Am 05.01.2018 um 05:30 schrieb Alex Deucher:
> Use the callback rather than open coding it.
>
> Signed-off-by: Alex Deucher <alexander.deucher at amd.com>
I would rather prefer if the caller of flush_gpu_tlb calls
amdgpu_asic_flush_hdp().
IIRC we have exactly one which calls this and the other one actually
wants to call amdgpu_asic_flush_hdp() only.
Regards,
Christian.
> ---
> drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
> index 8e28270d1ea9..e47a1ba57aae 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
> @@ -359,7 +359,7 @@ static int gmc_v6_0_mc_init(struct amdgpu_device *adev)
> static void gmc_v6_0_gart_flush_gpu_tlb(struct amdgpu_device *adev,
> uint32_t vmid)
> {
> - WREG32(mmHDP_MEM_COHERENCY_FLUSH_CNTL, 0);
> + amdgpu_asic_flush_hdp(adev);
>
> WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid);
> }
