AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
12,807 Commits 3 Branches 0 Tags
43e2bb61260a97c14aa4e3c8285b6bafbb920306
Commit Graph

3 Commits

Author SHA1 Message Date
Sameer Sahasrabuddhe 0421a35f71 SWDEV-295298: hide warp sync builtins in ROCm 6.2 (absent in 6.1) 
Change-Id: I6d59e4f3a133dce5ea0521fc9d09f93f9fd0f376
 2024-03-11 02:44:20 -04:00
Sameer Sahasrabuddhe d80168eb87 SWDEV-295298: hide warp sync builtins with a macro in ROCm 6.1 
Change-Id: Ie2efd233c0bcf8ad0e06223ee854fe4bd1060443
 2024-02-05 05:08:11 +00:00
Sameer Sahasrabuddhe c5ab5680b4 SWDEV-295298: introduce warp sync functions 
The following builtins from the CUDA spec are implemented:

  - __all_sync, __any_sync, __ballot_sync and __activemask
  - __match_any_sync and __match_all_sync
  - __shfl_sync, __shfl_up_sync, __shfl_down_sync, and __shfl_xor_sync

The following builtins are NOT implemented, pending support in the compiler:

  - __reduce_add_sync, __reduce_min_sync, __reduce_max_sync
  - __reduce_and_sync, __reduce_or_sync, __reduce_xor_sync

Change-Id: I07dedbbfe5449f4b5c9b040bed59f5603ccec8c3
 2024-01-11 22:31:27 -05:00