Dynamic wavefront creation for processing units using a hybrid compactor
A method, a non-transitory computer readable medium, and a processor for repacking dynamic wavefronts during program code execution on a processing unit, each dynamic wavefront including multiple threads are presented. If a branch instruction is detected, a determination is made whether all wavefronts following a same control path in the program code have reached a compaction point, which is the branch instruction. If no branch instruction is detected in executing the program code, a determination is made whether all wavefronts following the same control path have reached a reconvergence point, which is a beginning of a program code segment to be executed by both a taken branch and a not taken branch from a previous branch instruction. The dynamic wavefronts are repacked with all threads that follow the same control path, if all wavefronts following the same control path have reached the branch instruction or the reconvergence point.
- Research Organization:
- Advanced Micro Devices, Inc., Sunnyvale, CA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC52-07NA27344
- Assignee:
- Advanced Micro Devices, Inc. (Sunnyvale, CA)
- Patent Number(s):
- 9,898,287
- Application Number:
- 14/682,971
- OSTI ID:
- 1422752
- Resource Relation:
- Patent File Date: 2015 Apr 09
- Country of Publication:
- United States
- Language:
- English
Similar Records
Graphics processor unit with opportunistic inter-path reconvergence
System and method for hardware scheduling of conditional barriers and impatient barriers