]> www.infradead.org Git - users/jedix/linux-maple.git/commit
drm/xe/pcode: Treat pcode as per-tile rather than per-GT
authorMatt Roper <matthew.d.roper@intel.com>
Thu, 29 Aug 2024 22:06:21 +0000 (15:06 -0700)
committerMatt Roper <matthew.d.roper@intel.com>
Fri, 30 Aug 2024 15:56:20 +0000 (08:56 -0700)
commit3034cc8107b8d0c7d1b56584394e215dab57f8a3
tree894bb987beec8c164b14cdf4daed92067c47f334
parentcad08fa7760ace5e01b24c6f9e1f918fe72324bf
drm/xe/pcode: Treat pcode as per-tile rather than per-GT

There's only one instance of the pcode per tile, and for GT-related
accesses both the primary and media GT share the same register
interface.  Since Xe was using per-GT locking, the pcode mutex wasn't
actually protecting everything that it should since concurrent accesses
related to a tile's primary GT and media GT were possible.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240829220619.789159-5-matthew.d.roper@intel.com
drivers/gpu/drm/xe/compat-i915-headers/intel_pcode.h
drivers/gpu/drm/xe/compat-i915-headers/intel_uncore.h
drivers/gpu/drm/xe/xe_device_types.h
drivers/gpu/drm/xe/xe_gt.c
drivers/gpu/drm/xe/xe_gt_types.h
drivers/gpu/drm/xe/xe_guc_pc.c
drivers/gpu/drm/xe/xe_hwmon.c
drivers/gpu/drm/xe/xe_pcode.c
drivers/gpu/drm/xe/xe_pcode.h
drivers/gpu/drm/xe/xe_tile.c
drivers/gpu/drm/xe/xe_vram_freq.c