[patch 05/24] x86_64: Fix a race in the free_iommu path.

[Greg KH]

-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Mike Waychison <mikew@xxxxxxxxxx>

We do this by removing a micro-optimization that tries to avoid grabbing
the iommu_bitmap_lock spinlock and using a bus-locked operation.

This still races with other simultaneous alloc_iommu or free_iommu(size
> 1) which both use bus-unlocked operations. 

The end result of this race is eventually ending
up with an iommu_gart_bitmap that has bits errornously set all over,
making large contiguous iommu space allocations fail with 'PCI-DMA:
Out of IOMMU space'.

Signed-off-by: Mike Waychison <mikew@xxxxxxxxxx>
Signed-off-by: Andi Kleen <ak@xxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxx>

---
 arch/x86_64/kernel/pci-gart.c |    4 ----
 1 file changed, 4 deletions(-)

--- linux-2.6.16.11.orig/arch/x86_64/kernel/pci-gart.c
+++ linux-2.6.16.11/arch/x86_64/kernel/pci-gart.c
@@ -114,10 +114,6 @@ static unsigned long alloc_iommu(int siz
 static void free_iommu(unsigned long offset, int size)
 { 
 	unsigned long flags;
-	if (size == 1) { 
-		clear_bit(offset, iommu_gart_bitmap); 
-		return;
-	}
 	spin_lock_irqsave(&iommu_bitmap_lock, flags);
 	__clear_bit_string(iommu_gart_bitmap, offset, size);
 	spin_unlock_irqrestore(&iommu_bitmap_lock, flags);

--
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/