drm/amdgpu: disable concurrent flushes for Navi10 v2

Navi10 have a bug in the SDMA which can theoretically cause memory corruption with concurrent VMID flushes v2: explicitely check Navi10

Christian König committed 7 years ago

Commit a2bd77b1 parent 437298b1 file · +5 −2Browse files

Unified · Split

@@ -364,8 +364,11 @@ static int amdgpu_vmid_grab_used(struct amdgpu_vm *vm, · 363 unmodified lines
364	364	if (updates && (!flushed \|\| dma_fence_is_later(updates, flushed)))
365	365	needs_flush = true;
366	366
367		− /* Concurrent flushes are only possible starting with Vega10 */
368		− if (adev->asic_type < CHIP_VEGA10 && needs_flush)
	367	+ /* Concurrent flushes are only possible starting with Vega10 and
	368	+ * are broken on Navi10 and Navi14.
	369	+ */
	370	+ if (needs_flush && (adev->asic_type < CHIP_VEGA10 \|\|
	371	+ adev->asic_type == CHIP_NAVI10))
369	372	continue;
370	373
371	374	/* Good, we can use this VMID. Remember this submission as