From: Jake Moilanen Date: Tue, 31 Jan 2006 03:51:54 +0000 (-0600) Subject: [PATCH] powerpc: IOMMU SG paranoia X-Git-Tag: v2.6.16-rc3~67^2~16 X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=a958a2648602aaa8f98aeb3c1c914f1b8096bfdc;p=linux-2.6 [PATCH] powerpc: IOMMU SG paranoia This addresses two items, which are unlikely to be hit if we trust drivers. The first is moving a memory barrier below where the vmerged SG count is passed back, but before the list is set to end. If those instructions were reordered, there could be an issue in iommu_unmap_sg(). The second is making sure we terminate the list on the failure case of iommu_map_sg(). If a driver does not look at the failure return code, it could pass a ill-formed SG list to iommu_unmap_sg(). Signed-off-by: Jake Moilanen Acked-by: Olof Johansson Signed-off-by: Paul Mackerras --- diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c index 4d9b438891..946f3219fd 100644 --- a/arch/powerpc/kernel/iommu.c +++ b/arch/powerpc/kernel/iommu.c @@ -334,9 +334,6 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, spin_unlock_irqrestore(&(tbl->it_lock), flags); - /* Make sure updates are seen by hardware */ - mb(); - DBG("mapped %d elements:\n", outcount); /* For the sake of iommu_unmap_sg, we clear out the length in the @@ -347,6 +344,10 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, outs->dma_address = DMA_ERROR_CODE; outs->dma_length = 0; } + + /* Make sure updates are seen by hardware */ + mb(); + return outcount; failure: @@ -358,6 +359,8 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl, npages = (PAGE_ALIGN(s->dma_address + s->dma_length) - vaddr) >> PAGE_SHIFT; __iommu_free(tbl, vaddr, npages); + s->dma_address = DMA_ERROR_CODE; + s->dma_length = 0; } } spin_unlock_irqrestore(&(tbl->it_lock), flags);