From: Timur Kristóf Date: Wed, 13 May 2026 20:04:11 +0000 (+0200) Subject: drm/amdgpu/vce1: Check if VRAM address is lower than GART. X-Git-Tag: v7.1-rc5~19^2~3^2~17 X-Git-Url: http://git.ipfire.org/gitweb/index.cgi?a=commitdiff_plain;h=12b60cf345e84aa7546dd225b3ce3380c9ab97f5;p=thirdparty%2Fkernel%2Flinux.git drm/amdgpu/vce1: Check if VRAM address is lower than GART. Previously, I had assumed this was not possible so it was OK to not handle it, but now we got a report from a user who has a board that is configured this way. When the VCPU BO is already located in a low 32-bit address in VRAM (eg. when VRAM is mapped to the low address space), don't do the workaround. Fixes: 71aec08f80e7 ("amdgpu/vce: use amdgpu_gtt_mgr_alloc_entries") Signed-off-by: Timur Kristóf Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit f370ec9b164698a9ca1a7b59bfbea07f70df769d) --- diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v1_0.c index 884f24be36859..a49f11be74b20 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v1_0.c @@ -542,6 +542,9 @@ static int vce_v1_0_ensure_vcpu_bo_32bit_addr(struct amdgpu_device *adev) u64 vce_gart_start_offs; int r; + if (adev->gmc.vram_start < adev->gmc.gart_start) + return amdgpu_bo_gpu_offset(adev->vce.vcpu_bo) <= max_vcpu_bo_addr ? 0 : -EINVAL; + r = amdgpu_gtt_mgr_alloc_entries(&adev->mman.gtt_mgr, &adev->vce.gart_node, num_pages, DRM_MM_INSERT_LOW);