[tip:x86/dma] x86/dma/amd_gart: Look at dev->coherent_dma_mask instead of GFP_DMA
From: tip-bot for Christoph Hellwig
Date: Fri Mar 23 2018 - 15:50:45 EST
Commit-ID: f3c39d51043dcee5dd4d51dffcfe9ce01a263582
Gitweb: https://git.kernel.org/tip/f3c39d51043dcee5dd4d51dffcfe9ce01a263582
Author: Christoph Hellwig <hch@xxxxxx>
AuthorDate: Mon, 19 Mar 2018 11:38:17 +0100
Committer: Ingo Molnar <mingo@xxxxxxxxxx>
CommitDate: Tue, 20 Mar 2018 10:01:57 +0100
x86/dma/amd_gart: Look at dev->coherent_dma_mask instead of GFP_DMA
We want to phase out looking at the magic GFP_DMA flag in the DMA mapping
routines, so switch the gart driver to use the dev->coherent_dma_mask
instead, which is used to select the GFP_DMA flag in the caller.
Tested-by: Tom Lendacky <thomas.lendacky@xxxxxxx>
Signed-off-by: Christoph Hellwig <hch@xxxxxx>
Reviewed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
Cc: David Woodhouse <dwmw2@xxxxxxxxxxxxx>
Cc: Joerg Roedel <joro@xxxxxxxxxx>
Cc: Jon Mason <jdmason@xxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Muli Ben-Yehuda <mulix@xxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx
Link: http://lkml.kernel.org/r/20180319103826.12853-6-hch@xxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
---
arch/x86/kernel/amd_gart_64.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c
index 52e3abcf3e70..79ac6caaaabb 100644
--- a/arch/x86/kernel/amd_gart_64.c
+++ b/arch/x86/kernel/amd_gart_64.c
@@ -484,7 +484,7 @@ gart_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr,
unsigned long align_mask;
struct page *page;
- if (force_iommu && !(flag & GFP_DMA)) {
+ if (force_iommu && dev->coherent_dma_mask > DMA_BIT_MASK(24)) {
flag &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
page = alloc_pages(flag | __GFP_ZERO, get_order(size));
if (!page)