Am 25.04.22 um 13:14 schrieb Haohui Mai:
Sorry for the confusion. I misread the code, but it still seems to me
it is a valid issue. What the patch tries to do is to fix the
following pattern:
- WREG32(sdma_v5_2_get_reg_offset(adev, i,
mmSDMA0_GFX_RB_WPTR), lower_32_bits(rin
Sorry for the confusion. I misread the code, but it still seems to me
it is a valid issue. What the patch tries to do is to fix the
following pattern:
- WREG32(sdma_v5_2_get_reg_offset(adev, i,
mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr) << 2);
- WRE
Am 25.04.22 um 11:15 schrieb Haohui Mai:
Computing the address of the doorbell should be done before instead of after
separating the 64-bit address into the higher and lower half. The
current code sets the MMIO registers incorrectly if the address of the
doorbell is above 1G.
That doesn't make
Computing the address of the doorbell should be done before instead of after
separating the 64-bit address into the higher and lower half. The
current code sets the MMIO registers incorrectly if the address of the
doorbell is above 1G.
Signed-off-by: Haohui Mai
---
drivers/gpu/drm/amd/amdgpu/cik
