Merge branch 'oprofile-for-tip' of git://git.kernel.org/pub/scm/linux/kernel/git/rric/oprofile into x86/urgent
diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h
index 1a30c04..ac302a2 100644
--- a/arch/x86/include/asm/amd_iommu_types.h
+++ b/arch/x86/include/asm/amd_iommu_types.h
@@ -251,13 +251,6 @@
/* Pointer to PCI device of this IOMMU */
struct pci_dev *dev;
- /*
- * Capability pointer. There could be more than one IOMMU per PCI
- * device function if there are more than one AMD IOMMU capability
- * pointers.
- */
- u16 cap_ptr;
-
/* physical address of MMIO space */
u64 mmio_phys;
/* virtual address of MMIO space */
@@ -266,6 +259,13 @@
/* capabilities of that IOMMU read from ACPI */
u32 cap;
+ /*
+ * Capability pointer. There could be more than one IOMMU per PCI
+ * device function if there are more than one AMD IOMMU capability
+ * pointers.
+ */
+ u16 cap_ptr;
+
/* pci domain of this IOMMU */
u16 pci_seg;
@@ -284,19 +284,19 @@
/* size of command buffer */
u32 cmd_buf_size;
- /* event buffer virtual address */
- u8 *evt_buf;
/* size of event buffer */
u32 evt_buf_size;
+ /* event buffer virtual address */
+ u8 *evt_buf;
/* MSI number for event interrupt */
u16 evt_msi_num;
- /* if one, we need to send a completion wait command */
- int need_sync;
-
/* true if interrupts for this IOMMU are already enabled */
bool int_enabled;
+ /* if one, we need to send a completion wait command */
+ int need_sync;
+
/* default dma_ops domain for that IOMMU */
struct dma_ops_domain *default_dom;
};
diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h
index 7f225a4..097794f 100644
--- a/arch/x86/include/asm/dma-mapping.h
+++ b/arch/x86/include/asm/dma-mapping.h
@@ -71,15 +71,13 @@
/* Make sure we keep the same behaviour */
static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
{
-#ifdef CONFIG_X86_32
- return 0;
-#else
+#ifdef CONFIG_X86_64
struct dma_mapping_ops *ops = get_dma_ops(dev);
if (ops->mapping_error)
return ops->mapping_error(dev, dma_addr);
- return (dma_addr == bad_dma_address);
#endif
+ return (dma_addr == bad_dma_address);
}
#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index e4899e0..5662e22 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -187,6 +187,8 @@
spin_lock_irqsave(&iommu->lock, flags);
ret = __iommu_queue_command(iommu, cmd);
+ if (!ret)
+ iommu->need_sync = 1;
spin_unlock_irqrestore(&iommu->lock, flags);
return ret;
@@ -210,10 +212,13 @@
cmd.data[0] = CMD_COMPL_WAIT_INT_MASK;
CMD_SET_TYPE(&cmd, CMD_COMPL_WAIT);
- iommu->need_sync = 0;
-
spin_lock_irqsave(&iommu->lock, flags);
+ if (!iommu->need_sync)
+ goto out;
+
+ iommu->need_sync = 0;
+
ret = __iommu_queue_command(iommu, &cmd);
if (ret)
@@ -254,8 +259,6 @@
ret = iommu_queue_command(iommu, &cmd);
- iommu->need_sync = 1;
-
return ret;
}
@@ -281,8 +284,6 @@
ret = iommu_queue_command(iommu, &cmd);
- iommu->need_sync = 1;
-
return ret;
}
@@ -762,8 +763,6 @@
write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
iommu_queue_inv_dev_entry(iommu, devid);
-
- iommu->need_sync = 1;
}
/*****************************************************************************
@@ -858,6 +857,9 @@
print_devid(_bdf, 1);
}
+ if (domain_for_device(_bdf) == NULL)
+ set_device_domain(*iommu, *domain, _bdf);
+
return 1;
}
@@ -1031,8 +1033,7 @@
if (addr == bad_dma_address)
goto out;
- if (unlikely(iommu->need_sync))
- iommu_completion_wait(iommu);
+ iommu_completion_wait(iommu);
out:
spin_unlock_irqrestore(&domain->lock, flags);
@@ -1060,8 +1061,7 @@
__unmap_single(iommu, domain->priv, dma_addr, size, dir);
- if (unlikely(iommu->need_sync))
- iommu_completion_wait(iommu);
+ iommu_completion_wait(iommu);
spin_unlock_irqrestore(&domain->lock, flags);
}
@@ -1127,8 +1127,7 @@
goto unmap;
}
- if (unlikely(iommu->need_sync))
- iommu_completion_wait(iommu);
+ iommu_completion_wait(iommu);
out:
spin_unlock_irqrestore(&domain->lock, flags);
@@ -1173,8 +1172,7 @@
s->dma_address = s->dma_length = 0;
}
- if (unlikely(iommu->need_sync))
- iommu_completion_wait(iommu);
+ iommu_completion_wait(iommu);
spin_unlock_irqrestore(&domain->lock, flags);
}
@@ -1225,8 +1223,7 @@
goto out;
}
- if (unlikely(iommu->need_sync))
- iommu_completion_wait(iommu);
+ iommu_completion_wait(iommu);
out:
spin_unlock_irqrestore(&domain->lock, flags);
@@ -1257,8 +1254,7 @@
__unmap_single(iommu, domain->priv, dma_addr, size, DMA_BIDIRECTIONAL);
- if (unlikely(iommu->need_sync))
- iommu_completion_wait(iommu);
+ iommu_completion_wait(iommu);
spin_unlock_irqrestore(&domain->lock, flags);
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index a42b02b..ba7ad83 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -123,6 +123,8 @@
spin_lock_irqsave(&iommu_bitmap_lock, flags);
iommu_area_free(iommu_gart_bitmap, offset, size);
+ if (offset >= next_bit)
+ next_bit = offset + size;
spin_unlock_irqrestore(&iommu_bitmap_lock, flags);
}