The Documentation/DMA-API-HOWTO.txt states that dma_map_sg returns the numer of the created entries in the DMA address space. However the subsequent calls to dma_sync_sg_for_{device,cpu} and dma_unmap_sg must be called with the original number of entries passed to dma_map_sg. The sg_table->nents in turn holds the result of the dma_map_sg call as stated in include/linux/scatterlist.h. Adapt the code to obey those rules.
Signed-off-by: Marek Szyprowski m.szyprowski@samsung.com Reviewed-by: Christian König christian.koenig@amd.com --- For more information, see '[PATCH v2 00/21] DRM: fix struct sg_table nents vs. orig_nents misuse' thread: https://lkml.org/lkml/2020/5/4/373 --- drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 7 ++++--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 8 ++++---- 2 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index 43d8ed7..4df813e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -307,8 +307,9 @@ static struct sg_table *amdgpu_dma_buf_map(struct dma_buf_attachment *attach, if (IS_ERR(sgt)) return sgt;
- if (!dma_map_sg_attrs(attach->dev, sgt->sgl, sgt->nents, dir, - DMA_ATTR_SKIP_CPU_SYNC)) + sgt->nents = dma_map_sg_attrs(attach->dev, sgt->sgl, sgt->orig_nents, + dir, DMA_ATTR_SKIP_CPU_SYNC); + if (!sgt->nents) goto error_free; break;
@@ -349,7 +350,7 @@ static void amdgpu_dma_buf_unmap(struct dma_buf_attachment *attach, struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
if (sgt->sgl->page_link) { - dma_unmap_sg(attach->dev, sgt->sgl, sgt->nents, dir); + dma_unmap_sg(attach->dev, sgt->sgl, sgt->orig_nents, dir); sg_free_table(sgt); kfree(sgt); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index eff1f73..1f8c507 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -1043,7 +1043,6 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm) { struct amdgpu_device *adev = amdgpu_ttm_adev(ttm->bdev); struct amdgpu_ttm_tt *gtt = (void *)ttm; - unsigned nents; int r;
int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY); @@ -1059,8 +1058,9 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
/* Map SG to device */ r = -ENOMEM; - nents = dma_map_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction); - if (nents == 0) + ttm->sg->nents = dma_map_sg(adev->dev, ttm->sg->sgl, + ttm->sg->orig_nents, direction); + if (ttm->sg->nents == 0) goto release_sg;
/* convert SG to linear array of pages and dma addresses */ @@ -1091,7 +1091,7 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_tt *ttm) return;
/* unmap the pages mapped to the device */ - dma_unmap_sg(adev->dev, ttm->sg->sgl, ttm->sg->nents, direction); + dma_unmap_sg(adev->dev, ttm->sg->sgl, ttm->sg->orig_nents, direction);
sg_free_table(ttm->sg);