Re: [PATCH v7 00/10] Add CPER logging support for CRI

[Raag Jadav <raag.jadav@intel.com>]

On Thu, Oct 29, 2026 at 01:57:53PM +0530, Badal Nilawar wrote:
> This patch series adds CPER (Common Platform Error Record) logging support
> for RAS errors reported by Intel Xe GPUs. CPER records are emitted through
> a trace event.
> 
> v2:
>    - Extended CPER logging to Uncorrectable errors and xe_ras_get_counter
>      request
>    - Log CPER records via xe_log SIGID infra 
> 
> v3:
>   - Added the patches in incremental order (Raag)
>   - Fixed Sashiko warnings which are applicable 
> 
> v4:
>   - Split the patches further for incremental review (Raag)
>   - Refactor info queue handling (Raag)
>   - Fix applicable Sashiko warnings
> 
> v5:
>   - Fixed v4 comments
>   - Covered default paths in error handling flow
>   - Moved info queue related patches at top
>   - Added example logs
> 
> v6:
>   - Fixed v6 comments
>   - Dropped the Info Queue patches; they will be submitted separately
>   - Dropped "drm/xe/xe_ras: Report unknown errors using SIGID" as
>     SIGID logging is not needed for it (Riana/Raag)
> v7:
>   - Rebased
>   - Fixed Sashiko comments
>   - Added GUID of PCIID 6752

I think we missed a few comments[1][2][3] here. Please make sure to address
them.

Raag

[1] https://lore.kernel.org/intel-xe/auIoa-LxjETpt0LR@black.igk.intel.com/
[2] https://lore.kernel.org/intel-xe/auIrXSWHyxE2K_M6@black.igk.intel.com/
[3] https://lore.kernel.org/intel-xe/auIuFsYCyl5m2mrG@black.igk.intel.com/

> Badal Nilawar (10):
>   drm/xe/cper: Add CPER reporting entry point for xe_log
>   drm/xe/cper: Retrieve the error counter record for CPER reporting
>   drm/xe/cper: Initialize Intel-specific CPER header
>   drm/xe/cper: Construct CPER record for hardware errors
>   drm/xe/cper: Log CPER records for aggregate counter retrieval
>   drm/xe/cper: Emit cper record to trace buf
>   drm/xe/xe_ras: Report device memory errors using SIGID
>   drm/xe/xe_ras: Report core compute errors using SIGID
>   drm/xe/xe_ras: Report soc internal errors using SIGID
>   drm/xe/xe_ras: Report correctable errors using SIGID
> 
>  .../ABI/testing/sysfs-driver-intel-xe-cper    |  18 +
>  drivers/gpu/drm/xe/Makefile                   |   3 +
>  drivers/gpu/drm/xe/xe_cper.c                  | 321 ++++++++++++++++++
>  drivers/gpu/drm/xe/xe_cper.h                  |  37 ++
>  drivers/gpu/drm/xe/xe_cper_types.h            | 110 ++++++
>  drivers/gpu/drm/xe/xe_device_types.h          |   4 +
>  drivers/gpu/drm/xe/xe_log.c                   |   5 +-
>  drivers/gpu/drm/xe/xe_ras.c                   | 146 ++++++--
>  drivers/gpu/drm/xe/xe_ras.h                   |   2 +
>  drivers/gpu/drm/xe/xe_trace_cper.c            |   9 +
>  drivers/gpu/drm/xe/xe_trace_cper.h            |  65 ++++
>  11 files changed, 692 insertions(+), 28 deletions(-)
>  create mode 100644 Documentation/ABI/testing/sysfs-driver-intel-xe-cper
>  create mode 100644 drivers/gpu/drm/xe/xe_cper.c
>  create mode 100644 drivers/gpu/drm/xe/xe_cper.h
>  create mode 100644 drivers/gpu/drm/xe/xe_cper_types.h
>  create mode 100644 drivers/gpu/drm/xe/xe_trace_cper.c
>  create mode 100644 drivers/gpu/drm/xe/xe_trace_cper.h
> 
> -- 
> 2.56.0
>