diff options
| author | Joerg Roedel <jroedel@suse.de> | 2015-06-01 17:30:57 +0200 |
|---|---|---|
| committer | Joerg Roedel <jroedel@suse.de> | 2015-06-02 08:46:32 +0200 |
| commit | 2d0ec7a19b5796aa958636eeeff072b506460501 (patch) | |
| tree | f2b9357156dd8332baadb9f10db5e455af7787c7 /drivers/iommu/amd_iommu.c | |
| parent | Linux 4.1-rc6 (diff) | |
| download | linux-2d0ec7a19b5796aa958636eeeff072b506460501.tar.xz linux-2d0ec7a19b5796aa958636eeeff072b506460501.zip | |
Revert "iommu/amd: Don't allocate with __GFP_ZERO in alloc_coherent"
This reverts commit 5fc872c7323534e8f7dc21bab635e7a9b9659e07.
The DMA-API does not strictly require that the memory
returned by dma_alloc_coherent is zeroed out. For that
another function (dma_zalloc_coherent) should be used. But
all other x86 DMA-API implementation I checked zero out the
memory, so that some drivers rely on it and break when it is
not.
It seems the (driver-)world is not yet ready for this
change, so revert it.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Diffstat (limited to 'drivers/iommu/amd_iommu.c')
| -rw-r--r-- | drivers/iommu/amd_iommu.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index e43d48956dea..e1c7e9e51045 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -2930,6 +2930,7 @@ static void *alloc_coherent(struct device *dev, size_t size, size = PAGE_ALIGN(size); dma_mask = dev->coherent_dma_mask; flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); + flag |= __GFP_ZERO; page = alloc_pages(flag | __GFP_NOWARN, get_order(size)); if (!page) { |