469225bcaf
Merge remote-tracking branch 'origin/master' into develop
2023-06-14 08:01:50 -07:00
d16d1fb16b
fixing the error message for mpirun when number of requested GPUs exceeds the limits ( #37 )
2023-04-27 14:06:17 -04:00
e856fa720f
Revert "fixing the error message for mpirun when number of requested GPUs exceeds the limits ( #33 )" ( #36 )
...
This reverts commit e146460810
2023-04-25 13:44:43 -04:00
e146460810
fixing the error message for mpirun when number of requested GPUs exceeds the limits ( #33 )
2023-04-03 11:37:13 -04:00
0fc25d5b61
Merge pull request #32 from edgargabriel/topic/mpi-auto-compile
...
revamp cmake MPI detection
2023-03-03 07:55:52 -06:00
bdf58b1656
revamp cmake MPI detection
...
we honor user requested MPI installations using MPI_PATH first,
and check afterwards for MPICH and Open MPI in the default
Ubuntu and RHEL installation directories.
2023-03-02 19:40:13 +00:00
255750b094
Adding -pthread flag for linking issues into CMakeLists.txt and src/Makefile ( #31 )
2023-03-02 11:05:25 -05:00
5275aa5715
Adding -pthread flag for linking issues into src/Makefile ( #30 )
...
* Adding -pthread flag for linking issues into src/Makefile
* Adding -pthread flag for linking issues into CMakeLists.txt
2023-02-24 21:39:04 -05:00
453e72972b
Merge pull request #28 from edgargabriel/topic/mpi-auto-compile
...
auto-detect and enable MPI
2023-02-23 12:35:59 -06:00
2b2f23f42d
auto-detect and enable MPI
2023-02-23 18:27:08 +00:00
b3f0716190
Merge pull request #27 from edgargabriel/topic/half_prod_fix
...
fix algorithm assigning values in testsuite
2022-12-01 08:11:52 -06:00
e9f5be184c
fix algorithm assigning values in testsuite
...
avoid a division by zero which seems to only occur for op=prod and
datatype=half, since the maximum exponent is small (15) and can exceed
the number of ranks.
2022-11-30 23:01:46 +00:00
9d3a53dfa3
added std::max to avoid buffer overflow in printing ( #25 )
2022-11-01 11:34:55 -06:00
a8c920ca7a
Merge pull request #24 from edgargabriel/pr/cmake-fix
...
make cmake stage also pass in CI
2022-11-01 09:39:22 -05:00
377b28e5fb
make cmake stage also pass in CI
...
the subdir entry is not actually required for the compilation.
2022-10-31 22:07:15 +00:00
a80fbba12b
Merge pull request #23 from edgargabriel/pr/link-fix
...
add the rccl/lib directory to the link path
2022-10-31 15:54:55 -05:00
9c9746739a
add the rccl/lib directory to the link path
2022-10-31 19:01:22 +00:00
fb0d339c1b
Merge pull request #22 from edgargabriel/pr/compile-fix
...
fix a messing endif statement
2022-10-25 12:19:25 -05:00
8a754f15ad
fix a messing endif statement
...
error introduced with the web merger-resolution tool :-(
2022-10-25 16:31:57 +00:00
84e8be8e65
Merge pull request #21 from ROCmSoftwarePlatform/topic/v2.13.4-sync
...
Topic/v2.13.4 sync
2022-10-21 17:17:27 -05:00
4d7cd871c1
Merge branch 'develop' into topic/v2.13.4-sync
2022-10-21 17:12:45 -05:00
9a89c300b6
Allow more precise measurements of single operation ( #20 )
2022-10-21 22:07:41 +00:00
641e93e99c
make rccl-test compile again.
...
all files compile now.
mpi tests also pass
2022-10-21 22:07:33 +00:00
3ae371cce7
Merge remote-tracking branch 'nccl-tests/master' into topic/v2.13.4-sync
2022-10-14 16:02:54 -05:00
d22281cb3f
Allow more precise measurements of single operation ( #20 )
2022-10-12 17:28:04 -07:00
3fbd3280ce
removed hypercube from Makefile ( #19 )
2022-09-29 15:36:39 -06:00
d313d20a26
Update NCCL tests
2022-09-23 01:13:29 -07:00
749573f2d6
Fix preprocessor version check for ncclGetLastError()
...
ncclGetLastError() was added in NCCL 2.13.0
2022-09-07 16:10:41 -07:00
afa4c56b6a
Fix an issue with the last commit when data checking is disabled
2022-09-07 11:23:49 -07:00
a0a14911ee
Display N/A for error count in AlltoAll in-place test
...
AlltoAll does not support in-place buffers
2022-09-06 13:17:15 -07:00
bc5f7cfb0a
Changed top-level Makefile behavior so that BUILDDIR is interpreted
...
as relative to top-level directory. This done is by abspath'ing it before
passing it to subdirectory Makefile's.
The old behavior had two cases: with and without BUILDDIR being set by
the user. With BUILDDIR not set, the build dir would be named "build"
in the top-level directory. If BUILDDIR was set, then the build dir
would be placed at "src/${BUILDDIR}".
The new behavior is simpler, if BUILDDIR is not set then it defaults
to "build", and the directory holding the final build is always at just
"${BUILDDIR}" in the top level.
2022-08-23 10:08:49 -07:00
51af5572bf
Resync with NCCL 2.13
...
* Added "verifiable", a suite of kernels for generating and verifying reduction
input and output arrays in a bit-precise way.
* Data corruption errors now reported in number of wrong elements instead of max
deviation.
* Use ncclGetLastError.
* Don't run hypercube on non-powers of 2 ranks.
* Fix to hypercube data verification.
* Use "thread local" as the defaut CUDA capture mode.
* Replaced pthread_yield -> sched_yield()
* Bugfix to the cpu-side barrier/allreduce implementations.
2022-08-22 17:51:06 -07:00
45ec598ac4
Fix typo from previous merge
2022-08-12 14:42:17 +00:00
f6f3c44a7a
Enabling hipGraph codepath for future support ( #18 )
2022-08-09 16:45:27 -06:00
9025051bbb
Fix missing error checking for AllocateBuffs due to merge ( #17 )
2022-08-09 11:04:38 -07:00
d704668bf7
Add CMake files to build & package ( #15 )
...
* Add CMake files to build & package
* Change build technique on CI
* Correct CI build command
2022-08-09 11:17:07 -06:00
2af4f6bc3a
Allow gpu config override in CI ( #14 )
2022-07-28 09:19:16 -06:00
9925195afc
updated alltoallV test to not have any zero values ( #12 )
...
updated alltoallV test to not have any zero values between ranks
2022-07-21 10:28:53 -06:00
2a18737dc6
Merge pull request #11 from edgargabriel/ci-fix
...
update pytest before running CI
2022-06-13 09:52:40 -05:00
67544e2c34
update pytest before running CI
...
There seems to be in an incompatibility between the python installation
used in the CI and pytest. Update pytest before running CI.
2022-06-13 10:20:33 -04:00
937ea1926e
Merge pull request #10 from edgargabriel/multi-rank
...
Multi rank support
2022-06-10 14:03:33 -05:00
0500f2f132
implementation of multi-rank support in rccl-tests.
2022-06-10 14:54:10 -04:00
5cd2374edb
create branch up-to-date with rccl-test
2022-06-10 12:41:56 -04:00
3d6f70659a
Check for error code in install script ( #2 )
2022-06-10 12:37:53 -04:00
8274cb47b6
Merge pull request #96 from NVIDIA/nersc-linkage-fix
...
Add option to statically link cudart
2022-05-26 16:54:44 -07:00
6156759a40
Print GPU's full PCI bus ID
2022-04-06 16:46:17 +00:00
47238336d9
Update include path for custom RCCL build
2022-03-31 13:18:02 -04:00
698524e42e
move to a2a api ( #9 )
2022-02-18 08:31:40 -08:00
602b745ff4
Add missing hipStreamDestroy at test exit
2021-11-16 07:50:18 -08:00
de3ddbe261
Add option to statically link cudart
...
Build with CUDARTLIB=cudart_static to remove dynamic linkage
Also removed unused curand and nvToolsExt dependencies
BUG 95
2021-11-10 10:02:41 -08:00
Wenkai Du
Pedram Alizadeh
Edgar Gabriel
Edgar Gabriel
akolliasAMD
Wenkai Du
Sylvain Jeaugey
David Addison
John Bachan
gilbertlee-amd
Liam Wrubleski
Eiden Yoshida
amdkila
David Addison
Ziyue Yang