On 9/22/2020 1:08 PM, Logan Gunthorpe wrote:
dma_alloc_coherent() is called with a fixed SZ_2M size, but frees happen
with IOAT_CHUNK_SIZE. Recently, IOAT_CHUNK_SIZE was reduced to 512M but
the allocation did not change. To fix, change to using the
IOAT_CHUNK_SIZE define.
This was caught with the upcoming patchset for converting Intel platforms to the
dma-iommu implementation. It has a warning when the unmapped size differs from
the mapped size.
Fixes: a02254f8a676 ("dmaengine: ioat: Decreasing allocation chunk size 2M->512K")
Link: https://lore.kernel.org/intel-gfx/776771a2-247a-d1be-d882-bee02d919ae0@xxxxxxxxxxxx/
Suggested-by: Robin Murphy <robin.murphy@xxxxxxx>
Signed-off-by: Logan Gunthorpe <logang@xxxxxxxxxxxx>
Cc: Vinod Koul <vkoul@xxxxxxxxxx>
Cc: Dave Jiang <dave.jiang@xxxxxxxxx>
Cc: Dan Williams <dan.j.williams@xxxxxxxxx>
Acked-by: Dave Jiang <dave.jiang@xxxxxxxxx>
Thanks Logan.
---
drivers/dma/ioat/dma.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index a814b200299b..07296171e2bb 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -389,7 +389,7 @@ ioat_alloc_ring(struct dma_chan *c, int order, gfp_t flags)
struct ioat_descs *descs = &ioat_chan->descs[i];
descs->virt = dma_alloc_coherent(to_dev(ioat_chan),
- SZ_2M, &descs->hw, flags);
+ IOAT_CHUNK_SIZE, &descs->hw, flags);
if (!descs->virt) {
int idx;
base-commit: ba4f184e126b751d1bffad5897f263108befc780