aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2012-12-20 10:07:25 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2012-12-20 10:07:25 -0800
commit787314c35fbb97e02823a1b8eb8cfa58f366cd49 (patch)
tree3fe5a484c1846c80361217a726997484533e8344
parent6491d4d02893d9787ba67279595990217177b351 (diff)
parent9c6ecf6a3ade2dc4b03a239af68058b22897af41 (diff)
Merge tag 'iommu-updates-v3.8' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull IOMMU updates from Joerg Roedel: "A few new features this merge-window. The most important one is probably, that dma-debug now warns if a dma-handle is not checked with dma_mapping_error by the device driver. This requires minor changes to some architectures which make use of dma-debug. Most of these changes have the respective Acks by the Arch-Maintainers. Besides that there are updates to the AMD IOMMU driver for refactor the IOMMU-Groups support and to make sure it does not trigger a hardware erratum. The OMAP changes (for which I pulled in a branch from Tony Lindgren's tree) have a conflict in linux-next with the arm-soc tree. The conflict is in the file arch/arm/mach-omap2/clock44xx_data.c which is deleted in the arm-soc tree. It is safe to delete the file too so solve the conflict. Similar changes are done in the arm-soc tree in the common clock framework migration. A missing hunk from the patch in the IOMMU tree will be submitted as a seperate patch when the merge-window is closed." * tag 'iommu-updates-v3.8' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (29 commits) ARM: dma-mapping: support debug_dma_mapping_error ARM: OMAP4: hwmod data: ipu and dsp to use parent clocks instead of leaf clocks iommu/omap: Adapt to runtime pm iommu/omap: Migrate to hwmod framework iommu/omap: Keep mmu enabled when requested iommu/omap: Remove redundant clock handling on ISR iommu/amd: Remove obsolete comment iommu/amd: Don't use 512GB pages iommu/tegra: smmu: Move bus_set_iommu after probe for multi arch iommu/tegra: gart: Move bus_set_iommu after probe for multi arch iommu/tegra: smmu: Remove unnecessary PTC/TLB flush all tile: dma_debug: add debug_dma_mapping_error support sh: dma_debug: add debug_dma_mapping_error support powerpc: dma_debug: add debug_dma_mapping_error support mips: dma_debug: add debug_dma_mapping_error support microblaze: dma-mapping: support debug_dma_mapping_error ia64: dma_debug: add debug_dma_mapping_error support c6x: dma_debug: add debug_dma_mapping_error support ARM64: dma_debug: add debug_dma_mapping_error support intel-iommu: Prevent devices with RMRRs from being placed into SI Domain ...
-rw-r--r--Documentation/DMA-API-HOWTO.txt126
-rw-r--r--Documentation/DMA-API.txt12
-rw-r--r--arch/arm/include/asm/dma-mapping.h1
-rw-r--r--arch/arm/mach-omap2/devices.c2
-rw-r--r--arch/arm/mach-omap2/omap-iommu.c167
-rw-r--r--arch/arm/mach-omap2/omap_hwmod_44xx_data.c4
-rw-r--r--arch/arm64/include/asm/dma-mapping.h1
-rw-r--r--arch/c6x/include/asm/dma-mapping.h1
-rw-r--r--arch/ia64/include/asm/dma-mapping.h1
-rw-r--r--arch/microblaze/include/asm/dma-mapping.h2
-rw-r--r--arch/mips/include/asm/dma-mapping.h2
-rw-r--r--arch/powerpc/include/asm/dma-mapping.h1
-rw-r--r--arch/sh/include/asm/dma-mapping.h1
-rw-r--r--arch/sparc/include/asm/dma-mapping.h1
-rw-r--r--arch/tile/include/asm/dma-mapping.h1
-rw-r--r--arch/x86/include/asm/dma-mapping.h1
-rw-r--r--drivers/iommu/amd_iommu.c196
-rw-r--r--drivers/iommu/amd_iommu_types.h1
-rw-r--r--drivers/iommu/intel-iommu.c31
-rw-r--r--drivers/iommu/omap-iommu.c68
-rw-r--r--drivers/iommu/omap-iommu.h3
-rw-r--r--drivers/iommu/omap-iommu2.c36
-rw-r--r--drivers/iommu/tegra-gart.c2
-rw-r--r--drivers/iommu/tegra-smmu.c6
-rw-r--r--include/linux/dma-debug.h7
-rw-r--r--include/linux/platform_data/iommu-omap.h9
-rw-r--r--lib/dma-debug.c66
27 files changed, 477 insertions, 272 deletions
diff --git a/Documentation/DMA-API-HOWTO.txt b/Documentation/DMA-API-HOWTO.txt
index a0b6250add7..4a4fb295cee 100644
--- a/Documentation/DMA-API-HOWTO.txt
+++ b/Documentation/DMA-API-HOWTO.txt
@@ -468,11 +468,46 @@ To map a single region, you do:
size_t size = buffer->len;
dma_handle = dma_map_single(dev, addr, size, direction);
+ if (dma_mapping_error(dma_handle)) {
+ /*
+ * reduce current DMA mapping usage,
+ * delay and try again later or
+ * reset driver.
+ */
+ goto map_error_handling;
+ }
and to unmap it:
dma_unmap_single(dev, dma_handle, size, direction);
+You should call dma_mapping_error() as dma_map_single() could fail and return
+error. Not all dma implementations support dma_mapping_error() interface.
+However, it is a good practice to call dma_mapping_error() interface, which
+will invoke the generic mapping error check interface. Doing so will ensure
+that the mapping code will work correctly on all dma implementations without
+any dependency on the specifics of the underlying implementation. Using the
+returned address without checking for errors could result in failures ranging
+from panics to silent data corruption. Couple of example of incorrect ways to
+check for errors that make assumptions about the underlying dma implementation
+are as follows and these are applicable to dma_map_page() as well.
+
+Incorrect example 1:
+ dma_addr_t dma_handle;
+
+ dma_handle = dma_map_single(dev, addr, size, direction);
+ if ((dma_handle & 0xffff != 0) || (dma_handle >= 0x1000000)) {
+ goto map_error;
+ }
+
+Incorrect example 2:
+ dma_addr_t dma_handle;
+
+ dma_handle = dma_map_single(dev, addr, size, direction);
+ if (dma_handle == DMA_ERROR_CODE) {
+ goto map_error;
+ }
+
You should call dma_unmap_single when the DMA activity is finished, e.g.
from the interrupt which told you that the DMA transfer is done.
@@ -489,6 +524,14 @@ Specifically:
size_t size = buffer->len;
dma_handle = dma_map_page(dev, page, offset, size, direction);
+ if (dma_mapping_error(dma_handle)) {
+ /*
+ * reduce current DMA mapping usage,
+ * delay and try again later or
+ * reset driver.
+ */
+ goto map_error_handling;
+ }
...
@@ -496,6 +539,12 @@ Specifically:
Here, "offset" means byte offset within the given page.
+You should call dma_mapping_error() as dma_map_page() could fail and return
+error as outlined under the dma_map_single() discussion.
+
+You should call dma_unmap_page when the DMA activity is finished, e.g.
+from the interrupt which told you that the DMA transfer is done.
+
With scatterlists, you map a region gathered from several regions by:
int i, count = dma_map_sg(dev, sglist, nents, direction);
@@ -578,6 +627,14 @@ to use the dma_sync_*() interfaces.
dma_addr_t mapping;
mapping = dma_map_single(cp->dev, buffer, len, DMA_FROM_DEVICE);
+ if (dma_mapping_error(dma_handle)) {
+ /*
+ * reduce current DMA mapping usage,
+ * delay and try again later or
+ * reset driver.
+ */
+ goto map_error_handling;
+ }
cp->rx_buf = buffer;
cp->rx_len = len;
@@ -658,6 +715,75 @@ failure can be determined by:
* delay and try again later or
* reset driver.
*/
+ goto map_error_handling;
+ }
+
+- unmap pages that are already mapped, when mapping error occurs in the middle
+ of a multiple page mapping attempt. These example are applicable to
+ dma_map_page() as well.
+
+Example 1:
+ dma_addr_t dma_handle1;
+ dma_addr_t dma_handle2;
+
+ dma_handle1 = dma_map_single(dev, addr, size, direction);
+ if (dma_mapping_error(dev, dma_handle1)) {
+ /*
+ * reduce current DMA mapping usage,
+ * delay and try again later or
+ * reset driver.
+ */
+ goto map_error_handling1;
+ }
+ dma_handle2 = dma_map_single(dev, addr, size, direction);
+ if (dma_mapping_error(dev, dma_handle2)) {
+ /*
+ * reduce current DMA mapping usage,
+ * delay and try again later or
+ * reset driver.
+ */
+ goto map_error_handling2;
+ }
+
+ ...
+
+ map_error_handling2:
+ dma_unmap_single(dma_handle1);
+ map_error_handling1:
+
+Example 2: (if buffers are allocated a loop, unmap all mapped buffers when
+ mapping error is detected in the middle)
+
+ dma_addr_t dma_addr;
+ dma_addr_t array[DMA_BUFFERS];
+ int save_index = 0;
+
+ for (i = 0; i < DMA_BUFFERS; i++) {
+
+ ...
+
+ dma_addr = dma_map_single(dev, addr, size, direction);
+ if (dma_mapping_error(dev, dma_addr)) {
+ /*
+ * reduce current DMA mapping usage,
+ * delay and try again later or
+ * reset driver.
+ */
+ goto map_error_handling;
+ }
+ array[i].dma_addr = dma_addr;
+ save_index++;
+ }
+
+ ...
+
+ map_error_handling:
+
+ for (i = 0; i < save_index; i++) {
+
+ ...
+
+ dma_unmap_single(array[i].dma_addr);
}
Networking drivers must call dev_kfree_skb to free the socket buffer
diff --git a/Documentation/DMA-API.txt b/Documentation/DMA-API.txt
index 66bd97a95f1..78a6c569d20 100644
--- a/Documentation/DMA-API.txt
+++ b/Documentation/DMA-API.txt
@@ -678,3 +678,15 @@ out of dma_debug_entries. These entries are preallocated at boot. The number
of preallocated entries is defined per architecture. If it is too low for you
boot with 'dma_debug_entries=<your_desired_number>' to overwrite the
architectural default.
+
+void debug_dmap_mapping_error(struct device *dev, dma_addr_t dma_addr);
+
+dma-debug interface debug_dma_mapping_error() to debug drivers that fail
+to check dma mapping errors on addresses returned by dma_map_single() and
+dma_map_page() interfaces. This interface clears a flag set by
+debug_dma_map_page() to indicate that dma_mapping_error() has been called by
+the driver. When driver does unmap, debug_dma_unmap() checks the flag and if
+this flag is still set, prints warning message that includes call trace that
+leads up to the unmap. This interface can be called from dma_mapping_error()
+routines to enable dma mapping error check debugging.
+
diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h
index 67d06324e74..5b579b95150 100644
--- a/arch/arm/include/asm/dma-mapping.h
+++ b/arch/arm/include/asm/dma-mapping.h
@@ -91,6 +91,7 @@ static inline dma_addr_t virt_to_dma(struct device *dev, void *addr)
*/
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/arm/mach-omap2/devices.c b/arch/arm/mach-omap2/devices.c
index 4abb8b5e9bc..5e304d0719a 100644
--- a/arch/arm/mach-omap2/devices.c
+++ b/arch/arm/mach-omap2/devices.c
@@ -226,7 +226,7 @@ static struct platform_device omap3isp_device = {
};
static struct omap_iommu_arch_data omap3_isp_iommu = {
- .name = "isp",
+ .name = "mmu_isp",
};
int omap3_init_camera(struct isp_platform_data *pdata)
diff --git a/arch/arm/mach-omap2/omap-iommu.c b/arch/arm/mach-omap2/omap-iommu.c
index a6a4ff8744b..7642fc4672c 100644
--- a/arch/arm/mach-omap2/omap-iommu.c
+++ b/arch/arm/mach-omap2/omap-iommu.c
@@ -12,153 +12,60 @@
#include <linux/module.h>
#include <linux/platform_device.h>
+#include <linux/err.h>
+#include <linux/slab.h>
#include <linux/platform_data/iommu-omap.h>
+#include <plat/omap_hwmod.h>
+#include <plat/omap_device.h>
-#include "soc.h"
-#include "common.h"
-
-struct iommu_device {
- resource_size_t base;
- int irq;
- struct iommu_platform_data pdata;
- struct resource res[2];
-};
-static struct iommu_device *devices;
-static int num_iommu_devices;
-
-#ifdef CONFIG_ARCH_OMAP3
-static struct iommu_device omap3_devices[] = {
- {
- .base = 0x480bd400,
- .irq = 24 + OMAP_INTC_START,
- .pdata = {
- .name = "isp",
- .nr_tlb_entries = 8,
- .clk_name = "cam_ick",
- .da_start = 0x0,
- .da_end = 0xFFFFF000,
- },
- },
-#if defined(CONFIG_OMAP_IOMMU_IVA2)
- {
- .base = 0x5d000000,
- .irq = 28 + OMAP_INTC_START,
- .pdata = {
- .name = "iva2",
- .nr_tlb_entries = 32,
- .clk_name = "iva2_ck",
- .da_start = 0x11000000,
- .da_end = 0xFFFFF000,
- },
- },
-#endif
-};
-#define NR_OMAP3_IOMMU_DEVICES ARRAY_SIZE(omap3_devices)
-static struct platform_device *omap3_iommu_pdev[NR_OMAP3_IOMMU_DEVICES];
-#else
-#define omap3_devices NULL
-#define NR_OMAP3_IOMMU_DEVICES 0
-#define omap3_iommu_pdev NULL
-#endif
-
-#ifdef CONFIG_ARCH_OMAP4
-static struct iommu_device omap4_devices[] = {
- {
- .base = OMAP4_MMU1_BASE,
- .irq = 100 + OMAP44XX_IRQ_GIC_START,
- .pdata = {
- .name = "ducati",
- .nr_tlb_entries = 32,
- .clk_name = "ipu_fck",
- .da_start = 0x0,
- .da_end = 0xFFFFF000,
- },
- },
- {
- .base = OMAP4_MMU2_BASE,
- .irq = 28 + OMAP44XX_IRQ_GIC_START,
- .pdata = {
- .name = "tesla",
- .nr_tlb_entries = 32,
- .clk_name = "dsp_fck",
- .da_start = 0x0,
- .da_end = 0xFFFFF000,
- },
- },
-};
-#define NR_OMAP4_IOMMU_DEVICES ARRAY_SIZE(omap4_devices)
-static struct platform_device *omap4_iommu_pdev[NR_OMAP4_IOMMU_DEVICES];
-#else
-#define omap4_devices NULL
-#define NR_OMAP4_IOMMU_DEVICES 0
-#define omap4_iommu_pdev NULL
-#endif
-
-static struct platform_device **omap_iommu_pdev;
-
-static int __init omap_iommu_init(void)
+static int __init omap_iommu_dev_init(struct omap_hwmod *oh, void *unused)
{
- int i, err;
- struct resource res[] = {
- { .flags = IORESOURCE_MEM },
- { .flags = IORESOURCE_IRQ },
- };
+ struct platform_device *pdev;
+ struct iommu_platform_data *pdata;
+ struct omap_mmu_dev_attr *a = (struct omap_mmu_dev_attr *)oh->dev_attr;
+ static int i;
+
+ pdata = kzalloc(sizeof(*pdata), GFP_KERNEL);
+ if (!pdata)
+ return -ENOMEM;
+
+ pdata->name = oh->name;
+ pdata->nr_tlb_entries = a->nr_tlb_entries;
+ pdata->da_start = a->da_start;
+ pdata->da_end = a->da_end;
+
+ if (oh->rst_lines_cnt == 1) {
+ pdata->reset_name = oh->rst_lines->name;
+ pdata->assert_reset = omap_device_assert_hardreset;
+ pdata->deassert_reset = omap_device_deassert_hardreset;
+ }
- if (cpu_is_omap34xx()) {
- devices = omap3_devices;
- omap_iommu_pdev = omap3_iommu_pdev;
- num_iommu_devices = NR_OMAP3_IOMMU_DEVICES;
- } else if (cpu_is_omap44xx()) {
- devices = omap4_devices;
- omap_iommu_pdev = omap4_iommu_pdev;
- num_iommu_devices = NR_OMAP4_IOMMU_DEVICES;
- } else
- return -ENODEV;
+ pdev = omap_device_build("omap-iommu", i, oh, pdata, sizeof(*pdata),
+ NULL, 0, 0);
- for (i = 0; i < num_iommu_devices; i++) {
- struct platform_device *pdev;
- const struct iommu_device *d = &devices[i];
+ kfree(pdata);
- pdev = platform_device_alloc("omap-iommu", i);
- if (!pdev) {
- err = -ENOMEM;
- goto err_out;
- }
+ if (IS_ERR(pdev)) {
+ pr_err("%s: device build err: %ld\n", __func__, PTR_ERR(pdev));
+ return PTR_ERR(pdev);
+ }
- res[0].start = d->base;
- res[0].end = d->base + MMU_REG_SIZE - 1;
- res[1].start = res[1].end = d->irq;
+ i++;
- err = platform_device_add_resources(pdev, res,
- ARRAY_SIZE(res));
- if (err)
- goto err_out;
- err = platform_device_add_data(pdev, &d->pdata,
- sizeof(d->pdata));
- if (err)
- goto err_out;
- err = platform_device_add(pdev);
- if (err)
- goto err_out;
- omap_iommu_pdev[i] = pdev;
- }
return 0;
+}
-err_out:
- while (i--)
- platform_device_put(omap_iommu_pdev[i]);
- return err;
+static int __init omap_iommu_init(void)
+{
+ return omap_hwmod_for_each_by_class("mmu", omap_iommu_dev_init, NULL);
}
/* must be ready before omap3isp is probed */
subsys_initcall(omap_iommu_init);
static void __exit omap_iommu_exit(void)
{
- int i;
-
- for (i = 0; i < num_iommu_devices; i++)
- platform_device_unregister(omap_iommu_pdev[i]);
+ /* Do nothing */
}
module_exit(omap_iommu_exit);
diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c
index 272b0178dba..f9fab942d5b 100644
--- a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c
@@ -653,7 +653,7 @@ static struct omap_hwmod omap44xx_dsp_hwmod = {
.mpu_irqs = omap44xx_dsp_irqs,
.rst_lines = omap44xx_dsp_resets,
.rst_lines_cnt = ARRAY_SIZE(omap44xx_dsp_resets),
- .main_clk = "dsp_fck",
+ .main_clk = "dpll_iva_m4x2_ck",
.prcm = {
.omap4 = {
.clkctrl_offs = OMAP4_CM_TESLA_TESLA_CLKCTRL_OFFSET,
@@ -1679,7 +1679,7 @@ static struct omap_hwmod omap44xx_ipu_hwmod = {
.mpu_irqs = omap44xx_ipu_irqs,
.rst_lines = omap44xx_ipu_resets,
.rst_lines_cnt = ARRAY_SIZE(omap44xx_ipu_resets),
- .main_clk = "ipu_fck",
+ .main_clk = "ducati_clk_mux_ck",
.prcm = {
.omap4 = {
.clkctrl_offs = OMAP4_CM_DUCATI_DUCATI_CLKCTRL_OFFSET,
diff --git a/arch/arm64/include/asm/dma-mapping.h b/arch/arm64/include/asm/dma-mapping.h
index 538f4b44db5..99477689419 100644
--- a/arch/arm64/include/asm/dma-mapping.h
+++ b/arch/arm64/include/asm/dma-mapping.h
@@ -50,6 +50,7 @@ static inline phys_addr_t dma_to_phys(struct device *dev, dma_addr_t dev_addr)
static inline int dma_mapping_error(struct device *dev, dma_addr_t dev_addr)
{
struct dma_map_ops *ops = get_dma_ops(dev);
+ debug_dma_mapping_error(dev, dev_addr);
return ops->mapping_error(dev, dev_addr);
}
diff --git a/arch/c6x/include/asm/dma-mapping.h b/arch/c6x/include/asm/dma-mapping.h
index 03579fd99db..3c694065030 100644
--- a/arch/c6x/include/asm/dma-mapping.h
+++ b/arch/c6x/include/asm/dma-mapping.h
@@ -32,6 +32,7 @@ static inline int dma_set_mask(struct device *dev, u64 dma_mask)
*/
static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
+ debug_dma_mapping_error(dev, dma_addr);
return dma_addr == ~0;
}
diff --git a/arch/ia64/include/asm/dma-mapping.h b/arch/ia64/include/asm/dma-mapping.h
index 4f5e8148440..cf3ab7e784b 100644
--- a/arch/ia64/include/asm/dma-mapping.h
+++ b/arch/ia64/include/asm/dma-mapping.h
@@ -58,6 +58,7 @@ static inline void dma_free_attrs(struct device *dev, size_t size,
static inline int dma_mapping_error(struct device *dev, dma_addr_t daddr)
{
struct dma_map_ops *ops = platform_dma_get_ops(dev);
+ debug_dma_mapping_error(dev, daddr);
return ops->mapping_error(dev, daddr);
}
diff --git a/arch/microblaze/include/asm/dma-mapping.h b/arch/microblaze/include/asm/dma-mapping.h
index 01d228286cb..46460f1c49c 100644
--- a/arch/microblaze/include/asm/dma-mapping.h
+++ b/arch/microblaze/include/asm/dma-mapping.h
@@ -114,6 +114,8 @@ static inline void __dma_sync(unsigned long paddr,
static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
struct dma_map_ops *ops = get_dma_ops(dev);
+
+ debug_dma_mapping_error(dev, dma_addr);
if (ops->mapping_error)
return ops->mapping_error(dev, dma_addr);
diff --git a/arch/mips/include/asm/dma-mapping.h b/arch/mips/include/asm/dma-mapping.h
index be39a12901c..006b43e38a9 100644
--- a/arch/mips/include/asm/dma-mapping.h
+++ b/arch/mips/include/asm/dma-mapping.h
@@ -40,6 +40,8 @@ static inline int dma_supported(struct device *dev, u64 mask)
static inline int dma_mapping_error(struct device *dev, u64 mask)
{
struct dma_map_ops *ops = get_dma_ops(dev);
+
+ debug_dma_mapping_error(dev, mask);
return ops->mapping_error(dev, mask);
}
diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h
index 78160874809..e27e9ad6818 100644
--- a/arch/powerpc/include/asm/dma-mapping.h
+++ b/arch/powerpc/include/asm/dma-mapping.h
@@ -172,6 +172,7 @@ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
struct dma_map_ops *dma_ops = get_dma_ops(dev);
+ debug_dma_mapping_error(dev, dma_addr);
if (dma_ops->mapping_error)
return dma_ops->mapping_error(dev, dma_addr);
diff --git a/arch/sh/include/asm/dma-mapping.h b/arch/sh/include/asm/dma-mapping.h
index 8bd965e00a1..b437f2c780b 100644
--- a/arch/sh/include/asm/dma-mapping.h
+++ b/arch/sh/include/asm/dma-mapping.h
@@ -46,6 +46,7 @@ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
struct dma_map_ops *ops = get_dma_ops(dev);
+ debug_dma_mapping_error(dev, dma_addr);
if (ops->mapping_error)
return ops->mapping_error(dev, dma_addr);
diff --git a/arch/sparc/include/asm/dma-mapping.h b/arch/sparc/include/asm/dma-mapping.h
index 8493fd3c7ba..05fe53f5346 100644
--- a/arch/sparc/include/asm/dma-mapping.h
+++ b/arch/sparc/include/asm/dma-mapping.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/tile/include/asm/dma-mapping.h b/arch/tile/include/asm/dma-mapping.h
index 4b6247d1a31..f2ff191376b 100644
--- a/arch/tile/include/asm/dma-mapping.h
+++ b/arch/tile/include/asm/dma-mapping.h
@@ -72,6 +72,7 @@ static inline bool dma_capable(struct device *dev, dma_addr_t addr, 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 get_dma_ops(dev)->mapping_error(dev, dma_addr);
}
diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h
index f7b4c7903e7..808dae63eee 100644
--- a/arch/x86/include/asm/dma-mapping.h
+++ b/arch/x86/include/asm/dma-mapping.h
@@ -47,6 +47,7 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev)
static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
struct dma_map_ops *ops = get_dma_ops(dev);
+ debug_dma_mapping_error(dev, dma_addr);
if (ops->mapping_error)
return ops->mapping_error(dev, dma_addr);
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index 55074cba20e..c1c74e030a5 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -57,17 +57,9 @@
* physically contiguous memory regions it is mapping into page sizes
* that we support.
*
- * Traditionally the IOMMU core just handed us the mappings directly,
- * after making sure the size is an order of a 4KiB page and that the
- * mapping has natural alignment.
- *
- * To retain this behavior, we currently advertise that we support
- * all page sizes that are an order of 4KiB.
- *
- * If at some point we'd like to utilize the IOMMU core's new behavior,
- * we could change this to advertise the real page sizes we support.
+ * 512GB Pages are not supported due to a hardware bug
*/
-#define AMD_IOMMU_PGSIZES (~0xFFFUL)
+#define AMD_IOMMU_PGSIZES ((~0xFFFUL) & ~(2ULL << 38))
static DEFINE_RWLOCK(amd_iommu_devtable_lock);
@@ -140,6 +132,9 @@ static void free_dev_data(struct iommu_dev_data *dev_data)
list_del(&dev_data->dev_data_list);
spin_unlock_irqrestore(&dev_data_list_lock, flags);
+ if (dev_data->group)
+ iommu_group_put(dev_data->group);
+
kfree(dev_data);
}
@@ -274,41 +269,23 @@ static void swap_pci_ref(struct pci_dev **from, struct pci_dev *to)
*from = to;
}
-#define REQ_ACS_FLAGS (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF)
-
-static int iommu_init_device(struct device *dev)
+static struct pci_bus *find_hosted_bus(struct pci_bus *bus)
{
- struct pci_dev *dma_pdev = NULL, *pdev = to_pci_dev(dev);
- struct iommu_dev_data *dev_data;
- struct iommu_group *group;
- u16 alias;
- int ret;
-
- if (dev->archdata.iommu)
- return 0;
-
- dev_data = find_dev_data(get_device_id(dev));
- if (!dev_data)
- return -ENOMEM;
-
- alias = amd_iommu_alias_table[dev_data->devid];
- if (alias != dev_data->devid) {
- struct iommu_dev_data *alias_data;
+ while (!bus->self) {
+ if (!pci_is_root_bus(bus))
+ bus = bus->parent;
+ else
+ return ERR_PTR(-ENODEV);
+ }
- alias_data = find_dev_data(alias);
- if (alias_data == NULL) {
- pr_err("AMD-Vi: Warning: Unhandled device %s\n",
- dev_name(dev));
- free_dev_data(dev_data);
- return -ENOTSUPP;
- }
- dev_data->alias_data = alias_data;
+ return bus;
+}
- dma_pdev = pci_get_bus_and_slot(alias >> 8, alias & 0xff);
- }
+#define REQ_ACS_FLAGS (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF)
- if (dma_pdev == NULL)
- dma_pdev = pci_dev_get(pdev);
+static struct pci_dev *get_isolation_root(struct pci_dev *pdev)
+{
+ struct pci_dev *dma_pdev = pdev;
/* Account for quirked devices */
swap_pci_ref(&dma_pdev, pci_get_dma_source(dma_pdev));
@@ -330,14 +307,9 @@ static int iommu_init_device(struct device *dev)
* Finding the next device may require skipping virtual buses.
*/
while (!pci_is_root_bus(dma_pdev->bus)) {
- struct pci_bus *bus = dma_pdev->bus;
-
- while (!bus->self) {
- if (!pci_is_root_bus(bus))
- bus = bus->parent;
- else
- goto root_bus;
- }
+ struct pci_bus *bus = find_hosted_bus(dma_pdev->bus);
+ if (IS_ERR(bus))
+ break;
if (pci_acs_path_enabled(bus->self, NULL, REQ_ACS_FLAGS))
break;
@@ -345,19 +317,137 @@ static int iommu_init_device(struct device *dev)
swap_pci_ref(&dma_pdev, pci_dev_get(bus->self));
}
-root_bus:
- group = iommu_group_get(&dma_pdev->dev);
- pci_dev_put(dma_pdev);
+ return dma_pdev;
+}
+
+static int use_pdev_iommu_group(struct pci_dev *pdev, struct device *dev)
+{
+ struct iommu_group *group = iommu_group_get(&pdev->dev);
+ int ret;
+
if (!group) {
group = iommu_group_alloc();
if (IS_ERR(group))
return PTR_ERR(group);
+
+ WARN_ON(&pdev->dev != dev);
}
ret = iommu_group_add_device(group, dev);
-
iommu_group_put(group);
+ return ret;
+}
+
+static int use_dev_data_iommu_group(struct iommu_dev_data *dev_data,
+ struct device *dev)
+{
+ if (!dev_data->group) {
+ struct iommu_group *group = iommu_group_alloc();
+ if (IS_ERR(group))
+ return PTR_ERR(group);
+
+ dev_data->group = group;
+ }
+
+ return iommu_group_add_device(dev_data->group, dev);
+}
+
+static int init_iommu_group(struct device *dev)
+{
+ struct iommu_dev_data *dev_data;
+ struct iommu_group *group;
+ struct pci_dev *dma_pdev;
+ int ret;
+
+ group = iommu_group_get(dev);
+ if (group) {
+ iommu_group_put(group);
+ return 0;
+ }
+
+ dev_data = find_dev_data(get_device_id(dev));
+ if (!dev_data)
+ return -ENOMEM;
+
+ if (dev_data->alias_data) {
+ u16 alias;
+ struct pci_bus *bus;
+
+ if (dev_data->alias_data->group)
+ goto use_group;
+
+ /*
+ * If the alias device exists, it's effectively just a first
+ * level quirk for finding the DMA source.
+ */
+ alias = amd_iommu_alias_table[dev_data->devid];
+ dma_pdev = pci_get_bus_and_slot(alias >> 8, alias & 0xff);
+ if (dma_pdev) {
+ dma_pdev = get_isolation_root(dma_pdev);
+ goto use_pdev;
+ }
+
+ /*
+ * If the alias is virtual, try to find a parent device
+ * and test whether the IOMMU group is actualy rooted above
+ * the alias. Be careful to also test the parent device if
+ * we think the alias is the root of the group.
+ */
+ bus = pci_find_bus(0, alias >> 8);
+ if (!bus)
+ goto use_group;
+
+ bus = find_hosted_bus(bus);
+ if (IS_ERR(bus) || !bus->self)
+ goto use_group;
+
+ dma_pdev = get_isolation_root(pci_dev_get(bus->self));
+ if (dma_pdev != bus->self || (dma_pdev->multifunction &&
+ !pci_acs_enabled(dma_pdev, REQ_ACS_FLAGS)))
+ goto use_pdev;
+
+ pci_dev_put(dma_pdev);
+ goto use_group;
+ }
+
+ dma_pdev = get_isolation_root(pci_dev_get(to_pci_dev(dev)));
+use_pdev:
+ ret = use_pdev_iommu_group(dma_pdev, dev);
+ pci_dev_put(dma_pdev);
+ return ret;
+use_group:
+ return use_dev_data_iommu_group(dev_data->alias_data, dev);
+}
+
+static int iommu_init_device(struct device *dev)
+{
+ struct pci_dev *pdev = to_pci_dev(dev);
+ struct iommu_dev_data *dev_data;
+ u16 alias;
+ int ret;
+
+ if (dev->archdata.iommu)
+ return 0;
+
+ dev_data = find_dev_data(get_device_id(dev));
+ if (!dev_data)
+ return -ENOMEM;
+
+ alias = amd_iommu_alias_table[dev_data->devid];
+ if (alias != dev_data->devid) {
+ struct iommu_dev_data *alias_data;
+
+ alias_data = find_dev_data(alias);
+ if (alias_data == NULL) {
+ pr_err("AMD-Vi: Warning: Unhandled device %s\n",
+ dev_name(dev));
+ free_dev_data(dev_data);
+ return -ENOTSUPP;
+ }
+ dev_data->alias_data = alias_data;
+ }
+ ret = init_iommu_group(dev);
if (ret)
return ret;
diff --git a/drivers/iommu/amd_iommu_types.h b/drivers/iommu/amd_iommu_types.h
index c9aa3d079ff..e38ab438bb3 100644
--- a/drivers/iommu/amd_iommu_types.h
+++ b/drivers/iommu/amd_iommu_types.h
@@ -426,6 +426,7 @@ struct iommu_dev_data {
struct iommu_dev_data *alias_data;/* The alias dev_