[PATCH 1/2] drm/amdgpu: Move xgmi ras initialization from .late_init to early_init

[Zhou1, Tao]

[AMD Official Use Only]



> -----Original Message-----
> From: Chai, Thomas <YiPeng.Chai at amd.com>
> Sent: Wednesday, January 19, 2022 5:25 PM
> To: amd-gfx at lists.freedesktop.org
> Cc: Chai, Thomas <YiPeng.Chai at amd.com>; Zhang, Hawking
> <Hawking.Zhang at amd.com>; Zhou1, Tao <Tao.Zhou1 at amd.com>; Clements,
> John <John.Clements at amd.com>; Chai, Thomas <YiPeng.Chai at amd.com>
> Subject: [PATCH 1/2] drm/amdgpu: Move xgmi ras initialization from .late_init
> to early_init
> 
> Move xgmi ras initialization from .late_init to early_init, which let xgmi ras can
> be initialized only once.
> 
> Signed-off-by: yipechai <YiPeng.Chai at amd.com>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_device.c |  2 ++
>  drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c    |  5 -----
>  drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c    | 10 ++++++++++
>  drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h    |  3 +++
>  4 files changed, 15 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 22f50aadf694..ece6397f81de 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -2189,6 +2189,8 @@ static int amdgpu_device_ip_early_init(struct
> amdgpu_device *adev)
>  		}
>  	}
> 
> +	amdgpu_ras_early_init(adev);
> +
>  	adev->cg_flags &= amdgpu_cg_mask;
>  	adev->pg_flags &= amdgpu_pg_mask;
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> index 3483a82f5734..d83eee1984c8 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> @@ -452,11 +452,6 @@ int amdgpu_gmc_ras_late_init(struct amdgpu_device
> *adev)
>  			return r;
>  	}
> 
> -	if (!adev->gmc.xgmi.connected_to_cpu) {
> -		adev->gmc.xgmi.ras = &xgmi_ras;
> -		amdgpu_ras_register_ras_block(adev, &adev->gmc.xgmi.ras-
> >ras_block);
> -	}
> -
>  	if (adev->gmc.xgmi.ras && adev->gmc.xgmi.ras->ras_block.ras_late_init)
> {
>  		r = adev->gmc.xgmi.ras->ras_block.ras_late_init(adev, NULL);
>  		if (r)
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
> index 7a1d2bac698e..2b71611be388 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
> @@ -2238,6 +2238,16 @@ static void amdgpu_ras_counte_dw(struct
> work_struct *work)
>  	pm_runtime_put_autosuspend(dev->dev);
>  }
> 
> +int amdgpu_ras_early_init(struct amdgpu_device *adev) {
> +	if (!adev->gmc.xgmi.connected_to_cpu) {
> +		adev->gmc.xgmi.ras = &xgmi_ras;
> +		amdgpu_ras_register_ras_block(adev, &adev->gmc.xgmi.ras-
> >ras_block);
> +	}
> +
> +	return 0;
> +}
[Tao]: I suggest creating amdgpu_gmc_ras_late_init in amdgpu_gmc.c and call it in each gmc_vx.c.
Please make sure gpu reset and S3 tests pass.

> +
>  int amdgpu_ras_init(struct amdgpu_device *adev)  {
>  	struct amdgpu_ras *con = amdgpu_ras_get_context(adev); diff --git
> a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
> index a51a281bd91a..7d99e3736ab9 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
> @@ -600,6 +600,9 @@ amdgpu_ras_error_to_ta(enum
> amdgpu_ras_error_type error) {
>  	}
>  }
> 
> +/* called in ip_early_init */
> +int amdgpu_ras_early_init(struct amdgpu_device *adev);
> +
>  /* called in ip_init and ip_fini */
>  int amdgpu_ras_init(struct amdgpu_device *adev);  int amdgpu_ras_fini(struct
> amdgpu_device *adev);
> --
> 2.25.1