AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
43 Commits 3 Branches 0 Tags
95e498c939be8ec098fdda262952fc00f4012baa
Commit Graph

28 Commits

Author SHA1 Message Date
Wenkai Du 2813c96826 Add option to use alltoall, gather and scatter API 
These APIs launche RCCL kernel implementation by default. If environmental
variable RCCL_ALLTOALL_KERNEL_DISABLE=1, then the APIs use wrapper around
ncclSend and ncclRecv.
 2020-05-11 15:21:19 -07:00
Sourav Chakraborty 0d7c4db33e Add sendrecv benchmark 2020-04-10 14:45:17 -07:00
Sourav Chakraborty 07b8876277 Improve makefile to avoid LD_LIBRARY_PATH 2020-04-10 14:45:17 -07:00
Wenkai Du 6932a583e7 Add gather and scatter test 2020-04-10 14:45:10 -07:00
Wenkai Du c2dc7b613b Merge remote-tracking branch 'nccl/p2p' into p2p 2020-04-02 15:06:36 -07:00
Sylvain Jeaugey 119a0ecf60 Add alltoall perf test 2020-03-17 12:00:19 -07:00
Wenkai Du 6e9e05972b Add option for stress testing 2020-03-11 13:40:17 -07:00
Wenkai Du bd53e98df3 Fix build error with hip-clang 2019-11-22 10:31:31 -08:00
Wenkai Du 1cda2f52b6 Add bf16 support in rccl-tests 2019-11-18 13:18:34 -08:00
Wenkai Du 32399955af Fix build with RCCL 1.x API 2019-11-12 23:04:39 +00:00
Wenkai Du aa0f02bee0 Fix incorrect print out when data size is greater than 4GB 2019-11-11 11:39:59 -08:00
Wenkai Du 13d0ddd12e Init data for throughput iterations to avoid all zero data 2019-09-27 09:12:00 -07:00
Wenkai Du ab82f1af6f Fix memory leak and possible buffer overrun 2019-08-09 10:22:14 -07:00
Wenkai Du 23c374475f Allow call ncclCommAbort on Ctrl+C 2019-07-26 00:15:16 +00:00
Gilbert Lee 043eef6999 Checking that number of requested GPUs is not more than number of available GPUs 2019-07-11 15:36:21 +00:00
Stanley Tsang 924521ff57 Adding unit tests and files for CI (#4) 
* Adding initial unit test and Jenkins code.

Fixing scope of unit tests

Adding unit tests and files for CI

Fixing Jenkinsfile

* Removing typos from Jenkinsfile

* Making some fixes to the Jenkins file; temporarily disabling MPI

* Making corrections to Jenkinsfile

* Correcting dockerNodes entry in Jenkinsfile

* Fixed Jenkinsfile for CI

* Correcting Jenkinsfile for CI

* Updating README to include instructions on how to run unit tests.
 2019-06-28 08:52:44 -07:00
Wenkai Du 3fac1d679b Fix missing space in Makefile 2019-05-16 13:18:23 -07:00
Wenkai Du 86f053be84 enable building with mpich 
Use following command to build: MPICH=1 make
 2019-05-13 23:45:28 +00:00
Wenkai Du 3f89175af5 allow using different memory types for input and output buffers 2019-05-01 12:58:04 -07:00
Wenkai Du 7e80ea9d3a fix build with 1.0 library 2019-04-26 17:06:05 -07:00
Gilbert Lee 10e1572f72 Adding way to specify a custom RCCL shared library file to link against 2019-04-22 10:25:49 -07:00
Wenkai Du 4474fe168d workaround weak symbol issue 
hcc prints "error: alias must point to a defined variable or function"
 2019-04-18 10:34:55 -07:00
Stanley Tsang 71e663e62d Adding AMD copyright notices 2019-04-10 15:28:40 -07:00
Wenkai Du a15f771cb2 hipify nccl-tests to become rccl-tests 2019-04-10 13:43:58 -07:00
David Addison cbe7f65400 Resync all tests with test code from NCCL 2.4 
Major rework to merge most of the changes from the NCCL internal
tests into the public ones

Added "-m <agg_iters>" operation aggregation option.
Data integrity checking is now much more performant at scale.
Startup times at scale are improved.
Test latency units are now displayed in usec.
 2019-04-05 13:42:15 -07:00
Sylvain Jeaugey 925a70576e Print NCCL version at start 2017-12-21 15:10:09 -08:00
Sylvain Jeaugey 25016c8eeb Fix NCCL_HOME to be consistent with README 2017-08-09 10:41:31 -07:00
Sylvain Jeaugey b188a15299 Initial commit 2017-08-08 16:18:34 -07:00