RE: [PATCH 5/5] drm/amd/ras: add ras ce log switch for uniras

AMD General

> -----Original Message-----
> From: Chai, Thomas <[email protected]>
> Sent: Thursday, July 16, 2026 9:50 AM
> To: Zhou1, Tao <[email protected]>; Zhang, Hawking
> <[email protected]>; [email protected]
> Subject: RE: [PATCH 5/5] drm/amd/ras: add ras ce log switch for uniras
>
> AMD General
>
> Best Regards,
> Thomas
> -----Original Message-----
> From: amd-gfx <[email protected]> On Behalf Of Zhou1, Tao
> Sent: Wednesday, July 15, 2026 4:57 PM
> To: Zhang, Hawking <[email protected]>; [email protected]
> Subject: RE: [PATCH 5/5] drm/amd/ras: add ras ce log switch for uniras
>
> AMD General
>
> OK. Since Ce is working on it as well, he will refine his code per my patch and your
> suggestion.
>
> Regards,
> Tao
>
> > -----Original Message-----
> > From: Zhang, Hawking <[email protected]>
> > Sent: Wednesday, July 15, 2026 4:27 PM
> > To: Zhou1, Tao <[email protected]>; [email protected]
> > Cc: Zhou1, Tao <[email protected]>
> > Subject: RE: [PATCH 5/5] drm/amd/ras: add ras ce log switch for uniras
> >
> > AMD General
> >
> > +       init_config.debug_disable_ce_logs =
> > + adev->debug_disable_ce_logs;
> >
> > Let's pass the full amdgpu_debug_mask to ras core
> >
> > Regards,
> > Hawking
> >
> > -----Original Message-----
> > From: amd-gfx <[email protected]> On Behalf Of Tao
> > Zhou
> > Sent: Wednesday, July 15, 2026 3:48 PM
> > To: [email protected]
> > Cc: Zhou1, Tao <[email protected]>
> > Subject: [PATCH 5/5] drm/amd/ras: add ras ce log switch for uniras
> >
> > So we can disable ce log manually.
> >
> > Signed-off-by: Tao Zhou <[email protected]>
> > ---
> >  drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c | 1 +
> >  drivers/gpu/drm/amd/ras/rascore/ras.h            | 2 ++
> >  drivers/gpu/drm/amd/ras/rascore/ras_aca.c        | 7 ++++++-
> >  drivers/gpu/drm/amd/ras/rascore/ras_core.c       | 1 +
> >  4 files changed, 10 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c
> > b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c
> > index c55288c2df9a..1e95419989b4 100644
> > --- a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c
> > +++ b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c
> > @@ -310,6 +310,7 @@ static struct ras_core_context
> > *amdgpu_ras_mgr_create_ras_core(struct amdgpu_dev
> >                 amdgpu_ras_mgr_eeprom_is_supported(adev);
> >         init_config.poison_supported =
> >                 amdgpu_ras_is_poison_mode_supported(adev);
> > +       init_config.debug_disable_ce_logs =
> > + adev->debug_disable_ce_logs;
> >
> >         amdgpu_ras_mgr_init_aca_config(adev, &init_config);
> >         amdgpu_ras_mgr_init_eeprom_config(adev, &init_config); diff
> > --git a/drivers/gpu/drm/amd/ras/rascore/ras.h
> > b/drivers/gpu/drm/amd/ras/rascore/ras.h
> > index 5719bc1b6167..9d538a07c50e 100644
> > --- a/drivers/gpu/drm/amd/ras/rascore/ras.h
> > +++ b/drivers/gpu/drm/amd/ras/rascore/ras.h
> > @@ -301,6 +301,7 @@ struct ras_core_config {
> >
> >         bool poison_supported;
> >         bool ras_eeprom_supported;
> > +       bool debug_disable_ce_logs;
> >         const struct ras_sys_func *sys_fn;
> >
> >         struct ras_aca_config aca_cfg; @@ -348,6 +349,7 @@ struct
> > ras_core_context {
> >         bool ras_core_enabled;
> >
> >         u64 ras_fw_features;
> > +       bool debug_disable_ce_logs;
> >  };
> >
> >  struct ras_core_context *ras_core_create(struct ras_core_config
> > *init_config); diff -- git a/drivers/gpu/drm/amd/ras/rascore/ras_aca.c
> > b/drivers/gpu/drm/amd/ras/rascore/ras_aca.c
> > index 67a35409ff0e..9760894c4c66 100644
> > --- a/drivers/gpu/drm/amd/ras/rascore/ras_aca.c
> > +++ b/drivers/gpu/drm/amd/ras/rascore/ras_aca.c
> > @@ -98,7 +98,7 @@ static void aca_report_ecc_info(struct
> > ras_core_context *ras_core,
> >                         blk_name(blk));
> >         }
> >
> > -       if (ecc_count.new_ce_count) {
> > +       if (ecc_count.new_ce_count &&
> > + !ras_core->debug_disable_ce_logs) {
>
> [Thomas] Can we replace ras_core->debug_disable_ce_logs with the function
> interface in sun,ce patch?  The other changes are the same.

[Tao] sure, Ce will update it.

>
> >                 RAS_DEV_INFO(ras_core->dev,
> >                 "{%llu} socket: %d, die: %d, %u new correctable
> > hardware errors detected in %s block\n",
> >                         seq_no, skt, aid, ecc_count.new_ce_count,
> > blk_name(blk)); @@ -
> > 114,6 +114,11 @@ static void aca_bank_log(struct ras_core_context *ras_core,  {
> >         int i;
> >
> > +       if (ras_core->debug_disable_ce_logs &&
> > +           bank->ecc_type == RAS_ERR_TYPE__CE &&
> > +           !bank_ecc->real_de_count)
> > +               return;
> > +
> >         RAS_DEV_INFO(ras_core->dev,
> >                 "{%llu}" RAS_HW_ERR "Accelerator Check Architecture events
> logged\n",
> >                 bank->seq_no);
> > diff --git a/drivers/gpu/drm/amd/ras/rascore/ras_core.c
> > b/drivers/gpu/drm/amd/ras/rascore/ras_core.c
> > index 2346918c7736..cdef7727decb 100644
> > --- a/drivers/gpu/drm/amd/ras/rascore/ras_core.c
> > +++ b/drivers/gpu/drm/amd/ras/rascore/ras_core.c
> > @@ -377,6 +377,7 @@ int ras_core_hw_init(struct ras_core_context *ras_core)
> >                         ras_core->config->ras_eeprom_supported;
> >
> >         ras_core->poison_supported =
> > ras_core->config->poison_supported;
> > +       ras_core->debug_disable_ce_logs =
> > +ras_core->config->debug_disable_ce_logs;
> >
> >         ret = ras_psp_hw_init(ras_core);
> >         if (ret)
> > --
> > 2.34.1
> >
>