Re: [PATCH v3 2/5] dma-iommu: fix arch_sync_dma for map

From: Robin Murphy
Date: Thu Aug 12 2021 - 06:38:18 EST

On 2021-08-12 10:21, David Stevens wrote:
On Thu, Aug 12, 2021 at 3:47 AM Robin Murphy <robin.murphy@xxxxxxx> wrote:

On 2021-08-11 03:42, David Stevens wrote:
From: David Stevens <stevensd@xxxxxxxxxxxx>

When calling arch_sync_dma, we need to pass it the memory that's
actually being used for dma. When using swiotlb bounce buffers, this is
the bounce buffer. Move arch_sync_dma into the __iommu_dma_map_swiotlb
helper, so it can use the bounce buffer address if necessary. This also
means it is no longer necessary to call iommu_dma_sync_sg_for_device in
iommu_dma_map_sg for untrusted devices.

Fixes: 82612d66d51d ("iommu: Allow the dma-iommu api to use bounce buffers")
Signed-off-by: David Stevens <stevensd@xxxxxxxxxxxx>
drivers/iommu/dma-iommu.c | 16 +++++++---------
1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
index 54e103b989d9..4f0cc4a0a61f 100644
--- a/drivers/iommu/dma-iommu.c
+++ b/drivers/iommu/dma-iommu.c
@@ -576,6 +576,9 @@ static dma_addr_t __iommu_dma_map_swiotlb(struct device *dev, phys_addr_t phys,
memset(padding_start, 0, padding_size);

+ if (!coherent && !(attrs & DMA_ATTR_SKIP_CPU_SYNC))

Make that an "else if" - otherwise you're just reintroducing the same
thing that the third hunk is trying to clean up.

swiotlb_tbl_map_single only copies into the swiotlb buffer, it doesn't
do any architectural syncing. So this block is necessary in the
swiotlb case as well, for iommu_dma_map_page to work properly.

Urgh, apologies again - I had a mental short-circuit and was convinced that the SWIOTLB call already did its own cache maintenance. I really should have learned by now not to review fiddly patches like these while tired...
The third chunk is a separate optimization, so I'll split it out into
its own patch.

It's still a logical part of this change - cleaning up the potentially wrong sync while you make sure the right one happens. Even more so if you make the fix the better way by having iommu_dma_map_sg_swiotlb() call iommu_dma_map_page().