[PATCH Review 1/1] drm/amdgpu: print more correctable error info

[Zhou1, Tao]

[AMD Official Use Only]



> -----Original Message-----
> From: Stanley.Yang <Stanley.Yang at amd.com>
> Sent: Friday, April 8, 2022 10:18 AM
> To: amd-gfx at lists.freedesktop.org; Zhang, Hawking
> <Hawking.Zhang at amd.com>; Zhou1, Tao <Tao.Zhou1 at amd.com>; Li, Candice
> <Candice.Li at amd.com>
> Cc: Yang, Stanley <Stanley.Yang at amd.com>
> Subject: [PATCH Review 1/1] drm/amdgpu: print more correctable error info
> 

[Tao] it's better to add description for the patch.

> Change-Id: I09a2aae85cde3ab2cb6b042b973da6839ad024ec
> Signed-off-by: Stanley.Yang <Stanley.Yang at amd.com>
> ---
>  drivers/gpu/drm/amd/amdgpu/umc_v6_7.c | 62
> ++++++++++++++++++++++++++-
>  1 file changed, 60 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c
> b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c
> index c45d9c14ecbc..803119f75e39 100644
> --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c
> +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c
> @@ -70,15 +70,46 @@ static void
> umc_v6_7_ecc_info_query_correctable_error_count(struct amdgpu_device  {
>  	uint64_t mc_umc_status;
>  	uint32_t eccinfo_table_idx;
> +	uint32_t umc_reg_offset;
> +	uint32_t mc_umc_addr;
> +	uint64_t reg_value;
>  	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
> 
> +	umc_reg_offset = get_umc_v6_7_reg_offset(adev,
> +						umc_inst, ch_inst);
> +
>  	eccinfo_table_idx = umc_inst * adev->umc.channel_inst_num + ch_inst;
>  	/* check for SRAM correctable error
>  	  MCUMC_STATUS is a 64 bit register */
>  	mc_umc_status = ras-
> >umc_ecc.ecc[eccinfo_table_idx].mca_umc_status;
>  	if (REG_GET_FIELD(mc_umc_status,
> MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
> -	    REG_GET_FIELD(mc_umc_status,
> MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)
> +	    REG_GET_FIELD(mc_umc_status,
> MCA_UMC_UMC0_MCUMC_STATUST0, CECC) ==
> +1) {
>  		*error_count += 1;
> +
> +		if (mc_umc_status)
> +			dev_info(adev->dev, "MCA STATUS 0x%llx,
> umc_reg_offset 0x%x\n",
> +mc_umc_status, umc_reg_offset);
> +
> +		/* print IPID registers value */
> +		mc_umc_addr =
> +			SOC15_REG_OFFSET(UMC, 0,
> regMCA_UMC_UMC0_MCUMC_IPIDT0);
> +		reg_value = RREG64_PCIE((mc_umc_addr + umc_reg_offset) *
> 4);
> +		if (reg_value)
> +			dev_info(adev->dev, "MCA IPID 0x%llx, umc_reg_offset
> 0x%x\n",
> +reg_value, umc_reg_offset);
> +
> +		/* print SYND registers value */
> +		mc_umc_addr =
> +			SOC15_REG_OFFSET(UMC, 0,
> regMCA_UMC_UMC0_MCUMC_SYNDT0);
> +		reg_value = RREG64_PCIE((mc_umc_addr + umc_reg_offset) *
> 4);
> +		if (reg_value)
> +			dev_info(adev->dev, "MCA SYND 0x%llx,
> umc_reg_offset 0x%x\n",
> +reg_value, umc_reg_offset);
> +
> +		/* print MISC0 registers value */
> +		mc_umc_addr =
> +			SOC15_REG_OFFSET(UMC, 0,
> regMCA_UMC_UMC0_MCUMC_MISC0T0);
> +		reg_value = RREG64_PCIE((mc_umc_addr + umc_reg_offset) *
> 4);
> +		if (reg_value)
> +			dev_info(adev->dev, "MCA MISC0 0x%llx,
> umc_reg_offset 0x%x\n", reg_value, umc_reg_offset);
> +	}

[Tao] can we implement a query_error_status function and:

1. call query_error_status in xxx_error_count function, like this patch, or
2. register it to adev->umc.ras and call it in amdgpu_ras_get_ecc_info.

>  }
> 
>  static void umc_v6_7_ecc_info_querry_uncorrectable_error_count(struct
> amdgpu_device *adev, @@ -242,7 +273,9 @@ static void
> umc_v6_7_query_correctable_error_count(struct amdgpu_device *adev,
>  	uint32_t ecc_err_cnt_sel, ecc_err_cnt_sel_addr;
>  	uint32_t ecc_err_cnt, ecc_err_cnt_addr;
>  	uint64_t mc_umc_status;
> +	uint32_t mc_umc_addr;
>  	uint32_t mc_umc_status_addr;
> +	uint64_t reg_value;
> 
>  	/* UMC 6_1_1 registers */
>  	ecc_err_cnt_sel_addr =
> @@ -277,8 +310,33 @@ static void
> umc_v6_7_query_correctable_error_count(struct amdgpu_device *adev,
>  	  MCUMC_STATUS is a 64 bit register */
>  	mc_umc_status = RREG64_PCIE((mc_umc_status_addr +
> umc_reg_offset) * 4);
>  	if (REG_GET_FIELD(mc_umc_status,
> MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 &&
> -	    REG_GET_FIELD(mc_umc_status,
> MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)
> +	    REG_GET_FIELD(mc_umc_status,
> MCA_UMC_UMC0_MCUMC_STATUST0, CECC) ==
> +1) {
>  		*error_count += 1;
> +
> +		if (mc_umc_status)
> +			dev_info(adev->dev, "MCA STATUS 0x%llx,
> umc_reg_offset 0x%x\n",
> +mc_umc_status, umc_reg_offset);
> +
> +		/* print IPID registers value */
> +		mc_umc_addr =
> +			SOC15_REG_OFFSET(UMC, 0,
> regMCA_UMC_UMC0_MCUMC_IPIDT0);
> +		reg_value = RREG64_PCIE((mc_umc_addr + umc_reg_offset) *
> 4);
> +		if (reg_value)
> +			dev_info(adev->dev, "MCA IPID 0x%llx, umc_reg_offset
> 0x%x\n",
> +reg_value, umc_reg_offset);
> +
> +		/* print SYND registers value */
> +		mc_umc_addr =
> +			SOC15_REG_OFFSET(UMC, 0,
> regMCA_UMC_UMC0_MCUMC_SYNDT0);
> +		reg_value = RREG64_PCIE((mc_umc_addr + umc_reg_offset) *
> 4);
> +		if (reg_value)
> +			dev_info(adev->dev, "MCA SYND 0x%llx,
> umc_reg_offset 0x%x\n",
> +reg_value, umc_reg_offset);
> +
> +		/* print MISC0 registers value */
> +		mc_umc_addr =
> +			SOC15_REG_OFFSET(UMC, 0,
> regMCA_UMC_UMC0_MCUMC_MISC0T0);
> +		reg_value = RREG64_PCIE((mc_umc_addr + umc_reg_offset) *
> 4);
> +		if (reg_value)
> +			dev_info(adev->dev, "MCA MISC0 0x%llx,
> umc_reg_offset 0x%x\n", reg_value, umc_reg_offset);
> +	}
>  }
> 
>  static void umc_v6_7_querry_uncorrectable_error_count(struct amdgpu_device
> *adev,
> --
> 2.17.1