[PATCH 5/7] drm/amdgpu: add mmhub ras_late_init callback function (v2)

[Zhou1, Tao]

> -----Original Message-----
> From: Hawking Zhang <Hawking.Zhang at amd.com>
> Sent: 2019年8月29日 21:31
> To: amd-gfx at lists.freedesktop.org; Deucher, Alexander
> <Alexander.Deucher at amd.com>; Zhou1, Tao <Tao.Zhou1 at amd.com>; Chen,
> Guchun <Guchun.Chen at amd.com>
> Cc: Zhang, Hawking <Hawking.Zhang at amd.com>
> Subject: [PATCH 5/7] drm/amdgpu: add mmhub ras_late_init callback
> function (v2)
> 
> The function will be called in late init phase to do mmhub ras init
> 
> v2: check ras_late_init function pointer before invoking the function
> 
> Signed-off-by: Hawking Zhang <Hawking.Zhang at amd.com>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h |  1 +
>  drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c     | 26 ++++----------------------
>  drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c   | 28
> ++++++++++++++++++++++++++++
>  3 files changed, 33 insertions(+), 22 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
> index 2d75ecf..df04c71 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h
> @@ -23,6 +23,7 @@
> 
>  struct amdgpu_mmhub_funcs {
>  	void (*ras_init)(struct amdgpu_device *adev);
> +	int (*ras_late_init)(struct amdgpu_device *adev);
>  	void (*query_ras_error_count)(struct amdgpu_device *adev,
>  					void *ras_error_status);
>  };
> diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> index 8a7a56a..70a05e3 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
> @@ -762,7 +762,6 @@ static int gmc_v9_0_ecc_late_init(void *handle)  {
>  	int r;
>  	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
> -	struct ras_ih_if mmhub_ih_info;
>  	struct ras_fs_if umc_fs_info = {
>  		.sysfs_name = "umc_err_count",
>  		.debugfs_name = "umc_err_inject",
> @@ -770,10 +769,6 @@ static int gmc_v9_0_ecc_late_init(void *handle)
>  	struct ras_ih_if umc_ih_info = {
>  		.cb = gmc_v9_0_process_ras_data_cb,
>  	};
> -	struct ras_fs_if mmhub_fs_info = {
> -		.sysfs_name = "mmhub_err_count",
> -		.debugfs_name = "mmhub_err_inject",
> -	};
> 
>  	if (!adev->gmc.umc_ras_if) {
>  		adev->gmc.umc_ras_if = kmalloc(sizeof(struct
> ras_common_if), GFP_KERNEL); @@ -797,29 +792,16 @@ static int
> gmc_v9_0_ecc_late_init(void *handle)
>  			goto umc_late_fini;
>  	}
> 
> -	if (!adev->gmc.mmhub_ras_if) {
> -		adev->gmc.mmhub_ras_if = kmalloc(sizeof(struct
> ras_common_if), GFP_KERNEL);
> -		if (!adev->gmc.mmhub_ras_if)
> -			return -ENOMEM;
> -		adev->gmc.mmhub_ras_if->block =
> AMDGPU_RAS_BLOCK__MMHUB;
> -		adev->gmc.mmhub_ras_if->type =
> AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE;
> -		adev->gmc.mmhub_ras_if->sub_block_index = 0;
> -		strcpy(adev->gmc.mmhub_ras_if->name, "mmhub");
> +	if (adev->mmhub_funcs->ras_late_init) {
> +		r = adev->mmhub_funcs->ras_late_init(adev);
> +		if (r)
> +			return r;
>  	}
> -	mmhub_ih_info.head = mmhub_fs_info.head = *adev-
> >gmc.mmhub_ras_if;
[Tao] mmhub_ih_info.cb = NULL is recommended in case of random value

> -	r = amdgpu_ras_late_init(adev, adev->gmc.mmhub_ras_if,
> -				 &mmhub_fs_info, &mmhub_ih_info);
> -	if (r)
> -		goto mmhub_late_fini;
> -
>  	return 0;
> -mmhub_late_fini:
> -	amdgpu_ras_late_fini(adev, adev->gmc.mmhub_ras_if,
> &mmhub_ih_info);
>  umc_late_fini:
>  	amdgpu_ras_late_fini(adev, adev->gmc.umc_ras_if, &umc_ih_info);
>  free:
>  	kfree(adev->gmc.umc_ras_if);
> -	kfree(adev->gmc.mmhub_ras_if);
>  	return r;
>  }
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
> b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
> index 04cd4b6..9f7d5d1 100644
> --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
> +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
> @@ -31,6 +31,7 @@
>  #include "vega10_enum.h"
> 
>  #include "soc15_common.h"
> +#include "amdgpu_ras.h"
> 
>  #define mmDAGB0_CNTL_MISC2_RV 0x008f
>  #define mmDAGB0_CNTL_MISC2_RV_BASE_IDX 0 @@ -615,6 +616,33 @@
> static void mmhub_v1_0_query_ras_error_count(struct amdgpu_device
> *adev,
>  	}
>  }
> 
> +static int mmhub_v1_0_ras_late_init(struct amdgpu_device *adev) {
> +	int r;
> +	struct ras_ih_if mmhub_ih_info;
> +	struct ras_fs_if mmhub_fs_info = {
> +		.sysfs_name = "mmhub_err_count",
> +		.debugfs_name = "mmhub_err_inject",
> +	};
> +
> +	if (!adev->gmc.mmhub_ras_if) {
> +		adev->gmc.mmhub_ras_if = kmalloc(sizeof(struct
> ras_common_if), GFP_KERNEL);
> +		if (!adev->gmc.mmhub_ras_if)
> +			return -ENOMEM;
> +		adev->gmc.mmhub_ras_if->block =
> AMDGPU_RAS_BLOCK__MMHUB;
> +		adev->gmc.mmhub_ras_if->type =
> AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE;
> +		adev->gmc.mmhub_ras_if->sub_block_index = 0;
> +		strcpy(adev->gmc.mmhub_ras_if->name, "mmhub");
> +	}
> +	mmhub_ih_info.head = mmhub_fs_info.head = *adev-
> >gmc.mmhub_ras_if;
> +	r = amdgpu_ras_late_init(adev, adev->gmc.mmhub_ras_if,
> +				 &mmhub_fs_info, &mmhub_ih_info);
> +	if (r)
> +		kfree(adev->gmc.mmhub_ras_if);
> +	return r;
> +}
> +
>  const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs = {
> +	.ras_late_init = mmhub_v1_0_ras_late_init,
>  	.query_ras_error_count = mmhub_v1_0_query_ras_error_count,  };
> --
> 2.7.4