Yiltan 1347d5d628 [GDA] Alltoall optimization - single warp (#319) ...

* Remove testing of data types
As the collective is templated, we are just testing if sizeof(T) works

* Added single threaded varients

* Applied thread puts optimization to barrier

* Apply single threaded optimization to alltoall

* This optimization only works on bnxt, so place a switch to protect it

* Handle the edge case where the thread count is smaller than the number of PEs

2025-11-19 14:25:29 -05:00

..

amo_bitwise_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

amo_bitwise_tester.hpp

Update atomic functional tests (#262)

2025-10-06 10:50:50 -05:00

amo_extended_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

amo_extended_tester.hpp

Update atomic functional tests (#262)

2025-10-06 10:50:50 -05:00

amo_standard_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

amo_standard_tester.hpp

Update atomic functional tests (#262)

2025-10-06 10:50:50 -05:00

barrier_all_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

barrier_all_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

CMakeLists.txt

Tests/syncall (#291)

2025-10-23 13:40:41 -04:00

default_ctx_primitive_tester.cpp

add support for GPUs using wavefront size of 32 (#285)

2025-10-22 16:04:58 -05:00

default_ctx_primitive_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

empty_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

empty_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

ping_all_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

ping_all_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

ping_pong_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

ping_pong_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

primitive_mr_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

primitive_mr_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

primitive_tester.cpp

add support for GPUs using wavefront size of 32 (#285)

2025-10-22 16:04:58 -05:00

primitive_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

random_access_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

random_access_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

shmem_ptr_tester.cpp

add support for GPUs using wavefront size of 32 (#285)

2025-10-22 16:04:58 -05:00

shmem_ptr_tester.hpp

Implement rocshmem_ptr in IPC conduit (#197)

2025-07-28 12:01:02 -05:00

signaling_operations_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

signaling_operations_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

sync_all_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

sync_all_tester.hpp

Tests/syncall (#291)

2025-10-23 13:40:41 -04:00

team_alltoall_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

team_alltoall_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

team_barrier_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

team_barrier_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

team_broadcast_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

team_broadcast_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

team_ctx_infra_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

team_ctx_infra_tester.hpp

Add extended team tests (#207)

2025-08-01 08:50:14 -05:00

team_ctx_primitive_tester.cpp

add support for GPUs using wavefront size of 32 (#285)

2025-10-22 16:04:58 -05:00

team_ctx_primitive_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

team_fcollect_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

team_fcollect_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

team_reduction_tester.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

team_reduction_tester.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

team_sync_tester.cpp

Tests/syncall (#291)

2025-10-23 13:40:41 -04:00

team_sync_tester.hpp

Tests/syncall (#291)

2025-10-23 13:40:41 -04:00

test_driver.cpp

Tests/syncall (#291)

2025-10-23 13:40:41 -04:00

tester_arguments.cpp

functional_tests: n, nskip, nloop, nlarge options (#297)

2025-10-30 11:54:49 -04:00

tester_arguments.hpp

Tests/syncall (#291)

2025-10-23 13:40:41 -04:00

tester.cpp

[GDA] Alltoall optimization - single warp (#319)

2025-11-19 14:25:29 -05:00

tester.hpp

Tests/syncall (#291)

2025-10-23 13:40:41 -04:00

verify_results_kernels.hpp

Check RMA functional test data in GPU kernel (#91)

2025-04-28 16:06:05 -04:00

wavefront_primitives.cpp

gda ionic: use all threads in wave operations (#295)

2025-11-05 11:01:14 -06:00

wavefront_primitives.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00

workgroup_primitives.cpp

Cleanup/wg init (#260)

2025-10-07 14:34:18 -04:00

workgroup_primitives.hpp

functional_tests: use size_t for size variable (#190)

2025-07-03 13:26:54 -05:00