Wenju He 7b94b9ae13 [libclc] Refine generic __clc_get_sub_group_size with fast full sub-group path (#188895) ...

Add a fast path for the common case that total work-group size is
multiple of max sub-group size.

The fallback path is ported from amdgpu/workitem/clc_get_sub_group_size.cl.

Compiler can generate predicated instructions for the fallback path to
avoid branches.

2026-04-13 08:16:06 +08:00

..

include/clc

libclc: Partially implement nonuniform subgroup reduce functions (#188929)

2026-03-27 09:47:44 +00:00

lib

[libclc] Refine generic __clc_get_sub_group_size with fast full sub-group path (#188895)

2026-04-13 08:16:06 +08:00