1027 lines
58 KiB
Plaintext
1027 lines
58 KiB
Plaintext
* ea47bbe5 |||| 2025-02-03T18:23:30-08:00 |||| Merge branch 'release/3_2_5' into 'main'
|
|
|\
|
|
| * 9536c1ce |||| 2025-02-03T18:19:36-08:00 |||| Version 3.2.5.1
|
|
| * 125b24eb |||| 2025-02-03T18:17:37-08:00 |||| Merge branch 'sethh/llvm_ir_perf_regression' into 'devel'
|
|
| |\
|
|
| | * b4a66e51 |||| 2025-02-03T18:17:37-08:00 |||| src/llvm_bitcode: Performance fixes for bitcode library
|
|
| |/
|
|
| * 3cc6c8bc |||| 2025-02-03T11:47:22-08:00 |||| Merge branch 'release/merge_back_to_devel' into 'devel'
|
|
| |\
|
|
| | * 1924c7a3 |||| 2025-02-01T21:14:47-08:00 |||| NVSHMEM 3.2.4-2
|
|
| |/
|
|
|/|
|
|
* | 8065af2b |||| 2025-01-24T17:11:50-08:00 |||| Merge branch 'release/3_2_4' into 'main'
|
|
|\ \
|
|
| * | 2464d386 |||| 2025-01-24T17:07:40-08:00 |||| Version 3.2.4.1
|
|
| |/
|
|
| * a5b9fb7f |||| 2025-01-24T17:03:29-08:00 |||| Merge branch 'release/3_2_4' into 'devel'
|
|
| |\
|
|
| | * 4135e874 |||| 2025-01-24T17:03:28-08:00 |||| Bug fixes for 3.2.4
|
|
| |/
|
|
| * 9d4c0280 |||| 2025-01-24T09:53:00-08:00 |||| Merge branch 'release/merge_back_to_devel' into 'devel'
|
|
| |\
|
|
| | * 2b8415d6 |||| 2025-01-24T08:42:36-08:00 |||| NVSHMEM 3.2.3-2
|
|
| |/
|
|
|/|
|
|
* | c18a1c04 |||| 2025-01-17T22:03:18-08:00 |||| Merge branch 'release/3_2_3' into 'main'
|
|
|\ \
|
|
| * | b699e360 |||| 2025-01-17T21:59:54-08:00 |||| Version 3.2.3.1
|
|
| |/
|
|
| * 74879c80 |||| 2025-01-17T21:58:02-08:00 |||| Merge branch 'alanger/pr/use_cudadevicesync' into 'devel'
|
|
| |\
|
|
| | * fe18cfd0 |||| 2025-01-17T21:58:02-08:00 |||| NVSHMEM 3.2.3
|
|
| |/
|
|
| * be3e8b7b |||| 2025-01-17T15:39:01-08:00 |||| Merge branch 'release/merge_back_to_devel' into 'devel'
|
|
| |\
|
|
| | * eb2ad3c1 |||| 2025-01-17T15:33:09-08:00 |||| NVSHMEM 3.2.2-5
|
|
| |/
|
|
|/|
|
|
* | 142df88d |||| 2025-01-15T17:41:03-08:00 |||| Merge branch 'release/3_2_2_4' into 'main'
|
|
|\ \
|
|
| * | fe0e6f61 |||| 2025-01-15T17:39:40-08:00 |||| Version 3.2.2.4
|
|
| |/
|
|
| * 0e1db183 |||| 2025-01-15T17:36:50-08:00 |||| Merge branch 'sethh/perftest_fix_default_arguments' into 'devel'
|
|
| |\
|
|
| | * a9606f5e |||| 2025-01-15T17:36:50-08:00 |||| NVSHMEM Bug fixes for 3.2.2-3
|
|
| |/
|
|
| * e2fd0322 |||| 2025-01-15T08:14:54-08:00 |||| Merge branch 'release/merge_back_to_devel' into 'devel'
|
|
| |\
|
|
| | * ce90fea0 |||| 2025-01-14T14:11:25-08:00 |||| Version 3.2.2.2
|
|
| |/
|
|
|/|
|
|
* | 35d220b0 |||| 2025-01-13T19:14:50-08:00 |||| Merge branch 'release/3_2_2' into 'main'
|
|
|\ \
|
|
| * | 0fa893c2 |||| 2025-01-13T19:10:54-08:00 |||| NVSHMEM 3.2.2
|
|
| |/
|
|
| * 8e253699 |||| 2025-01-13T19:08:37-08:00 |||| Merge branch 'release/3_2_2_fixes' into 'devel'
|
|
| |\
|
|
| | * 2f38b5dd |||| 2025-01-13T19:08:37-08:00 |||| Release/3.2.2 fixes
|
|
| |/
|
|
| * a6753268 |||| 2025-01-09T09:08:23-08:00 |||| Merge branch 'release/merge_321_to_devel' into 'devel'
|
|
| |\
|
|
| | * 46aa130d |||| 2025-01-08T18:54:34-08:00 |||| Version 3.2.1-2
|
|
| |/
|
|
|/|
|
|
* | 59322595 |||| 2025-01-07T22:11:35-08:00 |||| Merge branch 'release/3_2_1' into 'main'
|
|
|\ \
|
|
| * \ eb3482ae |||| 2025-01-07T22:10:49-08:00 |||| Merge branch 'main' into 'release/3_2_1'
|
|
| |\ \
|
|
| |/ /
|
|
|/| |
|
|
* | | 34a7fab6 |||| 2025-01-01T08:19:33-08:00 |||| Merge branch 'release/3_2_0' into 'main'
|
|
|\ \ \
|
|
| * | | 6c9c33c3 |||| 2024-12-31T15:54:29-08:00 |||| NVSHMEM 3.2.0
|
|
| * | | d37dbb69 |||| 2024-12-31T16:00:07-08:00 |||| in-place clang-formatting
|
|
* | | | 8a6f4f1f |||| 2024-10-29T04:31:06-07:00 |||| Merge branch 'devel' into 'main'
|
|
|\ \ \ \
|
|
* \ \ \ \ 03a91ab9 |||| 2024-10-25T19:17:41-07:00 |||| Merge branch 'devel' into 'main'
|
|
|\ \ \ \ \
|
|
| | | | * | 85eb6252 |||| 2025-01-07T17:19:39-08:00 |||| NVSHMEM 3.2.1
|
|
| | | | |/
|
|
| | | | * d06cc3ef |||| 2025-01-07T17:16:03-08:00 |||| Merge branch 'sethh/devel_3_2_1' into 'devel'
|
|
| | | | |\
|
|
| | | | | * e5123937 |||| 2025-01-07T17:16:02-08:00 |||| NVSHMEM: Bug fixes for 3.2.1 RC
|
|
| | | | |/
|
|
| | | | * 80a85b3c |||| 2025-01-06T12:18:39-08:00 |||| Merge branch 'release/merge_320_to_devel' into 'devel'
|
|
| | | |/|
|
|
| | | | * 48408d43 |||| 2025-01-06T12:18:39-08:00 |||| NVSHMEM 3.2 Merge main back to devel
|
|
| | | |/
|
|
| | | * e833e1dd |||| 2024-12-31T15:51:43-08:00 |||| Merge branch 'sethh/perftest_examples_dlopen_mpi' into 'devel'
|
|
| | | |\
|
|
| | | | * 50f2a159 |||| 2024-12-31T15:51:43-08:00 |||| perftest and examples: Remove linker dependency on MPI and SHMEM.
|
|
| | | |/
|
|
| | | * 259f037e |||| 2024-12-31T14:44:54-08:00 |||| Merge branch 'pezhu/llvm_ir_library' into 'devel'
|
|
| | | |\
|
|
| | | | * 7faf2654 |||| 2024-12-31T14:44:54-08:00 |||| Tests for bitcode library
|
|
| | | |/
|
|
| | | * 92e4d0c3 |||| 2024-12-31T10:50:01-08:00 |||| Merge branch 'alanger/pr/coll_perftest_args' into 'devel'
|
|
| | | |\
|
|
| | | | * 3c9b8189 |||| 2024-12-31T10:50:01-08:00 |||| Command line args for collectives perftests
|
|
| | | * | fcfa91c0 |||| 2024-12-31T10:42:58-08:00 |||| Merge branch 'sethh/CI_speedup' into 'devel'
|
|
| | | |\ \
|
|
| | | | |/
|
|
| | | |/|
|
|
| | | | * 8c476231 |||| 2024-12-31T10:42:58-08:00 |||| CI/test: Move sysmem failing tests to skip
|
|
| | | |/
|
|
| | | * b1a9964b |||| 2024-12-30T16:24:46-08:00 |||| Merge branch 'sethh/CI_check_format_fix' into 'devel'
|
|
| | | |\
|
|
| | | | * 8add734e |||| 2024-12-30T16:24:46-08:00 |||| CI: Limit check-format scope.
|
|
| | | |/
|
|
| | | * f4417c11 |||| 2024-12-30T12:43:55-08:00 |||| Merge branch 'sethh/llvm_ir_library' into 'devel'
|
|
| | | |\
|
|
| | | | * ef378fb7 |||| 2024-12-30T12:43:54-08:00 |||| NVSHMEM: Add initial bitcode library
|
|
| | | |/
|
|
| | | * a64f7051 |||| 2024-12-25T00:54:06-08:00 |||| Merge branch 'arnavg/nvshmem/nvls-multi-cta-collectives' into 'devel'
|
|
| | | |\
|
|
| | | | * d2452cb5 |||| 2024-12-25T00:54:06-08:00 |||| Multi-CTA OnStream AllReduce and ReduceScatter
|
|
| | | |/
|
|
| | | * 83b7fc47 |||| 2024-12-24T06:27:11-08:00 |||| Merge branch 'sethh/moe_example' into 'devel'
|
|
| | | |\
|
|
| | | | * 2ca9287d |||| 2024-12-24T06:27:11-08:00 |||| examples: Add a performance example for MoE topK alltoallv
|
|
| | | |/
|
|
| | | * 18d562ee |||| 2024-12-23T17:33:39-08:00 |||| Merge branch 'arnavg/nvshmem/nccl-nvshmem-interop-test' into 'devel'
|
|
| | | |\
|
|
| | | | * 621d646d |||| 2024-12-23T17:33:39-08:00 |||| Adding NCCL & NVSHMEM Interop test for XLA
|
|
| | | * | 6b71334e |||| 2024-12-23T17:27:54-08:00 |||| Merge branch 'alanger/pr/ring_reduce_example' into 'devel'
|
|
| | | |\ \
|
|
| | | | |/
|
|
| | | |/|
|
|
| | | | * 3cecfd54 |||| 2024-12-23T17:27:54-08:00 |||| Ring based allreduce example code
|
|
| | | |/
|
|
| | | * f380662f |||| 2024-12-23T14:51:26-08:00 |||| Merge branch 'sethh/3_2_minor_bug_fixes' into 'devel'
|
|
| | | |\
|
|
| | | | * fc3eec65 |||| 2024-12-23T14:51:26-08:00 |||| NVSHMEM Version 3.2 Minor Bug Fixes
|
|
| | | |/
|
|
| | | * 2ab3f6bc |||| 2024-12-20T14:02:27-08:00 |||| Merge branch 'arnavg/nvshmem/nvls-nccl-check-fix' into 'devel'
|
|
| | | |\
|
|
| | | | * fe46ec1b |||| 2024-12-20T14:02:27-08:00 |||| Fix algo selection check for NVLS collectives
|
|
| | | |/
|
|
| | | * 7676aec0 |||| 2024-12-18T16:17:50-08:00 |||| Merge branch 'arnavg/nvshmem/fix-rs-minmax-overflow' into 'devel'
|
|
| | | |\
|
|
| | | | * ee9c12e2 |||| 2024-12-18T16:17:50-08:00 |||| Address RS min/max overflow issue
|
|
| | | |/
|
|
| | | * f4a5d86b |||| 2024-12-12T01:29:16-08:00 |||| Merge branch 'arnavg/nvshmem/barrier-sync-kval' into 'devel'
|
|
| | | |\
|
|
| | | | * ca697c00 |||| 2024-12-12T01:29:16-08:00 |||| Update heuristics for on-stream collectives
|
|
| | | |/
|
|
| | | * 8be7ee63 |||| 2024-12-11T02:15:13-08:00 |||| Merge branch 'monah/bug_4978102' into 'devel'
|
|
| | | |\
|
|
| | | | * 392afafb |||| 2024-12-11T02:15:13-08:00 |||| Fix typo for QA_BIND_TO environment parameter
|
|
| | | |/
|
|
| | | * 9ea7e4ab |||| 2024-12-10T10:48:14-08:00 |||| Merge branch 'sesu/nvbugs/4972941' into 'devel'
|
|
| | | |\
|
|
| | | | * 0a7c6562 |||| 2024-12-10T10:48:14-08:00 |||| Add a patch for NVLS reduce algo, resolve the issue when platform/reduction...
|
|
| | | |/
|
|
| | | * efb13c56 |||| 2024-12-02T03:22:20-08:00 |||| Merge branch 'sesu/nvbugs/4977495' into 'devel'
|
|
| | | |\
|
|
| | | | * f67b7503 |||| 2024-12-02T03:22:19-08:00 |||| Create folder before create char dev
|
|
| | | |/
|
|
| | | * cb08ea95 |||| 2024-11-21T17:12:02-08:00 |||| Merge branch 'sethh/eos_nvls_code' into 'devel'
|
|
| | | |\
|
|
| | | | * c3b81dae |||| 2024-11-21T17:12:02-08:00 |||| CI: add NVLS job for EOS.
|
|
| | | |/
|
|
| | | * 50af94a1 |||| 2024-11-19T00:29:05-08:00 |||| Merge branch 'kpunniyamurt/va_alloc_p2p_gpus' into 'devel'
|
|
| | | |\
|
|
| | | | * 13c63b1f |||| 2024-11-19T00:29:04-08:00 |||| Dynamically size the UC and MC heap size based on runtime number of P2P connected GPUs
|
|
| | | |/
|
|
| | | * 92105fda |||| 2024-11-18T03:02:45-08:00 |||| Merge branch 'sesu/bind-to-socket-for-bug-4903146' into 'devel'
|
|
| | | |\
|
|
| | | | * f67910ae |||| 2024-10-23T14:20:38+08:00 |||| Support --bind-to numa/none/socket in command lines; Add atomic cases in single node perftest.
|
|
| |_|_|/
|
|
|/| | |
|
|
| | | * ee8bc322 |||| 2024-11-15T23:23:57-08:00 |||| Merge branch 'arnavg/nvshmem/nvls-allreduce-bf16' into 'devel'
|
|
| | | |\
|
|
| | | | * e7ad5d9d |||| 2024-11-15T23:23:57-08:00 |||| Support for NVLS Half-Precision AR
|
|
| | | |/
|
|
| | | * 9b5dccfd |||| 2024-11-15T20:19:25-08:00 |||| Merge branch 'arnavg/nvshmem/shmem-perftest-breakage' into 'devel'
|
|
| | | |\
|
|
| | | | * 68c525af |||| 2024-11-15T20:19:24-08:00 |||| Use print_table_basic for pt-to-pt bw test
|
|
| | | |/
|
|
| | | * e09ddc97 |||| 2024-11-14T07:36:18-08:00 |||| Merge branch 'pmarkthub/nvbugs/4949016' into 'devel'
|
|
| | | |\
|
|
| | | | * 312dcca3 |||| 2024-11-14T07:36:18-08:00 |||| [NVBUG 4949016][IBGDA & IBDEVX] Fix bug in ibgda_rc_init2rtr when using RoCE v1 and add support for RoCE v2
|
|
| | | |/
|
|
| | | * daf8e45e |||| 2024-11-14T06:52:50-08:00 |||| Merge branch 'arnavg/nvshmem/imex-scripts' into 'devel'
|
|
| | | |\
|
|
| | | | * 7cb65bdd |||| 2024-11-14T06:52:50-08:00 |||| Add support for imex channel scripts
|
|
| | | * | de23f7e8 |||| 2024-11-14T06:50:18-08:00 |||| Merge branch 'sesu/nvbugs/4958070' into 'devel'
|
|
| | | |\ \
|
|
| | | | |/
|
|
| | | |/|
|
|
| | | | * 8fe6ef41 |||| 2024-11-14T06:50:17-08:00 |||| Fix for bug 4958070, Ignore the mc_ptr test when PE=1.
|
|
| | | |/
|
|
| | | * 60389bd1 |||| 2024-11-13T21:58:07-08:00 |||| Merge branch 'arnavg/nvshmem/nvls-oneshot-ar' into 'devel'
|
|
| | | |\
|
|
| | | | * 41a2ec63 |||| 2024-11-13T21:58:06-08:00 |||| Add NVLS OneShot AR implementation
|
|
| | | |/
|
|
| | | * 313b3bfe |||| 2024-11-12T10:31:31-08:00 |||| Merge branch 'arnavg/nvshmem/sm100_test_perftest' into 'devel'
|
|
| | | |\
|
|
| | | | * 56eec8df |||| 2024-11-12T10:31:31-08:00 |||| Update SM100 target to tests
|
|
| | | |/
|
|
| | | * 6abdd081 |||| 2024-11-07T16:31:04-08:00 |||| Merge branch 'vkhodel/better_pkg' into 'devel'
|
|
| | | |\
|
|
| | | | * 3a4517b0 |||| 2024-11-07T16:31:04-08:00 |||| Use ccache to get 3x speedup on CI/CD build jobs
|
|
| | | |/
|
|
| | | * 764df11d |||| 2024-11-07T07:52:44-08:00 |||| Merge branch 'arnavg/nvshmem/blackwell-enablement' into 'devel'
|
|
| | |/|
|
|
| | | * 77882bad |||| 2024-11-07T07:52:44-08:00 |||| Enable Blackwell SM100
|
|
| | |/
|
|
| | * 19095ef6 |||| 2024-10-29T04:11:56-07:00 |||| Merge branch 'sethh/fix_mnnvl_allgather' into 'devel'
|
|
| |/|
|
|
| | * 37932283 |||| 2024-10-29T04:11:56-07:00 |||| fcollect: disable LL in hybrid MNNVL environments.
|
|
| |/
|
|
| * 9197203c |||| 2024-10-25T19:03:34-07:00 |||| Merge branch 'sethh/fcollect_mnnvl_fix' into 'devel'
|
|
| |\
|
|
| | * e907c06b |||| 2024-10-25T19:03:32-07:00 |||| NVSHMEM: merge code for 3.1.6 RC
|
|
| |/
|
|
| * 9cf24b83 |||| 2024-10-25T18:34:57-07:00 |||| Merge branch 'main' into 'devel'
|
|
| |\
|
|
| |/
|
|
|/|
|
|
* | f75ccd73 |||| 2024-10-19T07:35:57-07:00 |||| Merge branch 'release/3_1_5' into 'main'
|
|
|\ \
|
|
| * \ b6e097ec |||| 2024-10-18T18:03:20-07:00 |||| Merge branch 'main' into release/3_1_5
|
|
| |\ \
|
|
| |/ /
|
|
|/| |
|
|
* | | aaab9820 |||| 2024-10-10T01:41:25-07:00 |||| Merge branch 'release/3_1_4' into 'main'
|
|
|\ \ \
|
|
| * | | b025b683 |||| 2024-10-10T01:41:25-07:00 |||| NVSHMEM 3.1.4 Release Candidate
|
|
|/ / /
|
|
* | | 7f152031 |||| 2024-10-05T11:28:05-07:00 |||| Merge branch 'release/3_1_1' into 'main'
|
|
|\ \ \
|
|
| * \ \ a00d31e9 |||| 2024-10-05T07:20:30-07:00 |||| Merge branch 'main' into release/3_1_1
|
|
| |\ \ \
|
|
| |/ / /
|
|
|/| | |
|
|
* | | | 6720e7b1 |||| 2024-10-02T17:45:08-07:00 |||| Merge branch 'release/3_1_0' into 'main'
|
|
|\ \ \ \
|
|
| * | | | 306498dc |||| 2024-10-02T17:45:07-07:00 |||| NVSHMEM: 3.1.0-0
|
|
|/ / / /
|
|
* | | | f9b69fb6 |||| 2024-06-18T18:10:08-07:00 |||| Merge branch 'sethh/v3_0_6_4' into 'main'
|
|
|\ \ \ \
|
|
| * | | | 3367a3ec |||| 2024-06-18T18:06:51-07:00 |||| version: Update to 3.0.6-4 for merge to main.
|
|
| | * | | 4848875c |||| 2024-10-05T07:10:01-07:00 |||| NVSHMEM 3.1.1 Release Candidate
|
|
| | | | * be845f4d |||| 2024-10-22T11:59:25-07:00 |||| Merge branch 'vkhodel/better_pkg' into 'devel'
|
|
| | | |/|
|
|
| | | | * 83356cb0 |||| 2024-10-22T11:59:25-07:00 |||| propagate errors from make to gitlab; fail if there is a container version...
|
|
| | | |/
|
|
| | | * 7ed0c555 |||| 2024-10-18T17:59:21-07:00 |||| Merge branch 'sethh/reduce_perf_fix' into 'devel'
|
|
| | | |\
|
|
| | | | * a47ca8f8 |||| 2024-10-18T17:59:21-07:00 |||| NVSHMEM Bug Fixes from 3.1.4 RC
|
|
| | | |/
|
|
| | | * bea63f72 |||| 2024-10-15T11:46:12-07:00 |||| Merge branch 'kartiki/ibgda-sysmem-heap-ci-on-aws' into 'devel'
|
|
| | | |\
|
|
| | | | * 4aa86920 |||| 2024-10-15T11:46:12-07:00 |||| Run IBGDA sysmem heap tests on AWS P5 cluster during CI
|
|
| | | |/
|
|
| | | * 1fb76150 |||| 2024-10-13T09:50:23-07:00 |||| Merge branch 'arnavg/nvshmem/clang-format-token' into 'devel'
|
|
| | | |\
|
|
| | | | * 9465e6f6 |||| 2024-10-13T09:50:22-07:00 |||| Update auto formatting clang-token
|
|
| | | |/
|
|
| | | * a81a76fa |||| 2024-10-11T17:46:18-07:00 |||| Merge branch 'vkhodel/docker' into 'devel'
|
|
| | | |\
|
|
| | | | * a4c67d1e |||| 2024-10-11T17:46:17-07:00 |||| minor changes to building and package scripts
|
|
| | | |/
|
|
| | | * 501a4a3d |||| 2024-10-10T01:36:33-07:00 |||| Merge branch 'sethh/fix_regcount_regression' into 'devel'
|
|
| | | |\
|
|
| | | | * 9d960290 |||| 2024-10-10T01:36:32-07:00 |||| NVSHMEM 3.1 Bug Fixes for Release Candidate
|
|
| | | |/
|
|
| | | * fcd34a53 |||| 2024-10-09T08:28:13-07:00 |||| Merge branch 'arnavg/nvshmem/nvbug4786955' into 'devel'
|
|
| | | |\
|
|
| | | | * 039dca3c |||| 2024-10-09T08:28:12-07:00 |||| Fix for NVSHMEM 3+ - nvmlDeviceGetGpuFabricInfoV segfaults on Eos with CUDA 12.5 and CUDA compatibility
|
|
| | | |/
|
|
| | | * c3f9b441 |||| 2024-10-08T09:18:24-07:00 |||| Merge branch 'arnavg/nvshmem/nvls-allreduce-pe-calc' into 'devel'
|
|
| | |/|
|
|
| | | * deffeebc |||| 2024-10-08T09:18:23-07:00 |||| Fix for [NVSHMEM 3.1][3.1.1-1] [NVLS feature] allreduce cases failed on single Viking H20 node.
|
|
| | |/
|
|
| | * 17c29f1a |||| 2024-10-05T07:07:49-07:00 |||| Merge branch 'arnavg/nvshmem/nvbug4891787' into 'devel'
|
|
| | |\
|
|
| | | * 278c5602 |||| 2024-10-05T07:07:48-07:00 |||| NVSHMEM: Fix bugs in 3.1.0 Release Candidate
|
|
| | |/
|
|
| | * 217814cc |||| 2024-10-04T17:59:05-07:00 |||| Merge branch 'release/merge_back_to_devel' into 'devel'
|
|
| | |\
|
|
| | | * fa74643d |||| 2024-10-04T17:59:05-07:00 |||| Merge back to devel after 3.1.0.0
|
|
| | |/
|
|
| | * c14419c2 |||| 2024-10-02T16:10:21-07:00 |||| Merge branch 'alanger/pr/add_half_precision_support' into 'devel'
|
|
| | |\
|
|
| | | * 53cd70a4 |||| 2024-10-02T16:10:21-07:00 |||| Add collective APIs for half and bfloat16 datatypes
|
|
| | * | 56628ccb |||| 2024-10-02T13:16:11-07:00 |||| Merge branch 'arnavg/nvshmem/nvls-default-selection' into 'devel'
|
|
| | |\ \
|
|
| | | * | 4efc0983 |||| 2024-10-02T13:16:10-07:00 |||| Tune default algo selection strategy for collectives
|
|
| | |/ /
|
|
| | * | 8fdf46a8 |||| 2024-10-02T07:13:22-07:00 |||| Merge branch 'sethh/LL128_implementation' into 'devel'
|
|
| | |\ \
|
|
| | | |/
|
|
| | |/|
|
|
| | | * 3e3c6347 |||| 2024-10-02T07:13:22-07:00 |||| FCollect LL8 optimizations and LL128 implementation
|
|
| | |/
|
|
| | * d69a86ea |||| 2024-09-30T17:15:17-07:00 |||| Merge branch 'sethh/CI_ninja_build_enable' into 'devel'
|
|
| | |\
|
|
| | | * 36e6357d |||| 2024-09-30T17:15:17-07:00 |||| CMake: fix regression in Ninja compilation.
|
|
| | |/
|
|
| | * 76e549e1 |||| 2024-09-30T15:17:08-07:00 |||| Merge branch 'alanger/pr/fix_shmem_get_bw_reporting' into 'devel'
|
|
| | |\
|
|
| | | * c9f60979 |||| 2024-09-30T15:17:08-07:00 |||| Fix shmem_get_bw reporting
|
|
| | |/
|
|
| | * f3fbaabd |||| 2024-09-30T11:34:39-07:00 |||| Merge branch 'alanger/pr/increase_reduce_scratch_size' into 'devel'
|
|
| | |\
|
|
| | | * 698d0797 |||| 2024-09-30T07:45:02-07:00 |||| Increase REDUCE SCRATCH SIZE
|
|
| | * | 9597ce6c |||| 2024-09-30T08:29:03-07:00 |||| Merge branch 'pmarkthub/gpu_verbs/dmabuf-for-control-bufs' into 'devel'
|
|
| | |\ \
|
|
| | | |/
|
|
| | |/|
|
|
| | | * ae11d71c |||| 2024-09-30T08:29:03-07:00 |||| [IBGDA] Implement DMABUF support for control buffers
|
|
| | |/
|
|
| | * 893b77cd |||| 2024-09-27T23:01:27-07:00 |||| Merge branch 'pmarkthub/gpu_verbs/a2a-with-signal-set' into 'devel'
|
|
| | |\
|
|
| | | * 33157a71 |||| 2024-09-27T23:01:27-07:00 |||| [IBGDA & A2A] Change to A2A implementation to use SIGNAL_SET when using P2P or IBGDA
|
|
| | |/
|
|
| | * 8d365154 |||| 2024-09-27T21:11:16-07:00 |||| Merge branch 'alanger/pr/dynamic_gid' into 'devel'
|
|
| | |\
|
|
| | | * 9d606c25 |||| 2024-09-27T21:11:15-07:00 |||| Dynamic GID detection
|
|
| | |/
|
|
| | * efd75b57 |||| 2024-09-27T15:23:35-07:00 |||| Merge branch 'sethh/kitmaker_fixes' into 'devel'
|
|
| | |\
|
|
| | | * 565e84e6 |||| 2024-09-27T15:23:34-07:00 |||| CMake: Prepare NVSHMEM Packages for external packaging.
|
|
| | |/
|
|
| | * 08b615b2 |||| 2024-09-25T10:29:33-07:00 |||| Merge branch 'arnavg/nvshmem/fcollect_multi_cta' into 'devel'
|
|
| | |\
|
|
| | | * 2d2adc32 |||| 2024-09-25T10:29:33-07:00 |||| NVLS Allgather Multi-CTA Support for DIRECT
|
|
| | |/
|
|
| | * 3119db98 |||| 2024-09-24T20:01:41-07:00 |||| Merge branch 'arnavg/nvshmem/nvls-allreduce' into 'devel'
|
|
| | |\
|
|
| | | * 949e5eec |||| 2024-09-24T20:01:41-07:00 |||| NVLS AllReduce TwoShot support
|
|
| | |/
|
|
| | * 30e5b4f8 |||| 2024-09-23T16:12:25-07:00 |||| Merge branch 'sethh/devicelib_env_rename' into 'devel'
|
|
| | |\
|
|
| | | * 8c670ffd |||| 2024-09-23T16:12:24-07:00 |||| CMake: Rename the devicelib cuda home variable
|
|
| | |/
|
|
| | * bf015085 |||| 2024-09-23T12:24:07-07:00 |||| Merge branch 'sethh/add_cmake_binaries' into 'devel'
|
|
| | |\
|
|
| | | * 550007b4 |||| 2024-09-23T12:24:07-07:00 |||| CMake: Prepare packages for 3.1
|
|
| | |/
|
|
| | * 8695ac11 |||| 2024-09-16T20:18:38-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
| | |\
|
|
| | | * 35dc6efb |||| 2024-09-16T20:18:38-07:00 |||| [coverity]: cid: 22649617,22649625,22649627
|
|
| | |/
|
|
| | * 29f96e58 |||| 2024-09-10T17:20:48-07:00 |||| Merge branch 'alanger/pr/fix_team_split_strided' into 'devel'
|
|
| | |\
|
|
| | | * 33558d66 |||| 2024-09-10T17:20:48-07:00 |||| Fix stride calculation in team split strided
|
|
| | |/
|
|
| | * 3548b1ac |||| 2024-09-06T18:13:28-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
| | |\
|
|
| | | * 7aa79dfa |||| 2024-09-06T18:13:28-07:00 |||| [coverity] CID: minor fix for CID 22649630 22650451 22650450
|
|
| | |/
|
|
| | * a84a5400 |||| 2024-09-05T17:18:12-07:00 |||| Merge branch 'arnavg/nvshmem/nvbug4655341' into 'devel'
|
|
| | |\
|
|
| | | * 91fcf39e |||| 2024-09-05T17:18:12-07:00 |||| [nvbug: 4655341] Fix build issue for DVS
|
|
| | |/
|
|
| | * f9f67ba7 |||| 2024-08-30T15:47:41-07:00 |||| Merge branch 'vectorize_packLL' into 'devel'
|
|
| | |\
|
|
| | | * 56cdd816 |||| 2024-08-30T15:47:41-07:00 |||| Vectorize LL8 PackLL
|
|
| | * | 14eb70a3 |||| 2024-08-30T13:24:09-07:00 |||| Merge branch 'ridgunji/mem_heap_test_development' into 'devel'
|
|
| | |\ \
|
|
| | | |/
|
|
| | |/|
|
|
| | | * 13b1fbbd |||| 2024-08-30T13:24:09-07:00 |||| Development of the NVSHMEM symmetric heap class tests
|
|
| | |/
|
|
| | * 90276aa8 |||| 2024-08-30T09:11:32-07:00 |||| Merge branch 'arnavg/nvshmem/nvls-rs-support' into 'devel'
|
|
| | |\
|
|
| | | * 73d74a0e |||| 2024-08-30T09:11:32-07:00 |||| NVLS ReduceScatter Support for internal teams
|
|
| | |/
|
|
| | * 3fc23cad |||| 2024-08-29T17:17:02-07:00 |||| Merge branch 'abaran/function_call_fix' into 'devel'
|
|
| | |\
|
|
| | | * a8af239a |||| 2024-08-29T17:17:02-07:00 |||| Ensure get psync len function uses get fcollect psync len function
|
|
| | |/
|
|
| | * 5a4186e9 |||| 2024-08-26T18:30:23-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
| | |\
|
|
| | | * fed7da5f |||| 2024-08-26T18:30:23-07:00 |||| [coverity] Logically dead code
|
|
| | |/
|
|
| | * 7663c0c8 |||| 2024-08-19T17:37:30-07:00 |||| Merge branch 'arnavg/nvshmem/uid-bootstrap-examples' into 'devel'
|
|
| | |\
|
|
| | | * 7991b526 |||| 2024-08-19T17:37:30-07:00 |||| Add UID bootstrap example
|
|
| | |/
|
|
| | * e2b23ccf |||| 2024-08-19T16:39:04-07:00 |||| Merge branch 'arnavg/nvshmem/perftest-bugs' into 'devel'
|
|
| | |\
|
|
| | | * f1f9bc69 |||| 2024-08-19T16:39:04-07:00 |||| OOB access in atomic_bw
|
|
| | |/
|
|
| | * 1104593c |||| 2024-08-15T12:08:38-07:00 |||| Merge branch 'arnavg/nvshmem/cuda-compat-issue' into 'devel'
|
|
| | |\
|
|
| | | * 360da707 |||| 2024-08-15T12:08:37-07:00 |||| WAR for CUDA 12.5 compat with MNNVL
|
|
| | |/
|
|
| | * 14c607ed |||| 2024-08-13T03:52:39-07:00 |||| Merge branch 'arnavg/nvshmem/nvls-mc-ptr' into 'devel'
|
|
| | |\
|
|
| | | * 06768a01 |||| 2024-08-13T03:52:38-07:00 |||| NVLS Multicast Heap API Support
|
|
| | |/
|
|
| | * 2f857b9c |||| 2024-08-12T18:05:00-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
| | |\
|
|
| | | * 013edf35 |||| 2024-08-12T18:04:59-07:00 |||| [Coverity] 22649610 22649594 Uninitialized scalar variable
|
|
| | * | 7a05d961 |||| 2024-08-12T13:38:05-07:00 |||| Merge branch 'alanger/pr/add_rs_api' into 'devel'
|
|
| | |\ \
|
|
| | | |/
|
|
| | |/|
|
|
| | | * 88766216 |||| 2024-08-12T13:38:04-07:00 |||| Add reducescatter collective API
|
|
| | |/
|
|
| | * 811ce7a3 |||| 2024-08-08T11:54:51-07:00 |||| Merge branch 'arnavg/nvshmem/nvls-support-ll-algo' into 'devel'
|
|
| | |\
|
|
| | | * b99bddf2 |||| 2024-08-08T11:54:51-07:00 |||| NVLS LL algo support for allgather onstream and device API
|
|
| | |/
|
|
| | * ff1f4058 |||| 2024-08-08T10:50:15-07:00 |||| Merge branch 'ridgunji/mem_unit_tests' into 'devel'
|
|
| | |\
|
|
| | | * 3fa0ad82 |||| 2024-08-08T10:50:15-07:00 |||| Added first unit test of mem_p2p_transport
|
|
| | |/
|
|
| | * f19ad92a |||| 2024-08-08T07:50:35-07:00 |||| Merge branch 'alanger/pr/revert_to_3.0_version' into 'devel'
|
|
| | |\
|
|
| | | * 417fa3aa |||| 2024-08-07T15:01:35-07:00 |||| clang format
|
|
| | | * 7adff972 |||| 2024-08-07T15:00:06-07:00 |||| Revert minor version; Set to 3.0.15.0
|
|
| | |/
|
|
| | * 8729ad70 |||| 2024-08-06T16:53:37-07:00 |||| Merge branch 'alanger/pr/use_custreamwaitvalue_when_bypassflush_set' into 'devel'
|
|
| | |\
|
|
| | | * 24169d66 |||| 2024-07-17T14:13:50-07:00 |||| use cuStreamWaitValue when BYPASS_FLUSH is set
|
|
| | |/
|
|
| | * 806cbff1 |||| 2024-08-06T16:35:48-07:00 |||| Merge branch 'alanger/pr/fix_custreamwrite_usage' into 'devel'
|
|
| | |\
|
|
| | | * e7b2df97 |||| 2024-07-17T14:15:28-07:00 |||| Update version to 3.1.5
|
|
| | | * 2426d962 |||| 2024-07-12T21:21:57-07:00 |||| Use cuStreamWriteValue only when job connectivity is LDST_ATOMICS
|
|
| | |/
|
|
| | * 8bf506a4 |||| 2024-08-02T17:37:14-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
| | |\
|
|
| | | * d0113f98 |||| 2024-08-02T17:37:14-07:00 |||| Coverity: 22649622 Resource leak
|
|
| | |/
|
|
| | * c3111f62 |||| 2024-07-31T10:09:19-07:00 |||| Merge branch 'abaran/perftest_bug_fix' into 'devel'
|
|
| | |\
|
|
| | | * 8fb7d54e |||| 2024-07-31T10:09:18-07:00 |||| Fix bug 4719294
|
|
| | |/
|
|
| | * 5455c98d |||| 2024-07-30T10:29:55-07:00 |||| Merge branch 'arnavg/nvshmem/nvbug4774872-2' into 'devel'
|
|
| | |\
|
|
| | | * 8a1c7cc5 |||| 2024-07-30T10:29:55-07:00 |||| Fix nvbug 4774872
|
|
| | |/
|
|
| | * 350b7431 |||| 2024-07-29T11:29:49-07:00 |||| Merge branch 'sethh/Fix_Cuda_symbols' into 'devel'
|
|
| | |\
|
|
| | | * bc1dcfd5 |||| 2024-07-29T11:29:49-07:00 |||| nvmlwrap: Fix cuda constant values.
|
|
| | |/
|
|
| | * 44ad5336 |||| 2024-07-29T11:27:29-07:00 |||| Merge branch 'arnavg/nvshmem/remote-perftest-ib-list' into 'devel'
|
|
| | |\
|
|
| | | * 5efaec26 |||| 2024-07-29T11:27:29-07:00 |||| Remove host/coll variants from perftest IB list
|
|
| | |/
|
|
| | * a03a73eb |||| 2024-07-26T10:57:36-07:00 |||| Merge branch 'arnavg/nvshmem/nvls-support-device-state' into 'devel'
|
|
| | |\
|
|
| | | * dd049429 |||| 2024-07-26T10:57:35-07:00 |||| NVLINK SHARP support for internal teams
|
|
| | |/
|
|
| | * 1ca262e3 |||| 2024-07-25T23:05:26-07:00 |||| Merge branch 'arnavg/nvshmem/enable_sm100' into 'devel'
|
|
| | |\
|
|
| | | * a32a64c1 |||| 2024-07-25T23:05:25-07:00 |||| Enable Blackwell build and improve CI & docs
|
|
| | |/
|
|
| | * c9e6d98b |||| 2024-07-24T23:42:50-07:00 |||| Merge branch 'ridgunji/nvshmem/fix_vmm_segfault' into 'devel'
|
|
| | |\
|
|
| | | * 9bdc6c2e |||| 2024-07-24T23:42:50-07:00 |||| Add missing remote transport initialization for segfault issue
|
|
| | |/
|
|
| | * 36f79014 |||| 2024-07-22T18:23:41-07:00 |||| Merge branch 'ridgunji/mem_transport_minorFixes' into 'devel'
|
|
| | |\
|
|
| | | * b75562b1 |||| 2024-07-22T18:23:41-07:00 |||| Made minor fixes to the mem_transport.cpp and .hpp files
|
|
| | |/
|
|
| | * 5ba20e7f |||| 2024-07-22T11:25:43-07:00 |||| Merge branch 'arnavg/nvshmem/remove-perftest-host' into 'devel'
|
|
| | |\
|
|
| | | * 228fa78b |||| 2024-07-22T11:25:43-07:00 |||| Deprecate host-side coll perftest
|
|
| | |/
|
|
| | * c920aefc |||| 2024-07-18T10:46:54-07:00 |||| Merge branch 'arnavg/nvshmem/reduction_bug' into 'devel'
|
|
| | |\
|
|
| | | * e71f7b84 |||| 2024-07-18T10:46:54-07:00 |||| Fix nvbug for cuda invalid resource handle for reduction
|
|
| | |/
|
|
| | * dd0a6c8c |||| 2024-07-15T17:28:01-07:00 |||| Merge branch 'arnavg/nvshmem/host-coll-bug' into 'devel'
|
|
| | |\
|
|
| | | * eb2f58ab |||| 2024-07-15T17:28:01-07:00 |||| Address bug in perftest/host/coll
|
|
| | |/
|
|
| | * d72718ed |||| 2024-07-12T14:58:45-07:00 |||| Merge branch 'arnavg/nvshmem/perftest-r2r-debug' into 'devel'
|
|
| | |\
|
|
| | | * 86497ec1 |||| 2024-07-12T14:58:45-07:00 |||| Remove memcpyAsync to minimize r2r variations
|
|
| | |/
|
|
| | * 18f37eda |||| 2024-07-09T13:05:05-07:00 |||| Merge branch 'sethh/symmetric_mem_refactor' into 'devel'
|
|
| | |\
|
|
| | | * 53c05e0f |||| 2024-07-03T12:53:38-07:00 |||| symmetric_heap: move functions called in class to protected.
|
|
| | * | b7854869 |||| 2024-07-09T13:04:40-07:00 |||| Merge branch 'sethh/env_disambiguation' into 'devel'
|
|
| | |\ \
|
|
| | | |/
|
|
| | |/|
|
|
| | | * 66ae3c24 |||| 2024-07-05T13:06:34-07:00 |||| in-place clang-formatting
|
|
| | | * 6b0cae05 |||| 2024-07-05T11:56:06-07:00 |||| env: Clarify the bootstrap plugin environment vars
|
|
| | * | 80ec6914 |||| 2024-07-05T23:56:35-07:00 |||| Merge branch 'alanger/pr/use_custreamwaitvalue' into 'devel'
|
|
| | |\ \
|
|
| | | |/
|
|
| | |/|
|
|
| | | * 14a36c27 |||| 2024-07-05T23:56:35-07:00 |||| Use cuStreamWriteValue64 instead of signal_op_on_stream in put_signal_on_stream API
|
|
| | |/
|
|
| | * 72e24da1 |||| 2024-07-02T13:02:24-07:00 |||| Merge branch 'sethh/CI_fix_timeouts' into 'devel'
|
|
| | |\
|
|
| | | * 2b9f4aaa |||| 2024-07-02T13:02:24-07:00 |||| CI: add timeouts to jobs missing them.
|
|
| | |/
|
|
| | * 73c12740 |||| 2024-06-28T15:56:42-07:00 |||| Merge branch 'arnavg/nvshmem/versioned-structs' into 'devel'
|
|
| | |\
|
|
| | | * c619b00d |||| 2024-06-28T15:56:42-07:00 |||| Fix missing initialization in gpu_coll_env_params_var
|
|
| | |/
|
|
| | * 5461c5e1 |||| 2024-06-28T14:25:57-07:00 |||| Merge branch 'alanger/pr/fix_timeout_initializer' into 'devel'
|
|
| | |\
|
|
| | | * 5200481a |||| 2024-06-25T17:22:25-07:00 |||| Update version to 3.0.8.0
|
|
| | | * 06fc6e5d |||| 2024-06-25T17:19:52-07:00 |||| Use correct init value for timeout
|
|
| | |/
|
|
| | * a1c97275 |||| 2024-06-28T14:09:25-07:00 |||| Merge branch 'alanger/pr/fix_sysmem_ci' into 'devel'
|
|
| | |\
|
|
| | | * 58b8f196 |||| 2024-06-28T14:07:42-07:00 |||| Update version to 3.0.7.2
|
|
| | | * b467dbb1 |||| 2024-06-18T15:34:33-07:00 |||| Fix sysmem CI skip test list
|
|
| | |/
|
|
| | * 4a281dd7 |||| 2024-06-26T14:48:51-07:00 |||| Merge branch 'abaran/nvshmem/perf_metrics' into 'devel'
|
|
| | |\
|
|
| | | * 6a63a200 |||| 2024-06-26T14:48:51-07:00 |||| improve readability and added metrics to perftest output
|
|
| | |/
|
|
| | * ec550184 |||| 2024-06-24T15:58:40-07:00 |||| Merge branch 'arnavg/nvshmem/fcollect-algo' into 'devel'
|
|
| |/|
|
|
| | * 50b73f60 |||| 2024-06-24T15:58:40-07:00 |||| Add algo selection for fcollect
|
|
| |/
|
|
| * 66fd88cb |||| 2024-06-18T18:03:01-07:00 |||| Merge branch 'arnavg/nvshmem/missing-cudawrap-11.0' into 'devel'
|
|
| |\
|
|
| | * 09de3888 |||| 2024-06-18T17:59:12-07:00 |||| version: bump for merge back to devel.
|
|
| | * 7f3fc350 |||| 2024-06-18T17:57:50-07:00 |||| Merge branch 'main' into arnavg/nvshmem/missing-cudawrap-11.0
|
|
| | |\
|
|
| |_|/
|
|
|/| |
|
|
* | | 31641700 |||| 2024-06-18T11:52:31-07:00 |||| Merge branch 'sethh/nvshmem_3_0_6_2' into 'main'
|
|
|\ \ \
|
|
| * | | cb5e81e5 |||| 2024-06-17T16:51:10-07:00 |||| version: Update to 3.0.6-2
|
|
| |/ /
|
|
| | * 0d5870d9 |||| 2024-06-18T17:47:25-07:00 |||| Bump up
|
|
| | * 7a0362cf |||| 2024-06-18T17:45:52-07:00 |||| Add missing include
|
|
| |/
|
|
| * eed0f38d |||| 2024-06-18T07:35:06-07:00 |||| Merge branch 'alanger/pr/fix_mnnvl_discovery_2' into 'devel'
|
|
| |\
|
|
| | * 4db06157 |||| 2024-06-17T18:55:59-07:00 |||| in-place clang-formatting
|
|
| | * f7db57e0 |||| 2024-06-17T18:52:42-07:00 |||| Fix typo
|
|
| | * fc3c8af4 |||| 2024-06-17T17:56:28-07:00 |||| Fix typo
|
|
| | * 930f934a |||| 2024-06-17T17:34:22-07:00 |||| Check if p2p list is of form (start, stride, size)
|
|
| |/
|
|
| * 1e2775b5 |||| 2024-06-17T16:46:34-07:00 |||| Merge branch 'alanger/pr/fix_mnnvl_discovery_2' into 'devel'
|
|
| |\
|
|
| | * 893a74bb |||| 2024-06-17T11:28:33-07:00 |||| NVSHMEM: Make MNNVL failures non-fatal.
|
|
| | * 57bb8b93 |||| 2024-06-17T15:18:22-07:00 |||| Fix MNNVL discovery code
|
|
| |/
|
|
| * 888a729e |||| 2024-06-17T14:10:57-07:00 |||| Merge branch 'pmarkthub/gpu_verbs/fix-roce-bug' into 'devel'
|
|
| |\
|
|
| | * 3924b1dc |||| 2024-06-17T09:22:26-07:00 |||| version: Update version for bug fix.
|
|
| | * 0a3b1178 |||| 2024-06-17T09:21:35-07:00 |||| Merge branch 'main' into pmarkthub/gpu_verbs/fix-roce-bug
|
|
| | |\
|
|
| |_|/
|
|
|/| |
|
|
* | | e710567d |||| 2024-06-13T21:32:55-07:00 |||| Merge branch 'sethh/3_0_respin_merge_main' into 'main'
|
|
|\ \ \
|
|
| * \ \ a103f4a1 |||| 2024-06-13T15:39:53-07:00 |||| Merge branch 'main' into sethh/3_0_respin_merge_main
|
|
| |\ \ \
|
|
| |/ / /
|
|
|/| | |
|
|
* | | | 8269a094 |||| 2024-06-04T16:22:59-07:00 |||| Merge branch 'alanger/pr/update_main_with_3_0_4' into 'main'
|
|
|\ \ \ \
|
|
| * \ \ \ d96d1b62 |||| 2024-06-04T16:21:20-07:00 |||| Merge branch 'main' into alanger/pr/update_main_with_3_0_4
|
|
| |\ \ \ \
|
|
| |/ / / /
|
|
|/| | | |
|
|
* | | | | 6dae0eba |||| 2024-06-01T17:37:04-07:00 |||| Merge branch '3_0_rc1' into 'main'
|
|
* | | | | 8851f6c2 |||| 2024-06-01T11:56:22-07:00 |||| version: Update package version formerge to main
|
|
| * | | | ef146a96 |||| 2024-06-04T14:37:15-07:00 |||| Update version to 3.0.4.1
|
|
| | * | | bebc18a5 |||| 2024-06-13T15:34:58-07:00 |||| version: Update tweak version for merge to main.
|
|
| | |/ /
|
|
| | | * 4c7bdbbb |||| 2024-06-16T23:34:21-07:00 |||| Fix an IBGDA bug showing up when using RoCE because the send_cq and ah objects are inside union
|
|
| | |/
|
|
| | * 6af79152 |||| 2024-06-13T15:13:42-07:00 |||| Merge branch 'sethh/3_0_respin' into 'devel'
|
|
| |/|
|
|
| | * bcc12304 |||| 2024-06-13T15:13:42-07:00 |||| Merge Request for all 3.0.5 bug fixes.
|
|
| |/
|
|
| * e141ffc9 |||| 2024-06-04T09:35:20-07:00 |||| Merge branch 'alanger/pr/team_shared_fix' into 'devel'
|
|
|/|
|
|
| * 61171fc6 |||| 2024-06-04T09:35:20-07:00 |||| Fix bug during team creation
|
|
|/
|
|
* 9c2208bf |||| 2024-06-01T11:54:06-07:00 |||| Merge branch 'sethh/test_new_config' into 'devel'
|
|
|\
|
|
| * 06617d40 |||| 2024-06-01T11:36:43-07:00 |||| CI: Fix config for merge to main
|
|
|/
|
|
* 38c852a7 |||| 2024-06-01T08:22:26-07:00 |||| Merge branch 'alanger/pr/optimize_small_msg_rdxn' into 'devel'
|
|
|\
|
|
| * 27cc23f7 |||| 2024-06-01T05:25:24-07:00 |||| libfabric: Add missing header
|
|
| * 57146112 |||| 2024-06-01T02:33:26-07:00 |||| include/device: remove uses of cuda.h from dev headers
|
|
| * 496a3bd5 |||| 2024-05-31T16:36:15-07:00 |||| Add cuda.h to reduce.cuh
|
|
| * 497f879d |||| 2024-05-31T16:27:49-07:00 |||| UPdate version to 3.0.3.0
|
|
| * 3126ce7b |||| 2024-05-31T15:20:10-07:00 |||| fix build-time error
|
|
| * 94a35361 |||| 2024-03-18T16:53:01-07:00 |||| in-place clang-formatting
|
|
| * a85da2c7 |||| 2024-05-31T14:26:23-07:00 |||| Guard constexpr usage with c++17 macro
|
|
| * 5c6cac1c |||| 2024-03-18T16:50:02-07:00 |||| Use cooperative groups for internode reduce
|
|
| * 7a2db71c |||| 2024-03-18T16:22:01-07:00 |||| Do parallel sends/recvs in small msg reduce block scoped
|
|
|/
|
|
* b3d38963 |||| 2024-06-01T05:18:59-07:00 |||| Merge branch 'alanger/pr/oberon_nw_support' into 'devel'
|
|
|\
|
|
| * 1388d173 |||| 2024-06-01T02:49:02-07:00 |||| Merge branch 'devel' into 'alanger/pr/oberon_nw_support'
|
|
| |\
|
|
| |/
|
|
|/|
|
|
* | 69b26c26 |||| 2024-06-01T02:36:22-07:00 |||| Merge branch 'abaran/team_max_fix' into 'devel'
|
|
|\ \
|
|
| * | e3df35ea |||| 2024-05-31T14:22:13-07:00 |||| updated project version to 3.0.1
|
|
| * | 4dad2ecb |||| 2024-05-31T14:17:33-07:00 |||| Merge branch 'devel' into abaran/team_max_fix
|
|
| |\ \
|
|
| |/ /
|
|
|/| |
|
|
| * | 8fcb42d3 |||| 2024-05-31T09:10:11-07:00 |||| changed log to NVSHMEM_INIT
|
|
| * | 8cf97c18 |||| 2024-05-30T13:49:18-07:00 |||| manual clang format fix
|
|
| * | d419cfcc |||| 2024-05-30T11:21:59-07:00 |||| collapsed prints to one statement
|
|
| * | 841d02a5 |||| 2024-05-30T09:45:02-07:00 |||| print heap size attr
|
|
| * | b7bb52ee |||| 2024-05-30T09:23:17-07:00 |||| changed print level
|
|
| * | 94ff8a7f |||| 2024-05-29T19:15:18-07:00 |||| debug statements and team max fix
|
|
| | * f4adc761 |||| 2024-06-01T02:16:40-07:00 |||| team_internal.cpp: Fix uninitialized variable list.
|
|
| | * d7a66619 |||| 2024-05-31T15:50:52-07:00 |||| in-place clang-formatting
|
|
| | * 6211de66 |||| 2024-05-31T15:45:59-07:00 |||| Fix rebase error
|
|
| | * 62b6acfd |||| 2024-05-31T15:41:54-07:00 |||| Update version to 3.0.2.0
|
|
| | * e9909bbb |||| 2024-05-29T19:20:31-07:00 |||| fix rebase issues
|
|
| | * d16a7307 |||| 2024-04-22T16:19:43-07:00 |||| Properly compose TEAM_SHARED
|
|
| | * a7584d60 |||| 2024-03-26T20:19:40-07:00 |||| Add Oberon Networking Support
|
|
| |/
|
|
|/|
|
|
* | af1b1459 |||| 2024-05-31T13:57:39-07:00 |||| Merge branch 'abi_compat_changes_uber_branch' into 'devel'
|
|
|\ \
|
|
| * | 058d503e |||| 2024-05-31T03:58:17-07:00 |||| ibgda: Add padding to state struct.
|
|
| * | b374ee6e |||| 2024-05-31T02:56:26-07:00 |||| Merge branch 'pmarkthub/gpu_verbs/async-postsend-cpu' into abi_compat_changes_uber_branch
|
|
| |\ \
|
|
| | * | 1b0ead25 |||| 2024-05-30T01:00:09-07:00 |||| Change the expected size of struct nvshmemi_ibgda_device_qp_v1 to 184
|
|
| | * | 199f3e1c |||| 2024-05-30T00:53:29-07:00 |||| Merge branch 'devel' into pmarkthub/gpu_verbs/async-postsend-cpu
|
|
| | |\|
|
|
| | * | bafc36f5 |||| 2024-05-23T19:01:44-07:00 |||| in-place clang-formatting
|
|
| | * | 8531a21b |||| 2024-05-23T18:58:07-07:00 |||| Change the error report regarding GPU UAR mapping failure to warning
|
|
| | * | 90cda940 |||| 2024-05-22T19:16:30-07:00 |||| in-place clang-formatting
|
|
| | * | 45b368d9 |||| 2024-05-22T19:12:28-07:00 |||| Fix compile errors
|
|
| | * | 90954dec |||| 2024-05-22T19:01:24-07:00 |||| Merge branch 'devel' into pmarkthub/gpu_verbs/async-postsend-cpu
|
|
| | |\ \
|
|
| | * | | af764de6 |||| 2024-05-19T22:31:14-07:00 |||| in-place clang-formatting
|
|
| | * | | 5f706be3 |||| 2024-05-19T22:27:55-07:00 |||| Rename NIC_POSTER_LOC to NIC_HANDLER
|
|
| | * | | ad6c398e |||| 2024-02-14T00:28:20-08:00 |||| in-place clang-formatting
|
|
| | * | | 027d6a04 |||| 2024-02-14T00:25:18-08:00 |||| Fix bug when PeerMappingOverride regkey is not set
|
|
| | * | | b0d7ed33 |||| 2024-02-13T20:27:22-08:00 |||| in-place clang-formatting
|
|
| | * | | 1f8fcefd |||| 2024-02-13T06:18:00-08:00 |||| in-place clang-formatting
|
|
| | * | | 805c9382 |||| 2024-02-13T20:23:53-08:00 |||| Fix bug in poll cq when the NIC poster is GPU
|
|
| | * | | 3b49b0c4 |||| 2024-02-13T06:14:30-08:00 |||| Fix bugs related to assigning and using the new prod_idx variable
|
|
| | * | | 0f31dc03 |||| 2024-02-13T01:32:47-08:00 |||| Use gdr_copy_from_mapping to take a snapshot of prod_idx_array
|
|
| | * | | 07795856 |||| 2024-02-13T01:21:17-08:00 |||| Modify the progress_transports proxy code to also run IBGDA
|
|
| | * | | c1f78fd0 |||| 2024-02-13T01:20:57-08:00 |||| Setup use_async_postsend flag
|
|
| | * | | ed2593ca |||| 2024-02-09T02:18:21-08:00 |||| Apply ibgda_nic_poster_loc to various parts to support both GPU and CPU
|
|
| | * | | c524650c |||| 2024-02-09T02:01:18-08:00 |||| Introduce NVSHMEM_IBGDA_NIC_POSTER_LOC envvar and implement an auto-discovery algorithm
|
|
| | * | | 12504702 |||| 2024-02-09T02:00:47-08:00 |||| Remove unused prod_idx_array
|
|
| | * | | 23779573 |||| 2024-02-08T01:11:47-08:00 |||| Add GDRCOPY compile-time guard
|
|
| | * | | b4c46b9a |||| 2024-02-08T01:11:09-08:00 |||| Fix compile error in ibgda_device.cuh
|
|
| | * | | c321bdd1 |||| 2024-01-22T06:10:35-08:00 |||| Aligned size when calling gdr_pin_buffer
|
|
| | * | | abd46257 |||| 2024-01-22T01:33:54-08:00 |||| Added gdrcopy support to ibgda_gpu_mem_alloc
|
|
| | * | | 8cda0a8c |||| 2024-01-21T23:34:38-08:00 |||| Implemented async post-send with CPU as the poster
|
|
| | * | | 99fc75aa |||| 2024-02-08T00:16:57-08:00 |||| Implement use_async_postsend on the device side
|
|
| | * | | 33899a7b |||| 2024-02-07T23:21:25-08:00 |||| Use prod_idx instead of ready_head in ibgda_quiet
|
|
| | * | | bcdcf483 |||| 2024-02-07T23:20:19-08:00 |||| Use on-demand ctrl_seg for DB ringing
|
|
| * | | | 7834bdae |||| 2024-05-31T02:48:58-07:00 |||| CI: fix issues with builds after merge.
|
|
| * | | | 43186f08 |||| 2024-05-30T13:54:34-07:00 |||| Apply clang-formatting
|
|
| * | | | 05253ab6 |||| 2024-05-29T19:21:57-07:00 |||| Build error fixes
|
|
| * | | | d4cffe21 |||| 2024-05-29T15:28:06-07:00 |||| Reorder initializer fields
|
|
| * | | | 19d163c1 |||| 2024-05-29T11:28:58-07:00 |||| Fix sizeof struct for TEAM_INITIALIZER
|
|
| * | | | 7b37601f |||| 2024-05-28T13:14:36-07:00 |||| Update misleading log message
|
|
| * | | | e073dc58 |||| 2024-05-28T01:53:16-07:00 |||| Apply clang-formatting
|
|
| * | | | a7f12b47 |||| 2024-05-28T01:22:20-07:00 |||| Fix usage of macro and parse version
|
|
| * | | | 0c5a7285 |||| 2024-05-27T22:23:27-07:00 |||| Apply clang-formatting
|
|
| * | | | 63d5c9db |||| 2024-05-27T22:12:13-07:00 |||| Address build error:
|
|
| * | | | a684bdfe |||| 2024-05-27T21:23:17-07:00 |||| Add proper versioning to uid_handle
|
|
| * | | | bbb1e909 |||| 2024-05-21T09:18:54-07:00 |||| CI: Address review comments and strengthen tests.
|
|
| * | | | 99ed25f7 |||| 2024-05-19T12:46:48-07:00 |||| CI: Update lib versions and add ABI change tests.
|
|
| * | | | dbec265e |||| 2024-05-18T15:31:01-07:00 |||| version: update all version checks in the repo.
|
|
| * | | | 14378e48 |||| 2024-05-18T08:37:18-07:00 |||| types: version all abi crossing types.
|
|
| * | | | 05798e87 |||| 2024-05-09T09:34:08-07:00 |||| NVSHMEM: Add sentinel values to all public enums.
|
|
* | | | | dc310261 |||| 2024-05-31T13:57:18-07:00 |||| Merge branch 'arnavg/nvshmem/sysmem-logs-stale' into 'devel'
|
|
|\ \ \ \ \
|
|
| * | | | | 4be4ba5d |||| 2024-05-31T13:57:17-07:00 |||| Remove stale sysmem shm file
|
|
| |/ / / /
|
|
* | | | | d46b4759 |||| 2024-05-31T10:01:44-07:00 |||| Merge branch 'alanger/pr/fix_bcast_threshold' into 'devel'
|
|
|\ \ \ \ \
|
|
| |/ / / /
|
|
|/| | | |
|
|
| * | | | 0c9c577a |||| 2024-05-30T18:06:35-07:00 |||| Fix bcast ll threshold
|
|
|/ / / /
|
|
* | | | 344c2ca9 |||| 2024-05-30T17:09:45-07:00 |||| Merge branch 'revert-23621e75' into 'devel'
|
|
|\ \ \ \
|
|
| |_|_|/
|
|
|/| | |
|
|
| * | | 781d752a |||| 2024-05-30T11:59:52-07:00 |||| Revert "Merge branch 'sethh/abi_compat_changes' into 'devel'"
|
|
|/ / /
|
|
* | | 23621e75 |||| 2024-05-27T13:21:52-07:00 |||| Merge branch 'sethh/abi_compat_changes' into 'devel'
|
|
|\ \ \
|
|
| * | | 8a9d1166 |||| 2024-05-21T09:18:54-07:00 |||| CI: Address review comments and strengthen tests.
|
|
| * | | 0bb276e8 |||| 2024-05-19T12:46:48-07:00 |||| CI: Update lib versions and add ABI change tests.
|
|
| * | | 609ce43a |||| 2024-05-18T15:31:01-07:00 |||| version: update all version checks in the repo.
|
|
| * | | ade7790c |||| 2024-05-18T08:37:18-07:00 |||| types: version all abi crossing types.
|
|
| * | | 48728f6d |||| 2024-05-09T09:34:08-07:00 |||| NVSHMEM: Add sentinel values to all public enums.
|
|
|/ / /
|
|
* | | 71ace3b1 |||| 2024-05-26T13:13:54-07:00 |||| Merge branch 'sethh/nvshmem_common_h_fix' into 'devel'
|
|
|\ \ \
|
|
| |_|/
|
|
|/| |
|
|
| * | 0cc41699 |||| 2024-05-22T11:05:49-07:00 |||| include: fix missing ptrdiff definition in gcc9+
|
|
|/ /
|
|
* | e1f44696 |||| 2024-05-22T12:03:36-07:00 |||| Merge branch 'alanger/pr/fix_ptr_test' into 'devel'
|
|
|\ \
|
|
| * | 7f5c2c68 |||| 2024-05-20T18:32:11-07:00 |||| Use atomicAdd_system
|
|
* | | 5be64b1b |||| 2024-05-22T08:28:34-07:00 |||| Merge branch 'pmarkthub/nvbugs/4272672' into 'devel'
|
|
|\ \ \
|
|
| * | | 6cfc75db |||| 2024-05-02T03:06:22-07:00 |||| Fix rebase bug
|
|
| * | | 533536d1 |||| 2023-12-13T11:59:48-08:00 |||| fix build error
|
|
| * | | 38a397b0 |||| 2023-12-13T11:08:41-08:00 |||| build error
|
|
| * | | 8562b249 |||| 2023-12-10T18:15:13-08:00 |||| Fixed clang format error
|
|
| * | | 0bab14d0 |||| 2023-12-10T18:14:43-08:00 |||| Changed the return type of ibgda_destroy_cq_shared_objects to void
|
|
| * | | e559cd79 |||| 2023-12-10T18:08:00-08:00 |||| Rearranged members in struct ibgda_device
|
|
| * | | 7d846a8f |||| 2023-11-30T09:49:12+00:00 |||| Fixed bug related to cq_offset
|
|
| * | | 116b825e |||| 2023-11-07T15:02:28-08:00 |||| Fixed clang formatting
|
|
| * | | 2200d39e |||| 2023-11-07T14:16:57-08:00 |||| Consolidated CQ and DBR buffer allocation and mapping
|
|
| * | | 2fab0e8e |||| 2023-11-07T13:46:52-08:00 |||| Consolidated WQ and DBR buffers allocation and mapping
|
|
| * | | 1a8e1502 |||| 2023-11-07T13:44:23-08:00 |||| Exposed wq_umem_offset field in CREATE_QP in mlx5_ifc.h
|
|
* | | | f0317cbc |||| 2024-05-21T17:41:34-07:00 |||| Merge branch 'arnavg/nvshmem/fix-build-issue' into 'devel'
|
|
|\ \ \ \
|
|
| |_|/ /
|
|
|/| | |
|
|
| * | | 9cd52f9e |||| 2024-05-21T14:30:47-07:00 |||| Address build issue with use of auto
|
|
|/ / /
|
|
* | | 236309bd |||| 2024-05-20T16:47:18-07:00 |||| Merge branch 'arnavg/nvshmem/build-without-remote' into 'devel'
|
|
|\ \ \
|
|
| * | | 78f311a6 |||| 2024-05-20T16:47:18-07:00 |||| Add macro for conditional compile of remote transports
|
|
| | |/
|
|
| |/|
|
|
* | | 2a17c228 |||| 2024-05-20T14:33:35-07:00 |||| Merge branch 'pmarkthub/gpu_verbs/change-default-qp-type' into 'devel'
|
|
|\ \ \
|
|
| * | | 0e9f734f |||| 2024-05-13T01:12:22-07:00 |||| Change the default IBGDA QP type to RC
|
|
| |/ /
|
|
* | | ae8c3445 |||| 2024-05-20T14:30:12-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
|\ \ \
|
|
| * | | da21d25d |||| 2024-05-20T14:30:12-07:00 |||| [Coverity]: CID 2261576[0,2,3] Missing check for errors after kernel launch
|
|
| |/ /
|
|
* | | cd126d69 |||| 2024-05-20T09:49:09-07:00 |||| Merge branch 'sethh/version_copy' into 'devel'
|
|
|\ \ \
|
|
| |/ /
|
|
|/| |
|
|
| * | c71934e9 |||| 2024-05-19T14:51:22-07:00 |||| cmake: Add additional version.txt for packaging.
|
|
* | | 96ac092e |||| 2024-05-17T16:39:12-07:00 |||| Merge branch 'pmarkthub/gpu_verbs/fix-sysmem-cst-dci-bug' into 'devel'
|
|
|\ \ \
|
|
| * | | 5ae2f5f9 |||| 2024-05-16T20:48:43-07:00 |||| Remove ibgda_get_cst_dci and use ibgda_get_dci instead
|
|
* | | | 159fec8f |||| 2024-05-17T16:37:51-07:00 |||| Merge branch 'alanger/pr/fix_p2p_atomic' into 'devel'
|
|
|\ \ \ \
|
|
| |_|/ /
|
|
|/| | |
|
|
| * | | cafa0252 |||| 2024-05-16T15:39:16-07:00 |||| in-place clang-formatting
|
|
| * | | c7f5048c |||| 2024-05-12T21:05:37-07:00 |||| Use system scoped atomics for P2P
|
|
* | | | 0c7ba4df |||| 2024-05-17T08:29:13-07:00 |||| Merge branch 'sethh/abi_reduction' into 'devel'
|
|
|\ \ \ \
|
|
| |_|/ /
|
|
|/| | |
|
|
| * | | 667d7bc6 |||| 2024-05-16T14:26:44-07:00 |||| MR: Address comments from Arnav
|
|
| * | | 50801287 |||| 2024-05-09T00:21:56-07:00 |||| nvshmem_host: Reduce .sym file to minimum set.
|
|
| * | | 46694f26 |||| 2024-05-07T08:25:51-07:00 |||| include: Fix order of options and defines.
|
|
| * | | 703337b4 |||| 2024-05-06T16:39:55-07:00 |||| include: remove deprecated functions.
|
|
| * | | bc756ca4 |||| 2024-05-06T13:29:14-07:00 |||| proxy: move shared channel structs to device_host header.
|
|
| * | | 8b830e95 |||| 2024-05-05T19:58:27-07:00 |||| src/include: Refactor includes to exclude internal defs from ABI.
|
|
| * | | f459d546 |||| 2024-05-05T00:31:00-07:00 |||| scripts: Generalize include_checker and add to CI
|
|
| * | | 51dbff2a |||| 2024-05-04T22:57:49-07:00 |||| nvshmem: Add nvshmem_build_options includes
|
|
| * | | ec80726d |||| 2024-05-03T06:37:01-07:00 |||| src/bootstrap: Add info to bootstrap and simplify ABI
|
|
| * | | 13bde9ae |||| 2024-05-03T02:21:53-07:00 |||| host: Factor out non_abi calls from host to dev headers
|
|
| * | | 7e134269 |||| 2024-04-28T19:25:32-07:00 |||| scripts: Add check for header include breakage.
|
|
| * | | ec4ebeec |||| 2024-04-28T19:24:12-07:00 |||| include: move uid types to their own file.
|
|
| * | | e6952560 |||| 2024-04-24T11:57:02-07:00 |||| Simplify host-side nvshmemi_state_t type.
|
|
* | | | 2e4d8918 |||| 2024-05-16T16:23:58-07:00 |||| Merge branch 'arnavg/nvshmem/coll-precision-issue' into 'devel'
|
|
|\ \ \ \
|
|
| |_|_|/
|
|
|/| | |
|
|
| * | | 2015747b |||| 2024-05-16T16:23:58-07:00 |||| Address perftest precision issue
|
|
* | | | 6163201b |||| 2024-05-14T10:22:04-07:00 |||| Merge branch 'alanger/pr/fix_ib_ro_compile_error' into 'devel'
|
|
|\ \ \ \
|
|
| |/ / /
|
|
|/| | |
|
|
| * | | 1eaf2c58 |||| 2024-05-13T12:07:06-07:00 |||| fix compile error
|
|
|/ / /
|
|
* | | 003394f6 |||| 2024-05-09T11:03:11-07:00 |||| Merge branch 'arnavg/nvshmem/NVSHMEM-1464' into 'devel'
|
|
|\ \ \
|
|
| |/ /
|
|
|/| |
|
|
| * | fa6f9309 |||| 2024-05-09T11:03:10-07:00 |||| Address non-reentrant function call in test
|
|
* | | 4575be3e |||| 2024-05-09T00:05:39-07:00 |||| Merge branch 'arnavg/nvshmem/sysmem-memory-corruption' into 'devel'
|
|
|\ \ \
|
|
| * | | 0349d8ac |||| 2024-05-09T00:05:38-07:00 |||| Fix sysmem + P2P or sysmem + IBRC memory corruption
|
|
* | | | 0bca3dc6 |||| 2024-05-08T13:32:43-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
|\ \ \ \
|
|
| |/ / /
|
|
|/| | |
|
|
| * | | 04182872 |||| 2024-05-08T13:32:43-07:00 |||| coverity fix: cid:22615765 cid:22615764
|
|
| |/ /
|
|
* | | 8f82a324 |||| 2024-05-06T16:45:13-07:00 |||| Merge branch 'arnavg/nvshmem/missing-handle-type' into 'devel'
|
|
|\ \ \
|
|
| |/ /
|
|
|/| |
|
|
| * | 04501d2d |||| 2024-05-06T16:45:13-07:00 |||| Missing enum definition for CUDA 11.0 builds
|
|
|/ /
|
|
* | e94ec50c |||| 2024-05-04T20:38:37-07:00 |||| Merge branch 'arnavg/nvshmem/oopfy_p2p_phase2' into 'devel'
|
|
|\ \
|
|
| * | 99d93548 |||| 2024-05-04T20:38:37-07:00 |||| [NVSHMEM-1555] Symmetric Memory Management OOPfyication [Phase-2]
|
|
|/ /
|
|
* | b5c0a6fe |||| 2024-05-03T06:32:42-07:00 |||| Merge branch 'sethh/IWYU_Macro_Fix' into 'devel'
|
|
|\ \
|
|
| * | 1de66e17 |||| 2024-05-03T02:30:02-07:00 |||| in-place clang-formatting
|
|
| * | 45418f00 |||| 2024-05-02T19:16:25-07:00 |||| src: Add check for IWYU quirks.
|
|
* | | 40965f5c |||| 2024-05-03T02:05:27-07:00 |||| Merge branch 'arnavg/nvshmem/sysmem_rail_issue' into 'devel'
|
|
|\ \ \
|
|
| * | | f578f107 |||| 2024-05-03T02:05:26-07:00 |||| Address SYSMEM + RAIL_OPT data corruption and incorrect memGran
|
|
| |/ /
|
|
* | | 7d400353 |||| 2024-05-03T00:58:06-07:00 |||| Merge branch 'arnavg/nvshmem/crash-in-unreg' into 'devel'
|
|
|\ \ \
|
|
| |/ /
|
|
|/| |
|
|
| * | 84af2d19 |||| 2024-05-03T00:58:06-07:00 |||| Add NULL check for cache of mem_handles in remote transports
|
|
|/ /
|
|
* | 063ab210 |||| 2024-04-30T08:26:41-07:00 |||| Merge branch 'sethh/build_info_fix' into 'devel'
|
|
|\ \
|
|
| * | f6ee0173 |||| 2024-04-29T21:03:02-07:00 |||| CMake: Add nvshmem version to git_commit.txt.
|
|
* | | b3064671 |||| 2024-04-30T08:03:41-07:00 |||| Merge branch 'arnavg/perftest/on_stream_issue' into 'devel'
|
|
|\ \ \
|
|
| |/ /
|
|
|/| |
|
|
| * | 62340517 |||| 2024-04-30T08:03:41-07:00 |||| Missing cudaEventCreate in perftest onstream collectives test
|
|
* | | eaf8dce7 |||| 2024-04-29T16:26:29-07:00 |||| Merge branch 'sethh/header_fixes' into 'devel'
|
|
|\ \ \
|
|
| |/ /
|
|
|/| |
|
|
| * | f9602364 |||| 2024-04-26T16:03:32-07:00 |||| include: Add package version to nvshmem_version.h.
|
|
| * | f0a4747e |||| 2024-04-25T18:04:30-07:00 |||| scripts: Add top-of-file docs to IWYU scripts
|
|
| * | f283211e |||| 2024-04-15T18:29:35-07:00 |||| src: remove using namespace std from headers
|
|
| * | d727ad26 |||| 2024-04-15T16:44:39-07:00 |||| include: move non_abi headers to a new directory
|
|
| * | 4574b474 |||| 2024-04-15T16:39:07-07:00 |||| src: move util directory under host.
|
|
| * | 47d423ac |||| 2024-04-15T15:38:33-07:00 |||| CI/IWYU: add cuda files to IWYU check.
|
|
| * | b8e106cb |||| 2024-04-15T13:15:42-07:00 |||| CI: update headers based on new IWYU output
|
|
| * | 195c9383 |||| 2024-04-11T19:42:08-07:00 |||| nvshmem: Add IWYU script for headers.
|
|
| * | f0af3bff |||| 2024-04-10T17:43:49-07:00 |||| include/common: Move common files to ABI folders.
|
|
| * | 6305a60e |||| 2024-04-10T16:39:44-07:00 |||| CI: Add back IWYU check to source build.
|
|
| * | e997cf4c |||| 2024-04-10T14:27:30-07:00 |||| src: Fix includes based on iwyu
|
|
| * | b35512a5 |||| 2024-04-09T14:08:09-07:00 |||| transport/ibgda: change header configuration
|
|
| * | 3d4032e4 |||| 2024-04-09T13:02:00-07:00 |||| transport: initial header relocation
|
|
| * | ae358a5b |||| 2024-04-09T11:06:39-07:00 |||| device: move rma.cu to host.
|
|
| |/
|
|
* | 167bc5c8 |||| 2024-04-29T10:23:15-07:00 |||| Merge branch 'sethh/add_ptx' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * 57fbc162 |||| 2024-04-19T17:39:58-07:00 |||| CI: add a host-side run to confirm ptx functionality.
|
|
| * 72f49168 |||| 2024-04-19T15:58:48-07:00 |||| CI: add test to confirm we are building PTX into our code.
|
|
* | 571ea82f |||| 2024-04-23T10:57:13-07:00 |||| Merge branch 'arnavg/nvshmem/broadcastmem_support' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * 52af6f66 |||| 2024-04-23T10:57:13-07:00 |||| [NVSHMEM-1431]: Add support for fcollectmem, broadcastmem and onstream APIs
|
|
* | 342bd384 |||| 2024-04-22T13:22:59-07:00 |||| Merge branch 'arnavg/nvshmem/nvbug-4613362' into 'devel'
|
|
|\ \
|
|
| * | d105d810 |||| 2024-04-22T02:25:42-07:00 |||| Address compile error on aarch64
|
|
| |/
|
|
* | bf995ae9 |||| 2024-04-22T11:42:00-07:00 |||| Merge branch 'arnavg/nvshmem/team-destroy-subteam' into 'devel'
|
|
|\ \
|
|
| * | 042e0269 |||| 2024-04-22T11:41:59-07:00 |||| NVSHMEM-1560: Fix leak in nvshmemi_team_destroy
|
|
| |/
|
|
* | 3c01a002 |||| 2024-04-22T09:26:55-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * c1d6a7f3 |||| 2024-04-22T09:26:54-07:00 |||| coverity:22635859 + 22625286
|
|
* | 46e4c0fb |||| 2024-04-17T07:25:30-07:00 |||| Merge branch 'sethh/abi_fixes' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * 5bdf4ed0 |||| 2024-04-09T14:13:14-07:00 |||| include: move cuda_interface_sync.cu to host.
|
|
| * 894394d5 |||| 2024-04-01T17:21:51-07:00 |||| device: Move proxy code to host and headers.
|
|
| * 4663d1fc |||| 2024-03-14T14:39:16-07:00 |||| device: remove cuda_interface_sync.cu
|
|
| * 7c378ff0 |||| 2024-03-14T14:30:14-07:00 |||| device: Remove quiet_on_stream
|
|
| * ffc24b21 |||| 2024-03-14T11:06:51-07:00 |||| cudawrap: Add additional missing enums.
|
|
| * f8979128 |||| 2024-03-14T10:27:07-07:00 |||| src: Move on_stream collectives to host lib
|
|
| * 3e45ea10 |||| 2024-03-14T08:24:39-07:00 |||| cudawrap: Add declarations for remaining symbols
|
|
| * cb097e30 |||| 2024-03-08T12:35:25-08:00 |||| nvshmem: fix remaining compiler bugs
|
|
| * fada9e6a |||| 2024-03-07T17:46:34-08:00 |||| NVSHMEM: Remove all CUDA and CUDART Compiletime Checks
|
|
| * ff42b6a8 |||| 2024-03-07T15:32:04-08:00 |||| nvshmem: remove compile-time flags for CUDA < 11.0
|
|
| * 36b73e85 |||| 2024-03-07T14:55:21-08:00 |||| nvshmemi_state_t: Remove scratch spaces used in debug paths.
|
|
| * 774b5e9c |||| 2024-03-07T10:50:47-08:00 |||| nvshmemi_state_t: remove cudevice
|
|
| * 705fb3bf |||| 2024-03-07T10:46:42-08:00 |||| nvshmemi_state_t: remove prefix variable
|
|
| * 185e542a |||| 2024-02-06T14:32:43-07:00 |||| src/include: remove shared_memory.h from public headers.
|
|
|/
|
|
* d09ab1f4 |||| 2024-04-15T13:20:09-07:00 |||| Merge branch 'arnavg/nvshmem/mpi_bootstrap_bug' into 'devel'
|
|
|\
|
|
| * 9bb10546 |||| 2024-04-15T13:20:09-07:00 |||| Fix bootstrap bugs and add debug messages
|
|
* | b52fbbab |||| 2024-04-15T08:30:54-07:00 |||| Merge branch 'sethh/ibdevx_atomics_fix' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * 42ad138d |||| 2024-04-12T21:30:27-07:00 |||| mlx5: Fix endianness check.
|
|
|/
|
|
* ce397460 |||| 2024-04-08T16:24:12-07:00 |||| Merge branch 'arnavg/nvshmem/nvbug4593392' into 'devel'
|
|
|\
|
|
| * 3d7684f9 |||| 2024-04-08T16:24:12-07:00 |||| [NVBUG] 4593392: Use empty ctor definition when using noexcept instead of default specifier
|
|
|/
|
|
* ae487acc |||| 2024-04-04T15:41:28-07:00 |||| Merge branch 'sethh/dlopen_use_soname' into 'devel'
|
|
|\
|
|
| * 70fa64fe |||| 2024-04-03T14:49:02-07:00 |||| cuda/nvmlwrap: dlopen SONAME instead of Linker Name
|
|
* | 2be8a094 |||| 2024-04-04T11:10:59-07:00 |||| Merge branch 'sethh/error_message_fixes' into 'devel'
|
|
|\ \
|
|
| * | d6445ca4 |||| 2024-04-03T15:39:57-07:00 |||| in-place clang-formatting
|
|
| * | a4532451 |||| 2024-04-03T15:35:02-07:00 |||| init: remove extraneous messages about cuda API failures
|
|
| |/
|
|
* | fdc4b13c |||| 2024-04-04T09:45:21-07:00 |||| Merge branch 'arnavg/nvshmem/malloc_noaction' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * b9dbc22d |||| 2024-04-04T09:45:20-07:00 |||| [NVSHMEM-1373] Add no action support for malloc/calloc/align/free
|
|
|/
|
|
* 243c2c72 |||| 2024-04-02T12:52:50-07:00 |||| Merge branch 'sesu/guarded_nvshmem_common_ibgda.h_header_in_mem_transport.cpp' into 'devel'
|
|
|\
|
|
| * ce40d074 |||| 2024-04-02T17:35:09+08:00 |||| header src/include/common/nvshmem_common_ibgda.h should be guarded by a macro in new source file mem_transport.cpp
|
|
* | d3b57540 |||| 2024-04-02T10:49:49-07:00 |||| Merge branch 'sethh/fix_bin_dir_permissions' into 'devel'
|
|
|\ \
|
|
| * | d1965473 |||| 2024-04-01T18:31:38-07:00 |||| CMake: Remove test_cxx11.sh
|
|
| * | 13fd429e |||| 2024-04-01T18:30:45-07:00 |||| CMake: Add execute permissions to bin and scripts.
|
|
| |/
|
|
* | 5595a9d9 |||| 2024-04-02T09:13:52-07:00 |||| Merge branch 'arnavg/nvshmem/fix_ci_build' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * 202752e5 |||| 2024-04-01T15:52:26-07:00 |||| Address path issue
|
|
|/
|
|
* 31546578 |||| 2024-04-01T09:49:04-07:00 |||| Merge branch 'arnavg/nvshmem/oopfy_mem_phase1' into 'devel'
|
|
|\
|
|
| * 319b827c |||| 2024-04-01T09:49:03-07:00 |||| [NVSHMEM-1556] Symmetric Memory Management OOPfyication [Phase-1]
|
|
|/
|
|
* 32a802ef |||| 2024-03-27T10:44:52-07:00 |||| Merge branch 'sesu/python_scr_enhancement' into 'devel'
|
|
|\
|
|
| * 4546b27f |||| 2024-03-27T18:10:49+08:00 |||| Write running cmd line into both out and err files.
|
|
* | 597a5704 |||| 2024-03-25T11:13:05-07:00 |||| Merge branch 'sesu/python_scr_enhancement' into 'devel'
|
|
|\|
|
|
| * 4a3ff413 |||| 2024-03-25T15:02:00+08:00 |||| Support show minor and max results on screen; Support run partial cases in cmdline mode.
|
|
| * 2d7f579c |||| 2024-03-25T14:11:45+08:00 |||| Support show_data and cases arguments in scri
|
|
| * c377211c |||| 2024-03-22T17:50:05+08:00 |||| Add a parameter. The results of the first and last row of each table can be displayed on the screen.
|
|
|/
|
|
* b4a3bbef |||| 2024-03-20T12:56:26-07:00 |||| Merge branch 'sesu/scr_enhancement' into 'devel'
|
|
|\
|
|
| * 0827ed92 |||| 2024-03-19T16:50:34+08:00 |||| Remove some redundant comment lines.
|
|
| * 8da12f6f |||| 2024-03-19T16:45:01+08:00 |||| Prints error output for failed test cases.
|
|
* | fec8a571 |||| 2024-03-20T10:08:38-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * 27b2b080 |||| 2024-03-20T10:08:38-07:00 |||| Coverity fix
|
|
|/
|
|
* 9d57c972 |||| 2024-03-14T08:11:42-07:00 |||| Merge branch 'sethh/remove_make_files' into 'devel'
|
|
|\
|
|
| * 741d696d |||| 2024-03-06T17:15:27-08:00 |||| NVSHMEM: remove the make build system.
|
|
* | 8290549f |||| 2024-03-14T07:40:01-07:00 |||| Merge branch 'sethh/fix_test_output' into 'devel'
|
|
|\ \
|
|
| * | 552da2ab |||| 2024-03-06T10:56:35-08:00 |||| in-place clang-formatting
|
|
| * | 68924c9e |||| 2024-03-06T08:27:43-08:00 |||| Perftest: Make test output format a runtime parameter.
|
|
| |/
|
|
* | 743e444a |||| 2024-03-11T16:54:50-07:00 |||| Merge branch 'pezhu/devel_coverity' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * df5a3d50 |||| 2024-03-11T16:54:49-07:00 |||| Pezhu/devel coverity
|
|
|/
|
|
* abaadb89 |||| 2024-03-06T08:16:45-08:00 |||| Merge branch 'pmarkthub/gpu_verbs/add-atomic-cap-check' into 'devel'
|
|
|\
|
|
| * ecf5dbc7 |||| 2024-03-05T19:43:34-08:00 |||| in-place clang-formatting
|
|
| * 09be957c |||| 2024-03-06T03:40:24+00:00 |||| Call nvshmemt_ib_common_check_nic_ext_atomic_support in ibdevx
|
|
| * df44adee |||| 2024-03-06T03:30:11+00:00 |||| Move the nic atomic support check to transport_mlx5_common
|
|
| * db30de70 |||| 2024-03-04T23:38:31-08:00 |||| in-place clang-formatting
|
|
| * 8dcf4e39 |||| 2024-03-05T07:34:42+00:00 |||| Add atomic cap query and check. Skip that HCA if it does not support all necessary atomics.
|
|
* | 4f733477 |||| 2024-03-05T13:45:06-08:00 |||| Merge branch 'sethh/ci_update_partitions' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * e8298836 |||| 2024-02-20T17:23:05-08:00 |||| CI: add A40 partition to IB-based jobs.
|
|
|/
|
|
* d4127ab6 |||| 2024-02-29T07:50:00-08:00 |||| Merge branch 'sethh/fix_non_ib_build' into 'devel'
|
|
|\
|
|
| * e025118e |||| 2024-02-27T15:56:44-08:00 |||| mem.cpp: conditionally include nvshmem_common_ibgda.h
|
|
* | 6343d2b2 |||| 2024-02-29T01:31:06-08:00 |||| Merge branch 'arnavg/nvshmem/uid_bootstrap/ci_stage' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * 18e8e828 |||| 2024-02-29T01:31:05-08:00 |||| NVSHMEM-1486: Introduce UID bootstrap CI stage
|
|
|/
|
|
* 5c983b14 |||| 2024-02-27T11:01:12-08:00 |||| Merge branch 'sethh/ci_remove_p9' into 'devel'
|
|
|\
|
|
| * 8e7b22c9 |||| 2024-02-26T18:19:27-07:00 |||| CI: Remove P9 Pipelines
|
|
|/
|
|
* 87a2f60d |||| 2024-02-23T14:57:23-08:00 |||| Merge branch 'sethh/add_debugger_delay' into 'devel'
|
|
|\
|
|
| * 4ed9a125 |||| 2024-02-20T13:41:43-08:00 |||| in-place clang-formatting
|
|
| * e25467c0 |||| 2024-02-16T20:25:08-08:00 |||| env: add option to delay init for debugger attach
|
|
* | 744b520d |||| 2024-02-23T14:03:51-08:00 |||| Merge branch 'sesu/scripts_enhancement' into 'devel'
|
|
|\ \
|
|
| * | a8bd687e |||| 2024-02-20T12:53:53+08:00 |||| Add /lib64 in mpirun cmdline.
|
|
| * | f72ffba5 |||| 2024-01-24T18:19:13+08:00 |||| Handle an expected error with the 'try' command
|
|
| * | a191b66f |||| 2024-01-22T17:09:03+08:00 |||| Update interact mode
|
|
| * | b58bfe87 |||| 2024-01-12T16:19:21+08:00 |||| Remove null -genv value in list, when it exists.
|
|
| |/
|
|
* | a72943d9 |||| 2024-02-21T10:59:35-08:00 |||| Merge branch 'arnavg/nvshmem/uninitialized_mutex_destroy' into 'devel'
|
|
|\ \
|
|
| * | efbc9ddf |||| 2024-02-21T10:59:35-08:00 |||| Use syscheck macro for better errno decoding of pthread_mutex APIs
|
|
| /
|
|
* | f66a16f5 |||| 2024-02-21T06:47:53-08:00 |||| Merge branch 'arnavg/nvshmem/bootstrap_versioning' into 'devel'
|
|
|\ \
|
|
| * | 5524f9aa |||| 2024-02-21T06:47:52-08:00 |||| Simplify bootstrap common in nvshmem host library
|
|
| |/
|
|
* | 55a9cbd8 |||| 2024-02-21T00:33:28-08:00 |||| Merge branch 'arnavg/nvshmem/address-coverage-dup-issue' into 'devel'
|
|
|\ \
|
|
| |/
|
|
|/|
|
|
| * b91f89d2 |||| 2024-02-21T00:33:28-08:00 |||| Rename simple-pmi to avoid coverage hit due to file dup issue
|
|
|/
|
|
* bc86efee |||| 2024-02-20T12:09:48-08:00 |||| Merge branch 'sethh/ci_cluster_update_fix' into 'devel'
|
|
|\
|
|
| * df21a3b0 |||| 2024-02-20T09:04:27-08:00 |||| CI: Temporarily disable the libfabric and ucx status jobs.
|
|
| * 65f12d1f |||| 2024-02-15T18:54:06-08:00 |||| CI: Update paths and arguments for gpu comms cluster
|
|
| * fc19a9d5 |||| 2024-02-11T19:52:05-08:00 |||| src/util: Fix strnlen calls in sockets.cpp.
|
|
* | 8d71a88d |||| 2024-02-20T11:57:57-08:00 |||| Merge branch 'pmarkthub/gpu_verbs/fix_qp_warp_mapping' into 'devel'
|
|
|/
|
|
* 60a6b1bd |||| 2024-02-13T12:22:48-08:00 |||| Merge branch 'jdinan/mr/nvhsmem' into 'devel' |