On 07/04/2025 15:32, Christian König wrote:
Am 07.04.25 um 16:18 schrieb Matthew Auld:
The page_link lower bits of the first sg could contain something like SG_END, if we are mapping a single VRAM page or contiguous blob which fits into one sg entry. Rather pull out the struct page, and use that in our check to know if we mapped struct pages vs VRAM.
Fixes: f44ffd677fb3 ("drm/amdgpu: add support for exporting VRAM using DMA-buf v3") Signed-off-by: Matthew Auld matthew.auld@intel.com Cc: Christian König christian.koenig@amd.com Cc: amd-gfx@lists.freedesktop.org Cc: stable@vger.kernel.org # v5.8+
Good point, haven't thought about that at all since we only abuse the sg table as DMA addr container.
Reviewed-by: Christian König christian.koenig@amd.com
Were is patch #1 from this series?
That one is xe specific: https://lore.kernel.org/intel-xe/20250407141823.44504-3-matthew.auld@intel.c...
I copied your approach with using page_link here, but with added sg_page().
Thanks, Christian.
drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index 9f627caedc3f..c9842a0e2a1c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -184,7 +184,7 @@ static void amdgpu_dma_buf_unmap(struct dma_buf_attachment *attach, struct sg_table *sgt, enum dma_data_direction dir) {
- if (sgt->sgl->page_link) {
- if (sg_page(sgt->sgl)) { dma_unmap_sgtable(attach->dev, sgt, dir, 0); sg_free_table(sgt); kfree(sgt);