Skip to contentSkip to conversation

Use the DMA32 zone for dma_alloc_coherent()/pci_alloc_consistent on x86-64

PATCH In review Andi Kleen 2 participants 3 messages Sep 11, 2005 – Sep 12, 2005

lore
Share & tools
lore

Andi Kleen

author
#0 lore raw

PATCH 3/3 Use the DMA32 zone for dma_alloc_coherent()/pci_alloc_consistent on x86-64

Use the DMA32 zone for dma_alloc_coherent()/pci_alloc_consistent on x86-64

arch/x86_64/kernel/pci-gart.c +3 -1
Index: linux/arch/x86_64/kernel/pci-gart.c
===================================================================
@@ -219,6 +219,8 @@ dma_alloc_coherent(struct device *dev, s
 	/* Kludge to make it bug-to-bug compatible with i386. i386
 	   uses the normal dma_mask for alloc_coherent. */
 	dma_mask &= *dev->dma_mask;
+	if (dma_mask <= 0xffffffff)
+		gfp |= GFP_DMA32;
 
  again:
 	memory = dma_alloc_pages(dev, gfp, get_order(size));
@@ -245,7 +247,7 @@ dma_alloc_coherent(struct device *dev, s
 				}
 
 				if (!(gfp & GFP_DMA)) { 
-					gfp |= GFP_DMA; 
+					gfp = (gfp & ~GFP_DMA32) | GFP_DMA;
 					goto again;
 				}
 				return NULL;
#1 lore raw
Quoted 10 lines · Andi Kleen

Use the DMA32 zone for dma_alloc_coherent()/pci_alloc_consistent on x86-64

Index: linux/arch/x86_64/kernel/pci-gart.c
===================================================================

arch/x86_64/kernel/pci-gart.c dma_alloc_coherent(struct device *dev, s · line 223

Quoted 5 lines
arch/x86_64/kernel/pci-gart.c +2 -0
@@ -219,6 +219,8 @@ dma_alloc_coherent(struct device *dev, s
 	/* Kludge to make it bug-to-bug compatible with i386. i386
 	   uses the normal dma_mask for alloc_coherent. */
 	dma_mask &= *dev->dma_mask;
+	if (dma_mask <= 0xffffffff)
+		gfp |= GFP_DMA32;

How can this be right?

Let's say that dma_mask is 0xfffff (ie 20-bit legacy DMA). It will trigger the test, and set the GFP_DMA32 bit. Which can't be right.

I'm going to drop the DMA32 parts of the patches, they seem to be pretty raw.

Linus

author
#2 lore raw
Quoted 6 lines · Linus Torvalds
Quoted 5 lines · Andi Kleen

Use the DMA32 zone for dma_alloc_coherent()/pci_alloc_consistent on x86-64

Quoted 12 lines

arch/x86_64/kernel/pci-gart.c dma_alloc_coherent(struct device *dev, s · line 223

Quoted 10 lines

Index: linux/arch/x86_64/kernel/pci-gart.c
===================================================================

arch/x86_64/kernel/pci-gart.c +2 -0
@@ -219,6 +219,8 @@ dma_alloc_coherent(struct device *dev, s
 	/* Kludge to make it bug-to-bug compatible with i386. i386
 	   uses the normal dma_mask for alloc_coherent. */
 	dma_mask &= *dev->dma_mask;
+	if (dma_mask <= 0xffffffff)
+		gfp |= GFP_DMA32;

How can this be right?

No that's intentional. The rationale is that usually the mask is bigger than 16MB, so you first try it in the higher zone where you already have some chance to get the right memory, and then only
if you don't get it fall back to the "precious" zone.

The cost is very low because the allocators are fast enough.

Given for a real 0xfffff allocation it is not very likely it will work, but most <4GB allocations are bigger than that.

I had the same algorithm previously with GFP_DMA.

Quoted 5 lines

Let's say that dma_mask is 0xfffff (ie 20-bit legacy DMA). It will trigger the test, and set the GFP_DMA32 bit. Which can't be right.

I'm going to drop the DMA32 parts of the patches, they seem to be pretty raw.

Please don't.

-Andi