 %f14, %f30 stda %f16, [%o0] ASI_BLK_P ldd [%o1 + 0x058], %f6 - fmovd %f0, %f16 + fsrc2 %f0, %f16 ldd [%o1 + 0x060], %f8 - fmovd %f2, %f18 + fsrc2 %f2, %f18 ldd [%o1 + 0x068], %f10 - fmovd %f4, %f20 + fsrc2 %f4, %f20 ldd [%o1 + 0x070], %f12 - fmovd %f6, %f22 + fsrc2 %f6, %f22 ldd [%o1 + 0x078], %f14 - fmovd %f8, %f24 + fsrc2 %f8, %f24 ldd [%o1 + 0x080], %f0 prefetch [%o1 + 0x180], #one_read - fmovd %f10, %f26 + fsrc2 %f10, %f26 subcc %o2, 1, %o2 add %o0, 0x40, %o0 bne,pt %xcc, 1b add %o1, 0x40, %o1 ldd [%o1 + 0x048], %f2 - fmovd %f12, %f28 + fsrc2 %f12, %f28 ldd [%o1 + 0x050], %f4 - fmovd %f14, %f30 + fsrc2 %f14, %f30 stda %f16, [%o0] ASI_BLK_P ldd [%o1 + 0x058], %f6 - fmovd %f0, %f16 + fsrc2 %f0, %f16 ldd [%o1 + 0x060], %f8 - fmovd %f2, %f18 + fsrc2 %f2, %f18 ldd [%o1 + 0x068], %f10 - fmovd %f4, %f20 + fsrc2 %f4, %f20 ldd [%o1 + 0x070], %f12 - fmovd %f6, %f22 + fsrc2 %f6, %f22 add %o0, 0x40, %o0 ldd [%o1 + 0x078], %f14 - fmovd %f8, %f24 - fmovd %f10, %f26 - fmovd %f12, %f28 - fmovd %f14, %f30 + fsrc2 %f8, %f24 + fsrc2 %f10, %f26 + fsrc2 %f12, %f28 + fsrc2 %f14, %f30 stda %f16, [%o0] ASI_BLK_P membar #Sync VISExitHalf diff --git a/arch/sparc/lib/ksyms.c b/arch/sparc/lib/ksyms.c index 3b31218cafc..323335b9cd2 100644 --- a/arch/sparc/lib/ksyms.c +++ b/arch/sparc/lib/ksyms.c @@ -98,15 +98,6 @@ EXPORT_SYMBOL(___copy_from_user); EXPORT_SYMBOL(___copy_in_user); EXPORT_SYMBOL(__clear_user); -/* RW semaphores */ -EXPORT_SYMBOL(__down_read); -EXPORT_SYMBOL(__down_read_trylock); -EXPORT_SYMBOL(__down_write); -EXPORT_SYMBOL(__down_write_trylock); -EXPORT_SYMBOL(__up_read); -EXPORT_SYMBOL(__up_write); -EXPORT_SYMBOL(__downgrade_write); - /* Atomic counter implementation. */ EXPORT_SYMBOL(atomic_add); EXPORT_SYMBOL(atomic_add_ret); @@ -116,6 +107,7 @@ EXPORT_SYMBOL(atomic64_add); EXPORT_SYMBOL(atomic64_add_ret); EXPORT_SYMBOL(atomic64_sub); EXPORT_SYMBOL(atomic64_sub_ret); +EXPORT_SYMBOL(atomic64_dec_if_positive); /* Atomic bit operations. */ EXPORT_SYMBOL(test_and_set_bit); @@ -134,6 +126,10 @@ EXPORT_SYMBOL(copy_user_page); void VISenter(void); EXPORT_SYMBOL(VISenter); +/* CRYPTO code needs this */ +void VISenterhalf(void); +EXPORT_SYMBOL(VISenterhalf); + extern void xor_vis_2(unsigned long, unsigned long *, unsigned long *); extern void xor_vis_3(unsigned long, unsigned long *, unsigned long *, unsigned long *); diff --git a/arch/sparc/lib/usercopy.c b/arch/sparc/lib/usercopy.c deleted file mode 100644 index 5c4284ce1c0..00000000000 --- a/arch/sparc/lib/usercopy.c +++ /dev/null @@ -1,9 +0,0 @@ -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/bug.h> - -void copy_from_user_overflow(void) -{ - WARN(1, "Buffer overflow detected!\n"); -} -EXPORT_SYMBOL(copy_from_user_overflow); diff --git a/arch/sparc/math-emu/math_64.c b/arch/sparc/math-emu/math_64.c index 1704068da92..034aadbff03 100644 --- a/arch/sparc/math-emu/math_64.c +++ b/arch/sparc/math-emu/math_64.c @@ -320,7 +320,7 @@ int do_mathemu(struct pt_regs *regs, struct fpustate *f, bool illegal_insn_trap) XR = 0; else if (freg < 16) XR = regs->u_regs[freg]; - else if (test_thread_flag(TIF_32BIT)) { + else if (!test_thread_64bit_stack(regs->u_regs[UREG_FP])) { struct reg_window32 __user *win32; flushw_user (); win32 = (struct reg_window32 __user *)((unsigned long)((u32)regs->u_regs[UREG_FP])); diff --git a/arch/sparc/math-emu/sfp-util_32.h b/arch/sparc/math-emu/sfp-util_32.h index d1b2aff3c25..bb587d5f3d9 100644 --- a/arch/sparc/math-emu/sfp-util_32.h +++ b/arch/sparc/math-emu/sfp-util_32.h @@ -4,20 +4,20 @@ #include <asm/byteorder.h> #define add_ssaaaa(sh, sl, ah, al, bh, bl) \ - __asm__ ("addcc %r4,%5,%1\n\t" \ + __asm__ ("addcc %r4,%5,%1\n\t" \ "addx %r2,%3,%0\n" \ - : "=r" ((USItype)(sh)), \ - "=&r" ((USItype)(sl)) \ + : "=r" (sh), \ + "=&r" (sl) \ : "%rJ" ((USItype)(ah)), \ "rI" ((USItype)(bh)), \ "%rJ" ((USItype)(al)), \ "rI" ((USItype)(bl)) \ : "cc") #define sub_ddmmss(sh, sl, ah, al, bh, bl) \ - __asm__ ("subcc %r4,%5,%1\n\t" \ + __asm__ ("subcc %r4,%5,%1\n\t" \ "subx %r2,%3,%0\n" \ - : "=r" ((USItype)(sh)), \ - "=&r" ((USItype)(sl)) \ + : "=r" (sh), \ + "=&r" (sl) \ : "rJ" ((USItype)(ah)), \ "rI" ((USItype)(bh)), \ "rJ" ((USItype)(al)), \ @@ -65,8 +65,8 @@ "mulscc %%g1,0,%%g1\n\t" \ "add %%g1,%%g2,%0\n\t" \ "rd %%y,%1\n" \ - : "=r" ((USItype)(w1)), \ - "=r" ((USItype)(w0)) \ + : "=r" (w1), \ + "=r" (w0) \ : "%rI" ((USItype)(u)), \ "r" ((USItype)(v)) \ : "%g1", "%g2", "cc") @@ -98,8 +98,8 @@ "sub %1,%2,%1\n\t" \ "3: xnor %0,0,%0\n\t" \ "! End of inline udiv_qrnnd\n" \ - : "=&r" ((USItype)(q)), \ - "=&r" ((USItype)(r)) \ + : "=&r" (q), \ + "=&r" (r) \ : "r" ((USItype)(d)), \ "1" ((USItype)(n1)), \ "0" ((USItype)(n0)) : "%g1", "cc") diff --git a/arch/sparc/math-emu/sfp-util_64.h b/arch/sparc/math-emu/sfp-util_64.h index 425d3cf01af..51320a861cc 100644 --- a/arch/sparc/math-emu/sfp-util_64.h +++ b/arch/sparc/math-emu/sfp-util_64.h @@ -17,8 +17,8 @@ "bcs,a,pn %%xcc, 1f\n\t" \ "add %0, 1, %0\n" \ "1:" \ - : "=r" ((UDItype)(sh)), \ - "=&r" ((UDItype)(sl)) \ + : "=r" (sh), \ + "=&r" (sl) \ : "r" ((UDItype)(ah)), \ "r" ((UDItype)(bh)), \ "r" ((UDItype)(al)), \ @@ -31,8 +31,8 @@ "bcs,a,pn %%xcc, 1f\n\t" \ "sub %0, 1, %0\n" \ "1:" \ - : "=r" ((UDItype)(sh)), \ - "=&r" ((UDItype)(sl)) \ + : "=r" (sh), \ + "=&r" (sl) \ : "r" ((UDItype)(ah)), \ "r" ((UDItype)(bh)), \ "r" ((UDItype)(al)), \ @@ -64,8 +64,8 @@ "sllx %3,32,%3\n\t" \ "add %1,%3,%1\n\t" \ "add %5,%2,%0" \ - : "=r" ((UDItype)(wh)), \ - "=&r" ((UDItype)(wl)), \ + : "=r" (wh), \ + "=&r" (wl), \ "=&r" (tmp1), "=&r" (tmp2), "=&r" (tmp3), "=&r" (tmp4) \ : "r" ((UDItype)(u)), \ "r" ((UDItype)(v)) \ diff --git a/arch/sparc/mm/fault_32.c b/arch/sparc/mm/fault_32.c index f46cf6be337..908e8c17c90 100644 --- a/arch/sparc/mm/fault_32.c +++ b/arch/sparc/mm/fault_32.c @@ -26,32 +26,14 @@ #include <asm/pgtable.h> #include <asm/openprom.h> #include <asm/oplib.h> +#include <asm/setup.h> #include <asm/smp.h> #include <asm/traps.h> #include <asm/uaccess.h> -int show_unhandled_signals = 1; - -/* At boot time we determine these two values necessary for setting - * up the segment maps and page table entries (pte's). - */ - -int num_contexts; - -/* Return how much physical memory we have. */ -unsigned long probe_memory(void) -{ - unsigned long total = 0; - int i; +#include "mm_32.h" - for (i = 0; sp_banks[i].num_bytes; i++) - total += sp_banks[i].num_bytes; - - return total; -} - -static void unhandled_fault(unsigned long, struct task_struct *, - struct pt_regs *) __attribute__ ((noreturn)); +int show_unhandled_signals = 1; static void __noreturn unhandled_fault(unsigned long address, struct task_struct *tsk, @@ -159,9 +141,6 @@ static void __do_fault_siginfo(int code, int sig, struct pt_regs *regs, force_sig_info (sig, &info, current); } -extern unsigned long safe_compute_effective_address(struct pt_regs *, - unsigned int); - static unsigned long compute_si_addr(struct pt_regs *regs, int text_fault) { unsigned int insn; @@ -195,8 +174,7 @@ asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write, unsigned long g2; int from_user = !(regs->psr & PSR_PS); int fault, code; - unsigned int flags = (FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE | - (write ? FAULT_FLAG_WRITE : 0)); + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; if (text_fault) address = regs->pc; @@ -253,6 +231,11 @@ good_area: goto bad_area; } + if (from_user) + flags |= FAULT_FLAG_USER; + if (write) + flags |= FAULT_FLAG_WRITE; + /* * If for any reason at all we couldn't handle the fault, * make sure we exit gracefully rather than endlessly redo @@ -283,6 +266,7 @@ good_area: } if (fault & VM_FAULT_RETRY) { flags &= ~FAULT_FLAG_ALLOW_RETRY; + flags |= FAULT_FLAG_TRIED; /* No need to up_read(&mm->mmap_sem) as we would * have already released it in __lock_page_or_retry @@ -400,6 +384,7 @@ static void force_user_fault(unsigned long address, int write) struct vm_area_struct *vma; struct task_struct *tsk = current; struct mm_struct *mm = tsk->mm; + unsigned int flags = FAULT_FLAG_USER; int code; code = SEGV_MAPERR; @@ -419,11 +404,12 @@ good_area: if (write) { if (!(vma->vm_flags & VM_WRITE)) goto bad_area; + flags |= FAULT_FLAG_WRITE; } else { if (!(vma->vm_flags & (VM_READ | VM_EXEC))) goto bad_area; } - switch (handle_mm_fault(mm, vma, address, write ? FAULT_FLAG_WRITE : 0)) { + switch (handle_mm_fault(mm, vma, address, flags)) { case VM_FAULT_SIGBUS: case VM_FAULT_OOM: goto do_sigbus; diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c index 1fe0429b631..587cd056512 100644 --- a/arch/sparc/mm/fault_64.c +++ b/arch/sparc/mm/fault_64.c @@ -21,6 +21,7 @@ #include <linux/kprobes.h> #include <linux/kdebug.h> #include <linux/percpu.h> +#include <linux/context_tracking.h> #include <asm/page.h> #include <asm/pgtable.h> @@ -31,6 +32,7 @@ #include <asm/lsu.h> #include <asm/sections.h> #include <asm/mmu_context.h> +#include <asm/setup.h> int show_unhandled_signals = 1; @@ -95,38 +97,51 @@ static unsigned int get_user_insn(unsigned long tpc) pte_t *ptep, pte; unsigned long pa; u32 insn = 0; - unsigned long pstate; - if (pgd_none(*pgdp)) - goto outret; + if (pgd_none(*pgdp) || unlikely(pgd_bad(*pgdp))) + goto out; pudp = pud_offset(pgdp, tpc); - if (pud_none(*pudp)) - goto outret; - pmdp = pmd_offset(pudp, tpc); - if (pmd_none(*pmdp)) - goto outret; - - /* This disables preemption for us as well. */ - __asm__ __volatile__("rdpr %%pstate, %0" : "=r" (pstate)); - __asm__ __volatile__("wrpr %0, %1, %%pstate" - : : "r" (pstate), "i" (PSTATE_IE)); - ptep = pte_offset_map(pmdp, tpc); - pte = *ptep; - if (!pte_present(pte)) + if (pud_none(*pudp) || unlikely(pud_bad(*pudp))) goto out; - pa = (pte_pfn(pte) << PAGE_SHIFT); - pa += (tpc & ~PAGE_MASK); - - /* Use phys bypass so we don't pollute dtlb/dcache. */ - __asm__ __volatile__("lduwa [%1] %2, %0" - : "=r" (insn) - : "r" (pa), "i" (ASI_PHYS_USE_EC)); + /* This disables preemption for us as well. */ + local_irq_disable(); + pmdp = pmd_offset(pudp, tpc); + if (pmd_none(*pmdp) || unlikely(pmd_bad(*pmdp))) + goto out_irq_enable; + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + if (pmd_trans_huge(*pmdp)) { + if (pmd_trans_splitting(*pmdp)) + goto out_irq_enable; + + pa = pmd_pfn(*pmdp) << PAGE_SHIFT; + pa += tpc & ~HPAGE_MASK; + + /* Use phys bypass so we don't pollute dtlb/dcache. */ + __asm__ __volatile__("lduwa [%1] %2, %0" + : "=r" (insn) + : "r" (pa), "i" (ASI_PHYS_USE_EC)); + } else +#endif + { + ptep = pte_offset_map(pmdp, tpc); + pte = *ptep; + if (pte_present(pte)) { + pa = (pte_pfn(pte) << PAGE_SHIFT); + pa += (tpc & ~PAGE_MASK); + + /* Use phys bypass so we don't pollute dtlb/dcache. */ + __asm__ __volatile__("lduwa [%1] %2, %0" + : "=r" (insn) + : "r" (pa), "i" (ASI_PHYS_USE_EC)); + } + pte_unmap(ptep); + } +out_irq_enable: + local_irq_enable(); out: - pte_unmap(ptep); - __asm__ __volatile__("wrpr %0, 0x0, %%pstate" : : "r" (pstate)); -outret: return insn; } @@ -151,10 +166,9 @@ show_signal_msg(struct pt_regs *regs, int sig, int code, printk(KERN_CONT "\n"); } -extern unsigned long compute_effective_address(struct pt_regs *, unsigned int, unsigned int); - static void do_fault_siginfo(int code, int sig, struct pt_regs *regs, - unsigned int insn, int fault_code) + unsigned long fault_addr, unsigned int insn, + int fault_code) { unsigned long addr; siginfo_t info; @@ -162,10 +176,18 @@ static void do_fault_siginfo(int code, int sig, struct pt_regs *regs, info.si_code = code; info.si_signo = sig; info.si_errno = 0; - if (fault_code & FAULT_CODE_ITLB) + if (fault_code & FAULT_CODE_ITLB) { addr = regs->tpc; - else - addr = compute_effective_address(regs, insn, 0); + } else { + /* If we were able to probe the faulting instruction, use it + * to compute a precise fault address. Otherwise use the fault + * time provided address which may only have page granularity. + */ + if (insn) + addr = compute_effective_address(regs, insn, 0); + else + addr = fault_addr; + } info.si_addr = (void __user *) addr; info.si_trapno = 0; @@ -175,9 +197,6 @@ static void do_fault_siginfo(int code, int sig, struct pt_regs *regs, force_sig_info(sig, &info, current); } -extern int handle_ldf_stq(u32, struct pt_regs *); -extern int handle_ld_nf(u32, struct pt_regs *); - static unsigned int get_fault_insn(struct pt_regs *regs, unsigned int insn) { if (!insn) { @@ -240,7 +259,7 @@ static void __kprobes do_kernel_fault(struct pt_regs *regs, int si_code, /* The si_code was set to make clear whether * this was a SEGV_MAPERR or SEGV_ACCERR fault. */ - do_fault_siginfo(si_code, SIGSEGV, regs, insn, fault_code); + do_fault_siginfo(si_code, SIGSEGV, regs, address, insn, fault_code); return; } @@ -260,20 +279,9 @@ static void noinline __kprobes bogus_32bit_fault_tpc(struct pt_regs *regs) show_regs(regs); } -static void noinline __kprobes bogus_32bit_fault_address(struct pt_regs *regs, - unsigned long addr) -{ - static int times; - - if (times++ < 10) - printk(KERN_ERR "FAULT[%s:%d]: 32-bit process " - "reports 64-bit fault address [%lx]\n", - current->comm, current->pid, addr); - show_regs(regs); -} - asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) { + enum ctx_state prev_state = exception_enter(); struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned int insn = 0; @@ -284,7 +292,7 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) fault_code = get_thread_fault_code(); if (notify_page_fault(regs)) - return; + goto exit_exception; si_code = SEGV_MAPERR; address = current_thread_info()->fault_address; @@ -300,10 +308,8 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) goto intr_or_no_mm; } } - if (unlikely((address >> 32) != 0)) { - bogus_32bit_fault_address(regs, address); + if (unlikely((address >> 32) != 0)) goto intr_or_no_mm; - } } if (regs->tstate & TSTATE_PRIV) { @@ -315,9 +321,10 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) /* Valid, no problems... */ } else { bad_kernel_pc(regs, address); - return; + goto exit_exception; } - } + } else + flags |= FAULT_FLAG_USER; /* * If we're in an interrupt or have no user @@ -420,17 +427,18 @@ good_area: vma->vm_file != NULL) set_thread_fault_code(fault_code | FAULT_CODE_BLKCOMMIT); + + flags |= FAULT_FLAG_WRITE; } else { /* Allow reads even for write-only mappings */ if (!(vma->vm_flags & (VM_READ | VM_EXEC))) goto bad_area; } - flags |= ((fault_code & FAULT_CODE_WRITE) ? FAULT_FLAG_WRITE : 0); fault = handle_mm_fault(mm, vma, address, flags); if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) - return; + goto exit_exception; if (unlikely(fault & VM_FAULT_ERROR)) { if (fault & VM_FAULT_OOM) @@ -452,6 +460,7 @@ good_area: } if (fault & VM_FAULT_RETRY) { flags &= ~FAULT_FLAG_ALLOW_RETRY; + flags |= FAULT_FLAG_TRIED; /* No need to up_read(&mm->mmap_sem) as we would * have already released it in __lock_page_or_retry @@ -464,18 +473,25 @@ good_area: up_read(&mm->mmap_sem); mm_rss = get_mm_rss(mm); -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) mm_rss -= (mm->context.huge_pte_count * (HPAGE_SIZE / PAGE_SIZE)); #endif if (unlikely(mm_rss > mm->context.tsb_block[MM_TSB_BASE].tsb_rss_limit)) tsb_grow(mm, MM_TSB_BASE, mm_rss); -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) mm_rss = mm->context.huge_pte_count; if (unlikely(mm_rss > - mm->context.tsb_block[MM_TSB_HUGE].tsb_rss_limit)) - tsb_grow(mm, MM_TSB_HUGE, mm_rss); + mm->context.tsb_block[MM_TSB_HUGE].tsb_rss_limit)) { + if (mm->context.tsb_block[MM_TSB_HUGE].tsb) + tsb_grow(mm, MM_TSB_HUGE, mm_rss); + else + hugetlb_setup(regs); + + } #endif +exit_exception: + exception_exit(prev_state); return; /* @@ -488,7 +504,7 @@ bad_area: handle_kernel_fault: do_kernel_fault(regs, si_code, fault_code, insn, address); - return; + goto exit_exception; /* * We ran out of memory, or some other thing happened to us that made @@ -499,7 +515,7 @@ out_of_memory: up_read(&mm->mmap_sem); if (!(regs->tstate & TSTATE_PRIV)) { pagefault_out_of_memory(); - return; + goto exit_exception; } goto handle_kernel_fault; @@ -515,7 +531,7 @@ do_sigbus: * Send a sigbus, regardless of whether we were in kernel * or user mode. */ - do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, insn, fault_code); + do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, address, insn, fault_code); /* Kernel mode? Handle exceptions or die */ if (regs->tstate & TSTATE_PRIV) diff --git a/arch/sparc/mm/gup.c b/arch/sparc/mm/gup.c index 42c55df3aec..1aed0432c64 100644 --- a/arch/sparc/mm/gup.c +++ b/arch/sparc/mm/gup.c @@ -66,6 +66,55 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr, return 1; } +static int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr, + unsigned long end, int write, struct page **pages, + int *nr) +{ + struct page *head, *page, *tail; + int refs; + + if (!(pmd_val(pmd) & _PAGE_VALID)) + return 0; + + if (write && !pmd_write(pmd)) + return 0; + + refs = 0; + head = pmd_page(pmd); + page = head + ((addr & ~PMD_MASK) >> PAGE_SHIFT); + tail = page; + do { + VM_BUG_ON(compound_head(page) != head); + pages[*nr] = page; + (*nr)++; + page++; + refs++; + } while (addr += PAGE_SIZE, addr != end); + + if (!page_cache_add_speculative(head, refs)) { + *nr -= refs; + return 0; + } + + if (unlikely(pmd_val(pmd) != pmd_val(*pmdp))) { + *nr -= refs; + while (refs--) + put_page(head); + return 0; + } + + /* Any tail page need their mapcount reference taken before we + * return. + */ + while (refs--) { + if (PageTail(tail)) + get_huge_page_tail(tail); + tail++; + } + + return 1; +} + static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end, int write, struct page **pages, int *nr) { @@ -77,9 +126,14 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end, pmd_t pmd = *pmdp; next = pmd_addr_end(addr, end); - if (pmd_none(pmd)) + if (pmd_none(pmd) || pmd_trans_splitting(pmd)) return 0; - if (!gup_pte_range(pmd, addr, next, write, pages, nr)) + if (unlikely(pmd_large(pmd))) { + if (!gup_huge_pmd(pmdp, pmd, addr, next, + write, pages, nr)) + return 0; + } else if (!gup_pte_range(pmd, addr, next, write, + pages, nr)) return 0; } while (pmdp++, addr = next, addr != end); diff --git a/arch/sparc/mm/highmem.c b/arch/sparc/mm/highmem.c index 055c66cf1bf..449f864f0ce 100644 --- a/arch/sparc/mm/highmem.c +++ b/arch/sparc/mm/highmem.c @@ -22,13 +22,31 @@ * shared by CPUs, and so precious, and establishing them requires IPI. * Atomic kmaps are lightweight and we may have NCPUS more of them. */ -#include <linux/mm.h> #include <linux/highmem.h> #include <linux/export.h> -#include <asm/pgalloc.h> +#include <linux/mm.h> + #include <asm/cacheflush.h> #include <asm/tlbflush.h> -#include <asm/fixmap.h> +#include <asm/pgalloc.h> +#include <asm/vaddrs.h> + +pgprot_t kmap_prot; + +static pte_t *kmap_pte; + +void __init kmap_init(void) +{ + unsigned long address; + pmd_t *dir; + + address = __fix_to_virt(FIX_KMAP_BEGIN); + dir = pmd_offset(pgd_offset_k(address), address); + + /* cache the first kmap pte */ + kmap_pte = pte_offset_kernel(dir, address); + kmap_prot = __pgprot(SRMMU_ET_PTE | SRMMU_PRIV | SRMMU_CACHE); +} void *kmap_atomic(struct page *page) { @@ -110,21 +128,3 @@ void __kunmap_atomic(void *kvaddr) pagefault_enable(); } EXPORT_SYMBOL(__kunmap_atomic); - -/* We may be fed a pagetable here by ptep_to_xxx and others. */ -struct page *kmap_atomic_to_page(void *ptr) -{ - unsigned long idx, vaddr = (unsigned long)ptr; - pte_t *pte; - - if (vaddr < SRMMU_NOCACHE_VADDR) - return virt_to_page(ptr); - if (vaddr < PKMAP_BASE) - return pfn_to_page(__nocache_pa(vaddr) >> PAGE_SHIFT); - BUG_ON(vaddr < FIXADDR_START); - BUG_ON(vaddr > FIXADDR_TOP); - - idx = virt_to_fix(vaddr); - pte = kmap_pte - (idx - FIX_KMAP_BEGIN); - return pte_page(*pte); -} diff --git a/arch/sparc/mm/hugetlbpage.c b/arch/sparc/mm/hugetlbpage.c index 07e14535375..d329537739c 100644 --- a/arch/sparc/mm/hugetlbpage.c +++ b/arch/sparc/mm/hugetlbpage.c @@ -4,7 +4,6 @@ * Copyright (C) 2002, 2003, 2006 David S. Miller (davem@davemloft.net) */ -#include <linux/init.h> #include <linux/fs.h> #include <linux/mm.h> #include <linux/hugetlb.h> @@ -21,8 +20,6 @@ /* Slightly simplified from the non-hugepage variant because by * definition we don't have to worry about any page coloring stuff */ -#define VA_EXCLUDE_START (0x0000080000000000UL - (1UL << 32UL)) -#define VA_EXCLUDE_END (0xfffff80000000000UL + (1UL << 32UL)) static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp, unsigned long addr, @@ -30,55 +27,28 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *filp, unsigned long pgoff, unsigned long flags) { - struct mm_struct *mm = current->mm; - struct vm_area_struct * vma; unsigned long task_size = TASK_SIZE; - unsigned long start_addr; + struct vm_unmapped_area_info info; if (test_thread_flag(TIF_32BIT)) task_size = STACK_TOP32; - if (unlikely(len >= VA_EXCLUDE_START)) - return -ENOMEM; - if (len > mm->cached_hole_size) { - start_addr = addr = mm->free_area_cache; - } else { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; + info.flags = 0; + info.length = len; + info.low_limit = TASK_UNMAPPED_BASE; + info.high_limit = min(task_size, VA_EXCLUDE_START); + info.align_mask = PAGE_MASK & ~HPAGE_MASK; + info.align_offset = 0; + addr = vm_unmapped_area(&info); + + if ((addr & ~PAGE_MASK) && task_size > VA_EXCLUDE_END) { + VM_BUG_ON(addr != -ENOMEM); + info.low_limit = VA_EXCLUDE_END; + info.high_limit = task_size; + addr = vm_unmapped_area(&info); } - task_size -= len; - -full_search: - addr = ALIGN(addr, HPAGE_SIZE); - - for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { - /* At this point: (!vma || addr < vma->vm_end). */ - if (addr < VA_EXCLUDE_START && - (addr + len) >= VA_EXCLUDE_START) { - addr = VA_EXCLUDE_END; - vma = find_vma(mm, VA_EXCLUDE_END); - } - if (unlikely(task_size < addr)) { - if (start_addr != TASK_UNMAPPED_BASE) { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; - goto full_search; - } - return -ENOMEM; - } - if (likely(!vma || addr + len <= vma->vm_start)) { - /* - * Remember the place where we stopped the search: - */ - mm->free_area_cache = addr + len; - return addr; - } - if (addr + mm->cached_hole_size < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; - - addr = ALIGN(vma->vm_end, HPAGE_SIZE); - } + return addr; } static unsigned long @@ -87,71 +57,34 @@ hugetlb_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, const unsigned long pgoff, const unsigned long flags) { - struct vm_area_struct *vma; struct mm_struct *mm = current->mm; unsigned long addr = addr0; + struct vm_unmapped_area_info info; /* This should only ever run for 32-bit processes. */ BUG_ON(!test_thread_flag(TIF_32BIT)); - /* check if free_area_cache is useful for us */ - if (len <= mm->cached_hole_size) { - mm->cached_hole_size = 0; - mm->free_area_cache = mm->mmap_base; - } - - /* either no address requested or can't fit in requested address hole */ - addr = mm->free_area_cache & HPAGE_MASK; - - /* make sure it can fit in the remaining address space */ - if (likely(addr > len)) { - vma = find_vma(mm, addr-len); - if (!vma || addr <= vma->vm_start) { - /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr-len); - } - } - - if (unlikely(mm->mmap_base < len)) - goto bottomup; - - addr = (mm->mmap_base-len) & HPAGE_MASK; - - do { - /* - * Lookup failure means no vma is above this address, - * else if new region fits below vma->vm_start, - * return with success: - */ - vma = find_vma(mm, addr); - if (likely(!vma || addr+len <= vma->vm_start)) { - /* remember the address as a hint for next time */ - return (mm->free_area_cache = addr); - } + info.flags = VM_UNMAPPED_AREA_TOPDOWN; + info.length = len; + info.low_limit = PAGE_SIZE; + info.high_limit = mm->mmap_base; + info.align_mask = PAGE_MASK & ~HPAGE_MASK; + info.align_offset = 0; + addr = vm_unmapped_area(&info); - /* remember the largest hole we saw so far */ - if (addr + mm->cached_hole_size < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; - - /* try just below the current vma->vm_start */ - addr = (vma->vm_start-len) & HPAGE_MASK; - } while (likely(len < vma->vm_start)); - -bottomup: /* * A failed mmap() very likely causes application failure, * so fall back to the bottom-up function here. This scenario * can happen with large stack limits and large mmap() * allocations. */ - mm->cached_hole_size = ~0UL; - mm->free_area_cache = TASK_UNMAPPED_BASE; - addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); - /* - * Restore the topdown base: - */ - mm->free_area_cache = mm->mmap_base; - mm->cached_hole_size = ~0UL; + if (addr & ~PAGE_MASK) { + VM_BUG_ON(addr != -ENOMEM); + info.flags = 0; + info.low_limit = TASK_UNMAPPED_BASE; + info.high_limit = STACK_TOP32; + addr = vm_unmapped_area(&info); + } return addr; } @@ -303,53 +236,3 @@ struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, { return NULL; } - -static void context_reload(void *__data) -{ - struct mm_struct *mm = __data; - - if (mm == current->mm) - load_secondary_context(mm); -} - -void hugetlb_prefault_arch_hook(struct mm_struct *mm) -{ - struct tsb_config *tp = &mm->context.tsb_block[MM_TSB_HUGE]; - - if (likely(tp->tsb != NULL)) - return; - - tsb_grow(mm, MM_TSB_HUGE, 0); - tsb_context_switch(mm); - smp_tsb_sync(mm); - - /* On UltraSPARC-III+ and later, configure the second half of - * the Data-TLB for huge pages. - */ - if (tlb_type == cheetah_plus) { - unsigned long ctx; - - spin_lock(&ctx_alloc_lock); - ctx = mm->context.sparc64_ctx_val; - ctx &= ~CTX_PGSZ_MASK; - ctx |= CTX_PGSZ_BASE << CTX_PGSZ0_SHIFT; - ctx |= CTX_PGSZ_HUGE << CTX_PGSZ1_SHIFT; - - if (ctx != mm->context.sparc64_ctx_val) { - /* When changing the page size fields, we - * must perform a context flush so that no - * stale entries match. This flush must - * occur with the original context register - * settings. - */ - do_flush_tlb_mm(mm); - - /* Reload the context register of all processors - * also executing in this address space. - */ - mm->context.sparc64_ctx_val = ctx; - on_each_cpu(context_reload, mm, 0); - } - spin_unlock(&ctx_alloc_lock); - } -} diff --git a/arch/sparc/mm/hypersparc.S b/arch/sparc/mm/hypersparc.S index 44aad32eeb4..969f96450f6 100644 --- a/arch/sparc/mm/hypersparc.S +++ b/arch/sparc/mm/hypersparc.S @@ -74,7 +74,7 @@ hypersparc_flush_cache_mm_out: /* The things we do for performance... */ hypersparc_flush_cache_range: - ld [%o0 + 0x0], %o0 /* XXX vma->vm_mm, GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 #ifndef CONFIG_SMP ld [%o0 + AOFF_mm_context], %g1 cmp %g1, -1 @@ -163,7 +163,7 @@ hypersparc_flush_cache_range_out: */ /* Verified, my ass... */ hypersparc_flush_cache_page: - ld [%o0 + 0x0], %o0 /* XXX vma->vm_mm, GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 ld [%o0 + AOFF_mm_context], %g2 #ifndef CONFIG_SMP cmp %g2, -1 @@ -284,7 +284,7 @@ hypersparc_flush_tlb_mm_out: sta %g5, [%g1] ASI_M_MMUREGS hypersparc_flush_tlb_range: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 mov SRMMU_CTX_REG, %g1 ld [%o0 + AOFF_mm_context], %o3 lda [%g1] ASI_M_MMUREGS, %g5 @@ -307,7 +307,7 @@ hypersparc_flush_tlb_range_out: sta %g5, [%g1] ASI_M_MMUREGS hypersparc_flush_tlb_page: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 mov SRMMU_CTX_REG, %g1 ld [%o0 + AOFF_mm_context], %o3 andn %o1, (PAGE_SIZE - 1), %o1 diff --git a/arch/sparc/mm/init_32.c b/arch/sparc/mm/init_32.c index ef5c779ec85..eb828715527 100644 --- a/arch/sparc/mm/init_32.c +++ b/arch/sparc/mm/init_32.c @@ -31,10 +31,13 @@ #include <asm/pgtable.h> #include <asm/vaddrs.h> #include <asm/pgalloc.h> /* bug in asm-generic/tlb.h: check_pgt_cache */ +#include <asm/setup.h> #include <asm/tlb.h> #include <asm/prom.h> #include <asm/leon.h> +#include "mm_32.h" + unsigned long *sparc_valid_addr_bitmap; EXPORT_SYMBOL(sparc_valid_addr_bitmap); @@ -45,9 +48,6 @@ unsigned long pfn_base; EXPORT_SYMBOL(pfn_base); struct sparc_phys_banks sp_banks[SPARC_PHYS_BANKS+1]; -unsigned long sparc_unmapped_base; - -struct pgtable_cache_struct pgt_quicklists; /* Initial ramdisk setup */ extern unsigned int sparc_ramdisk_image; @@ -55,56 +55,17 @@ extern unsigned int sparc_ramdisk_size; unsigned long highstart_pfn, highend_pfn; -pte_t *kmap_pte; -pgprot_t kmap_prot; - -#define kmap_get_fixmap_pte(vaddr) \ - pte_offset_kernel(pmd_offset(pgd_offset_k(vaddr), (vaddr)), (vaddr)) - -void __init kmap_init(void) -{ - /* cache the first kmap pte */ - kmap_pte = kmap_get_fixmap_pte(__fix_to_virt(FIX_KMAP_BEGIN)); - kmap_prot = __pgprot(SRMMU_ET_PTE | SRMMU_PRIV | SRMMU_CACHE); -} - void show_mem(unsigned int filter) { printk("Mem-info:\n"); show_free_areas(filter); printk("Free swap: %6ldkB\n", - nr_swap_pages << (PAGE_SHIFT-10)); + get_nr_swap_pages() << (PAGE_SHIFT-10)); printk("%ld pages of RAM\n", totalram_pages); printk("%ld free pages\n", nr_free_pages()); -#if 0 /* undefined pgtable_cache_size, pgd_cache_size */ - printk("%ld pages in page table cache\n",pgtable_cache_size); -#ifndef CONFIG_SMP - if (sparc_cpu_model == sun4m || sparc_cpu_model == sun4d) - printk("%ld entries in page dir cache\n",pgd_cache_size); -#endif -#endif } -void __init sparc_context_init(int numctx) -{ - int ctx; - - ctx_list_pool = __alloc_bootmem(numctx * sizeof(struct ctx_list), SMP_CACHE_BYTES, 0UL); - - for(ctx = 0; ctx < numctx; ctx++) { - struct ctx_list *clist; - clist = (ctx_list_pool + ctx); - clist->ctx_number = ctx; - clist->ctx_mm = NULL; - } - ctx_free.next = ctx_free.prev = &ctx_free; - ctx_used.next = ctx_used.prev = &ctx_used; - for(ctx = 0; ctx < numctx; ctx++) - add_to_free_ctxlist(ctx_list_pool + ctx); -} - -extern unsigned long cmdline_memory_size; unsigned long last_valid_pfn; unsigned long calc_highpages(void) @@ -287,27 +248,9 @@ unsigned long __init bootmem_init(unsigned long *pages_avail) * init routine based upon the Sun model type on the Sparc. * */ -extern void srmmu_paging_init(void); -extern void device_scan(void); - void __init paging_init(void) { - switch(sparc_cpu_model) { - case sparc_leon: - leon_init(); - /* fall through */ - case sun4m: - case sun4d: - srmmu_paging_init(); - sparc_unmapped_base = 0x50000000; - break; - default: - prom_printf("paging_init: Cannot init paging on this Sparc\n"); - prom_printf("paging_init: sparc_cpu_model = %d\n", sparc_cpu_model); - prom_printf("paging_init: Halting...\n"); - prom_halt(); - } - + srmmu_paging_init(); prom_build_devicetree(); of_fill_in_cpu_data(); device_scan(); @@ -338,22 +281,12 @@ static void map_high_region(unsigned long start_pfn, unsigned long end_pfn) printk("mapping high region %08lx - %08lx\n", start_pfn, end_pfn); #endif - for (tmp = start_pfn; tmp < end_pfn; tmp++) { - struct page *page = pfn_to_page(tmp); - - ClearPageReserved(page); - init_page_count(page); - __free_page(page); - totalhigh_pages++; - } + for (tmp = start_pfn; tmp < end_pfn; tmp++) + free_highmem_page(pfn_to_page(tmp)); } void __init mem_init(void) { - int codepages = 0; - int datapages = 0; - int initpages = 0; - int reservedpages = 0; int i; if (PKMAP_BASE+LAST_PKMAP*PAGE_SIZE >= FIXADDR_START) { @@ -385,15 +318,12 @@ void __init mem_init(void) max_mapnr = last_valid_pfn - pfn_base; high_memory = __va(max_low_pfn << PAGE_SHIFT); - - totalram_pages = free_all_bootmem(); + free_all_bootmem(); for (i = 0; sp_banks[i].num_bytes != 0; i++) { unsigned long start_pfn = sp_banks[i].base_addr >> PAGE_SHIFT; unsigned long end_pfn = (sp_banks[i].base_addr + sp_banks[i].num_bytes) >> PAGE_SHIFT; - num_physpages += sp_banks[i].num_bytes >> PAGE_SHIFT; - if (end_pfn <= highstart_pfn) continue; @@ -403,72 +333,19 @@ void __init mem_init(void) map_high_region(start_pfn, end_pfn); } - totalram_pages += totalhigh_pages; - - codepages = (((unsigned long) &_etext) - ((unsigned long)&_start)); - codepages = PAGE_ALIGN(codepages) >> PAGE_SHIFT; - datapages = (((unsigned long) &_edata) - ((unsigned long)&_etext)); - datapages = PAGE_ALIGN(datapages) >> PAGE_SHIFT; - initpages = (((unsigned long) &__init_end) - ((unsigned long) &__init_begin)); - initpages = PAGE_ALIGN(initpages) >> PAGE_SHIFT; - - /* Ignore memory holes for the purpose of counting reserved pages */ - for (i=0; i < max_low_pfn; i++) - if (test_bit(i >> (20 - PAGE_SHIFT), sparc_valid_addr_bitmap) - && PageReserved(pfn_to_page(i))) - reservedpages++; - - printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init, %ldk highmem)\n", - nr_free_pages() << (PAGE_SHIFT-10), - num_physpages << (PAGE_SHIFT - 10), - codepages << (PAGE_SHIFT-10), - reservedpages << (PAGE_SHIFT - 10), - datapages << (PAGE_SHIFT-10), - initpages << (PAGE_SHIFT-10), - totalhigh_pages << (PAGE_SHIFT-10)); + mem_init_print_info(NULL); } void free_initmem (void) { - unsigned long addr; - unsigned long freed; - - addr = (unsigned long)(&__init_begin); - freed = (unsigned long)(&__init_end) - addr; - for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { - struct page *p; - - memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE); - p = virt_to_page(addr); - - ClearPageReserved(p); - init_page_count(p); - __free_page(p); - totalram_pages++; - num_physpages++; - } - printk(KERN_INFO "Freeing unused kernel memory: %ldk freed\n", - freed >> 10); + free_initmem_default(POISON_FREE_INITMEM); } #ifdef CONFIG_BLK_DEV_INITRD void free_initrd_mem(unsigned long start, unsigned long end) { - if (start < end) - printk(KERN_INFO "Freeing initrd memory: %ldk freed\n", - (end - start) >> 10); - for (; start < end; start += PAGE_SIZE) { - struct page *p; - - memset((void *)start, POISON_FREE_INITMEM, PAGE_SIZE); - p = virt_to_page(start); - - ClearPageReserved(p); - init_page_count(p); - __free_page(p); - totalram_pages++; - num_physpages++; - } + free_reserved_area((void *)start, (void *)end, POISON_FREE_INITMEM, + "initrd"); } #endif diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index 6026fdd1b2e..16b58ff11e6 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -47,30 +47,49 @@ #include <asm/prom.h> #include <asm/mdesc.h> #include <asm/cpudata.h> +#include <asm/setup.h> #include <asm/irq.h> #include "init_64.h" -unsigned long kern_linear_pte_xor[2] __read_mostly; +unsigned long kern_linear_pte_xor[4] __read_mostly; -/* A bitmap, one bit for every 256MB of physical memory. If the bit - * is clear, we should use a 4MB page (via kern_linear_pte_xor[0]) else - * if set we should use a 256MB page (via kern_linear_pte_xor[1]). +/* A bitmap, two bits for every 256MB of physical memory. These two + * bits determine what page size we use for kernel linear + * translations. They form an index into kern_linear_pte_xor[]. The + * value in the indexed slot is XOR'd with the TLB miss virtual + * address to form the resulting TTE. The mapping is: + * + * 0 ==> 4MB + * 1 ==> 256MB + * 2 ==> 2GB + * 3 ==> 16GB + * + * All sun4v chips support 256MB pages. Only SPARC-T4 and later + * support 2GB pages, and hopefully future cpus will support the 16GB + * pages as well. For slots 2 and 3, we encode a 256MB TTE xor there + * if these larger page sizes are not supported by the cpu. + * + * It would be nice to determine this from the machine description + * 'cpu' properties, but we need to have this table setup before the + * MDESC is initialized. */ unsigned long kpte_linear_bitmap[KPTE_BITMAP_BYTES / sizeof(unsigned long)]; #ifndef CONFIG_DEBUG_PAGEALLOC -/* A special kernel TSB for 4MB and 256MB linear mappings. - * Space is allocated for this right after the trap table - * in arch/sparc64/kernel/head.S +/* A special kernel TSB for 4MB, 256MB, 2GB and 16GB linear mappings. + * Space is allocated for this right after the trap table in + * arch/sparc64/kernel/head.S */ extern struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES]; #endif +static unsigned long cpu_pgsz_mask; + #define MAX_BANKS 32 -static struct linux_prom64_registers pavail[MAX_BANKS] __devinitdata; -static int pavail_ents __devinitdata; +static struct linux_prom64_registers pavail[MAX_BANKS]; +static int pavail_ents; static int cmp_p64(const void *a, const void *b) { @@ -101,7 +120,8 @@ static void __init read_obp_memory(const char *property, ret = prom_getproperty(node, property, (char *) regs, prop_size); if (ret == -1) { - prom_printf("Couldn't get %s property from /memory.\n"); + prom_printf("Couldn't get %s property from /memory.\n", + property); prom_halt(); } @@ -257,7 +277,6 @@ static inline void tsb_insert(struct tsb *ent, unsigned long tag, unsigned long } unsigned long _PAGE_ALL_SZ_BITS __read_mostly; -unsigned long _PAGE_SZBITS __read_mostly; static void flush_dcache(unsigned long pfn) { @@ -288,12 +307,39 @@ static void flush_dcache(unsigned long pfn) } } +/* mm->context.lock must be held */ +static void __update_mmu_tsb_insert(struct mm_struct *mm, unsigned long tsb_index, + unsigned long tsb_hash_shift, unsigned long address, + unsigned long tte) +{ + struct tsb *tsb = mm->context.tsb_block[tsb_index].tsb; + unsigned long tag; + + if (unlikely(!tsb)) + return; + + tsb += ((address >> tsb_hash_shift) & + (mm->context.tsb_block[tsb_index].tsb_nentries - 1UL)); + tag = (address >> 22UL); + tsb_insert(tsb, tag, tte); +} + +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) +static inline bool is_hugetlb_pte(pte_t pte) +{ + if ((tlb_type == hypervisor && + (pte_val(pte) & _PAGE_SZALL_4V) == _PAGE_SZHUGE_4V) || + (tlb_type != hypervisor && + (pte_val(pte) & _PAGE_SZALL_4U) == _PAGE_SZHUGE_4U)) + return true; + return false; +} +#endif + void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *ptep) { struct mm_struct *mm; - struct tsb *tsb; - unsigned long tag, flags; - unsigned long tsb_index, tsb_hash_shift; + unsigned long flags; pte_t pte = *ptep; if (tlb_type != hypervisor) { @@ -305,28 +351,16 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t * mm = vma->vm_mm; - tsb_index = MM_TSB_BASE; - tsb_hash_shift = PAGE_SHIFT; - spin_lock_irqsave(&mm->context.lock, flags); -#ifdef CONFIG_HUGETLB_PAGE - if (mm->context.tsb_block[MM_TSB_HUGE].tsb != NULL) { - if ((tlb_type == hypervisor && - (pte_val(pte) & _PAGE_SZALL_4V) == _PAGE_SZHUGE_4V) || - (tlb_type != hypervisor && - (pte_val(pte) & _PAGE_SZALL_4U) == _PAGE_SZHUGE_4U)) { - tsb_index = MM_TSB_HUGE; - tsb_hash_shift = HPAGE_SHIFT; - } - } +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) + if (mm->context.huge_pte_count && is_hugetlb_pte(pte)) + __update_mmu_tsb_insert(mm, MM_TSB_HUGE, REAL_HPAGE_SHIFT, + address, pte_val(pte)); + else #endif - - tsb = mm->context.tsb_block[tsb_index].tsb; - tsb += ((address >> tsb_hash_shift) & - (mm->context.tsb_block[tsb_index].tsb_nentries - 1UL)); - tag = (address >> 22UL); - tsb_insert(tsb, tag, pte_val(pte)); + __update_mmu_tsb_insert(mm, MM_TSB_BASE, PAGE_SHIFT, + address, pte_val(pte)); spin_unlock_irqrestore(&mm->context.lock, flags); } @@ -403,6 +437,12 @@ EXPORT_SYMBOL(flush_icache_range); void mmu_info(struct seq_file *m) { + static const char *pgsz_strings[] = { + "8K", "64K", "512K", "4MB", "32MB", + "256MB", "2GB", "16GB", + }; + int i, printed; + if (tlb_type == cheetah) seq_printf(m, "MMU Type\t: Cheetah\n"); else if (tlb_type == cheetah_plus) @@ -414,6 +454,17 @@ void mmu_info(struct seq_file *m) else seq_printf(m, "MMU Type\t: ???\n"); + seq_printf(m, "MMU PGSZs\t: "); + printed = 0; + for (i = 0; i < ARRAY_SIZE(pgsz_strings); i++) { + if (cpu_pgsz_mask & (1UL << i)) { + seq_printf(m, "%s%s", + printed ? "," : "", pgsz_strings[i]); + printed++; + } + } + seq_putc(m, '\n'); + #ifdef CONFIG_DEBUG_DCFLUSH seq_printf(m, "DCPageFlushes\t: %d\n", atomic_read(&dcpage_flushes)); @@ -462,7 +513,7 @@ static void __init read_obp_translations(void) prom_halt(); } if (unlikely(n > sizeof(prom_trans))) { - prom_printf("prom_mappings: Size %Zd is too big.\n", n); + prom_printf("prom_mappings: Size %d is too big.\n", n); prom_halt(); } @@ -524,7 +575,7 @@ static void __init hypervisor_tlb_lock(unsigned long vaddr, unsigned long ret = sun4v_mmu_map_perm_addr(vaddr, 0, pte, mmu); if (ret != 0) { - prom_printf("hypervisor_tlb_lock[%lx:%lx:%lx:%lx]: " + prom_printf("hypervisor_tlb_lock[%lx:%x:%lx:%lx]: " "errors with %lx\n", vaddr, 0, pte, mmu, ret); prom_halt(); } @@ -538,7 +589,7 @@ static void __init remap_kernel(void) int i, tlb_ent = sparc64_highest_locked_tlbent(); tte_vaddr = (unsigned long) KERNBASE; - phys_page = (prom_boot_mapping_phys_low >> 22UL) << 22UL; + phys_page = (prom_boot_mapping_phys_low >> ILOG2_4MB) << ILOG2_4MB; tte_data = kern_large_tte(phys_page); kern_locked_tte_data = tte_data; @@ -580,7 +631,7 @@ static void __init inherit_prom_mappings(void) void prom_world(int enter) { if (!enter) - set_fs((mm_segment_t) { get_thread_current_ds() }); + set_fs(get_fs()); __asm__ __volatile__("flushw"); } @@ -631,10 +682,9 @@ void get_new_mmu_context(struct mm_struct *mm) { unsigned long ctx, new_ctx; unsigned long orig_pgsz_bits; - unsigned long flags; int new_version; - spin_lock_irqsave(&ctx_alloc_lock, flags); + spin_lock(&ctx_alloc_lock); orig_pgsz_bits = (mm->context.sparc64_ctx_val & CTX_PGSZ_MASK); ctx = (tlb_context_cache + 1) & CTX_NR_MASK; new_ctx = find_next_zero_bit(mmu_context_bmap, 1 << CTX_NR_BITS, ctx); @@ -670,7 +720,7 @@ void get_new_mmu_context(struct mm_struct *mm) out: tlb_context_cache = new_ctx; mm->context.sparc64_ctx_val = new_ctx | orig_pgsz_bits; - spin_unlock_irqrestore(&ctx_alloc_lock, flags); + spin_unlock(&ctx_alloc_lock); if (unlikely(new_version)) smp_new_mmu_context_version(); @@ -745,11 +795,11 @@ struct node_mem_mask { static struct node_mem_mask node_masks[MAX_NUMNODES]; static int num_node_masks; +#ifdef CONFIG_NEED_MULTIPLE_NODES + int numa_cpu_lookup_table[NR_CPUS]; cpumask_t numa_cpumask_lookup_table[MAX_NUMNODES]; -#ifdef CONFIG_NEED_MULTIPLE_NODES - struct mdesc_mblock { u64 base; u64 size; @@ -838,17 +888,21 @@ static void __init allocate_node_data(int nid) static void init_node_masks_nonnuma(void) { +#ifdef CONFIG_NEED_MULTIPLE_NODES int i; +#endif numadbg("Initializing tables for non-numa.\n"); node_masks[0].mask = node_masks[0].val = 0; num_node_masks = 1; +#ifdef CONFIG_NEED_MULTIPLE_NODES for (i = 0; i < NR_CPUS; i++) numa_cpu_lookup_table[i] = 0; cpumask_setall(&numa_cpumask_lookup_table[0]); +#endif } #ifdef CONFIG_NEED_MULTIPLE_NODES @@ -972,7 +1026,8 @@ static void __init add_node_ranges(void) "start[%lx] end[%lx]\n", nid, start, this_end); - memblock_set_node(start, this_end - start, nid); + memblock_set_node(start, this_end - start, + &memblock.memory, nid); start = this_end; } } @@ -1049,7 +1104,14 @@ static int __init grab_mblocks(struct mdesc_handle *md) m->size = *val; val = mdesc_get_property(md, node, "address-congruence-offset", NULL); - m->offset = *val; + + /* The address-congruence-offset property is optional. + * Explicity zero it be identifty this. + */ + if (val) + m->offset = *val; + else + m->offset = 0UL; numadbg("MBLOCK[%d]: base[%llx] size[%llx] offset[%llx]\n", count - 1, m->base, m->size, m->offset); @@ -1269,7 +1331,7 @@ static void __init bootmem_init_nonnuma(void) (top_of_ram - total_ram) >> 20); init_node_masks_nonnuma(); - memblock_set_node(0, (phys_addr_t)ULLONG_MAX, 0); + memblock_set_node(0, (phys_addr_t)ULLONG_MAX, &memblock.memory, 0); allocate_node_data(0); node_set_online(0); } @@ -1358,32 +1420,75 @@ static unsigned long __ref kernel_map_range(unsigned long pstart, extern unsigned int kvmap_linear_patch[1]; #endif /* CONFIG_DEBUG_PAGEALLOC */ -static void __init mark_kpte_bitmap(unsigned long start, unsigned long end) +static void __init kpte_set_val(unsigned long index, unsigned long val) { - const unsigned long shift_256MB = 28; - const unsigned long mask_256MB = ((1UL << shift_256MB) - 1UL); - const unsigned long size_256MB = (1UL << shift_256MB); + unsigned long *ptr = kpte_linear_bitmap; - while (start < end) { - long remains; + val <<= ((index % (BITS_PER_LONG / 2)) * 2); + ptr += (index / (BITS_PER_LONG / 2)); - remains = end - start; - if (remains < size_256MB) - break; + *ptr |= val; +} - if (start & mask_256MB) { - start = (start + size_256MB) & ~mask_256MB; - continue; - } +static const unsigned long kpte_shift_min = 28; /* 256MB */ +static const unsigned long kpte_shift_max = 34; /* 16GB */ +static const unsigned long kpte_shift_incr = 3; + +static unsigned long kpte_mark_using_shift(unsigned long start, unsigned long end, + unsigned long shift) +{ + unsigned long size = (1UL << shift); + unsigned long mask = (size - 1UL); + unsigned long remains = end - start; + unsigned long val; + + if (remains < size || (start & mask)) + return start; + + /* VAL maps: + * + * shift 28 --> kern_linear_pte_xor index 1 + * shift 31 --> kern_linear_pte_xor index 2 + * shift 34 --> kern_linear_pte_xor index 3 + */ + val = ((shift - kpte_shift_min) / kpte_shift_incr) + 1; + + remains &= ~mask; + if (shift != kpte_shift_max) + remains = size; + + while (remains) { + unsigned long index = start >> kpte_shift_min; + + kpte_set_val(index, val); + + start += 1UL << kpte_shift_min; + remains -= 1UL << kpte_shift_min; + } + + return start; +} + +static void __init mark_kpte_bitmap(unsigned long start, unsigned long end) +{ + unsigned long smallest_size, smallest_mask; + unsigned long s; - while (remains >= size_256MB) { - unsigned long index = start >> shift_256MB; + smallest_size = (1UL << kpte_shift_min); + smallest_mask = (smallest_size - 1UL); - __set_bit(index, kpte_linear_bitmap); + while (start < end) { + unsigned long orig_start = start; + + for (s = kpte_shift_max; s >= kpte_shift_min; s -= kpte_shift_incr) { + start = kpte_mark_using_shift(start, end, s); - start += size_256MB; - remains -= size_256MB; + if (start != orig_start) + break; } + + if (start == orig_start) + start = (start + smallest_size) & ~smallest_mask; } } @@ -1458,6 +1563,96 @@ unsigned long __init find_ecache_flush_span(unsigned long size) return ~0UL; } +unsigned long PAGE_OFFSET; +EXPORT_SYMBOL(PAGE_OFFSET); + +static void __init page_offset_shift_patch_one(unsigned int *insn, unsigned long phys_bits) +{ + unsigned long final_shift; + unsigned int val = *insn; + unsigned int cnt; + + /* We are patching in ilog2(max_supported_phys_address), and + * we are doing so in a manner similar to a relocation addend. + * That is, we are adding the shift value to whatever value + * is in the shift instruction count field already. + */ + cnt = (val & 0x3f); + val &= ~0x3f; + + /* If we are trying to shift >= 64 bits, clear the destination + * register. This can happen when phys_bits ends up being equal + * to MAX_PHYS_ADDRESS_BITS. + */ + final_shift = (cnt + (64 - phys_bits)); + if (final_shift >= 64) { + unsigned int rd = (val >> 25) & 0x1f; + + val = 0x80100000 | (rd << 25); + } else { + val |= final_shift; + } + *insn = val; + + __asm__ __volatile__("flush %0" + : /* no outputs */ + : "r" (insn)); +} + +static void __init page_offset_shift_patch(unsigned long phys_bits) +{ + extern unsigned int __page_offset_shift_patch; + extern unsigned int __page_offset_shift_patch_end; + unsigned int *p; + + p = &__page_offset_shift_patch; + while (p < &__page_offset_shift_patch_end) { + unsigned int *insn = (unsigned int *)(unsigned long)*p; + + page_offset_shift_patch_one(insn, phys_bits); + + p++; + } +} + +static void __init setup_page_offset(void) +{ + unsigned long max_phys_bits = 40; + + if (tlb_type == cheetah || tlb_type == cheetah_plus) { + max_phys_bits = 42; + } else if (tlb_type == hypervisor) { + switch (sun4v_chip_type) { + case SUN4V_CHIP_NIAGARA1: + case SUN4V_CHIP_NIAGARA2: + max_phys_bits = 39; + break; + case SUN4V_CHIP_NIAGARA3: + max_phys_bits = 43; + break; + case SUN4V_CHIP_NIAGARA4: + case SUN4V_CHIP_NIAGARA5: + case SUN4V_CHIP_SPARC64X: + default: + max_phys_bits = 47; + break; + } + } + + if (max_phys_bits > MAX_PHYS_ADDRESS_BITS) { + prom_printf("MAX_PHYS_ADDRESS_BITS is too small, need %lu\n", + max_phys_bits); + prom_halt(); + } + + PAGE_OFFSET = PAGE_OFFSET_BY_BITS(max_phys_bits); + + pr_info("PAGE_OFFSET is 0x%016lx (max_phys_bits == %lu)\n", + PAGE_OFFSET, max_phys_bits); + + page_offset_shift_patch(max_phys_bits); +} + static void __init tsb_phys_patch(void) { struct tsb_ldquad_phys_patch_entry *pquad; @@ -1577,13 +1772,16 @@ static void __init sun4v_ktsb_init(void) ktsb_descr[0].resv = 0; #ifndef CONFIG_DEBUG_PAGEALLOC - /* Second KTSB for 4MB/256MB mappings. */ + /* Second KTSB for 4MB/256MB/2GB/16GB mappings. */ ktsb_pa = (kern_base + ((unsigned long)&swapper_4m_tsb[0] - KERNBASE)); ktsb_descr[1].pgsz_idx = HV_PGSZ_IDX_4MB; - ktsb_descr[1].pgsz_mask = (HV_PGSZ_MASK_4MB | - HV_PGSZ_MASK_256MB); + ktsb_descr[1].pgsz_mask = ((HV_PGSZ_MASK_4MB | + HV_PGSZ_MASK_256MB | + HV_PGSZ_MASK_2GB | + HV_PGSZ_MASK_16GB) & + cpu_pgsz_mask); ktsb_descr[1].assoc = 1; ktsb_descr[1].num_ttes = KERNEL_TSB4M_NENTRIES; ktsb_descr[1].ctx_idx = 0; @@ -1592,7 +1790,7 @@ static void __init sun4v_ktsb_init(void) #endif } -void __cpuinit sun4v_ktsb_register(void) +void sun4v_ktsb_register(void) { unsigned long pa, ret; @@ -1606,10 +1804,51 @@ void __cpuinit sun4v_ktsb_register(void) } } +static void __init sun4u_linear_pte_xor_finalize(void) +{ +#ifndef CONFIG_DEBUG_PAGEALLOC + /* This is where we would add Panther support for + * 32MB and 256MB pages. + */ +#endif +} + +static void __init sun4v_linear_pte_xor_finalize(void) +{ +#ifndef CONFIG_DEBUG_PAGEALLOC + if (cpu_pgsz_mask & HV_PGSZ_MASK_256MB) { + kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZ256MB_4V) ^ + PAGE_OFFSET; + kern_linear_pte_xor[1] |= (_PAGE_CP_4V | _PAGE_CV_4V | + _PAGE_P_4V | _PAGE_W_4V); + } else { + kern_linear_pte_xor[1] = kern_linear_pte_xor[0]; + } + + if (cpu_pgsz_mask & HV_PGSZ_MASK_2GB) { + kern_linear_pte_xor[2] = (_PAGE_VALID | _PAGE_SZ2GB_4V) ^ + PAGE_OFFSET; + kern_linear_pte_xor[2] |= (_PAGE_CP_4V | _PAGE_CV_4V | + _PAGE_P_4V | _PAGE_W_4V); + } else { + kern_linear_pte_xor[2] = kern_linear_pte_xor[1]; + } + + if (cpu_pgsz_mask & HV_PGSZ_MASK_16GB) { + kern_linear_pte_xor[3] = (_PAGE_VALID | _PAGE_SZ16GB_4V) ^ + PAGE_OFFSET; + kern_linear_pte_xor[3] |= (_PAGE_CP_4V | _PAGE_CV_4V | + _PAGE_P_4V | _PAGE_W_4V); + } else { + kern_linear_pte_xor[3] = kern_linear_pte_xor[2]; + } +#endif +} + /* paging_init() sets up the page tables */ static unsigned long last_valid_pfn; -pgd_t swapper_pg_dir[2048]; +pgd_t swapper_pg_dir[PTRS_PER_PGD]; static void sun4u_pgprot_init(void); static void sun4v_pgprot_init(void); @@ -1620,6 +1859,8 @@ void __init paging_init(void) unsigned long real_end, i; int node; + setup_page_offset(); + /* These build time checkes make sure that the dcache_dirty_cpu() * page->flags usage will work. * @@ -1645,7 +1886,7 @@ void __init paging_init(void) BUILD_BUG_ON(NR_CPUS > 4096); - kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL; + kern_base = (prom_boot_mapping_phys_low >> ILOG2_4MB) << ILOG2_4MB; kern_size = (unsigned long)&_end - (unsigned long)KERNBASE; /* Invalidate both kernel TSBs. */ @@ -1665,10 +1906,8 @@ void __init paging_init(void) ktsb_phys_patch(); } - if (tlb_type == hypervisor) { + if (tlb_type == hypervisor) sun4v_patch_tlb_handlers(); - sun4v_ktsb_init(); - } /* Find available physical memory... * @@ -1703,7 +1942,7 @@ void __init paging_init(void) shift = kern_base + PAGE_OFFSET - ((unsigned long)KERNBASE); real_end = (unsigned long)_end; - num_kernel_image_mappings = DIV_ROUND_UP(real_end - KERNBASE, 1 << 22); + num_kernel_image_mappings = DIV_ROUND_UP(real_end - KERNBASE, 1 << ILOG2_4MB); printk("Kernel: Using %d locked TLB entries for main kernel image.\n", num_kernel_image_mappings); @@ -1727,9 +1966,6 @@ void __init paging_init(void) __flush_tlb_all(); - if (tlb_type == hypervisor) - sun4v_ktsb_register(); - prom_build_devicetree(); of_populate_present_mask(); #ifndef CONFIG_SMP @@ -1742,8 +1978,36 @@ void __init paging_init(void) #ifndef CONFIG_SMP mdesc_fill_in_cpu_data(cpu_all_mask); #endif + mdesc_get_page_sizes(cpu_all_mask, &cpu_pgsz_mask); + + sun4v_linear_pte_xor_finalize(); + + sun4v_ktsb_init(); + sun4v_ktsb_register(); + } else { + unsigned long impl, ver; + + cpu_pgsz_mask = (HV_PGSZ_MASK_8K | HV_PGSZ_MASK_64K | + HV_PGSZ_MASK_512K | HV_PGSZ_MASK_4MB); + + __asm__ __volatile__("rdpr %%ver, %0" : "=r" (ver)); + impl = ((ver >> 32) & 0xffff); + if (impl == PANTHER_IMPL) + cpu_pgsz_mask |= (HV_PGSZ_MASK_32MB | + HV_PGSZ_MASK_256MB); + + sun4u_linear_pte_xor_finalize(); } + /* Flush the TLBs and the 4M TSB so that the updated linear + * pte XOR settings are realized for all mappings. + */ + __flush_tlb_all(); +#ifndef CONFIG_DEBUG_PAGEALLOC + memset(swapper_4m_tsb, 0x40, sizeof(swapper_4m_tsb)); +#endif + __flush_tlb_all(); + /* Setup bootmem... */ last_valid_pfn = end_pfn = bootmem_init(phys_base); @@ -1777,7 +2041,7 @@ void __init paging_init(void) printk("Booting Linux...\n"); } -int __devinit page_in_phys_avail(unsigned long paddr) +int page_in_phys_avail(unsigned long paddr) { int i; @@ -1835,7 +2099,7 @@ static void __init setup_valid_addr_bitmap_from_pavail(unsigned long *bitmap) if (new_start <= old_start && new_end >= (old_start + PAGE_SIZE)) { - set_bit(old_start >> 22, bitmap); + set_bit(old_start >> ILOG2_4MB, bitmap); goto do_next_page; } } @@ -1867,15 +2131,24 @@ static void __init patch_tlb_miss_handler_bitmap(void) flushi(&valid_addr_bitmap_insn[0]); } +static void __init register_page_bootmem_info(void) +{ +#ifdef CONFIG_NEED_MULTIPLE_NODES + int i; + + for_each_online_node(i) + if (NODE_DATA(i)->node_spanned_pages) + register_page_bootmem_info_node(NODE_DATA(i)); +#endif +} void __init mem_init(void) { - unsigned long codepages, datapages, initpages; unsigned long addr, last; addr = PAGE_OFFSET + kern_base; last = PAGE_ALIGN(kern_size) + addr; while (addr < last) { - set_bit(__pa(addr) >> 22, sparc64_valid_addr_bitmap); + set_bit(__pa(addr) >> ILOG2_4MB, sparc64_valid_addr_bitmap); addr += PAGE_SIZE; } @@ -1884,26 +2157,8 @@ void __init mem_init(void) high_memory = __va(last_valid_pfn << PAGE_SHIFT); -#ifdef CONFIG_NEED_MULTIPLE_NODES - { - int i; - for_each_online_node(i) { - if (NODE_DATA(i)->node_spanned_pages != 0) { - totalram_pages += - free_all_bootmem_node(NODE_DATA(i)); - } - } - totalram_pages += free_low_memory_core_early(MAX_NUMNODES); - } -#else - totalram_pages = free_all_bootmem(); -#endif - - /* We subtract one to account for the mem_map_zero page - * allocated below. - */ - totalram_pages -= 1; - num_physpages = totalram_pages; + register_page_bootmem_info(); + free_all_bootmem(); /* * Set up the zero page, mark it reserved, so that page count @@ -1914,21 +2169,9 @@ void __init mem_init(void) prom_printf("paging_init: Cannot alloc zero page.\n"); prom_halt(); } - SetPageReserved(mem_map_zero); + mark_page_reserved(mem_map_zero); - codepages = (((unsigned long) _etext) - ((unsigned long) _start)); - codepages = PAGE_ALIGN(codepages) >> PAGE_SHIFT; - datapages = (((unsigned long) _edata) - ((unsigned long) _etext)); - datapages = PAGE_ALIGN(datapages) >> PAGE_SHIFT; - initpages = (((unsigned long) __init_end) - ((unsigned long) __init_begin)); - initpages = PAGE_ALIGN(initpages) >> PAGE_SHIFT; - - printk("Memory: %luk available (%ldk kernel code, %ldk data, %ldk init) [%016lx,%016lx]\n", - nr_free_pages() << (PAGE_SHIFT-10), - codepages << (PAGE_SHIFT-10), - datapages << (PAGE_SHIFT-10), - initpages << (PAGE_SHIFT-10), - PAGE_OFFSET, (last_valid_pfn << PAGE_SHIFT)); + mem_init_print_info(NULL); if (tlb_type == cheetah || tlb_type == cheetah_plus) cheetah_ecache_flush_init(); @@ -1954,39 +2197,22 @@ void free_initmem(void) initend = (unsigned long)(__init_end) & PAGE_MASK; for (; addr < initend; addr += PAGE_SIZE) { unsigned long page; - struct page *p; page = (addr + ((unsigned long) __va(kern_base)) - ((unsigned long) KERNBASE)); memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE); - if (do_free) { - p = virt_to_page(page); - - ClearPageReserved(p); - init_page_count(p); - __free_page(p); - num_physpages++; - totalram_pages++; - } + if (do_free) + free_reserved_page(virt_to_page(page)); } } #ifdef CONFIG_BLK_DEV_INITRD void free_initrd_mem(unsigned long start, unsigned long end) { - if (start < end) - printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); - for (; start < end; start += PAGE_SIZE) { - struct page *p = virt_to_page(start); - - ClearPageReserved(p); - init_page_count(p); - __free_page(p); - num_physpages++; - totalram_pages++; - } + free_reserved_area((void *)start, (void *)end, POISON_FREE_INITMEM, + "initrd"); } #endif @@ -2020,10 +2246,12 @@ EXPORT_SYMBOL(_PAGE_CACHE); #ifdef CONFIG_SPARSEMEM_VMEMMAP unsigned long vmemmap_table[VMEMMAP_SIZE]; -int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node) +static long __meminitdata addr_start, addr_end; +static int __meminitdata node_start; + +int __meminit vmemmap_populate(unsigned long vstart, unsigned long vend, + int node) { - unsigned long vstart = (unsigned long) start; - unsigned long vend = (unsigned long) (start + nr); unsigned long phys_start = (vstart - VMEMMAP_BASE); unsigned long phys_end = (vend - VMEMMAP_BASE); unsigned long addr = phys_start & VMEMMAP_CHUNK_MASK; @@ -2044,21 +2272,41 @@ int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node) void *block; if (!(*vmem_pp & _PAGE_VALID)) { - block = vmemmap_alloc_block(1UL << 22, node); + block = vmemmap_alloc_block(1UL << ILOG2_4MB, node); if (!block) return -ENOMEM; *vmem_pp = pte_base | __pa(block); - printk(KERN_INFO "[%p-%p] page_structs=%lu " - "node=%d entry=%lu/%lu\n", start, block, nr, - node, - addr >> VMEMMAP_CHUNK_SHIFT, - VMEMMAP_SIZE); + /* check to see if we have contiguous blocks */ + if (addr_end != addr || node_start != node) { + if (addr_start) + printk(KERN_DEBUG " [%lx-%lx] on node %d\n", + addr_start, addr_end-1, node_start); + addr_start = addr; + node_start = node; + } + addr_end = addr + VMEMMAP_CHUNK; } } return 0; } + +void __meminit vmemmap_populate_print_last(void) +{ + if (addr_start) { + printk(KERN_DEBUG " [%lx-%lx] on node %d\n", + addr_start, addr_end-1, node_start); + addr_start = 0; + addr_end = 0; + node_start = 0; + } +} + +void vmemmap_free(unsigned long start, unsigned long end) +{ +} + #endif /* CONFIG_SPARSEMEM_VMEMMAP */ static void prot_init_common(unsigned long page_none, @@ -2092,6 +2340,7 @@ static void __init sun4u_pgprot_init(void) { unsigned long page_none, page_shared, page_copy, page_readonly; unsigned long page_exec_bit; + int i; PAGE_KERNEL = __pgprot (_PAGE_PRESENT_4U | _PAGE_VALID | _PAGE_CACHE_4U | _PAGE_P_4U | @@ -2110,19 +2359,17 @@ static void __init sun4u_pgprot_init(void) __ACCESS_BITS_4U | _PAGE_E_4U); #ifdef CONFIG_DEBUG_PAGEALLOC - kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4U) ^ - 0xfffff80000000000UL; + kern_linear_pte_xor[0] = _PAGE_VALID ^ PAGE_OFFSET; #else kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4U) ^ - 0xfffff80000000000UL; + PAGE_OFFSET; #endif kern_linear_pte_xor[0] |= (_PAGE_CP_4U | _PAGE_CV_4U | _PAGE_P_4U | _PAGE_W_4U); - /* XXX Should use 256MB on Panther. XXX */ - kern_linear_pte_xor[1] = kern_linear_pte_xor[0]; + for (i = 1; i < 4; i++) + kern_linear_pte_xor[i] = kern_linear_pte_xor[0]; - _PAGE_SZBITS = _PAGE_SZBITS_4U; _PAGE_ALL_SZ_BITS = (_PAGE_SZ4MB_4U | _PAGE_SZ512K_4U | _PAGE_SZ64K_4U | _PAGE_SZ8K_4U | _PAGE_SZ32MB_4U | _PAGE_SZ256MB_4U); @@ -2146,6 +2393,7 @@ static void __init sun4v_pgprot_init(void) { unsigned long page_none, page_shared, page_copy, page_readonly; unsigned long page_exec_bit; + int i; PAGE_KERNEL = __pgprot (_PAGE_PRESENT_4V | _PAGE_VALID | _PAGE_CACHE_4V | _PAGE_P_4V | @@ -2158,29 +2406,20 @@ static void __init sun4v_pgprot_init(void) _PAGE_CACHE = _PAGE_CACHE_4V; #ifdef CONFIG_DEBUG_PAGEALLOC - kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^ - 0xfffff80000000000UL; + kern_linear_pte_xor[0] = _PAGE_VALID ^ PAGE_OFFSET; #else kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4V) ^ - 0xfffff80000000000UL; + PAGE_OFFSET; #endif kern_linear_pte_xor[0] |= (_PAGE_CP_4V | _PAGE_CV_4V | _PAGE_P_4V | _PAGE_W_4V); -#ifdef CONFIG_DEBUG_PAGEALLOC - kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^ - 0xfffff80000000000UL; -#else - kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZ256MB_4V) ^ - 0xfffff80000000000UL; -#endif - kern_linear_pte_xor[1] |= (_PAGE_CP_4V | _PAGE_CV_4V | - _PAGE_P_4V | _PAGE_W_4V); + for (i = 1; i < 4; i++) + kern_linear_pte_xor[i] = kern_linear_pte_xor[0]; pg_iobits = (_PAGE_VALID | _PAGE_PRESENT_4V | __DIRTY_BITS_4V | __ACCESS_BITS_4V | _PAGE_E_4V); - _PAGE_SZBITS = _PAGE_SZBITS_4V; _PAGE_ALL_SZ_BITS = (_PAGE_SZ16GB_4V | _PAGE_SZ2GB_4V | _PAGE_SZ256MB_4V | _PAGE_SZ32MB_4V | _PAGE_SZ4MB_4V | _PAGE_SZ512K_4V | @@ -2313,3 +2552,150 @@ void __flush_tlb_all(void) __asm__ __volatile__("wrpr %0, 0, %%pstate" : : "r" (pstate)); } + +pte_t *pte_alloc_one_kernel(struct mm_struct *mm, + unsigned long address) +{ + struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK | + __GFP_REPEAT | __GFP_ZERO); + pte_t *pte = NULL; + + if (page) + pte = (pte_t *) page_address(page); + + return pte; +} + +pgtable_t pte_alloc_one(struct mm_struct *mm, + unsigned long address) +{ + struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK | + __GFP_REPEAT | __GFP_ZERO); + if (!page) + return NULL; + if (!pgtable_page_ctor(page)) { + free_hot_cold_page(page, 0); + return NULL; + } + return (pte_t *) page_address(page); +} + +void pte_free_kernel(struct mm_struct *mm, pte_t *pte) +{ + free_page((unsigned long)pte); +} + +static void __pte_free(pgtable_t pte) +{ + struct page *page = virt_to_page(pte); + + pgtable_page_dtor(page); + __free_page(page); +} + +void pte_free(struct mm_struct *mm, pgtable_t pte) +{ + __pte_free(pte); +} + +void pgtable_free(void *table, bool is_page) +{ + if (is_page) + __pte_free(table); + else + kmem_cache_free(pgtable_cache, table); +} + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr, + pmd_t *pmd) +{ + unsigned long pte, flags; + struct mm_struct *mm; + pmd_t entry = *pmd; + + if (!pmd_large(entry) || !pmd_young(entry)) + return; + + pte = pmd_val(entry); + + /* We are fabricating 8MB pages using 4MB real hw pages. */ + pte |= (addr & (1UL << REAL_HPAGE_SHIFT)); + + mm = vma->vm_mm; + + spin_lock_irqsave(&mm->context.lock, flags); + + if (mm->context.tsb_block[MM_TSB_HUGE].tsb != NULL) + __update_mmu_tsb_insert(mm, MM_TSB_HUGE, REAL_HPAGE_SHIFT, + addr, pte); + + spin_unlock_irqrestore(&mm->context.lock, flags); +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ + +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) +static void context_reload(void *__data) +{ + struct mm_struct *mm = __data; + + if (mm == current->mm) + load_secondary_context(mm); +} + +void hugetlb_setup(struct pt_regs *regs) +{ + struct mm_struct *mm = current->mm; + struct tsb_config *tp; + + if (in_atomic() || !mm) { + const struct exception_table_entry *entry; + + entry = search_exception_tables(regs->tpc); + if (entry) { + regs->tpc = entry->fixup; + regs->tnpc = regs->tpc + 4; + return; + } + pr_alert("Unexpected HugeTLB setup in atomic context.\n"); + die_if_kernel("HugeTSB in atomic", regs); + } + + tp = &mm->context.tsb_block[MM_TSB_HUGE]; + if (likely(tp->tsb == NULL)) + tsb_grow(mm, MM_TSB_HUGE, 0); + + tsb_context_switch(mm); + smp_tsb_sync(mm); + + /* On UltraSPARC-III+ and later, configure the second half of + * the Data-TLB for huge pages. + */ + if (tlb_type == cheetah_plus) { + unsigned long ctx; + + spin_lock(&ctx_alloc_lock); + ctx = mm->context.sparc64_ctx_val; + ctx &= ~CTX_PGSZ_MASK; + ctx |= CTX_PGSZ_BASE << CTX_PGSZ0_SHIFT; + ctx |= CTX_PGSZ_HUGE << CTX_PGSZ1_SHIFT; + + if (ctx != mm->context.sparc64_ctx_val) { + /* When changing the page size fields, we + * must perform a context flush so that no + * stale entries match. This flush must + * occur with the original context register + * settings. + */ + do_flush_tlb_mm(mm); + + /* Reload the context register of all processors + * also executing in this address space. + */ + mm->context.sparc64_ctx_val = ctx; + on_each_cpu(context_reload, mm, 0); + } + spin_unlock(&ctx_alloc_lock); + } +} +#endif diff --git a/arch/sparc/mm/init_64.h b/arch/sparc/mm/init_64.h index 3e1ac8b96ca..0668b364f44 100644 --- a/arch/sparc/mm/init_64.h +++ b/arch/sparc/mm/init_64.h @@ -1,25 +1,27 @@ #ifndef _SPARC64_MM_INIT_H #define _SPARC64_MM_INIT_H +#include <asm/page.h> + /* Most of the symbols in this file are defined in init.c and * marked non-static so that assembler code can get at them. */ -#define MAX_PHYS_ADDRESS (1UL << 41UL) +#define MAX_PHYS_ADDRESS (1UL << MAX_PHYS_ADDRESS_BITS) #define KPTE_BITMAP_CHUNK_SZ (256UL * 1024UL * 1024UL) #define KPTE_BITMAP_BYTES \ - ((MAX_PHYS_ADDRESS / KPTE_BITMAP_CHUNK_SZ) / 8) + ((MAX_PHYS_ADDRESS / KPTE_BITMAP_CHUNK_SZ) / 4) #define VALID_ADDR_BITMAP_CHUNK_SZ (4UL * 1024UL * 1024UL) #define VALID_ADDR_BITMAP_BYTES \ ((MAX_PHYS_ADDRESS / VALID_ADDR_BITMAP_CHUNK_SZ) / 8) -extern unsigned long kern_linear_pte_xor[2]; +extern unsigned long kern_linear_pte_xor[4]; extern unsigned long kpte_linear_bitmap[KPTE_BITMAP_BYTES / sizeof(unsigned long)]; extern unsigned int sparc64_highest_unlocked_tlb_ent; extern unsigned long sparc64_kern_pri_context; extern unsigned long sparc64_kern_pri_nuc_bits; extern unsigned long sparc64_kern_sec_context; -extern void mmu_info(struct seq_file *m); +void mmu_info(struct seq_file *m); struct linux_prom_translation { unsigned long virt; @@ -34,7 +36,7 @@ extern unsigned int prom_trans_ents; /* Exported for SMP bootup purposes. */ extern unsigned long kern_locked_tte_data; -extern void prom_world(int enter); +void prom_world(int enter); #ifdef CONFIG_SPARSEMEM_VMEMMAP #define VMEMMAP_CHUNK_SHIFT 22 diff --git a/arch/sparc/mm/io-unit.c b/arch/sparc/mm/io-unit.c index eb99862e965..f311bf21901 100644 --- a/arch/sparc/mm/io-unit.c +++ b/arch/sparc/mm/io-unit.c @@ -25,6 +25,8 @@ #include <asm/dma.h> #include <asm/oplib.h> +#include "mm_32.h" + /* #define IOUNIT_DEBUG */ #ifdef IOUNIT_DEBUG #define IOD(x) printk(x) @@ -38,7 +40,8 @@ static void __init iounit_iommu_init(struct platform_device *op) { struct iounit_struct *iounit; - iopte_t *xpt, *xptend; + iopte_t __iomem *xpt; + iopte_t __iomem *xptend; iounit = kzalloc(sizeof(struct iounit_struct), GFP_ATOMIC); if (!iounit) { @@ -62,10 +65,10 @@ static void __init iounit_iommu_init(struct platform_device *op) op->dev.archdata.iommu = iounit; iounit->page_table = xpt; spin_lock_init(&iounit->lock); - - for (xptend = iounit->page_table + (16 * PAGE_SIZE) / sizeof(iopte_t); - xpt < xptend;) - iopte_val(*xpt++) = 0; + + xptend = iounit->page_table + (16 * PAGE_SIZE) / sizeof(iopte_t); + for (; xpt < xptend; xpt++) + sbus_writel(0, xpt); } static int __init iounit_init(void) @@ -130,7 +133,7 @@ nexti: scan = find_next_zero_bit(iounit->bmap, limit, scan); vaddr = IOUNIT_DMA_BASE + (scan << PAGE_SHIFT) + (vaddr & ~PAGE_MASK); for (k = 0; k < npages; k++, iopte = __iopte(iopte_val(iopte) + 0x100), scan++) { set_bit(scan, iounit->bmap); - iounit->page_table[scan] = iopte; + sbus_writel(iopte, &iounit->page_table[scan]); } IOD(("%08lx\n", vaddr)); return vaddr; @@ -202,7 +205,7 @@ static int iounit_map_dma_area(struct device *dev, dma_addr_t *pba, unsigned lon struct iounit_struct *iounit = dev->archdata.iommu; unsigned long page, end; pgprot_t dvma_prot; - iopte_t *iopte; + iopte_t __iomem *iopte; *pba = addr; @@ -224,8 +227,8 @@ static int iounit_map_dma_area(struct device *dev, dma_addr_t *pba, unsigned lon i = ((addr - IOUNIT_DMA_BASE) >> PAGE_SHIFT); - iopte = (iopte_t *)(iounit->page_table + i); - *iopte = MKIOPTE(__pa(page)); + iopte = iounit->page_table + i; + sbus_writel(MKIOPTE(__pa(page)), iopte); } addr += PAGE_SIZE; va += PAGE_SIZE; diff --git a/arch/sparc/mm/iommu.c b/arch/sparc/mm/iommu.c index a8a58cad9d2..491511d37e3 100644 --- a/arch/sparc/mm/iommu.c +++ b/arch/sparc/mm/iommu.c @@ -27,6 +27,8 @@ #include <asm/iommu.h> #include <asm/dma.h> +#include "mm_32.h" + /* * This can be sized dynamically, but we will do this * only when we have a guidance about actual I/O pressures. @@ -34,12 +36,9 @@ #define IOMMU_RNGE IOMMU_RNGE_256MB #define IOMMU_START 0xF0000000 #define IOMMU_WINSIZE (256*1024*1024U) -#define IOMMU_NPTES (IOMMU_WINSIZE/PAGE_SIZE) /* 64K PTEs, 265KB */ +#define IOMMU_NPTES (IOMMU_WINSIZE/PAGE_SIZE) /* 64K PTEs, 256KB */ #define IOMMU_ORDER 6 /* 4096 * (1<<6) */ -/* srmmu.c */ -extern int viking_mxcc_present; -extern int flush_page_for_dma_global; static int viking_flush; /* viking.S */ extern void viking_flush_page(unsigned long page); @@ -59,6 +58,8 @@ static void __init sbus_iommu_init(struct platform_device *op) struct iommu_struct *iommu; unsigned int impl, vers; unsigned long *bitmap; + unsigned long control; + unsigned long base; unsigned long tmp; iommu = kmalloc(sizeof(struct iommu_struct), GFP_KERNEL); @@ -73,12 +74,14 @@ static void __init sbus_iommu_init(struct platform_device *op) prom_printf("Cannot map IOMMU registers\n"); prom_halt(); } - impl = (iommu->regs->control & IOMMU_CTRL_IMPL) >> 28; - vers = (iommu->regs->control & IOMMU_CTRL_VERS) >> 24; - tmp = iommu->regs->control; - tmp &= ~(IOMMU_CTRL_RNGE); - tmp |= (IOMMU_RNGE_256MB | IOMMU_CTRL_ENAB); - iommu->regs->control = tmp; + + control = sbus_readl(&iommu->regs->control); + impl = (control & IOMMU_CTRL_IMPL) >> 28; + vers = (control & IOMMU_CTRL_VERS) >> 24; + control &= ~(IOMMU_CTRL_RNGE); + control |= (IOMMU_RNGE_256MB | IOMMU_CTRL_ENAB); + sbus_writel(control, &iommu->regs->control); + iommu_invalidate(iommu->regs); iommu->start = IOMMU_START; iommu->end = 0xffffffff; @@ -90,8 +93,8 @@ static void __init sbus_iommu_init(struct platform_device *op) it to us. */ tmp = __get_free_pages(GFP_KERNEL, IOMMU_ORDER); if (!tmp) { - prom_printf("Unable to allocate iommu table [0x%08x]\n", - IOMMU_NPTES*sizeof(iopte_t)); + prom_printf("Unable to allocate iommu table [0x%lx]\n", + IOMMU_NPTES * sizeof(iopte_t)); prom_halt(); } iommu->page_table = (iopte_t *)tmp; @@ -100,7 +103,9 @@ static void __init sbus_iommu_init(struct platform_device *op) memset(iommu->page_table, 0, IOMMU_NPTES*sizeof(iopte_t)); flush_cache_all(); flush_tlb_all(); - iommu->regs->base = __pa((unsigned long) iommu->page_table) >> 4; + + base = __pa((unsigned long)iommu->page_table) >> 4; + sbus_writel(base, &iommu->regs->base); iommu_invalidate(iommu->regs); bitmap = kmalloc(IOMMU_NPTES>>3, GFP_KERNEL); diff --git a/arch/sparc/mm/leon_mm.c b/arch/sparc/mm/leon_mm.c index 5bed085a2c1..3b17b6f7895 100644 --- a/arch/sparc/mm/leon_mm.c +++ b/arch/sparc/mm/leon_mm.c @@ -15,10 +15,10 @@ #include <asm/leon.h> #include <asm/tlbflush.h> -#include "srmmu.h" +#include "mm_32.h" int leon_flush_during_switch = 1; -int srmmu_swprobe_trace; +static int srmmu_swprobe_trace; static inline unsigned long leon_get_ctable_ptr(void) { diff --git a/arch/sparc/mm/mm_32.h b/arch/sparc/mm/mm_32.h new file mode 100644 index 00000000000..a6c27ca9a72 --- /dev/null +++ b/arch/sparc/mm/mm_32.h @@ -0,0 +1,24 @@ +/* fault_32.c - visible as they are called from assembler */ +asmlinkage int lookup_fault(unsigned long pc, unsigned long ret_pc, + unsigned long address); +asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write, + unsigned long address); + +void window_overflow_fault(void); +void window_underflow_fault(unsigned long sp); +void window_ret_fault(struct pt_regs *regs); + +/* srmmu.c */ +extern char *srmmu_name; +extern int viking_mxcc_present; +extern int flush_page_for_dma_global; + +extern void (*poke_srmmu)(void); + +void __init srmmu_paging_init(void); + +/* iommu.c */ +void ld_mmu_iommu(void); + +/* io-unit.c */ +void ld_mmu_iounit(void); diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c index 62e3f577330..be65f035d18 100644 --- a/arch/sparc/mm/srmmu.c +++ b/arch/sparc/mm/srmmu.c @@ -8,57 +8,54 @@ * Copyright (C) 1999,2000 Anton Blanchard (anton@samba.org) */ -#include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/vmalloc.h> -#include <linux/pagemap.h> -#include <linux/init.h> +#include <linux/seq_file.h> #include <linux/spinlock.h> #include <linux/bootmem.h> -#include <linux/fs.h> -#include <linux/seq_file.h> +#include <linux/pagemap.h> +#include <linux/vmalloc.h> #include <linux/kdebug.h> +#include <linux/export.h> +#include <linux/kernel.h> +#include <linux/init.h> #include <linux/log2.h> #include <linux/gfp.h> +#include <linux/fs.h> +#include <linux/mm.h> -#include <asm/bitext.h> -#include <asm/page.h> +#include <asm/mmu_context.h> +#include <asm/cacheflush.h> +#include <asm/tlbflush.h> +#include <asm/io-unit.h> #include <asm/pgalloc.h> #include <asm/pgtable.h> -#include <asm/io.h> +#include <asm/bitext.h> #include <asm/vaddrs.h> -#include <asm/traps.h> -#include <asm/smp.h> -#include <asm/mbus.h> #include <asm/cache.h> +#include <asm/traps.h> #include <asm/oplib.h> +#include <asm/mbus.h> +#include <asm/page.h> #include <asm/asi.h> #include <asm/msi.h> -#include <asm/mmu_context.h> -#include <asm/io-unit.h> -#include <asm/cacheflush.h> -#include <asm/tlbflush.h> +#include <asm/smp.h> +#include <asm/io.h> /* Now the cpu specific definitions. */ -#include <asm/viking.h> -#include <asm/mxcc.h> -#include <asm/ross.h> +#include <asm/turbosparc.h> #include <asm/tsunami.h> +#include <asm/viking.h> #include <asm/swift.h> -#include <asm/turbosparc.h> #include <asm/leon.h> +#include <asm/mxcc.h> +#include <asm/ross.h> -#include "srmmu.h" +#include "mm_32.h" enum mbus_module srmmu_modtype; static unsigned int hwbug_bitmask; int vac_cache_size; int vac_line_size; -struct ctx_list *ctx_list_pool; -struct ctx_list ctx_free; -struct ctx_list ctx_used; - extern struct resource sparc_iomap; extern unsigned long last_valid_pfn; @@ -66,6 +63,7 @@ extern unsigned long last_valid_pfn; static pgd_t *srmmu_swapper_pg_dir; const struct sparc32_cachetlb_ops *sparc32_cachetlb_ops; +EXPORT_SYMBOL(sparc32_cachetlb_ops); #ifdef CONFIG_SMP const struct sparc32_cachetlb_ops *local_ops; @@ -102,7 +100,6 @@ static unsigned long srmmu_nocache_end; #define SRMMU_NOCACHE_ALIGN_MAX (sizeof(ctxd_t)*SRMMU_MAX_CONTEXTS) void *srmmu_nocache_pool; -void *srmmu_nocache_bitmap; static struct bit_map srmmu_nocache_map; static inline int srmmu_pmd_none(pmd_t pmd) @@ -136,8 +133,8 @@ void pmd_populate(struct mm_struct *mm, pmd_t *pmdp, struct page *ptep) } } -/* Find an entry in the third-level page table.. */ -pte_t *pte_offset_kernel(pmd_t * dir, unsigned long address) +/* Find an entry in the third-level page table.. */ +pte_t *pte_offset_kernel(pmd_t *dir, unsigned long address) { void *pte; @@ -151,55 +148,61 @@ pte_t *pte_offset_kernel(pmd_t * dir, unsigned long address) * align: bytes, number to align at. * Returns the virtual address of the allocated area. */ -static unsigned long __srmmu_get_nocache(int size, int align) +static void *__srmmu_get_nocache(int size, int align) { int offset; + unsigned long addr; if (size < SRMMU_NOCACHE_BITMAP_SHIFT) { - printk("Size 0x%x too small for nocache request\n", size); + printk(KERN_ERR "Size 0x%x too small for nocache request\n", + size); size = SRMMU_NOCACHE_BITMAP_SHIFT; } - if (size & (SRMMU_NOCACHE_BITMAP_SHIFT-1)) { - printk("Size 0x%x unaligned int nocache request\n", size); - size += SRMMU_NOCACHE_BITMAP_SHIFT-1; + if (size & (SRMMU_NOCACHE_BITMAP_SHIFT - 1)) { + printk(KERN_ERR "Size 0x%x unaligned int nocache request\n", + size); + size += SRMMU_NOCACHE_BITMAP_SHIFT - 1; } BUG_ON(align > SRMMU_NOCACHE_ALIGN_MAX); offset = bit_map_string_get(&srmmu_nocache_map, - size >> SRMMU_NOCACHE_BITMAP_SHIFT, - align >> SRMMU_NOCACHE_BITMAP_SHIFT); + size >> SRMMU_NOCACHE_BITMAP_SHIFT, + align >> SRMMU_NOCACHE_BITMAP_SHIFT); if (offset == -1) { - printk("srmmu: out of nocache %d: %d/%d\n", - size, (int) srmmu_nocache_size, - srmmu_nocache_map.used << SRMMU_NOCACHE_BITMAP_SHIFT); - return 0; + printk(KERN_ERR "srmmu: out of nocache %d: %d/%d\n", + size, (int) srmmu_nocache_size, + srmmu_nocache_map.used << SRMMU_NOCACHE_BITMAP_SHIFT); + return NULL; } - return (SRMMU_NOCACHE_VADDR + (offset << SRMMU_NOCACHE_BITMAP_SHIFT)); + addr = SRMMU_NOCACHE_VADDR + (offset << SRMMU_NOCACHE_BITMAP_SHIFT); + return (void *)addr; } -unsigned long srmmu_get_nocache(int size, int align) +void *srmmu_get_nocache(int size, int align) { - unsigned long tmp; + void *tmp; tmp = __srmmu_get_nocache(size, align); if (tmp) - memset((void *)tmp, 0, size); + memset(tmp, 0, size); return tmp; } -void srmmu_free_nocache(unsigned long vaddr, int size) +void srmmu_free_nocache(void *addr, int size) { + unsigned long vaddr; int offset; + vaddr = (unsigned long)addr; if (vaddr < SRMMU_NOCACHE_VADDR) { printk("Vaddr %lx is smaller than nocache base 0x%lx\n", vaddr, (unsigned long)SRMMU_NOCACHE_VADDR); BUG(); } - if (vaddr+size > srmmu_nocache_end) { + if (vaddr + size > srmmu_nocache_end) { printk("Vaddr %lx is bigger than nocache end 0x%lx\n", vaddr, srmmu_nocache_end); BUG(); @@ -212,7 +215,7 @@ void srmmu_free_nocache(unsigned long vaddr, int size) printk("Size 0x%x is too small\n", size); BUG(); } - if (vaddr & (size-1)) { + if (vaddr & (size - 1)) { printk("Vaddr %lx is not aligned to size 0x%x\n", vaddr, size); BUG(); } @@ -226,13 +229,23 @@ void srmmu_free_nocache(unsigned long vaddr, int size) static void srmmu_early_allocate_ptable_skeleton(unsigned long start, unsigned long end); -extern unsigned long probe_memory(void); /* in fault.c */ +/* Return how much physical memory we have. */ +static unsigned long __init probe_memory(void) +{ + unsigned long total = 0; + int i; + + for (i = 0; sp_banks[i].num_bytes; i++) + total += sp_banks[i].num_bytes; + + return total; +} /* * Reserve nocache dynamically proportionally to the amount of * system RAM. -- Tomas Szepe <szepe@pinerecords.com>, June 2002 */ -static void srmmu_nocache_calcsize(void) +static void __init srmmu_nocache_calcsize(void) { unsigned long sysmemavail = probe_memory() / 1024; int srmmu_nocache_npages; @@ -255,6 +268,7 @@ static void srmmu_nocache_calcsize(void) static void __init srmmu_nocache_init(void) { + void *srmmu_nocache_bitmap; unsigned int bitmap_bits; pgd_t *pgd; pmd_t *pmd; @@ -268,10 +282,12 @@ static void __init srmmu_nocache_init(void) SRMMU_NOCACHE_ALIGN_MAX, 0UL); memset(srmmu_nocache_pool, 0, srmmu_nocache_size); - srmmu_nocache_bitmap = __alloc_bootmem(bitmap_bits >> 3, SMP_CACHE_BYTES, 0UL); + srmmu_nocache_bitmap = + __alloc_bootmem(BITS_TO_LONGS(bitmap_bits) * sizeof(long), + SMP_CACHE_BYTES, 0UL); bit_map_init(&srmmu_nocache_map, srmmu_nocache_bitmap, bitmap_bits); - srmmu_swapper_pg_dir = (pgd_t *)__srmmu_get_nocache(SRMMU_PGD_TABLE_SIZE, SRMMU_PGD_TABLE_SIZE); + srmmu_swapper_pg_dir = __srmmu_get_nocache(SRMMU_PGD_TABLE_SIZE, SRMMU_PGD_TABLE_SIZE); memset(__nocache_fix(srmmu_swapper_pg_dir), 0, SRMMU_PGD_TABLE_SIZE); init_mm.pgd = srmmu_swapper_pg_dir; @@ -304,7 +320,7 @@ pgd_t *get_pgd_fast(void) { pgd_t *pgd = NULL; - pgd = (pgd_t *)__srmmu_get_nocache(SRMMU_PGD_TABLE_SIZE, SRMMU_PGD_TABLE_SIZE); + pgd = __srmmu_get_nocache(SRMMU_PGD_TABLE_SIZE, SRMMU_PGD_TABLE_SIZE); if (pgd) { pgd_t *init = pgd_offset_k(0); memset(pgd, 0, USER_PTRS_PER_PGD * sizeof(pgd_t)); @@ -330,8 +346,11 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) if ((pte = (unsigned long)pte_alloc_one_kernel(mm, address)) == 0) return NULL; - page = pfn_to_page( __nocache_pa(pte) >> PAGE_SHIFT ); - pgtable_page_ctor(page); + page = pfn_to_page(__nocache_pa(pte) >> PAGE_SHIFT); + if (!pgtable_page_ctor(page)) { + __free_page(page); + return NULL; + } return page; } @@ -344,18 +363,50 @@ void pte_free(struct mm_struct *mm, pgtable_t pte) if (p == 0) BUG(); p = page_to_pfn(pte) << PAGE_SHIFT; /* Physical address */ - p = (unsigned long) __nocache_va(p); /* Nocached virtual */ - srmmu_free_nocache(p, PTE_SIZE); + + /* free non cached virtual address*/ + srmmu_free_nocache(__nocache_va(p), PTE_SIZE); } -/* - */ +/* context handling - a dynamically sized pool is used */ +#define NO_CONTEXT -1 + +struct ctx_list { + struct ctx_list *next; + struct ctx_list *prev; + unsigned int ctx_number; + struct mm_struct *ctx_mm; +}; + +static struct ctx_list *ctx_list_pool; +static struct ctx_list ctx_free; +static struct ctx_list ctx_used; + +/* At boot time we determine the number of contexts */ +static int num_contexts; + +static inline void remove_from_ctx_list(struct ctx_list *entry) +{ + entry->next->prev = entry->prev; + entry->prev->next = entry->next; +} + +static inline void add_to_ctx_list(struct ctx_list *head, struct ctx_list *entry) +{ + entry->next = head; + (entry->prev = head->prev)->next = entry; + head->prev = entry; +} +#define add_to_free_ctxlist(entry) add_to_ctx_list(&ctx_free, entry) +#define add_to_used_ctxlist(entry) add_to_ctx_list(&ctx_used, entry) + + static inline void alloc_context(struct mm_struct *old_mm, struct mm_struct *mm) { struct ctx_list *ctxp; ctxp = ctx_free.next; - if(ctxp != &ctx_free) { + if (ctxp != &ctx_free) { remove_from_ctx_list(ctxp); add_to_used_ctxlist(ctxp); mm->context = ctxp->ctx_number; @@ -363,9 +414,9 @@ static inline void alloc_context(struct mm_struct *old_mm, struct mm_struct *mm) return; } ctxp = ctx_used.next; - if(ctxp->ctx_mm == old_mm) + if (ctxp->ctx_mm == old_mm) ctxp = ctxp->next; - if(ctxp == &ctx_used) + if (ctxp == &ctx_used) panic("out of mmu contexts"); flush_cache_mm(ctxp->ctx_mm); flush_tlb_mm(ctxp->ctx_mm); @@ -385,11 +436,31 @@ static inline void free_context(int context) add_to_free_ctxlist(ctx_old); } +static void __init sparc_context_init(int numctx) +{ + int ctx; + unsigned long size; + + size = numctx * sizeof(struct ctx_list); + ctx_list_pool = __alloc_bootmem(size, SMP_CACHE_BYTES, 0UL); + + for (ctx = 0; ctx < numctx; ctx++) { + struct ctx_list *clist; + + clist = (ctx_list_pool + ctx); + clist->ctx_number = ctx; + clist->ctx_mm = NULL; + } + ctx_free.next = ctx_free.prev = &ctx_free; + ctx_used.next = ctx_used.prev = &ctx_used; + for (ctx = 0; ctx < numctx; ctx++) + add_to_free_ctxlist(ctx_list_pool + ctx); +} void switch_mm(struct mm_struct *old_mm, struct mm_struct *mm, struct task_struct *tsk) { - if(mm->context == NO_CONTEXT) { + if (mm->context == NO_CONTEXT) { spin_lock(&srmmu_context_spinlock); alloc_context(old_mm, mm); spin_unlock(&srmmu_context_spinlock); @@ -407,7 +478,7 @@ void switch_mm(struct mm_struct *old_mm, struct mm_struct *mm, /* Low level IO area allocation on the SRMMU. */ static inline void srmmu_mapioaddr(unsigned long physaddr, - unsigned long virt_addr, int bus_type) + unsigned long virt_addr, int bus_type) { pgd_t *pgdp; pmd_t *pmdp; @@ -420,8 +491,7 @@ static inline void srmmu_mapioaddr(unsigned long physaddr, ptep = pte_offset_kernel(pmdp, virt_addr); tmp = (physaddr >> 4) | SRMMU_ET_PTE; - /* - * I need to test whether this is consistent over all + /* I need to test whether this is consistent over all * sun4m's. The bus_type represents the upper 4 bits of * 36-bit physical address on the I/O space lines... */ @@ -591,10 +661,10 @@ static void __init srmmu_early_allocate_ptable_skeleton(unsigned long start, pmd_t *pmdp; pte_t *ptep; - while(start < end) { + while (start < end) { pgdp = pgd_offset_k(start); if (pgd_none(*(pgd_t *)__nocache_fix(pgdp))) { - pmdp = (pmd_t *) __srmmu_get_nocache( + pmdp = __srmmu_get_nocache( SRMMU_PMD_TABLE_SIZE, SRMMU_PMD_TABLE_SIZE); if (pmdp == NULL) early_pgtable_allocfail("pmd"); @@ -602,8 +672,8 @@ static void __init srmmu_early_allocate_ptable_skeleton(unsigned long start, pgd_set(__nocache_fix(pgdp), pmdp); } pmdp = pmd_offset(__nocache_fix(pgdp), start); - if(srmmu_pmd_none(*(pmd_t *)__nocache_fix(pmdp))) { - ptep = (pte_t *)__srmmu_get_nocache(PTE_SIZE, PTE_SIZE); + if (srmmu_pmd_none(*(pmd_t *)__nocache_fix(pmdp))) { + ptep = __srmmu_get_nocache(PTE_SIZE, PTE_SIZE); if (ptep == NULL) early_pgtable_allocfail("pte"); memset(__nocache_fix(ptep), 0, PTE_SIZE); @@ -622,18 +692,18 @@ static void __init srmmu_allocate_ptable_skeleton(unsigned long start, pmd_t *pmdp; pte_t *ptep; - while(start < end) { + while (start < end) { pgdp = pgd_offset_k(start); if (pgd_none(*pgdp)) { - pmdp = (pmd_t *)__srmmu_get_nocache(SRMMU_PMD_TABLE_SIZE, SRMMU_PMD_TABLE_SIZE); + pmdp = __srmmu_get_nocache(SRMMU_PMD_TABLE_SIZE, SRMMU_PMD_TABLE_SIZE); if (pmdp == NULL) early_pgtable_allocfail("pmd"); memset(pmdp, 0, SRMMU_PMD_TABLE_SIZE); pgd_set(pgdp, pmdp); } pmdp = pmd_offset(pgdp, start); - if(srmmu_pmd_none(*pmdp)) { - ptep = (pte_t *) __srmmu_get_nocache(PTE_SIZE, + if (srmmu_pmd_none(*pmdp)) { + ptep = __srmmu_get_nocache(PTE_SIZE, PTE_SIZE); if (ptep == NULL) early_pgtable_allocfail("pte"); @@ -658,7 +728,7 @@ static inline unsigned long srmmu_probe(unsigned long vaddr) "=r" (retval) : "r" (vaddr | 0x400), "i" (ASI_M_FLUSH_PROBE)); } else { - retval = leon_swprobe(vaddr, 0); + retval = leon_swprobe(vaddr, NULL); } return retval; } @@ -671,72 +741,76 @@ static inline unsigned long srmmu_probe(unsigned long vaddr) static void __init srmmu_inherit_prom_mappings(unsigned long start, unsigned long end) { + unsigned long probed; + unsigned long addr; pgd_t *pgdp; pmd_t *pmdp; pte_t *ptep; - int what = 0; /* 0 = normal-pte, 1 = pmd-level pte, 2 = pgd-level pte */ - unsigned long prompte; + int what; /* 0 = normal-pte, 1 = pmd-level pte, 2 = pgd-level pte */ - while(start <= end) { + while (start <= end) { if (start == 0) break; /* probably wrap around */ - if(start == 0xfef00000) + if (start == 0xfef00000) start = KADB_DEBUGGER_BEGVM; - if(!(prompte = srmmu_probe(start))) { + probed = srmmu_probe(start); + if (!probed) { + /* continue probing until we find an entry */ start += PAGE_SIZE; continue; } - + /* A red snapper, see what it really is. */ what = 0; - - if(!(start & ~(SRMMU_REAL_PMD_MASK))) { - if(srmmu_probe((start-PAGE_SIZE) + SRMMU_REAL_PMD_SIZE) == prompte) + addr = start - PAGE_SIZE; + + if (!(start & ~(SRMMU_REAL_PMD_MASK))) { + if (srmmu_probe(addr + SRMMU_REAL_PMD_SIZE) == probed) what = 1; } - - if(!(start & ~(SRMMU_PGDIR_MASK))) { - if(srmmu_probe((start-PAGE_SIZE) + SRMMU_PGDIR_SIZE) == - prompte) + + if (!(start & ~(SRMMU_PGDIR_MASK))) { + if (srmmu_probe(addr + SRMMU_PGDIR_SIZE) == probed) what = 2; } - + pgdp = pgd_offset_k(start); - if(what == 2) { - *(pgd_t *)__nocache_fix(pgdp) = __pgd(prompte); + if (what == 2) { + *(pgd_t *)__nocache_fix(pgdp) = __pgd(probed); start += SRMMU_PGDIR_SIZE; continue; } if (pgd_none(*(pgd_t *)__nocache_fix(pgdp))) { - pmdp = (pmd_t *)__srmmu_get_nocache(SRMMU_PMD_TABLE_SIZE, SRMMU_PMD_TABLE_SIZE); + pmdp = __srmmu_get_nocache(SRMMU_PMD_TABLE_SIZE, + SRMMU_PMD_TABLE_SIZE); if (pmdp == NULL) early_pgtable_allocfail("pmd"); memset(__nocache_fix(pmdp), 0, SRMMU_PMD_TABLE_SIZE); pgd_set(__nocache_fix(pgdp), pmdp); } pmdp = pmd_offset(__nocache_fix(pgdp), start); - if(srmmu_pmd_none(*(pmd_t *)__nocache_fix(pmdp))) { - ptep = (pte_t *) __srmmu_get_nocache(PTE_SIZE, - PTE_SIZE); + if (srmmu_pmd_none(*(pmd_t *)__nocache_fix(pmdp))) { + ptep = __srmmu_get_nocache(PTE_SIZE, PTE_SIZE); if (ptep == NULL) early_pgtable_allocfail("pte"); memset(__nocache_fix(ptep), 0, PTE_SIZE); pmd_set(__nocache_fix(pmdp), ptep); } - if(what == 1) { - /* - * We bend the rule where all 16 PTPs in a pmd_t point + if (what == 1) { + /* We bend the rule where all 16 PTPs in a pmd_t point * inside the same PTE page, and we leak a perfectly * good hardware PTE piece. Alternatives seem worse. */ unsigned int x; /* Index of HW PMD in soft cluster */ + unsigned long *val; x = (start >> PMD_SHIFT) & 15; - *(unsigned long *)__nocache_fix(&pmdp->pmdv[x]) = prompte; + val = &pmdp->pmdv[x]; + *(unsigned long *)__nocache_fix(val) = probed; start += SRMMU_REAL_PMD_SIZE; continue; } ptep = pte_offset_kernel(__nocache_fix(pmdp), start); - *(pte_t *)__nocache_fix(ptep) = __pte(prompte); + *(pte_t *)__nocache_fix(ptep) = __pte(probed); start += PAGE_SIZE; } } @@ -765,18 +839,18 @@ static unsigned long __init map_spbank(unsigned long vbase, int sp_entry) if (vstart < min_vaddr || vstart >= max_vaddr) return vstart; - + if (vend > max_vaddr || vend < min_vaddr) vend = max_vaddr; - while(vstart < vend) { + while (vstart < vend) { do_large_mapping(vstart, pstart); vstart += SRMMU_PGDIR_SIZE; pstart += SRMMU_PGDIR_SIZE; } return vstart; } -static inline void map_kernel(void) +static void __init map_kernel(void) { int i; @@ -789,12 +863,7 @@ static inline void map_kernel(void) } } -/* Paging initialization on the Sparc Reference MMU. */ -extern void sparc_context_init(int); - -void (*poke_srmmu)(void) __cpuinitdata = NULL; - -extern unsigned long bootmem_init(unsigned long *pages_avail); +void (*poke_srmmu)(void) = NULL; void __init srmmu_paging_init(void) { @@ -806,6 +875,7 @@ void __init srmmu_paging_init(void) pte_t *pte; unsigned long pages_avail; + init_mm.context = (unsigned long) NO_CONTEXT; sparc_iomap.start = SUN4M_IOBASE_VADDR; /* 16MB of IOSPACE on all sun4m's. */ if (sparc_cpu_model == sun4d) @@ -814,9 +884,9 @@ void __init srmmu_paging_init(void) /* Find the number of contexts on the srmmu. */ cpunode = prom_getchild(prom_root_node); num_contexts = 0; - while(cpunode != 0) { + while (cpunode != 0) { prom_getstring(cpunode, "device_type", node_str, sizeof(node_str)); - if(!strcmp(node_str, "cpu")) { + if (!strcmp(node_str, "cpu")) { num_contexts = prom_getintdefault(cpunode, "mmu-nctx", 0x8); break; } @@ -824,7 +894,7 @@ void __init srmmu_paging_init(void) } } - if(!num_contexts) { + if (!num_contexts) { prom_printf("Something wrong, can't find cpu node in paging_init.\n"); prom_halt(); } @@ -834,14 +904,14 @@ void __init srmmu_paging_init(void) srmmu_nocache_calcsize(); srmmu_nocache_init(); - srmmu_inherit_prom_mappings(0xfe400000,(LINUX_OPPROM_ENDVM-PAGE_SIZE)); + srmmu_inherit_prom_mappings(0xfe400000, (LINUX_OPPROM_ENDVM - PAGE_SIZE)); map_kernel(); /* ctx table has to be physically aligned to its size */ - srmmu_context_table = (ctxd_t *)__srmmu_get_nocache(num_contexts*sizeof(ctxd_t), num_contexts*sizeof(ctxd_t)); + srmmu_context_table = __srmmu_get_nocache(num_contexts * sizeof(ctxd_t), num_contexts * sizeof(ctxd_t)); srmmu_ctx_table_phys = (ctxd_t *)__nocache_pa((unsigned long)srmmu_context_table); - for(i = 0; i < num_contexts; i++) + for (i = 0; i < num_contexts; i++) srmmu_ctxd_set((ctxd_t *)__nocache_fix(&srmmu_context_table[i]), srmmu_swapper_pg_dir); flush_cache_all(); @@ -897,7 +967,7 @@ void __init srmmu_paging_init(void) void mmu_info(struct seq_file *m) { - seq_printf(m, + seq_printf(m, "MMU type\t: %s\n" "contexts\t: %d\n" "nocache total\t: %ld\n" @@ -908,10 +978,16 @@ void mmu_info(struct seq_file *m) srmmu_nocache_map.used << SRMMU_NOCACHE_BITMAP_SHIFT); } +int init_new_context(struct task_struct *tsk, struct mm_struct *mm) +{ + mm->context = NO_CONTEXT; + return 0; +} + void destroy_context(struct mm_struct *mm) { - if(mm->context != NO_CONTEXT) { + if (mm->context != NO_CONTEXT) { flush_cache_mm(mm); srmmu_ctxd_set(&srmmu_context_table[mm->context], srmmu_swapper_pg_dir); flush_tlb_mm(mm); @@ -941,13 +1017,12 @@ static void __init init_vac_layout(void) #endif nd = prom_getchild(prom_root_node); - while((nd = prom_getsibling(nd)) != 0) { + while ((nd = prom_getsibling(nd)) != 0) { prom_getstring(nd, "device_type", node_str, sizeof(node_str)); - if(!strcmp(node_str, "cpu")) { + if (!strcmp(node_str, "cpu")) { vac_line_size = prom_getint(nd, "cache-line-size"); if (vac_line_size == -1) { - prom_printf("can't determine cache-line-size, " - "halting.\n"); + prom_printf("can't determine cache-line-size, halting.\n"); prom_halt(); } cache_lines = prom_getint(nd, "cache-nlines"); @@ -958,9 +1033,9 @@ static void __init init_vac_layout(void) vac_cache_size = cache_lines * vac_line_size; #ifdef CONFIG_SMP - if(vac_cache_size > max_size) + if (vac_cache_size > max_size) max_size = vac_cache_size; - if(vac_line_size < min_line_size) + if (vac_line_size < min_line_size) min_line_size = vac_line_size; //FIXME: cpus not contiguous!! cpu++; @@ -971,7 +1046,7 @@ static void __init init_vac_layout(void) #endif } } - if(nd == 0) { + if (nd == 0) { prom_printf("No CPU nodes found, halting.\n"); prom_halt(); } @@ -983,7 +1058,7 @@ static void __init init_vac_layout(void) (int)vac_cache_size, (int)vac_line_size); } -static void __cpuinit poke_hypersparc(void) +static void poke_hypersparc(void) { volatile unsigned long clear; unsigned long mreg = srmmu_get_mmureg(); @@ -1035,7 +1110,7 @@ static void __init init_hypersparc(void) hypersparc_setup_blockops(); } -static void __cpuinit poke_swift(void) +static void poke_swift(void) { unsigned long mreg; @@ -1082,7 +1157,7 @@ static void __init init_swift(void) "=r" (swift_rev) : "r" (SWIFT_MASKID_ADDR), "i" (ASI_M_BYPASS)); srmmu_name = "Fujitsu Swift"; - switch(swift_rev) { + switch (swift_rev) { case 0x11: case 0x20: case 0x23: @@ -1215,17 +1290,18 @@ static void turbosparc_flush_tlb_page(struct vm_area_struct *vma, unsigned long } -static void __cpuinit poke_turbosparc(void) +static void poke_turbosparc(void) { unsigned long mreg = srmmu_get_mmureg(); unsigned long ccreg; /* Clear any crap from the cache or else... */ turbosparc_flush_cache_all(); - mreg &= ~(TURBOSPARC_ICENABLE | TURBOSPARC_DCENABLE); /* Temporarily disable I & D caches */ + /* Temporarily disable I & D caches */ + mreg &= ~(TURBOSPARC_ICENABLE | TURBOSPARC_DCENABLE); mreg &= ~(TURBOSPARC_PCENABLE); /* Don't check parity */ srmmu_set_mmureg(mreg); - + ccreg = turbosparc_get_ccreg(); #ifdef TURBOSPARC_WRITEBACK @@ -1248,7 +1324,7 @@ static void __cpuinit poke_turbosparc(void) default: ccreg |= (TURBOSPARC_SCENABLE); } - turbosparc_set_ccreg (ccreg); + turbosparc_set_ccreg(ccreg); mreg |= (TURBOSPARC_ICENABLE | TURBOSPARC_DCENABLE); /* I & D caches on */ mreg |= (TURBOSPARC_ICSNOOP); /* Icache snooping on */ @@ -1277,7 +1353,7 @@ static void __init init_turbosparc(void) poke_srmmu = poke_turbosparc; } -static void __cpuinit poke_tsunami(void) +static void poke_tsunami(void) { unsigned long mreg = srmmu_get_mmureg(); @@ -1318,7 +1394,7 @@ static void __init init_tsunami(void) tsunami_setup_blockops(); } -static void __cpuinit poke_viking(void) +static void poke_viking(void) { unsigned long mreg = srmmu_get_mmureg(); static int smp_catch; @@ -1342,7 +1418,7 @@ static void __cpuinit poke_viking(void) unsigned long bpreg; mreg &= ~(VIKING_TCENABLE); - if(smp_catch++) { + if (smp_catch++) { /* Must disable mixed-cmd mode here for other cpu's. */ bpreg = viking_get_bpreg(); bpreg &= ~(VIKING_ACTION_MIX); @@ -1411,7 +1487,7 @@ static void __init init_viking(void) unsigned long mreg = srmmu_get_mmureg(); /* Ahhh, the viking. SRMMU VLSI abortion number two... */ - if(mreg & VIKING_MMODE) { + if (mreg & VIKING_MMODE) { srmmu_name = "TI Viking"; viking_mxcc_present = 0; msi_set_sync(); @@ -1467,8 +1543,8 @@ static void __init get_srmmu_type(void) } /* Second, check for HyperSparc or Cypress. */ - if(mod_typ == 1) { - switch(mod_rev) { + if (mod_typ == 1) { + switch (mod_rev) { case 7: /* UP or MP Hypersparc */ init_hypersparc(); @@ -1488,9 +1564,8 @@ static void __init get_srmmu_type(void) } return; } - - /* - * Now Fujitsu TurboSparc. It might happen that it is + + /* Now Fujitsu TurboSparc. It might happen that it is * in Swift emulation mode, so we will check later... */ if (psr_typ == 0 && psr_vers == 5) { @@ -1499,15 +1574,15 @@ static void __init get_srmmu_type(void) } /* Next check for Fujitsu Swift. */ - if(psr_typ == 0 && psr_vers == 4) { + if (psr_typ == 0 && psr_vers == 4) { phandle cpunode; char node_str[128]; /* Look if it is not a TurboSparc emulating Swift... */ cpunode = prom_getchild(prom_root_node); - while((cpunode = prom_getsibling(cpunode)) != 0) { + while ((cpunode = prom_getsibling(cpunode)) != 0) { prom_getstring(cpunode, "device_type", node_str, sizeof(node_str)); - if(!strcmp(node_str, "cpu")) { + if (!strcmp(node_str, "cpu")) { if (!prom_getintdefault(cpunode, "psr-implementation", 1) && prom_getintdefault(cpunode, "psr-version", 1) == 5) { init_turbosparc(); @@ -1516,13 +1591,13 @@ static void __init get_srmmu_type(void) break; } } - + init_swift(); return; } /* Now the Viking family of srmmu. */ - if(psr_typ == 4 && + if (psr_typ == 4 && ((psr_vers == 0) || ((psr_vers == 1) && (mod_typ == 0) && (mod_rev == 0)))) { init_viking(); @@ -1530,7 +1605,7 @@ static void __init get_srmmu_type(void) } /* Finally the Tsunami. */ - if(psr_typ == 4 && psr_vers == 1 && (mod_typ || mod_rev)) { + if (psr_typ == 4 && psr_vers == 1 && (mod_typ || mod_rev)) { init_tsunami(); return; } @@ -1694,9 +1769,6 @@ static struct sparc32_cachetlb_ops smp_cachetlb_ops = { /* Load up routines and constants for sun4m and sun4d mmu */ void __init load_mmu(void) { - extern void ld_mmu_iommu(void); - extern void ld_mmu_iounit(void); - /* Functions */ get_srmmu_type(); diff --git a/arch/sparc/mm/srmmu.h b/arch/sparc/mm/srmmu.h deleted file mode 100644 index 5703274ccf8..00000000000 --- a/arch/sparc/mm/srmmu.h +++ /dev/null @@ -1,4 +0,0 @@ -/* srmmu.c */ -extern char *srmmu_name; - -extern void (*poke_srmmu)(void); diff --git a/arch/sparc/mm/swift.S b/arch/sparc/mm/swift.S index c801c3953a0..5d2b88d3942 100644 --- a/arch/sparc/mm/swift.S +++ b/arch/sparc/mm/swift.S @@ -105,7 +105,7 @@ swift_flush_cache_mm_out: .globl swift_flush_cache_range swift_flush_cache_range: - ld [%o0 + 0x0], %o0 /* XXX vma->vm_mm, GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 sub %o2, %o1, %o2 sethi %hi(4096), %o3 cmp %o2, %o3 @@ -116,7 +116,7 @@ swift_flush_cache_range: .globl swift_flush_cache_page swift_flush_cache_page: - ld [%o0 + 0x0], %o0 /* XXX vma->vm_mm, GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 70: ld [%o0 + AOFF_mm_context], %g2 cmp %g2, -1 @@ -219,7 +219,7 @@ swift_flush_sig_insns: .globl swift_flush_tlb_range .globl swift_flush_tlb_all swift_flush_tlb_range: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 swift_flush_tlb_mm: ld [%o0 + AOFF_mm_context], %g2 cmp %g2, -1 @@ -233,7 +233,7 @@ swift_flush_tlb_all_out: .globl swift_flush_tlb_page swift_flush_tlb_page: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 mov SRMMU_CTX_REG, %g1 ld [%o0 + AOFF_mm_context], %o3 andn %o1, (PAGE_SIZE - 1), %o1 diff --git a/arch/sparc/mm/tlb.c b/arch/sparc/mm/tlb.c index b1f279cd00b..b89aba217e3 100644 --- a/arch/sparc/mm/tlb.c +++ b/arch/sparc/mm/tlb.c @@ -4,7 +4,6 @@ */ #include <linux/kernel.h> -#include <linux/init.h> #include <linux/percpu.h> #include <linux/mm.h> #include <linux/swap.h> @@ -24,11 +23,17 @@ static DEFINE_PER_CPU(struct tlb_batch, tlb_batch); void flush_tlb_pending(void) { struct tlb_batch *tb = &get_cpu_var(tlb_batch); + struct mm_struct *mm = tb->mm; - if (tb->tlb_nr) { - flush_tsb_user(tb); + if (!tb->tlb_nr) + goto out; - if (CTX_VALID(tb->mm->context)) { + flush_tsb_user(tb); + + if (CTX_VALID(mm->context)) { + if (tb->tlb_nr == 1) { + global_flush_tlb_page(mm, tb->vaddrs[0]); + } else { #ifdef CONFIG_SMP smp_flush_tlb_pending(tb->mm, tb->tlb_nr, &tb->vaddrs[0]); @@ -37,22 +42,68 @@ void flush_tlb_pending(void) tb->tlb_nr, &tb->vaddrs[0]); #endif } - tb->tlb_nr = 0; } + tb->tlb_nr = 0; + +out: put_cpu_var(tlb_batch); } -void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr, - pte_t *ptep, pte_t orig, int fullmm) +void arch_enter_lazy_mmu_mode(void) +{ + struct tlb_batch *tb = &__get_cpu_var(tlb_batch); + + tb->active = 1; +} + +void arch_leave_lazy_mmu_mode(void) +{ + struct tlb_batch *tb = &__get_cpu_var(tlb_batch); + + if (tb->tlb_nr) + flush_tlb_pending(); + tb->active = 0; +} + +static void tlb_batch_add_one(struct mm_struct *mm, unsigned long vaddr, + bool exec) { struct tlb_batch *tb = &get_cpu_var(tlb_batch); unsigned long nr; vaddr &= PAGE_MASK; - if (pte_exec(orig)) + if (exec) vaddr |= 0x1UL; + nr = tb->tlb_nr; + + if (unlikely(nr != 0 && mm != tb->mm)) { + flush_tlb_pending(); + nr = 0; + } + + if (!tb->active) { + flush_tsb_user_page(mm, vaddr); + global_flush_tlb_page(mm, vaddr); + goto out; + } + + if (nr == 0) + tb->mm = mm; + + tb->vaddrs[nr] = vaddr; + tb->tlb_nr = ++nr; + if (nr >= TLB_BATCH_NR) + flush_tlb_pending(); + +out: + put_cpu_var(tlb_batch); +} + +void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr, + pte_t *ptep, pte_t orig, int fullmm) +{ if (tlb_type != hypervisor && pte_dirty(orig)) { unsigned long paddr, pfn = pte_pfn(orig); @@ -77,26 +128,116 @@ void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr, } no_cache_flush: + if (!fullmm) + tlb_batch_add_one(mm, vaddr, pte_exec(orig)); +} + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static void tlb_batch_pmd_scan(struct mm_struct *mm, unsigned long vaddr, + pmd_t pmd) +{ + unsigned long end; + pte_t *pte; + + pte = pte_offset_map(&pmd, vaddr); + end = vaddr + HPAGE_SIZE; + while (vaddr < end) { + if (pte_val(*pte) & _PAGE_VALID) { + bool exec = pte_exec(*pte); + + tlb_batch_add_one(mm, vaddr, exec); + } + pte++; + vaddr += PAGE_SIZE; + } + pte_unmap(pte); +} + +void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t pmd) +{ + pmd_t orig = *pmdp; + + *pmdp = pmd; - if (fullmm) { - put_cpu_var(tlb_batch); + if (mm == &init_mm) return; + + if ((pmd_val(pmd) ^ pmd_val(orig)) & _PAGE_PMD_HUGE) { + if (pmd_val(pmd) & _PAGE_PMD_HUGE) + mm->context.huge_pte_count++; + else + mm->context.huge_pte_count--; + + /* Do not try to allocate the TSB hash table if we + * don't have one already. We have various locks held + * and thus we'll end up doing a GFP_KERNEL allocation + * in an atomic context. + * + * Instead, we let the first TLB miss on a hugepage + * take care of this. + */ } - nr = tb->tlb_nr; + if (!pmd_none(orig)) { + addr &= HPAGE_MASK; + if (pmd_trans_huge(orig)) { + pte_t orig_pte = __pte(pmd_val(orig)); + bool exec = pte_exec(orig_pte); - if (unlikely(nr != 0 && mm != tb->mm)) { - flush_tlb_pending(); - nr = 0; + tlb_batch_add_one(mm, addr, exec); + tlb_batch_add_one(mm, addr + REAL_HPAGE_SIZE, exec); + } else { + tlb_batch_pmd_scan(mm, addr, orig); + } } +} - if (nr == 0) - tb->mm = mm; +void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) +{ + pmd_t entry = *pmdp; - tb->vaddrs[nr] = vaddr; - tb->tlb_nr = ++nr; - if (nr >= TLB_BATCH_NR) - flush_tlb_pending(); + pmd_val(entry) &= ~_PAGE_VALID; - put_cpu_var(tlb_batch); + set_pmd_at(vma->vm_mm, address, pmdp, entry); + flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE); +} + +void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp, + pgtable_t pgtable) +{ + struct list_head *lh = (struct list_head *) pgtable; + + assert_spin_locked(&mm->page_table_lock); + + /* FIFO */ + if (!pmd_huge_pte(mm, pmdp)) + INIT_LIST_HEAD(lh); + else + list_add(lh, (struct list_head *) pmd_huge_pte(mm, pmdp)); + pmd_huge_pte(mm, pmdp) = pgtable; +} + +pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp) +{ + struct list_head *lh; + pgtable_t pgtable; + + assert_spin_locked(&mm->page_table_lock); + + /* FIFO */ + pgtable = pmd_huge_pte(mm, pmdp); + lh = (struct list_head *) pgtable; + if (list_empty(lh)) + pmd_huge_pte(mm, pmdp) = NULL; + else { + pmd_huge_pte(mm, pmdp) = (pgtable_t) lh->next; + list_del(lh); + } + pte_val(pgtable[0]) = 0; + pte_val(pgtable[1]) = 0; + + return pgtable; } +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c index c52add79b83..a06576683c3 100644 --- a/arch/sparc/mm/tsb.c +++ b/arch/sparc/mm/tsb.c @@ -7,11 +7,11 @@ #include <linux/preempt.h> #include <linux/slab.h> #include <asm/page.h> -#include <asm/tlbflush.h> -#include <asm/tlb.h> -#include <asm/mmu_context.h> #include <asm/pgtable.h> +#include <asm/mmu_context.h> +#include <asm/setup.h> #include <asm/tsb.h> +#include <asm/tlb.h> #include <asm/oplib.h> extern struct tsb swapper_tsb[KERNEL_TSB_NENTRIES]; @@ -46,28 +46,56 @@ void flush_tsb_kernel_range(unsigned long start, unsigned long end) } } +static void __flush_tsb_one_entry(unsigned long tsb, unsigned long v, + unsigned long hash_shift, + unsigned long nentries) +{ + unsigned long tag, ent, hash; + + v &= ~0x1UL; + hash = tsb_hash(v, hash_shift, nentries); + ent = tsb + (hash * sizeof(struct tsb)); + tag = (v >> 22UL); + + tsb_flush(ent, tag); +} + static void __flush_tsb_one(struct tlb_batch *tb, unsigned long hash_shift, unsigned long tsb, unsigned long nentries) { unsigned long i; - for (i = 0; i < tb->tlb_nr; i++) { - unsigned long v = tb->vaddrs[i]; - unsigned long tag, ent, hash; + for (i = 0; i < tb->tlb_nr; i++) + __flush_tsb_one_entry(tsb, tb->vaddrs[i], hash_shift, nentries); +} - v &= ~0x1UL; +void flush_tsb_user(struct tlb_batch *tb) +{ + struct mm_struct *mm = tb->mm; + unsigned long nentries, base, flags; - hash = tsb_hash(v, hash_shift, nentries); - ent = tsb + (hash * sizeof(struct tsb)); - tag = (v >> 22UL); + spin_lock_irqsave(&mm->context.lock, flags); - tsb_flush(ent, tag); + base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb; + nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries; + if (tlb_type == cheetah_plus || tlb_type == hypervisor) + base = __pa(base); + __flush_tsb_one(tb, PAGE_SHIFT, base, nentries); + +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) + if (mm->context.tsb_block[MM_TSB_HUGE].tsb) { + base = (unsigned long) mm->context.tsb_block[MM_TSB_HUGE].tsb; + nentries = mm->context.tsb_block[MM_TSB_HUGE].tsb_nentries; + if (tlb_type == cheetah_plus || tlb_type == hypervisor) + base = __pa(base); + __flush_tsb_one(tb, REAL_HPAGE_SHIFT, base, nentries); } +#endif + spin_unlock_irqrestore(&mm->context.lock, flags); } -void flush_tsb_user(struct tlb_batch *tb) +void flush_tsb_user_page(struct mm_struct *mm, unsigned long vaddr) { - struct mm_struct *mm = tb->mm; unsigned long nentries, base, flags; spin_lock_irqsave(&mm->context.lock, flags); @@ -76,43 +104,26 @@ void flush_tsb_user(struct tlb_batch *tb) nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries; if (tlb_type == cheetah_plus || tlb_type == hypervisor) base = __pa(base); - __flush_tsb_one(tb, PAGE_SHIFT, base, nentries); + __flush_tsb_one_entry(base, vaddr, PAGE_SHIFT, nentries); -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) if (mm->context.tsb_block[MM_TSB_HUGE].tsb) { base = (unsigned long) mm->context.tsb_block[MM_TSB_HUGE].tsb; nentries = mm->context.tsb_block[MM_TSB_HUGE].tsb_nentries; if (tlb_type == cheetah_plus || tlb_type == hypervisor) base = __pa(base); - __flush_tsb_one(tb, HPAGE_SHIFT, base, nentries); + __flush_tsb_one_entry(base, vaddr, REAL_HPAGE_SHIFT, nentries); } #endif spin_unlock_irqrestore(&mm->context.lock, flags); } -#if defined(CONFIG_SPARC64_PAGE_SIZE_8KB) #define HV_PGSZ_IDX_BASE HV_PGSZ_IDX_8K #define HV_PGSZ_MASK_BASE HV_PGSZ_MASK_8K -#elif defined(CONFIG_SPARC64_PAGE_SIZE_64KB) -#define HV_PGSZ_IDX_BASE HV_PGSZ_IDX_64K -#define HV_PGSZ_MASK_BASE HV_PGSZ_MASK_64K -#else -#error Broken base page size setting... -#endif -#ifdef CONFIG_HUGETLB_PAGE -#if defined(CONFIG_HUGETLB_PAGE_SIZE_64K) -#define HV_PGSZ_IDX_HUGE HV_PGSZ_IDX_64K -#define HV_PGSZ_MASK_HUGE HV_PGSZ_MASK_64K -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_512K) -#define HV_PGSZ_IDX_HUGE HV_PGSZ_IDX_512K -#define HV_PGSZ_MASK_HUGE HV_PGSZ_MASK_512K -#elif defined(CONFIG_HUGETLB_PAGE_SIZE_4MB) +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) #define HV_PGSZ_IDX_HUGE HV_PGSZ_IDX_4MB #define HV_PGSZ_MASK_HUGE HV_PGSZ_MASK_4MB -#else -#error Broken huge page size setting... -#endif #endif static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsigned long tsb_bytes) @@ -123,7 +134,19 @@ static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsign mm->context.tsb_block[tsb_idx].tsb_nentries = tsb_bytes / sizeof(struct tsb); - base = TSBMAP_BASE; + switch (tsb_idx) { + case MM_TSB_BASE: + base = TSBMAP_8K_BASE; + break; +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) + case MM_TSB_HUGE: + base = TSBMAP_4M_BASE; + break; +#endif + default: + BUG(); + } + tte = pgprot_val(PAGE_KERNEL_LOCKED); tsb_paddr = __pa(mm->context.tsb_block[tsb_idx].tsb); BUG_ON(tsb_paddr & (tsb_bytes - 1UL)); @@ -207,7 +230,7 @@ static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsign case MM_TSB_BASE: hp->pgsz_idx = HV_PGSZ_IDX_BASE; break; -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) case MM_TSB_HUGE: hp->pgsz_idx = HV_PGSZ_IDX_HUGE; break; @@ -222,7 +245,7 @@ static void setup_tsb_params(struct mm_struct *mm, unsigned long tsb_idx, unsign case MM_TSB_BASE: hp->pgsz_mask = HV_PGSZ_MASK_BASE; break; -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) case MM_TSB_HUGE: hp->pgsz_mask = HV_PGSZ_MASK_HUGE; break; @@ -263,7 +286,7 @@ void __init pgtable_cache_init(void) prom_halt(); } - for (i = 0; i < 8; i++) { + for (i = 0; i < ARRAY_SIZE(tsb_cache_names); i++) { unsigned long size = 8192 << i; const char *name = tsb_cache_names[i]; @@ -331,7 +354,7 @@ void tsb_grow(struct mm_struct *mm, unsigned long tsb_index, unsigned long rss) retry_tsb_alloc: gfp_flags = GFP_KERNEL; if (new_size > (PAGE_SIZE * 2)) - gfp_flags = __GFP_NOWARN | __GFP_NORETRY; + gfp_flags |= __GFP_NOWARN | __GFP_NORETRY; new_tsb = kmem_cache_alloc_node(tsb_caches[new_cache_index], gfp_flags, numa_node_id()); @@ -444,7 +467,7 @@ retry_tsb_alloc: int init_new_context(struct task_struct *tsk, struct mm_struct *mm) { -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) unsigned long huge_pte_count; #endif unsigned int i; @@ -453,7 +476,7 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm) mm->context.sparc64_ctx_val = 0UL; -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) /* We reset it to zero because the fork() page copying * will re-increment the counters as the parent PTEs are * copied into the child address space. @@ -474,7 +497,7 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm) */ tsb_grow(mm, MM_TSB_BASE, get_mm_rss(mm)); -#ifdef CONFIG_HUGETLB_PAGE +#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE) if (unlikely(huge_pte_count)) tsb_grow(mm, MM_TSB_HUGE, huge_pte_count); #endif diff --git a/arch/sparc/mm/tsunami.S b/arch/sparc/mm/tsunami.S index 4e55e8f7664..bf10a345fa8 100644 --- a/arch/sparc/mm/tsunami.S +++ b/arch/sparc/mm/tsunami.S @@ -24,7 +24,7 @@ /* Sliiick... */ tsunami_flush_cache_page: tsunami_flush_cache_range: - ld [%o0 + 0x0], %o0 /* XXX vma->vm_mm, GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 tsunami_flush_cache_mm: ld [%o0 + AOFF_mm_context], %g2 cmp %g2, -1 @@ -46,7 +46,7 @@ tsunami_flush_sig_insns: /* More slick stuff... */ tsunami_flush_tlb_range: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 tsunami_flush_tlb_mm: ld [%o0 + AOFF_mm_context], %g2 cmp %g2, -1 @@ -65,7 +65,7 @@ tsunami_flush_tlb_out: /* This one can be done in a fine grained manner... */ tsunami_flush_tlb_page: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 mov SRMMU_CTX_REG, %g1 ld [%o0 + AOFF_mm_context], %o3 andn %o1, (PAGE_SIZE - 1), %o1 diff --git a/arch/sparc/mm/ultra.S b/arch/sparc/mm/ultra.S index 874162a11ce..b4f4733abc6 100644 --- a/arch/sparc/mm/ultra.S +++ b/arch/sparc/mm/ultra.S @@ -53,6 +53,33 @@ __flush_tlb_mm: /* 18 insns */ nop .align 32 + .globl __flush_tlb_page +__flush_tlb_page: /* 22 insns */ + /* %o0 = context, %o1 = vaddr */ + rdpr %pstate, %g7 + andn %g7, PSTATE_IE, %g2 + wrpr %g2, %pstate + mov SECONDARY_CONTEXT, %o4 + ldxa [%o4] ASI_DMMU, %g2 + stxa %o0, [%o4] ASI_DMMU + andcc %o1, 1, %g0 + andn %o1, 1, %o3 + be,pn %icc, 1f + or %o3, 0x10, %o3 + stxa %g0, [%o3] ASI_IMMU_DEMAP +1: stxa %g0, [%o3] ASI_DMMU_DEMAP + membar #Sync + stxa %g2, [%o4] ASI_DMMU + sethi %hi(KERNBASE), %o4 + flush %o4 + retl + wrpr %g7, 0x0, %pstate + nop + nop + nop + nop + + .align 32 .globl __flush_tlb_pending __flush_tlb_pending: /* 26 insns */ /* %o0 = context, %o1 = nr, %o2 = vaddrs[] */ @@ -126,10 +153,10 @@ __spitfire_flush_tlb_mm_slow: .globl __flush_icache_page __flush_icache_page: /* %o0 = phys_page */ srlx %o0, PAGE_SHIFT, %o0 - sethi %uhi(PAGE_OFFSET), %g1 + sethi %hi(PAGE_OFFSET), %g1 sllx %o0, PAGE_SHIFT, %o0 sethi %hi(PAGE_SIZE), %g2 - sllx %g1, 32, %g1 + ldx [%g1 + %lo(PAGE_OFFSET)], %g1 add %o0, %g1, %o0 1: subcc %g2, 32, %g2 bne,pt %icc, 1b @@ -151,8 +178,8 @@ __flush_icache_page: /* %o0 = phys_page */ .align 64 .globl __flush_dcache_page __flush_dcache_page: /* %o0=kaddr, %o1=flush_icache */ - sethi %uhi(PAGE_OFFSET), %g1 - sllx %g1, 32, %g1 + sethi %hi(PAGE_OFFSET), %g1 + ldx [%g1 + %lo(PAGE_OFFSET)], %g1 sub %o0, %g1, %o0 ! physical address srlx %o0, 11, %o0 ! make D-cache TAG sethi %hi(1 << 14), %o2 ! D-cache size @@ -203,6 +230,31 @@ __cheetah_flush_tlb_mm: /* 19 insns */ retl wrpr %g7, 0x0, %pstate +__cheetah_flush_tlb_page: /* 22 insns */ + /* %o0 = context, %o1 = vaddr */ + rdpr %pstate, %g7 + andn %g7, PSTATE_IE, %g2 + wrpr %g2, 0x0, %pstate + wrpr %g0, 1, %tl + mov PRIMARY_CONTEXT, %o4 + ldxa [%o4] ASI_DMMU, %g2 + srlx %g2, CTX_PGSZ1_NUC_SHIFT, %o3 + sllx %o3, CTX_PGSZ1_NUC_SHIFT, %o3 + or %o0, %o3, %o0 /* Preserve nucleus page size fields */ + stxa %o0, [%o4] ASI_DMMU + andcc %o1, 1, %g0 + be,pn %icc, 1f + andn %o1, 1, %o3 + stxa %g0, [%o3] ASI_IMMU_DEMAP +1: stxa %g0, [%o3] ASI_DMMU_DEMAP + membar #Sync + stxa %g2, [%o4] ASI_DMMU + sethi %hi(KERNBASE), %o4 + flush %o4 + wrpr %g0, 0, %tl + retl + wrpr %g7, 0x0, %pstate + __cheetah_flush_tlb_pending: /* 27 insns */ /* %o0 = context, %o1 = nr, %o2 = vaddrs[] */ rdpr %pstate, %g7 @@ -235,8 +287,8 @@ __cheetah_flush_tlb_pending: /* 27 insns */ #ifdef DCACHE_ALIASING_POSSIBLE __cheetah_flush_dcache_page: /* 11 insns */ - sethi %uhi(PAGE_OFFSET), %g1 - sllx %g1, 32, %g1 + sethi %hi(PAGE_OFFSET), %g1 + ldx [%g1 + %lo(PAGE_OFFSET)], %g1 sub %o0, %g1, %o0 sethi %hi(PAGE_SIZE), %o4 1: subcc %o4, (1 << 5), %o4 @@ -269,6 +321,20 @@ __hypervisor_flush_tlb_mm: /* 10 insns */ retl nop +__hypervisor_flush_tlb_page: /* 11 insns */ + /* %o0 = context, %o1 = vaddr */ + mov %o0, %g2 + mov %o1, %o0 /* ARG0: vaddr + IMMU-bit */ + mov %g2, %o1 /* ARG1: mmu context */ + mov HV_MMU_ALL, %o2 /* ARG2: flags */ + srlx %o0, PAGE_SHIFT, %o0 + sllx %o0, PAGE_SHIFT, %o0 + ta HV_MMU_UNMAP_ADDR_TRAP + brnz,pn %o0, __hypervisor_tlb_tl0_error + mov HV_MMU_UNMAP_ADDR_TRAP, %o1 + retl + nop + __hypervisor_flush_tlb_pending: /* 16 insns */ /* %o0 = context, %o1 = nr, %o2 = vaddrs[] */ sllx %o1, 3, %g1 @@ -339,6 +405,13 @@ cheetah_patch_cachetlbops: call tlb_patch_one mov 19, %o2 + sethi %hi(__flush_tlb_page), %o0 + or %o0, %lo(__flush_tlb_page), %o0 + sethi %hi(__cheetah_flush_tlb_page), %o1 + or %o1, %lo(__cheetah_flush_tlb_page), %o1 + call tlb_patch_one + mov 22, %o2 + sethi %hi(__flush_tlb_pending), %o0 or %o0, %lo(__flush_tlb_pending), %o0 sethi %hi(__cheetah_flush_tlb_pending), %o1 @@ -397,10 +470,9 @@ xcall_flush_tlb_mm: /* 21 insns */ nop nop - .globl xcall_flush_tlb_pending -xcall_flush_tlb_pending: /* 21 insns */ - /* %g5=context, %g1=nr, %g7=vaddrs[] */ - sllx %g1, 3, %g1 + .globl xcall_flush_tlb_page +xcall_flush_tlb_page: /* 17 insns */ + /* %g5=context, %g1=vaddr */ mov PRIMARY_CONTEXT, %g4 ldxa [%g4] ASI_DMMU, %g2 srlx %g2, CTX_PGSZ1_NUC_SHIFT, %g4 @@ -408,20 +480,16 @@ xcall_flush_tlb_pending: /* 21 insns */ or %g5, %g4, %g5 mov PRIMARY_CONTEXT, %g4 stxa %g5, [%g4] ASI_DMMU -1: sub %g1, (1 << 3), %g1 - ldx [%g7 + %g1], %g5 - andcc %g5, 0x1, %g0 + andcc %g1, 0x1, %g0 be,pn %icc, 2f - - andn %g5, 0x1, %g5 + andn %g1, 0x1, %g5 stxa %g0, [%g5] ASI_IMMU_DEMAP 2: stxa %g0, [%g5] ASI_DMMU_DEMAP membar #Sync - brnz,pt %g1, 1b - nop stxa %g2, [%g4] ASI_DMMU retry nop + nop .globl xcall_flush_tlb_kernel_range xcall_flush_tlb_kernel_range: /* 25 insns */ @@ -481,8 +549,8 @@ xcall_sync_tick: .globl xcall_fetch_glob_regs xcall_fetch_glob_regs: - sethi %hi(global_reg_snapshot), %g1 - or %g1, %lo(global_reg_snapshot), %g1 + sethi %hi(global_cpu_snapshot), %g1 + or %g1, %lo(global_cpu_snapshot), %g1 __GET_CPUID(%g2) sllx %g2, 6, %g3 add %g1, %g3, %g1 @@ -509,6 +577,66 @@ xcall_fetch_glob_regs: stx %g3, [%g1 + GR_SNAP_THREAD] retry + .globl xcall_fetch_glob_pmu +xcall_fetch_glob_pmu: + sethi %hi(global_cpu_snapshot), %g1 + or %g1, %lo(global_cpu_snapshot), %g1 + __GET_CPUID(%g2) + sllx %g2, 6, %g3 + add %g1, %g3, %g1 + rd %pic, %g7 + stx %g7, [%g1 + (4 * 8)] + rd %pcr, %g7 + stx %g7, [%g1 + (0 * 8)] + retry + + .globl xcall_fetch_glob_pmu_n4 +xcall_fetch_glob_pmu_n4: + sethi %hi(global_cpu_snapshot), %g1 + or %g1, %lo(global_cpu_snapshot), %g1 + __GET_CPUID(%g2) + sllx %g2, 6, %g3 + add %g1, %g3, %g1 + + ldxa [%g0] ASI_PIC, %g7 + stx %g7, [%g1 + (4 * 8)] + mov 0x08, %g3 + ldxa [%g3] ASI_PIC, %g7 + stx %g7, [%g1 + (5 * 8)] + mov 0x10, %g3 + ldxa [%g3] ASI_PIC, %g7 + stx %g7, [%g1 + (6 * 8)] + mov 0x18, %g3 + ldxa [%g3] ASI_PIC, %g7 + stx %g7, [%g1 + (7 * 8)] + + mov %o0, %g2 + mov %o1, %g3 + mov %o5, %g7 + + mov HV_FAST_VT_GET_PERFREG, %o5 + mov 3, %o0 + ta HV_FAST_TRAP + stx %o1, [%g1 + (3 * 8)] + mov HV_FAST_VT_GET_PERFREG, %o5 + mov 2, %o0 + ta HV_FAST_TRAP + stx %o1, [%g1 + (2 * 8)] + mov HV_FAST_VT_GET_PERFREG, %o5 + mov 1, %o0 + ta HV_FAST_TRAP + stx %o1, [%g1 + (1 * 8)] + mov HV_FAST_VT_GET_PERFREG, %o5 + mov 0, %o0 + ta HV_FAST_TRAP + stx %o1, [%g1 + (0 * 8)] + + mov %g2, %o0 + mov %g3, %o1 + mov %g7, %o5 + + retry + #ifdef DCACHE_ALIASING_POSSIBLE .align 32 .globl xcall_flush_dcache_page_cheetah @@ -596,15 +724,13 @@ __hypervisor_xcall_flush_tlb_mm: /* 21 insns */ membar #Sync retry - .globl __hypervisor_xcall_flush_tlb_pending -__hypervisor_xcall_flush_tlb_pending: /* 21 insns */ - /* %g5=ctx, %g1=nr, %g7=vaddrs[], %g2,%g3,%g4,g6=scratch */ - sllx %g1, 3, %g1 + .globl __hypervisor_xcall_flush_tlb_page +__hypervisor_xcall_flush_tlb_page: /* 17 insns */ + /* %g5=ctx, %g1=vaddr */ mov %o0, %g2 mov %o1, %g3 mov %o2, %g4 -1: sub %g1, (1 << 3), %g1 - ldx [%g7 + %g1], %o0 /* ARG0: virtual address */ + mov %g1, %o0 /* ARG0: virtual address */ mov %g5, %o1 /* ARG1: mmu context */ mov HV_MMU_ALL, %o2 /* ARG2: flags */ srlx %o0, PAGE_SHIFT, %o0 @@ -613,8 +739,6 @@ __hypervisor_xcall_flush_tlb_pending: /* 21 insns */ mov HV_MMU_UNMAP_ADDR_TRAP, %g6 brnz,a,pn %o0, __hypervisor_tlb_xcall_error mov %o0, %g5 - brnz,pt %g1, 1b - nop mov %g2, %o0 mov %g3, %o1 mov %g4, %o2 @@ -697,6 +821,13 @@ hypervisor_patch_cachetlbops: call tlb_patch_one mov 10, %o2 + sethi %hi(__flush_tlb_page), %o0 + or %o0, %lo(__flush_tlb_page), %o0 + sethi %hi(__hypervisor_flush_tlb_page), %o1 + or %o1, %lo(__hypervisor_flush_tlb_page), %o1 + call tlb_patch_one + mov 11, %o2 + sethi %hi(__flush_tlb_pending), %o0 or %o0, %lo(__flush_tlb_pending), %o0 sethi %hi(__hypervisor_flush_tlb_pending), %o1 @@ -728,12 +859,12 @@ hypervisor_patch_cachetlbops: call tlb_patch_one mov 21, %o2 - sethi %hi(xcall_flush_tlb_pending), %o0 - or %o0, %lo(xcall_flush_tlb_pending), %o0 - sethi %hi(__hypervisor_xcall_flush_tlb_pending), %o1 - or %o1, %lo(__hypervisor_xcall_flush_tlb_pending), %o1 + sethi %hi(xcall_flush_tlb_page), %o0 + or %o0, %lo(xcall_flush_tlb_page), %o0 + sethi %hi(__hypervisor_xcall_flush_tlb_page), %o1 + or %o1, %lo(__hypervisor_xcall_flush_tlb_page), %o1 call tlb_patch_one - mov 21, %o2 + mov 17, %o2 sethi %hi(xcall_flush_tlb_kernel_range), %o0 or %o0, %lo(xcall_flush_tlb_kernel_range), %o0 diff --git a/arch/sparc/mm/viking.S b/arch/sparc/mm/viking.S index bf8ee0613ae..852257fcc82 100644 --- a/arch/sparc/mm/viking.S +++ b/arch/sparc/mm/viking.S @@ -108,7 +108,7 @@ viking_mxcc_flush_page: viking_flush_cache_page: viking_flush_cache_range: #ifndef CONFIG_SMP - ld [%o0 + 0x0], %o0 /* XXX vma->vm_mm, GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 #endif viking_flush_cache_mm: #ifndef CONFIG_SMP @@ -148,7 +148,7 @@ viking_flush_tlb_mm: #endif viking_flush_tlb_range: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 mov SRMMU_CTX_REG, %g1 ld [%o0 + AOFF_mm_context], %o3 lda [%g1] ASI_M_MMUREGS, %g5 @@ -173,7 +173,7 @@ viking_flush_tlb_range: #endif viking_flush_tlb_page: - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 mov SRMMU_CTX_REG, %g1 ld [%o0 + AOFF_mm_context], %o3 lda [%g1] ASI_M_MMUREGS, %g5 @@ -239,7 +239,7 @@ sun4dsmp_flush_tlb_range: tst %g5 bne 3f mov SRMMU_CTX_REG, %g1 - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 ld [%o0 + AOFF_mm_context], %o3 lda [%g1] ASI_M_MMUREGS, %g5 sethi %hi(~((1 << SRMMU_PGDIR_SHIFT) - 1)), %o4 @@ -265,7 +265,7 @@ sun4dsmp_flush_tlb_page: tst %g5 bne 2f mov SRMMU_CTX_REG, %g1 - ld [%o0 + 0x00], %o0 /* XXX vma->vm_mm GROSS XXX */ + ld [%o0 + VMA_VM_MM], %o0 ld [%o0 + AOFF_mm_context], %o3 lda [%g1] ASI_M_MMUREGS, %g5 and %o1, PAGE_MASK, %o1 diff --git a/arch/sparc/net/bpf_jit_comp.c b/arch/sparc/net/bpf_jit_comp.c index 1a69244e785..892a102671a 100644 --- a/arch/sparc/net/bpf_jit_comp.c +++ b/arch/sparc/net/bpf_jit_comp.c @@ -3,6 +3,7 @@ #include <linux/netdevice.h> #include <linux/filter.h> #include <linux/cache.h> +#include <linux/if_vlan.h> #include <asm/cacheflush.h> #include <asm/ptrace.h> @@ -82,9 +83,9 @@ static void bpf_flush_icache(void *start_, void *end_) #define BNE (F2(0, 2) | CONDNE) #ifdef CONFIG_SPARC64 -#define BNE_PTR (F2(0, 1) | CONDNE | (2 << 20)) +#define BE_PTR (F2(0, 1) | CONDE | (2 << 20)) #else -#define BNE_PTR BNE +#define BE_PTR BE #endif #define SETHI(K, REG) \ @@ -96,6 +97,7 @@ static void bpf_flush_icache(void *start_, void *end_) #define AND F3(2, 0x01) #define ANDCC F3(2, 0x11) #define OR F3(2, 0x02) +#define XOR F3(2, 0x03) #define SUB F3(2, 0x04) #define SUBCC F3(2, 0x14) #define MUL F3(2, 0x0a) /* umul */ @@ -311,6 +313,12 @@ do { *prog++ = BR_OPC | WDISP22(OFF); \ #define emit_addi(R1, IMM, R3) \ *prog++ = (ADD | IMMED | RS1(R1) | S13(IMM) | RD(R3)) +#define emit_and(R1, R2, R3) \ + *prog++ = (AND | RS1(R1) | RS2(R2) | RD(R3)) + +#define emit_andi(R1, IMM, R3) \ + *prog++ = (AND | IMMED | RS1(R1) | S13(IMM) | RD(R3)) + #define emit_alloc_stack(SZ) \ *prog++ = (SUB | IMMED | RS1(SP) | S13(SZ) | RD(SP)) @@ -407,18 +415,11 @@ void bpf_jit_compile(struct sk_filter *fp) emit_reg_move(O7, r_saved_O7); switch (filter[0].code) { - case BPF_S_RET_K: - case BPF_S_LD_W_LEN: - case BPF_S_ANC_PROTOCOL: - case BPF_S_ANC_PKTTYPE: - case BPF_S_ANC_IFINDEX: - case BPF_S_ANC_MARK: - case BPF_S_ANC_RXHASH: - case BPF_S_ANC_CPU: - case BPF_S_ANC_QUEUE: - case BPF_S_LD_W_ABS: - case BPF_S_LD_H_ABS: - case BPF_S_LD_B_ABS: + case BPF_RET | BPF_K: + case BPF_LD | BPF_W | BPF_LEN: + case BPF_LD | BPF_W | BPF_ABS: + case BPF_LD | BPF_H | BPF_ABS: + case BPF_LD | BPF_B | BPF_ABS: /* The first instruction sets the A register (or is * a "RET 'constant'") */ @@ -435,56 +436,75 @@ void bpf_jit_compile(struct sk_filter *fp) unsigned int t_offset; unsigned int f_offset; u32 t_op, f_op; + u16 code = bpf_anc_helper(&filter[i]); int ilen; - switch (filter[i].code) { - case BPF_S_ALU_ADD_X: /* A += X; */ + switch (code) { + case BPF_ALU | BPF_ADD | BPF_X: /* A += X; */ emit_alu_X(ADD); break; - case BPF_S_ALU_ADD_K: /* A += K; */ + case BPF_ALU | BPF_ADD | BPF_K: /* A += K; */ emit_alu_K(ADD, K); break; - case BPF_S_ALU_SUB_X: /* A -= X; */ + case BPF_ALU | BPF_SUB | BPF_X: /* A -= X; */ emit_alu_X(SUB); break; - case BPF_S_ALU_SUB_K: /* A -= K */ + case BPF_ALU | BPF_SUB | BPF_K: /* A -= K */ emit_alu_K(SUB, K); break; - case BPF_S_ALU_AND_X: /* A &= X */ + case BPF_ALU | BPF_AND | BPF_X: /* A &= X */ emit_alu_X(AND); break; - case BPF_S_ALU_AND_K: /* A &= K */ + case BPF_ALU | BPF_AND | BPF_K: /* A &= K */ emit_alu_K(AND, K); break; - case BPF_S_ALU_OR_X: /* A |= X */ + case BPF_ALU | BPF_OR | BPF_X: /* A |= X */ emit_alu_X(OR); break; - case BPF_S_ALU_OR_K: /* A |= K */ + case BPF_ALU | BPF_OR | BPF_K: /* A |= K */ emit_alu_K(OR, K); break; - case BPF_S_ALU_LSH_X: /* A <<= X */ + case BPF_ANC | SKF_AD_ALU_XOR_X: /* A ^= X; */ + case BPF_ALU | BPF_XOR | BPF_X: + emit_alu_X(XOR); + break; + case BPF_ALU | BPF_XOR | BPF_K: /* A ^= K */ + emit_alu_K(XOR, K); + break; + case BPF_ALU | BPF_LSH | BPF_X: /* A <<= X */ emit_alu_X(SLL); break; - case BPF_S_ALU_LSH_K: /* A <<= K */ + case BPF_ALU | BPF_LSH | BPF_K: /* A <<= K */ emit_alu_K(SLL, K); break; - case BPF_S_ALU_RSH_X: /* A >>= X */ + case BPF_ALU | BPF_RSH | BPF_X: /* A >>= X */ emit_alu_X(SRL); break; - case BPF_S_ALU_RSH_K: /* A >>= K */ + case BPF_ALU | BPF_RSH | BPF_K: /* A >>= K */ emit_alu_K(SRL, K); break; - case BPF_S_ALU_MUL_X: /* A *= X; */ + case BPF_ALU | BPF_MUL | BPF_X: /* A *= X; */ emit_alu_X(MUL); break; - case BPF_S_ALU_MUL_K: /* A *= K */ + case BPF_ALU | BPF_MUL | BPF_K: /* A *= K */ emit_alu_K(MUL, K); break; - case BPF_S_ALU_DIV_K: /* A /= K */ - emit_alu_K(MUL, K); - emit_read_y(r_A); + case BPF_ALU | BPF_DIV | BPF_K: /* A /= K with K != 0*/ + if (K == 1) + break; + emit_write_y(G0); +#ifdef CONFIG_SPARC32 + /* The Sparc v8 architecture requires + * three instructions between a %y + * register write and the first use. + */ + emit_nop(); + emit_nop(); + emit_nop(); +#endif + emit_alu_K(DIV, K); break; - case BPF_S_ALU_DIV_X: /* A /= X; */ + case BPF_ALU | BPF_DIV | BPF_X: /* A /= X; */ emit_cmpi(r_X, 0); if (pc_ret0 > 0) { t_offset = addrs[pc_ret0 - 1]; @@ -516,10 +536,10 @@ void bpf_jit_compile(struct sk_filter *fp) #endif emit_alu_X(DIV); break; - case BPF_S_ALU_NEG: + case BPF_ALU | BPF_NEG: emit_neg(); break; - case BPF_S_RET_K: + case BPF_RET | BPF_K: if (!K) { if (pc_ret0 == -1) pc_ret0 = i; @@ -528,7 +548,7 @@ void bpf_jit_compile(struct sk_filter *fp) emit_loadimm(K, r_A); } /* Fallthrough */ - case BPF_S_RET_A: + case BPF_RET | BPF_A: if (seen_or_pass0) { if (i != flen - 1) { emit_jump(cleanup_addr); @@ -545,18 +565,18 @@ void bpf_jit_compile(struct sk_filter *fp) emit_jmpl(r_saved_O7, 8, G0); emit_reg_move(r_A, O0); /* delay slot */ break; - case BPF_S_MISC_TAX: + case BPF_MISC | BPF_TAX: seen |= SEEN_XREG; emit_reg_move(r_A, r_X); break; - case BPF_S_MISC_TXA: + case BPF_MISC | BPF_TXA: seen |= SEEN_XREG; emit_reg_move(r_X, r_A); break; - case BPF_S_ANC_CPU: + case BPF_ANC | SKF_AD_CPU: emit_load_cpu(r_A); break; - case BPF_S_ANC_PROTOCOL: + case BPF_ANC | SKF_AD_PROTOCOL: emit_skb_load16(protocol, r_A); break; #if 0 @@ -564,73 +584,83 @@ void bpf_jit_compile(struct sk_filter *fp) * a bit field even though we very much * know what we are doing here. */ - case BPF_S_ANC_PKTTYPE: + case BPF_ANC | SKF_AD_PKTTYPE: __emit_skb_load8(pkt_type, r_A); emit_alu_K(SRL, 5); break; #endif - case BPF_S_ANC_IFINDEX: + case BPF_ANC | SKF_AD_IFINDEX: emit_skb_loadptr(dev, r_A); emit_cmpi(r_A, 0); - emit_branch(BNE_PTR, cleanup_addr + 4); + emit_branch(BE_PTR, cleanup_addr + 4); emit_nop(); emit_load32(r_A, struct net_device, ifindex, r_A); break; - case BPF_S_ANC_MARK: + case BPF_ANC | SKF_AD_MARK: emit_skb_load32(mark, r_A); break; - case BPF_S_ANC_QUEUE: + case BPF_ANC | SKF_AD_QUEUE: emit_skb_load16(queue_mapping, r_A); break; - case BPF_S_ANC_HATYPE: + case BPF_ANC | SKF_AD_HATYPE: emit_skb_loadptr(dev, r_A); emit_cmpi(r_A, 0); - emit_branch(BNE_PTR, cleanup_addr + 4); + emit_branch(BE_PTR, cleanup_addr + 4); emit_nop(); emit_load16(r_A, struct net_device, type, r_A); break; - case BPF_S_ANC_RXHASH: - emit_skb_load32(rxhash, r_A); + case BPF_ANC | SKF_AD_RXHASH: + emit_skb_load32(hash, r_A); + break; + case BPF_ANC | SKF_AD_VLAN_TAG: + case BPF_ANC | SKF_AD_VLAN_TAG_PRESENT: + emit_skb_load16(vlan_tci, r_A); + if (code == (BPF_ANC | SKF_AD_VLAN_TAG)) { + emit_andi(r_A, VLAN_VID_MASK, r_A); + } else { + emit_loadimm(VLAN_TAG_PRESENT, r_TMP); + emit_and(r_A, r_TMP, r_A); + } break; - case BPF_S_LD_IMM: + case BPF_LD | BPF_IMM: emit_loadimm(K, r_A); break; - case BPF_S_LDX_IMM: + case BPF_LDX | BPF_IMM: emit_loadimm(K, r_X); break; - case BPF_S_LD_MEM: + case BPF_LD | BPF_MEM: emit_ldmem(K * 4, r_A); break; - case BPF_S_LDX_MEM: + case BPF_LDX | BPF_MEM: emit_ldmem(K * 4, r_X); break; - case BPF_S_ST: + case BPF_ST: emit_stmem(K * 4, r_A); break; - case BPF_S_STX: + case BPF_STX: emit_stmem(K * 4, r_X); break; #define CHOOSE_LOAD_FUNC(K, func) \ ((int)K < 0 ? ((int)K >= SKF_LL_OFF ? func##_negative_offset : func) : func##_positive_offset) - case BPF_S_LD_W_ABS: + case BPF_LD | BPF_W | BPF_ABS: func = CHOOSE_LOAD_FUNC(K, bpf_jit_load_word); common_load: seen |= SEEN_DATAREF; emit_loadimm(K, r_OFF); emit_call(func); break; - case BPF_S_LD_H_ABS: + case BPF_LD | BPF_H | BPF_ABS: func = CHOOSE_LOAD_FUNC(K, bpf_jit_load_half); goto common_load; - case BPF_S_LD_B_ABS: + case BPF_LD | BPF_B | BPF_ABS: func = CHOOSE_LOAD_FUNC(K, bpf_jit_load_byte); goto common_load; - case BPF_S_LDX_B_MSH: + case BPF_LDX | BPF_B | BPF_MSH: func = CHOOSE_LOAD_FUNC(K, bpf_jit_load_byte_msh); goto common_load; - case BPF_S_LD_W_IND: + case BPF_LD | BPF_W | BPF_IND: func = bpf_jit_load_word; common_load_ind: seen |= SEEN_DATAREF | SEEN_XREG; if (K) { @@ -645,13 +675,13 @@ common_load_ind: seen |= SEEN_DATAREF | SEEN_XREG; } emit_call(func); break; - case BPF_S_LD_H_IND: + case BPF_LD | BPF_H | BPF_IND: func = bpf_jit_load_half; goto common_load_ind; - case BPF_S_LD_B_IND: + case BPF_LD | BPF_B | BPF_IND: func = bpf_jit_load_byte; goto common_load_ind; - case BPF_S_JMP_JA: + case BPF_JMP | BPF_JA: emit_jump(addrs[i + K]); emit_nop(); break; @@ -662,14 +692,14 @@ common_load_ind: seen |= SEEN_DATAREF | SEEN_XREG; f_op = FOP; \ goto cond_branch - COND_SEL(BPF_S_JMP_JGT_K, BGU, BLEU); - COND_SEL(BPF_S_JMP_JGE_K, BGEU, BLU); - COND_SEL(BPF_S_JMP_JEQ_K, BE, BNE); - COND_SEL(BPF_S_JMP_JSET_K, BNE, BE); - COND_SEL(BPF_S_JMP_JGT_X, BGU, BLEU); - COND_SEL(BPF_S_JMP_JGE_X, BGEU, BLU); - COND_SEL(BPF_S_JMP_JEQ_X, BE, BNE); - COND_SEL(BPF_S_JMP_JSET_X, BNE, BE); + COND_SEL(BPF_JMP | BPF_JGT | BPF_K, BGU, BLEU); + COND_SEL(BPF_JMP | BPF_JGE | BPF_K, BGEU, BLU); + COND_SEL(BPF_JMP | BPF_JEQ | BPF_K, BE, BNE); + COND_SEL(BPF_JMP | BPF_JSET | BPF_K, BNE, BE); + COND_SEL(BPF_JMP | BPF_JGT | BPF_X, BGU, BLEU); + COND_SEL(BPF_JMP | BPF_JGE | BPF_X, BGEU, BLU); + COND_SEL(BPF_JMP | BPF_JEQ | BPF_X, BE, BNE); + COND_SEL(BPF_JMP | BPF_JSET | BPF_X, BNE, BE); cond_branch: f_offset = addrs[i + filter[i].jf]; t_offset = addrs[i + filter[i].jt]; @@ -681,20 +711,20 @@ cond_branch: f_offset = addrs[i + filter[i].jf]; break; } - switch (filter[i].code) { - case BPF_S_JMP_JGT_X: - case BPF_S_JMP_JGE_X: - case BPF_S_JMP_JEQ_X: + switch (code) { + case BPF_JMP | BPF_JGT | BPF_X: + case BPF_JMP | BPF_JGE | BPF_X: + case BPF_JMP | BPF_JEQ | BPF_X: seen |= SEEN_XREG; emit_cmp(r_A, r_X); break; - case BPF_S_JMP_JSET_X: + case BPF_JMP | BPF_JSET | BPF_X: seen |= SEEN_XREG; emit_btst(r_A, r_X); break; - case BPF_S_JMP_JEQ_K: - case BPF_S_JMP_JGT_K: - case BPF_S_JMP_JGE_K: + case BPF_JMP | BPF_JEQ | BPF_K: + case BPF_JMP | BPF_JGT | BPF_K: + case BPF_JMP | BPF_JGE | BPF_K: if (is_simm13(K)) { emit_cmpi(r_A, K); } else { @@ -702,7 +732,7 @@ cond_branch: f_offset = addrs[i + filter[i].jf]; emit_cmp(r_A, r_TMP); } break; - case BPF_S_JMP_JSET_K: + case BPF_JMP | BPF_JSET | BPF_K: if (is_simm13(K)) { emit_btsti(r_A, K); } else { @@ -758,9 +788,7 @@ cond_branch: f_offset = addrs[i + filter[i].jf]; break; } if (proglen == oldproglen) { - image = module_alloc(max_t(unsigned int, - proglen, - sizeof(struct work_struct))); + image = module_alloc(proglen); if (!image) goto out; } @@ -768,35 +796,21 @@ cond_branch: f_offset = addrs[i + filter[i].jf]; } if (bpf_jit_enable > 1) - pr_err("flen=%d proglen=%u pass=%d image=%p\n", - flen, proglen, pass, image); + bpf_jit_dump(flen, proglen, pass, image); if (image) { - if (bpf_jit_enable > 1) - print_hex_dump(KERN_ERR, "JIT code: ", DUMP_PREFIX_ADDRESS, - 16, 1, image, proglen, false); bpf_flush_icache(image, image + proglen); fp->bpf_func = (void *)image; + fp->jited = 1; } out: kfree(addrs); return; } -static void jit_free_defer(struct work_struct *arg) -{ - module_free(NULL, arg); -} - -/* run from softirq, we must use a work_struct to call - * module_free() from process context - */ void bpf_jit_free(struct sk_filter *fp) { - if (fp->bpf_func != sk_run_filter) { - struct work_struct *work = (struct work_struct *)fp->bpf_func; - - INIT_WORK(work, jit_free_defer); - schedule_work(work); - } + if (fp->jited) + module_free(NULL, fp->bpf_func); + kfree(fp); } diff --git a/arch/sparc/power/Makefile b/arch/sparc/power/Makefile new file mode 100644 index 00000000000..3201ace0ddb --- /dev/null +++ b/arch/sparc/power/Makefile @@ -0,0 +1,3 @@ +# Makefile for Sparc-specific hibernate files. + +obj-$(CONFIG_HIBERNATION) += hibernate.o hibernate_asm.o diff --git a/arch/sparc/power/hibernate.c b/arch/sparc/power/hibernate.c new file mode 100644 index 00000000000..42b0b8ce699 --- /dev/null +++ b/arch/sparc/power/hibernate.c @@ -0,0 +1,42 @@ +/* + * hibernate.c: Hibernaton support specific for sparc64. + * + * Copyright (C) 2013 Kirill V Tkhai (tkhai@yandex.ru) + */ + +#include <linux/mm.h> + +#include <asm/hibernate.h> +#include <asm/visasm.h> +#include <asm/page.h> +#include <asm/tlb.h> + +/* References to section boundaries */ +extern const void __nosave_begin, __nosave_end; + +struct saved_context saved_context; + +/* + * pfn_is_nosave - check if given pfn is in the 'nosave' section + */ + +int pfn_is_nosave(unsigned long pfn) +{ + unsigned long nosave_begin_pfn = PFN_DOWN((unsigned long)&__nosave_begin); + unsigned long nosave_end_pfn = PFN_DOWN((unsigned long)&__nosave_end); + + return (pfn >= nosave_begin_pfn) && (pfn < nosave_end_pfn); +} + +void save_processor_state(void) +{ + save_and_clear_fpu(); +} + +void restore_processor_state(void) +{ + struct mm_struct *mm = current->active_mm; + + load_secondary_context(mm); + tsb_context_switch(mm); +} diff --git a/arch/sparc/power/hibernate_asm.S b/arch/sparc/power/hibernate_asm.S new file mode 100644 index 00000000000..79942166df8 --- /dev/null +++ b/arch/sparc/power/hibernate_asm.S @@ -0,0 +1,131 @@ +/* + * hibernate_asm.S: Hibernaton support specific for sparc64. + * + * Copyright (C) 2013 Kirill V Tkhai (tkhai@yandex.ru) + */ + +#include <linux/linkage.h> + +#include <asm/asm-offsets.h> +#include <asm/cpudata.h> +#include <asm/page.h> + +ENTRY(swsusp_arch_suspend) + save %sp, -128, %sp + save %sp, -128, %sp + flushw + + setuw saved_context, %g3 + + /* Save window regs */ + rdpr %cwp, %g2 + stx %g2, [%g3 + SC_REG_CWP] + rdpr %wstate, %g2 + stx %g2, [%g3 + SC_REG_WSTATE] + stx %fp, [%g3 + SC_REG_FP] + + /* Save state regs */ + rdpr %tick, %g2 + stx %g2, [%g3 + SC_REG_TICK] + rdpr %pstate, %g2 + stx %g2, [%g3 + SC_REG_PSTATE] + + /* Save global regs */ + stx %g4, [%g3 + SC_REG_G4] + stx %g5, [%g3 + SC_REG_G5] + stx %g6, [%g3 + SC_REG_G6] + + call swsusp_save + nop + + mov %o0, %i0 + restore + + mov %o0, %i0 + ret + restore + +ENTRY(swsusp_arch_resume) + /* Write restore_pblist to %l0 */ + sethi %hi(restore_pblist), %l0 + ldx [%l0 + %lo(restore_pblist)], %l0 + + call __flush_tlb_all + nop + + /* Write PAGE_OFFSET to %g7 */ + sethi %uhi(PAGE_OFFSET), %g7 + sllx %g7, 32, %g7 + + setuw (PAGE_SIZE-8), %g3 + + /* Use MMU Bypass */ + rd %asi, %g1 + wr %g0, ASI_PHYS_USE_EC, %asi + + ba fill_itlb + nop + +pbe_loop: + cmp %l0, %g0 + be restore_ctx + sub %l0, %g7, %l0 + + ldxa [%l0 ] %asi, %l1 /* address */ + ldxa [%l0 + 8] %asi, %l2 /* orig_address */ + + /* phys addr */ + sub %l1, %g7, %l1 + sub %l2, %g7, %l2 + + mov %g3, %l3 /* PAGE_SIZE-8 */ +copy_loop: + ldxa [%l1 + %l3] ASI_PHYS_USE_EC, %g2 + stxa %g2, [%l2 + %l3] ASI_PHYS_USE_EC + cmp %l3, %g0 + bne copy_loop + sub %l3, 8, %l3 + + /* next pbe */ + ba pbe_loop + ldxa [%l0 + 16] %asi, %l0 + +restore_ctx: + setuw saved_context, %g3 + + /* Restore window regs */ + wrpr %g0, 0, %canrestore + wrpr %g0, 0, %otherwin + wrpr %g0, 6, %cansave + wrpr %g0, 0, %cleanwin + + ldxa [%g3 + SC_REG_CWP] %asi, %g2 + wrpr %g2, %cwp + ldxa [%g3 + SC_REG_WSTATE] %asi, %g2 + wrpr %g2, %wstate + ldxa [%g3 + SC_REG_FP] %asi, %fp + + /* Restore state regs */ + ldxa [%g3 + SC_REG_PSTATE] %asi, %g2 + wrpr %g2, %pstate + ldxa [%g3 + SC_REG_TICK] %asi, %g2 + wrpr %g2, %tick + + /* Restore global regs */ + ldxa [%g3 + SC_REG_G4] %asi, %g4 + ldxa [%g3 + SC_REG_G5] %asi, %g5 + ldxa [%g3 + SC_REG_G6] %asi, %g6 + + wr %g1, %g0, %asi + + restore + restore + + wrpr %g0, 14, %pil + + retl + mov %g0, %o0 + +fill_itlb: + ba pbe_loop + wrpr %g0, 15, %pil diff --git a/arch/sparc/prom/bootstr_32.c b/arch/sparc/prom/bootstr_32.c index f5ec32e0d41..d2b49d2365e 100644 --- a/arch/sparc/prom/bootstr_32.c +++ b/arch/sparc/prom/bootstr_32.c @@ -23,23 +23,25 @@ prom_getbootargs(void) return barg_buf; } - switch(prom_vers) { + switch (prom_vers) { case PROM_V0: cp = barg_buf; /* Start from 1 and go over fd(0,0,0)kernel */ - for(iter = 1; iter < 8; iter++) { + for (iter = 1; iter < 8; iter++) { arg = (*(romvec->pv_v0bootargs))->argv[iter]; if (arg == NULL) break; - while(*arg != 0) { + while (*arg != 0) { /* Leave place for space and null. */ - if(cp >= barg_buf + BARG_LEN-2){ + if (cp >= barg_buf + BARG_LEN - 2) /* We might issue a warning here. */ break; - } *cp++ = *arg++; } *cp++ = ' '; + if (cp >= barg_buf + BARG_LEN - 1) + /* We might issue a warning here. */ + break; } *cp = 0; break; diff --git a/arch/sparc/prom/init_32.c b/arch/sparc/prom/init_32.c index 26c64cea3c9..9ac30c2b7db 100644 --- a/arch/sparc/prom/init_32.c +++ b/arch/sparc/prom/init_32.c @@ -27,13 +27,10 @@ EXPORT_SYMBOL(prom_root_node); struct linux_nodeops *prom_nodeops; /* You must call prom_init() before you attempt to use any of the - * routines in the prom library. It returns 0 on success, 1 on - * failure. It gets passed the pointer to the PROM vector. + * routines in the prom library. + * It gets passed the pointer to the PROM vector. */ -extern void prom_meminit(void); -extern void prom_ranges_init(void); - void __init prom_init(struct linux_romvec *rp) { romvec = rp; diff --git a/arch/sparc/prom/init_64.c b/arch/sparc/prom/init_64.c index 5016c5e2057..d95db755828 100644 --- a/arch/sparc/prom/init_64.c +++ b/arch/sparc/prom/init_64.c @@ -22,8 +22,8 @@ int prom_stdout; phandle prom_chosen_node; /* You must call prom_init() before you attempt to use any of the - * routines in the prom library. It returns 0 on success, 1 on - * failure. It gets passed the pointer to the PROM vector. + * routines in the prom library. + * It gets passed the pointer to the PROM vector. */ extern void prom_cif_init(void *, void *); diff --git a/arch/sparc/prom/misc_64.c b/arch/sparc/prom/misc_64.c index f178b9dcc7b..53a696d3eb3 100644 --- a/arch/sparc/prom/misc_64.c +++ b/arch/sparc/prom/misc_64.c @@ -81,11 +81,6 @@ void prom_feval(const char *fstring) } EXPORT_SYMBOL(prom_feval); -#ifdef CONFIG_SMP -extern void smp_capture(void); -extern void smp_release(void); -#endif - /* Drop into the prom, with the chance to continue with the 'go' * prom command. */ diff --git a/arch/sparc/prom/p1275.c b/arch/sparc/prom/p1275.c index 04a4540509d..e58b8172631 100644 --- a/arch/sparc/prom/p1275.c +++ b/arch/sparc/prom/p1275.c @@ -5,7 +5,6 @@ */ #include <linux/kernel.h> -#include <linux/init.h> #include <linux/sched.h> #include <linux/smp.h> #include <linux/string.h> diff --git a/arch/sparc/prom/tree_64.c b/arch/sparc/prom/tree_64.c index 92204c3800b..bd1b2a3ac34 100644 --- a/arch/sparc/prom/tree_64.c +++ b/arch/sparc/prom/tree_64.c @@ -39,7 +39,7 @@ inline phandle __prom_getchild(phandle node) return prom_node_to_node("child", node); } -inline phandle prom_getchild(phandle node) +phandle prom_getchild(phandle node) { phandle cnode; @@ -72,7 +72,7 @@ inline phandle __prom_getsibling(phandle node) return prom_node_to_node(prom_peer_name, node); } -inline phandle prom_getsibling(phandle node) +phandle prom_getsibling(phandle node) { phandle sibnode; @@ -89,7 +89,7 @@ EXPORT_SYMBOL(prom_getsibling); /* Return the length in bytes of property 'prop' at node 'node'. * Return -1 on error. */ -inline int prom_getproplen(phandle node, const char *prop) +int prom_getproplen(phandle node, const char *prop) { unsigned long args[6]; @@ -113,8 +113,8 @@ EXPORT_SYMBOL(prom_getproplen); * 'buffer' which has a size of 'bufsize'. If the acquisition * was successful the length will be returned, else -1 is returned. */ -inline int prom_getproperty(phandle node, const char *prop, - char *buffer, int bufsize) +int prom_getproperty(phandle node, const char *prop, + char *buffer, int bufsize) { unsigned long args[8]; int plen; @@ -141,7 +141,7 @@ EXPORT_SYMBOL(prom_getproperty); /* Acquire an integer property and return its value. Returns -1 * on failure. */ -inline int prom_getint(phandle node, const char *prop) +int prom_getint(phandle node, const char *prop) { int intprop; @@ -235,7 +235,7 @@ static const char *prom_nextprop_name = "nextprop"; /* Return the first property type for node 'node'. * buffer should be at least 32B in length */ -inline char *prom_firstprop(phandle node, char *buffer) +char *prom_firstprop(phandle node, char *buffer) { unsigned long args[7]; @@ -261,7 +261,7 @@ EXPORT_SYMBOL(prom_firstprop); * at node 'node' . Returns NULL string if no more * property types for this node. */ -inline char *prom_nextprop(phandle node, const char *oprop, char *buffer) +char *prom_nextprop(phandle node, const char *oprop, char *buffer) { unsigned long args[7]; char buf[32]; |