RE: [PATCH v2] drm/amdgpu: Enable second VCN for certain Navy Flounder.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



[Public]

Reviewed-by: Guchun Chen <guchun.chen@xxxxxxx>

Next I will figure out which navy flounder boards have the bad harvest table info from VBIOS, and add a quirk to enable this workaround.

Regards,
Guchun

-----Original Message-----
From: amd-gfx <amd-gfx-bounces@xxxxxxxxxxxxxxxxxxxxx> On Behalf Of Deucher, Alexander
Sent: Friday, January 7, 2022 1:21 AM
To: Zhou, Peng Ju <PengJu.Zhou@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx
Subject: RE: [PATCH v2] drm/amdgpu: Enable second VCN for certain Navy Flounder.

[Public]

> -----Original Message-----
> From: amd-gfx <amd-gfx-bounces@xxxxxxxxxxxxxxxxxxxxx> On Behalf Of 
> Peng Ju Zhou
> Sent: Thursday, January 6, 2022 2:30 AM
> To: amd-gfx@xxxxxxxxxxxxxxxxxxxxx
> Subject: [PATCH v2] drm/amdgpu: Enable second VCN for certain Navy 
> Flounder.
> 
> Certain Navy Flounder cards have 2 VCNs, enable it.
> 
> Signed-off-by: Peng Ju Zhou <PengJu.Zhou@xxxxxxx>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
> index 580a5b387122..57e001d73ec9 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
> @@ -550,7 +550,8 @@ void amdgpu_discovery_harvest_ip(struct
> amdgpu_device *adev)
>  	}
>  	/* some IP discovery tables on Navy Flounder don't have this set 
> correctly */
>  	if ((adev->ip_versions[UVD_HWIP][1] == IP_VERSION(3, 0, 1)) &&
> -	    (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 2)))
> +	    (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 3, 2)) &&
> +	    (adev->pdev->revision != 0xFF))

We added this check because some navy flounder boards did not correctly set the harvesting in the IP discovery table.  It would be nice to sort that out so we only enable this workaround on the boards with the bad harvest table.  I suppose that would probably come down to the PCI revision anyway, so this is probably fine.

Acked-by: Alex Deucher <alexander.deucher@xxxxxxx>


>  		adev->vcn.harvest_config |=
> AMDGPU_VCN_HARVEST_VCN1;
>  	if (vcn_harvest_count == adev->vcn.num_vcn_inst) {
>  		adev->harvest_ip_mask |= AMD_HARVEST_IP_VCN_MASK;
> --
> 2.33.1




[Index of Archives]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux