Jan Wassenberg 701841897b Default to disabling per-socket parallelization ... weights: default to Read for small-batch (only look at qbatch, not the larger prefill tbatch) PiperOrigin-RevId: 790787643		2025-08-04 09:49:14 -07:00
..
allocator.cc	Cleanup: remove unused kCyclic, remove 2 suffix	2025-05-13 01:06:41 -07:00
allocator.h	1.31x batch prefill, 1.24x batch decode speedup: NUMA binding	2025-05-16 07:42:13 -07:00
args.h	Move fields, io* and blob* from compression/ into io/	2025-05-06 11:17:19 -07:00
basics.h	Default to disabling per-socket parallelization	2025-08-04 09:49:14 -07:00
mat.cc	De-singleton ThreadingContext so callers can pass in their own	2025-07-22 02:08:46 -07:00
mat.h	De-singleton ThreadingContext so callers can pass in their own	2025-07-22 02:08:46 -07:00
test_util.h	Minor cleanup/fixes:	2024-09-09 06:58:09 -07:00
threading.cc	Avoid affinity related warnings on Apple. Refs #625	2025-07-03 08:22:31 -07:00
threading.h	Default to disabling per-socket parallelization	2025-08-04 09:49:14 -07:00
threading_context.cc	Pre-tune thread pool before matmul	2025-07-31 08:45:26 -07:00
threading_context.h	Default to disabling per-socket parallelization	2025-08-04 09:49:14 -07:00
threading_test.cc	De-singleton ThreadingContext so callers can pass in their own	2025-07-22 02:08:46 -07:00
topology.cc	Avoid affinity related warnings on Apple. Refs #625	2025-07-03 08:22:31 -07:00
topology.h	Fix thread name when skipping packages/clusters	2025-06-01 23:50:11 -07:00