2006-08-01 12:59:37

by Muli Ben-Yehuda

[permalink] [raw]
Subject: [PATCH x86-64]: remove superflous BUG_ON's in nommu and gart

There's no need to check for invalid DMA data direction in nommu and
gart since we do it in dma-mapping.h anyway before calling the
individual dma-ops.

Signed-off-by: Muli Ben-Yehuda <[email protected]>

diff -r b90678f884c4 arch/x86_64/kernel/pci-gart.c
--- a/arch/x86_64/kernel/pci-gart.c Mon Jul 31 21:01:32 2006 +0000
+++ b/arch/x86_64/kernel/pci-gart.c Tue Aug 01 15:52:52 2006 +0300
@@ -239,8 +239,6 @@ dma_addr_t gart_map_single(struct device
{
unsigned long phys_mem, bus;

- BUG_ON(dir == DMA_NONE);
-
if (!dev)
dev = &fallback_dev;

@@ -383,7 +381,6 @@ int gart_map_sg(struct device *dev, stru
unsigned long pages = 0;
int need = 0, nextneed;

- BUG_ON(dir == DMA_NONE);
if (nents == 0)
return 0;

diff -r b90678f884c4 arch/x86_64/kernel/pci-nommu.c
--- a/arch/x86_64/kernel/pci-nommu.c Mon Jul 31 21:01:32 2006 +0000
+++ b/arch/x86_64/kernel/pci-nommu.c Tue Aug 01 15:52:58 2006 +0300
@@ -59,7 +59,6 @@ int nommu_map_sg(struct device *hwdev, s
{
int i;

- BUG_ON(direction == DMA_NONE);
for (i = 0; i < nents; i++ ) {
struct scatterlist *s = &sg[i];
BUG_ON(!s->page);



2006-08-01 17:26:38

by Andi Kleen

[permalink] [raw]
Subject: Re: [PATCH x86-64]: remove superflous BUG_ON's in nommu and gart

On Tuesday 01 August 2006 14:59, Muli Ben-Yehuda wrote:
> There's no need to check for invalid DMA data direction in nommu and
> gart since we do it in dma-mapping.h anyway before calling the
> individual dma-ops.

Applied thanks

-Andi