[PATCH] drm/amd/amdgpu: reserve vm invalidation engine for firmware

[Christian König]

Am 17.11.22 um 10:11 schrieb Jack Xiao:
> If mes enabled, reserve VM invalidation engine 5 for firmware.
>
> Signed-off-by: Jack Xiao <Jack.Xiao at amd.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 6 ++++++
>   1 file changed, 6 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> index 4365ede42855..e970e3760cec 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> @@ -479,6 +479,12 @@ int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev)
>   	unsigned i;
>   	unsigned vmhub, inv_eng;
>   
> +	if (adev->enable_mes) {
> +		/* reserve engine 5 for firmware */
> +		for (vmhub = 0; vmhub < AMDGPU_MAX_VMHUBS; vmhub++)
> +			vm_inv_engs[vmhub] &= ~(1 << 5);
> +	}
> +

Is that fixed and nailed down with the fw team? If yes the patch is rb 
by me.

Regards,
Christian.

>   	for (i = 0; i < adev->num_rings; ++i) {
>   		ring = adev->rings[i];
>   		vmhub = ring->funcs->vmhub;