2007-10-17 08:49:56

by FUJITA Tomonori

Subject: [PATCH] IA64: iommu uses sg_next with an invalid sg element

sg list elements might not be continuous.

Signed-off-by: FUJITA Tomonori <[email protected]>
---
arch/ia64/hp/common/sba_iommu.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index 4338f41..3c95f41 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -1179,7 +1179,6 @@ sba_fill_pdir(
u64 *pdirp = NULL;
unsigned long dma_offset = 0;

- dma_sg--;
while (nents-- > 0) {
int cnt = startsg->dma_length;
startsg->dma_length = 0;
@@ -1201,7 +1200,8 @@ sba_fill_pdir(
u32 pide = startsg->dma_address & ~PIDE_FLAG;
dma_offset = (unsigned long) pide & ~iovp_mask;
startsg->dma_address = 0;
- dma_sg = sg_next(dma_sg);
+ if (n_mappings)
+ dma_sg = sg_next(dma_sg);
dma_sg->dma_address = pide | ioc->ibase;
pdirp = &(ioc->pdir_base[pide >> iovp_shift]);
n_mappings++;
--
1.5.2.4

2007-10-17 08:52:24

by Jens Axboe

Subject: Re: [PATCH] IA64: iommu uses sg_next with an invalid sg element

On Wed, Oct 17 2007, FUJITA Tomonori wrote:
> sg list elements might not be continuous.

Thanks, applied.

--
Jens Axboe