<html> <head> <meta http-equiv="Content-Type" content="text/html; charset=us-ascii"> <style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style> </head> <body dir="ltr"> [AMD Official Use Only - AMD Internal Distribution Only] <div> <div class="elementToProof" style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);"> Thanks, as the discuss offline, will improve it.</div> Best Regards, Dean <div id="appendonsend"></div> <hr style="display:inline-block;width:98%" tabindex="-1"> <div id="divRplyFwdMsg" dir="ltr">From: Wang, Yang(Kevin) <KevinYang.Wang@amd.com> Sent: Thursday, March 20, 2025 3:05 PM To: Liu, Xiang(Dean) <Xiang.Liu@amd.com>; amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org> Cc: Zhang, Hawking <Hawking.Zhang@amd.com>; Zhou1, Tao <Tao.Zhou1@amd.com>; Chai, Thomas <YiPeng.Chai@amd.com> Subject: RE: [PATCH] drm/amdgpu: Decode deferred error type in gfx aca bank parser <div> </div> </div> <div class="BodyFragment"><a name="BM_BEGIN"></a> <div>[AMD Official Use Only - AMD Internal Distribution Only] + if (type == ACA_ERROR_TYPE_UE) + aca_log_aca_error(handle, ACA_ERROR_TYPE_DEFERRED, err_data); + return aca_log_aca_error(handle, type, err_data); } It seems that the above code is incorrect, which may lead to error statistics of different types of errors. In the aca parser interface, I think there is enough information to identify the types of errors and correctly count errors, Please double check it, thanks. Best Regards, Kevin -----Original Message----- From: Liu, Xiang(Dean) <Xiang.Liu@amd.com> Sent: Thursday, March 20, 2025 14:15 To: amd-gfx@lists.freedesktop.org Cc: Zhang, Hawking <Hawking.Zhang@amd.com>; Wang, Yang(Kevin) <KevinYang.Wang@amd.com>; Zhou1, Tao <Tao.Zhou1@amd.com>; Chai, Thomas <YiPeng.Chai@amd.com>; Liu, Xiang(Dean) <Xiang.Liu@amd.com> Subject: [PATCH] drm/amdgpu: Decode deferred error type in gfx aca bank parser In the case of injecting uncorrected error with background workload, the deferred error among uncorrected errors need to be specified by checking the deferred and poison bits of status register. Signed-off-by: Xiang Liu <xiang.liu@amd.com> --- drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c | 3 +++ drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 11 +++++++---- 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c index ffd4c64e123c..3f45a600f547 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c @@ -541,6 +541,9 @@ static int __aca_get_error_data(struct amdgpu_device *adev, struct aca_handle *h if (ret) return ret; + if (type == ACA_ERROR_TYPE_UE) + aca_log_aca_error(handle, ACA_ERROR_TYPE_DEFERRED, err_data); + return aca_log_aca_error(handle, type, err_data); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index c0de682b7774..b21d784a7f9c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -876,7 +876,7 @@ static int gfx_v9_4_3_aca_bank_parser(struct aca_handle *handle, void *data) { struct aca_bank_info info; - u64 misc0; + u64 misc0, status; u32 instlo; int ret; @@ -890,12 +890,15 @@ static int gfx_v9_4_3_aca_bank_parser(struct aca_handle *handle, info.die_id = instlo == mmSMNAID_XCD0_MCA_SMU ? 0 : 1; misc0 = bank->regs[ACA_REG_IDX_MISC0]; + status = bank->regs[ACA_REG_IDX_STATUS]; switch (type) { case ACA_SMU_TYPE_UE: - bank->aca_err_type = ACA_ERROR_TYPE_UE; - ret = aca_error_cache_log_bank_error(handle, &info, - ACA_ERROR_TYPE_UE, 1ULL); + bank->aca_err_type = (ACA_REG__STATUS__POISON(status) || + ACA_REG__STATUS__DEFERRED(status)) ? + ACA_ERROR_TYPE_DEFERRED : + ACA_ERROR_TYPE_UE; + ret = aca_error_cache_log_bank_error(handle, &info, +bank->aca_err_type, 1ULL); break; case ACA_SMU_TYPE_CE: bank->aca_err_type = ACA_BANK_ERR_CE_DE_DECODE(bank); -- 2.34.1 </div> </div> </div> </body> </html>