ff97af6529
Fix typo in rank assignment ( #59 )
...
[ROCm/rccl-tests commit: 7715a0cf1f
2024-02-15 12:04:38 -08:00
73f1f3cb3a
Merge branch 'develop'
...
[ROCm/rccl-tests commit: 56a5bb0486
2024-01-30 12:50:38 -05:00
26b1b0b822
Add option to disable out-of-place
...
[ROCm/rccl-tests commit: a2bec5d2f6
2024-01-04 16:43:50 -06:00
cffd823582
Update default GPUs and build for AMDGPU_TARGETS ( #55 )
...
* Update default GPUs and build for AMDGPU_TARGETS
* Make GPU_TARGETS a cache variable
---------
Signed-off-by: nileshnegi <Nilesh.Negi@amd.com >
[ROCm/rccl-tests commit: b1f86ea6eb
2023-12-06 17:24:37 -06:00
5070d67f9a
Offload arch linking ( #54 )
...
* Update CMakeLists.txt
* Update CMakeLists.txt
* Link rccl_common object against hip::device
Previously the tests were compiled with `--amdgpu-target` to compile for multiple architectures, As rccl_common was not compiled against those architectures, this didn't work. Linking it against hip::device automatically links against all architectures in `AMDGPU_TARGETS`, and so are the test executables.
[ROCm/rccl-tests commit: e1a816b869
2023-12-05 19:20:46 -06:00
c07676388b
Fixing hipcc location for develop CI ( #52 )
...
[ROCm/rccl-tests commit: 8bfb67faf3
2023-10-19 13:29:42 -06:00
ccad358bc9
Warm up both out-of-place and in-place collectives ( #51 )
...
[ROCm/rccl-tests commit: 5ee7a08994
2023-10-16 12:13:50 -07:00
fa4dd9dbf0
Fixing hipcc location for CI ( #47 )
...
[ROCm/rccl-tests commit: 46375b1c52
2023-09-22 14:38:31 -06:00
0232bf4300
Update Makefile - HIPCC Path Updated to latest ( #46 )
...
[ROCm/rccl-tests commit: c1ec0c8aaf
2023-08-04 19:42:33 -04:00
2e12f0cfce
Update Makefile - HIPCC Path Updated to latest ( #45 )
...
[ROCm/rccl-tests commit: a6593375bc
2023-08-04 19:33:39 -04:00
aaae644862
search SLES install paths for MPI
...
[ROCm/rccl-tests commit: efdd4ad40b
2023-07-25 19:29:13 +00:00
84e9a3bbfc
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.
[ROCm/rccl-tests commit: 8fc00ec32e
2023-07-25 19:28:35 +00:00
3ff4890b6f
auto-detect and enable MPI
...
[ROCm/rccl-tests commit: c96ff57ac7
2023-07-25 19:27:13 +00:00
1b21a6fdd8
search SLES install paths for MPI
...
[ROCm/rccl-tests commit: 6048078be2
2023-07-24 12:02:44 -07:00
431f21ca11
Remove hardcoded number of GPUs limit for alltoallv ( #41 )
...
[ROCm/rccl-tests commit: fcd0888d53
2023-06-18 18:07:29 -07:00
a87b64d7c0
Fix merge error
...
[ROCm/rccl-tests commit: 652a24d38d
2023-06-14 20:26:33 +00:00
a2e8a44adb
Merge remote-tracking branch 'nccl/master' into develop
...
[ROCm/rccl-tests commit: bb0f15d407
2023-06-14 08:21:02 -07:00
7820b9f63f
Merge remote-tracking branch 'origin/master' into develop
...
[ROCm/rccl-tests commit: 469225bcaf
2023-06-14 08:01:50 -07:00
cba8bfd093
makefile: remove extra space
...
[ROCm/rccl-tests commit: 52ea1b2148
2023-06-06 09:47:50 +00:00
aa5f75aa9c
fixing the error message for mpirun when number of requested GPUs exceeds the limits ( #37 )
...
[ROCm/rccl-tests commit: d16d1fb16b
2023-04-27 14:06:17 -04:00
b7054b500a
Revert "fixing the error message for mpirun when number of requested GPUs exceeds the limits ( #33 )" ( #36 )
...
This reverts commit 39f83d5bb7e856fa720f
2023-04-25 13:44:43 -04:00
39f83d5bb7
fixing the error message for mpirun when number of requested GPUs exceeds the limits ( #33 )
...
[ROCm/rccl-tests commit: e146460810
2023-04-03 11:37:13 -04:00
4fd5ceeba8
fix handling of variable NVCC. Permit overriding the variable using environment variables
...
[ROCm/rccl-tests commit: 7ccda3c97b
2023-03-25 16:56:16 -03:00
b3db782c3f
Update README.md
...
[ROCm/rccl-tests commit: 17d0a42d5a
2023-03-23 09:05:41 -07:00
e86966907d
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.
[ROCm/rccl-tests commit: bdf58b1656
2023-03-02 19:40:13 +00:00
cd256748c0
Adding -pthread flag for linking issues into CMakeLists.txt and src/Makefile ( #31 )
...
[ROCm/rccl-tests commit: 255750b094
2023-03-02 11:05:25 -05:00
ed7707fa1b
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
[ROCm/rccl-tests commit: 5275aa5715
2023-02-24 21:39:04 -05:00
cc584c4825
auto-detect and enable MPI
...
[ROCm/rccl-tests commit: 2b2f23f42d
2023-02-23 18:27:08 +00:00
b70cac2b33
Update README.md
...
Improve MPI example to avoid confusion of number of processes / total number of GPUs.
https://github.com/NVIDIA/nccl-tests/issues/54#issuecomment-1212023369
[ROCm/rccl-tests commit: 2cbb968101
2023-01-03 08:47:43 +01:00
129a1b4b78
Add boot_id to the hostname hash due to collisions on Azure
...
Fixes #60
[ROCm/rccl-tests commit: 0b4c4cb99f
2022-12-12 01:16:46 -08:00
5ba670d551
Use DJB2a hash algorithm in getHostHash()
...
[ROCm/rccl-tests commit: 0aeba157db
2022-12-12 01:16:38 -08:00
84ee112d50
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.
[ROCm/rccl-tests commit: e9f5be184c
2022-11-30 23:01:46 +00:00
6313530fcc
Call cudaFreeHost() on wrongPerGpu not cudaFree()
...
[ROCm/rccl-tests commit: 24fcf64ed1
2022-11-22 11:18:37 -08:00
04b5c40b1c
Add fflush(stdout) before perf output
...
[ROCm/rccl-tests commit: 3bd2bd292b
2022-11-22 11:16:47 -08:00
1db5cf9986
added std::max to avoid buffer overflow in printing ( #25 )
...
[ROCm/rccl-tests commit: 9d3a53dfa3
2022-11-01 11:34:55 -06:00
3c2ff83a6a
make cmake stage also pass in CI
...
the subdir entry is not actually required for the compilation.
[ROCm/rccl-tests commit: 377b28e5fb
2022-10-31 22:07:15 +00:00
78706b1a95
add the rccl/lib directory to the link path
...
[ROCm/rccl-tests commit: 9c9746739a
2022-10-31 19:01:22 +00:00
9b06fe6fd5
fix a messing endif statement
...
error introduced with the web merger-resolution tool :-(
[ROCm/rccl-tests commit: 8a754f15ad
2022-10-25 16:31:57 +00:00
d76b2db507
Merge branch 'develop' into topic/v2.13.4-sync
...
[ROCm/rccl-tests commit: 4d7cd871c1
2022-10-21 17:12:45 -05:00
036145c75f
Allow more precise measurements of single operation ( #20 )
...
[ROCm/rccl-tests commit: 9a89c300b6
2022-10-21 22:07:41 +00:00
fab39367ae
make rccl-test compile again.
...
all files compile now.
mpi tests also pass
[ROCm/rccl-tests commit: 641e93e99c
2022-10-21 22:07:33 +00:00
08f9435e5a
Merge remote-tracking branch 'nccl-tests/master' into topic/v2.13.4-sync
...
[ROCm/rccl-tests commit: 3ae371cce7
2022-10-14 16:02:54 -05:00
d6264a0f11
Allow more precise measurements of single operation ( #20 )
...
[ROCm/rccl-tests commit: d22281cb3f
2022-10-12 17:28:04 -07:00
c0e3f4d443
Fix build on RHEL7 with GCC 4.8
...
Add -std=c++11 to CXXFLAGS.
Fixes #116 .
[ROCm/rccl-tests commit: 365b92a1ea
2022-10-12 01:24:14 -07:00
61fd36c006
removed hypercube from Makefile ( #19 )
...
[ROCm/rccl-tests commit: 3fbd3280ce
2022-09-29 15:36:39 -06:00
fdaa88710b
Update NCCL tests
...
[ROCm/rccl-tests commit: d313d20a26
2022-09-23 01:13:29 -07:00
35ee4ec3eb
Fix preprocessor version check for ncclGetLastError()
...
ncclGetLastError() was added in NCCL 2.13.0
[ROCm/rccl-tests commit: 749573f2d6
2022-09-07 16:10:41 -07:00
a43863e1a7
Fix an issue with the last commit when data checking is disabled
...
[ROCm/rccl-tests commit: afa4c56b6a
2022-09-07 11:23:49 -07:00
59ed17798f
Display N/A for error count in AlltoAll in-place test
...
AlltoAll does not support in-place buffers
[ROCm/rccl-tests commit: a0a14911ee
2022-09-06 13:17:15 -07:00
70b6c0f5e5
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.
[ROCm/rccl-tests commit: bc5f7cfb0a
2022-08-23 10:08:49 -07:00
Wenkai Du
akolliasAMD
Nusrat Islam
Nilesh M Negi
Lauren Wrubleski
Bertan Dogancay
gilbertlee-amd
arvindcheru
Edgar Gabriel
Wenkai Du
yangxingwu
Pedram Alizadeh
alan.souza
Felix Abecassis
Sylvain Jeaugey
David Addison
Jithin Jose
akolliasAMD
Edgar Gabriel
John Bachan