drm/xe: Split H2G and G2H into separate buffer objects

Browse source

H2G and G2H buffers have different access patterns (H2G is CPU-write,
GuC-read, while G2H is GPU-write, CPU-read). On dGPU, these patterns
benefit from different memory placements: H2G in VRAM and G2H in system
memory. Split the CT buffer into two separate buffers—one for H2G and
one for G2H—and select the optimal placement for each.

This provides a significant performance improvement on the G2H read
path, reducing a single read from ~20 µs to under 1 µs on BMG.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patch.msgid.link/20260218043319.809548-2-matthew.brost@intel.com

...

This commit is contained in:

Matthew Brost 2026-02-17 20:33:17 -08:00

parent 2bcbf2dcde

commit af3de6cf06

2 changed files with 47 additions and 24 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

67

drivers/gpu/drm/xe/xe_guc_ct.c

View file

4

drivers/gpu/drm/xe/xe_guc_ct_types.h

View file