Files
compute-runtime/shared/source/xe_hpc_core
Michal Mrozek ed897c302d performance: Implement V2 version of tg dispatch size algorithm.
Signed-off-by: Michal Mrozek <michal.mrozek@intel.com>
Related-To: NEO-6989

-Prevent imbalance in multi dimensional dispatches
-Make sure to utilize as much Eus as possible
-Prefer highest possible tg dspatch count possible
-Make sure that xe_core doesn't have uneven workgroups
2023-11-03 15:54:04 +01:00
..
2023-10-06 15:22:40 +02:00
2022-07-07 12:17:53 +02:00
2023-02-15 13:21:14 +01:00
2023-01-03 11:52:59 +01:00
2021-12-07 10:22:41 +01:00
2021-12-07 10:22:41 +01:00
2023-02-10 20:26:13 +01:00
2021-12-07 10:22:41 +01:00
2022-12-23 10:46:34 +01:00