microsoft/mscclpp
1
0
Fork 0
mirror of https://github.com/microsoft/mscclpp.git synced 2026-04-20 06:49:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

updates

Browse Source
This commit is contained in:
Changho Hwang
2026-03-18 20:30:18 +00:00
parent 6082648f80
commit 79a014976d
18 changed files with 68 additions and 68 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
.azure-pipelines/codecov.yml
View File

@@ -43,7 +43,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/codecov.yaml
- template: templates/codecov.yml
parameters:
subscription: mscclpp-ci
vmssName: mscclpp-ci
@@ -64,7 +64,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/codecov.yaml
- template: templates/codecov.yml
parameters:
subscription: mscclpp-ci-h100
vmssName: mscclpp-h100-ci
@@ -85,7 +85,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/codecov.yaml
- template: templates/codecov.yml
parameters:
subscription: mscclpp-ci-mi300x
vmssName: mscclpp-mi300x-ci

4
.azure-pipelines/integration-test.yml
View File

@@ -41,7 +41,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/integration-test.yaml
- template: templates/integration-test.yml
parameters:
subscription: mscclpp-ci
vmssName: mscclpp-ci
@@ -60,7 +60,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/integration-test.yaml
- template: templates/integration-test.yml
parameters:
subscription: mscclpp-ci-h100
vmssName: mscclpp-h100-ci

12
.azure-pipelines/multi-nodes-test.yml
View File

@@ -50,13 +50,13 @@ jobs:
echo "Entry already exists, nothing to do."
fi
- template: templates/deploy.yaml
- template: templates/deploy.yml
parameters:
subscription: msccl-it
vmssName: mscclit-vmss
resourceGroup: msccl-IT
- template: templates/run-remote-task.yaml
- template: templates/run-remote-task.yml
parameters:
name: RunMscclppTest
displayName: Run multi-nodes mscclpp-test
@@ -64,7 +64,7 @@ jobs:
remoteScript: |
bash /root/mscclpp/test/deploy/run_tests.sh mscclpp-test
- template: templates/run-remote-task.yaml
- template: templates/run-remote-task.yml
parameters:
name: RunMultiNodeUnitTest
displayName: Run multi-nodes unit tests
@@ -72,7 +72,7 @@ jobs:
remoteScript: |
bash /root/mscclpp/test/deploy/run_tests.sh mp-ut
- template: templates/run-remote-task.yaml
- template: templates/run-remote-task.yml
parameters:
name: RunMultiNodePythonTests
displayName: Run multi-nodes python tests
@@ -80,7 +80,7 @@ jobs:
remoteScript: |
bash /root/mscclpp/test/deploy/run_tests.sh pytests
- template: templates/run-remote-task.yaml
- template: templates/run-remote-task.yml
parameters:
name: RunMultiNodePythonBenchmark
displayName: Run multi-nodes python benchmark
@@ -88,7 +88,7 @@ jobs:
remoteScript: |
bash /root/mscclpp/test/deploy/run_tests.sh py-benchmark
- template: templates/stop.yaml
- template: templates/stop.yml
parameters:
subscription: msccl-it
vmssName: mscclit-vmss

4
.azure-pipelines/nccl-api-test.yaml → .azure-pipelines/nccl-api-test.yml
View File

@@ -40,7 +40,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/nccl-test.yaml
- template: templates/nccl-test.yml
parameters:
subscription: mscclpp-ci
vmssName: mscclpp-ci
@@ -60,7 +60,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/nccl-test.yaml
- template: templates/nccl-test.yml
parameters:
subscription: mscclpp-ci-h100
vmssName: mscclpp-h100-ci

2
.azure-pipelines/rccl-api-test.yml
View File

@@ -40,7 +40,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/rccl-test.yaml
- template: templates/rccl-test.yml
parameters:
subscription: mscclpp-ci-mi300x
vmssName: mscclpp-mi300x-ci

8
.azure-pipelines/templates/codecov.yaml → .azure-pipelines/templates/codecov.yml
View File

@@ -10,7 +10,7 @@ parameters:
type: string
steps:
- template: deploy.yaml
- template: deploy.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}
@@ -22,7 +22,7 @@ steps:
buildName: BuildCoverage
deployArgs: 'single-node-test true ${{ parameters.platform }}'
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: TestsCoverageNonPerf
displayName: Run unit_tests + mp_unit_tests (non-perf) with coverage
@@ -44,7 +44,7 @@ steps:
mpirun --allow-run-as-root -tag-output -np 4 ./build/bin/mp_unit_tests --exclude-perf-tests
echo "mp_unit_tests -np 4: PASSED"
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: CaptureCoverage
displayName: Capture coverage data with lcov
@@ -104,7 +104,7 @@ steps:
./codecov upload-process --disable-search -t $(CODECOV_TOKEN) -f coverage.info --flag ${{ parameters.platform }}-${{ parameters.gpuArch }}
workingDirectory: '$(System.DefaultWorkingDirectory)'
- template: stop.yaml
- template: stop.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}

0
.azure-pipelines/templates/deploy.yaml → .azure-pipelines/templates/deploy.yml
View File

16
.azure-pipelines/templates/integration-test.yaml → .azure-pipelines/templates/integration-test.yml
View File

@@ -10,14 +10,14 @@ parameters:
type: string
steps:
- template: deploy.yaml
- template: deploy.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}
gpuArch: ${{ parameters.gpuArch }}
deployArgs: 'single-node-test'
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: AllGatherTest
displayName: Run mscclpp AllGather test
@@ -27,14 +27,14 @@ steps:
mpirun --allow-run-as-root -np 8 --bind-to numa -x MSCCLPP_DEBUG=WARN ./build/bin/mscclpp-test/allgather_test_perf -b 1K -e 1G -f 2 -k 2 -o output.jsonl
mpirun --allow-run-as-root -np 8 --bind-to numa -x MSCCLPP_DEBUG=WARN ./build/bin/mscclpp-test/allgather_test_perf -b 1K -e 1G -f 2 -k 3 -o output.jsonl
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: SendRecvTest
displayName: Run mscclpp SendRecv test
remoteScript: |
mpirun --allow-run-as-root -np 8 --bind-to numa -x MSCCLPP_DEBUG=WARN ./build/bin/mscclpp-test/sendrecv_test_perf -b 1K -e 1G -f 2 -o output.jsonl
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: AllReduceTest
displayName: Run mscclpp AllReduce test
@@ -47,7 +47,7 @@ steps:
mpirun --allow-run-as-root -np 8 --bind-to numa -x MSCCLPP_DEBUG=WARN ./build/bin/mscclpp-test/allreduce_test_perf -b 12M -e 48M -i 3145728 2 -k 5 -o output.jsonl
mpirun --allow-run-as-root -np 8 --bind-to numa -x MSCCLPP_DEBUG=WARN ./build/bin/mscclpp-test/allreduce_test_perf -b 24K -e 768K -i 24576 -k 6 -w 100 -n 100 -o output.jsonl
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: AllToAll
displayName: Run mscclpp AllToAll test
@@ -55,14 +55,14 @@ steps:
mpirun --allow-run-as-root -np 8 --bind-to numa -x MSCCLPP_DEBUG=WARN ./build/bin/mscclpp-test/alltoall_test_perf -b 1K -e 1G -f 2 -o output.jsonl
mpirun --allow-run-as-root -np 8 --bind-to numa -x MSCCLPP_DEBUG=WARN ./build/bin/mscclpp-test/alltoall_test_perf -b 1K -e 1G -f 2 -k 1 -o output.jsonl
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: CheckPerfNumber
displayName: Check collective primitives performance
remoteScript: |
python3 test/mscclpp-test/check_perf_result.py --perf-file output.jsonl --baseline-file ${{ parameters.perfBaselineFile }}
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: PythonAllReduceBenchmark
displayName: Python Allreduce Benchmark
@@ -70,7 +70,7 @@ steps:
python3 -m pip install .
mpirun --allow-run-as-root -tag-output -x MSCCLPP_HOME=/root/mscclpp -np 8 python3 ./python/mscclpp_benchmark/allreduce_bench.py
- template: stop.yaml
- template: stop.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}

16
.azure-pipelines/templates/nccl-test.yaml → .azure-pipelines/templates/nccl-test.yml
View File

@@ -1,4 +1,4 @@
# .azure-pipelines/templates/nccl-test.yaml
# .azure-pipelines/templates/nccl-test.yml
# ----------------------------------------
# A step‐template that runs the entire MSCCLPP→NCCL test suite on one pool/container.
#
@@ -15,13 +15,13 @@ parameters:
default: "-gencode=arch=compute_80,code=sm_80"
steps:
- template: deploy.yaml
- template: deploy.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}
deployArgs: 'nccltest-single-node'
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: InstallNcclTests
displayName: Install NCCL Tests
@@ -31,7 +31,7 @@ steps:
cd nccl-tests
MPI=1 MPI_HOME=/usr/local/mpi make -j
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: InstallNccl
displayName: Install NCCL
@@ -46,7 +46,7 @@ steps:
cd nccl
make -j src.build NVCC_GENCODE=${{ parameters.nvccGencode }}
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: RunNcclAllGatherFallbaclkToNcclTest
displayName: Run NCCL AllGather Test with or without Fallback to NCCL operation
@@ -54,7 +54,7 @@ steps:
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN -x MSCCLPP_ENABLE_NCCL_FALLBACK=TRUE -x MSCCLPP_NCCL_LIB_PATH=/root/nccl/build/lib/libnccl.so -x MSCCLPP_FORCE_NCCL_FALLBACK_OPERATION="allgather" /root/nccl-tests/build/all_gather_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN -x MSCCLPP_ENABLE_NCCL_FALLBACK=TRUE -x MSCCLPP_NCCL_LIB_PATH=/root/nccl/build/lib/libnccl.so -x MSCCLPP_FORCE_NCCL_FALLBACK_OPERATION="allreduce" /root/nccl-tests/build/all_gather_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: RunNcclAllReduceFallbaclkToNcclTest
displayName: Run NCCL AllReduce Test with or without Fallback to NCCL operation
@@ -62,7 +62,7 @@ steps:
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN -x MSCCLPP_ENABLE_NCCL_FALLBACK=TRUE -x MSCCLPP_NCCL_LIB_PATH=/root/nccl/build/lib/libnccl.so -x MSCCLPP_FORCE_NCCL_FALLBACK_OPERATION="allreduce" /root/nccl-tests/build/all_reduce_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN -x MSCCLPP_ENABLE_NCCL_FALLBACK=TRUE -x MSCCLPP_NCCL_LIB_PATH=/root/nccl/build/lib/libnccl.so -x MSCCLPP_FORCE_NCCL_FALLBACK_OPERATION="allgather" /root/nccl-tests/build/all_reduce_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: RunNcclBroadcastFallbaclkToNcclTest
displayName: Run NCCL Broadcast Test with or without Fallback to NCCL operation
@@ -70,7 +70,7 @@ steps:
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN -x MSCCLPP_ENABLE_NCCL_FALLBACK=TRUE -x MSCCLPP_NCCL_LIB_PATH=/root/nccl/build/lib/libnccl.so -x MSCCLPP_FORCE_NCCL_FALLBACK_OPERATION="broadcast" /root/nccl-tests/build/broadcast_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN -x MSCCLPP_ENABLE_NCCL_FALLBACK=TRUE -x MSCCLPP_NCCL_LIB_PATH=/root/nccl/build/lib/libnccl.so -x MSCCLPP_FORCE_NCCL_FALLBACK_OPERATION="allreduce" /root/nccl-tests/build/broadcast_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
- template: stop.yaml
- template: stop.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}

12
.azure-pipelines/templates/rccl-test.yaml → .azure-pipelines/templates/rccl-test.yml
View File

@@ -1,4 +1,4 @@
# .azure-pipelines/templates/rccl-test.yaml
# .azure-pipelines/templates/rccl-test.yml
# ------------------------------------------------
# A step-template that runs the entire MSCCLPP→RCCL test suite on one pool/container.
#
@@ -17,7 +17,7 @@ parameters:
default: "gfx942"
steps:
- template: deploy.yaml
- template: deploy.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}
@@ -27,7 +27,7 @@ steps:
deployArgs: 'single-node-test true rocm'
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: InstallRcclTests
displayName: Install RCCL Tests
@@ -41,7 +41,7 @@ steps:
cd projects/rccl-tests
MPI=1 MPI_HOME=/usr/local/mpi make -j
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: RunRcclAllGatherTest
displayName: Run RCCL AllGather Test with or without MSCCLPP Lib
@@ -49,7 +49,7 @@ steps:
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN /root/rocm-systems/projects/rccl-tests/build/all_gather_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
mpirun -np 8 --bind-to numa --allow-run-as-root /root/rocm-systems/projects/rccl-tests/build/all_gather_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: RunRcclAllReduceTest
displayName: Run RCCL AllReduce Test with or without MSCCLPP Lib
@@ -57,7 +57,7 @@ steps:
mpirun -np 8 --bind-to numa --allow-run-as-root -x LD_PRELOAD=/root/mscclpp/build/lib/libmscclpp_nccl.so -x MSCCLPP_NCCL_SYMMETRIC_MEMORY=1 -x NCCL_DEBUG=WARN /root/rocm-systems/projects/rccl-tests/build/all_reduce_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
mpirun -np 8 --bind-to numa --allow-run-as-root /root/rocm-systems/projects/rccl-tests/build/all_reduce_perf -b 1K -e 1G -f 2 -d half -G 20 -w 10 -n 20
- template: stop.yaml
- template: stop.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}

0
.azure-pipelines/templates/run-remote-task.yaml → .azure-pipelines/templates/run-remote-task.yml
View File

0
.azure-pipelines/templates/stop.yaml → .azure-pipelines/templates/stop.yml
View File

14
.azure-pipelines/templates/ut-no-ib-env.yaml → .azure-pipelines/templates/ut-no-ib-env.yml
View File

@@ -7,7 +7,7 @@ parameters:
type: string
steps:
- template: deploy.yaml
- template: deploy.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}
@@ -15,14 +15,14 @@ steps:
cmakeArgs: '-DMSCCLPP_USE_IB=OFF'
deployArgs: 'single-node-test false'
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: UnitTests
displayName: Run mscclpp unit tests
remoteScript: |
./build/bin/unit_tests
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: MpUnitTests
displayName: Run mscclpp multi-process unit tests
@@ -31,14 +31,14 @@ steps:
mpirun --allow-run-as-root -tag-output -np 4 ./build/bin/mp_unit_tests
mpirun --allow-run-as-root -tag-output -np 8 ./build/bin/mp_unit_tests
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: PyTests
displayName: Run pytests
remoteScript: |
mpirun --allow-run-as-root -tag-output -x MSCCLPP_HOME=/root/mscclpp -x MSCCLPP_DISABLE_IB_TESTS=1 -np 8 python3 -m pytest ./python/test/test_mscclpp.py -x
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: StopContainer
displayName: Stop existing container
@@ -82,14 +82,14 @@ steps:
arguments: single-node-test false
workingDirectory: $(System.DefaultWorkingDirectory)
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: PyTestsWithIbBuildDisableIb
displayName: Run pytests (IB build, IB tests disabled)
remoteScript: |
mpirun --allow-run-as-root -tag-output -x MSCCLPP_HOME=/root/mscclpp -x MSCCLPP_DISABLE_IB_TESTS=1 -np 8 python3 -m pytest ./python/test/test_mscclpp.py -x
- template: stop.yaml
- template: stop.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}

8
.azure-pipelines/templates/ut-npkit.yaml → .azure-pipelines/templates/ut-npkit.yml
View File

@@ -8,7 +8,7 @@ parameters:
steps:
- template: deploy.yaml
- template: deploy.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}
@@ -16,7 +16,7 @@ steps:
cmakeArgs: '-DMSCCLPP_NPKIT_FLAGS="-DENABLE_NPKIT -DENABLE_NPKIT_EVENT_TIME_SYNC_CPU -DENABLE_NPKIT_EVENT_TIME_SYNC_GPU -DENABLE_NPKIT_EVENT_EXECUTOR_INIT_ENTRY -DENABLE_NPKIT_EVENT_EXECUTOR_INIT_EXIT -DENABLE_NPKIT_EVENT_EXECUTOR_OP_BASE_ENTRY -DENABLE_NPKIT_EVENT_EXECUTOR_OP_BASE_EXIT"'
deployArgs: 'single-node-test'
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: MpUnitTests
displayName: Run mscclpp multi-process unit tests
@@ -30,7 +30,7 @@ steps:
grep -q NPKIT_EVENT_EXECUTOR_WAIT_ENTRY ./npkit_output/npkit_event_trace.json
grep -q NPKIT_EVENT_EXECUTOR_READ_REDUCE_COPY_SEND_ENTRY ./npkit_output/npkit_event_trace.json
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: PyTests
displayName: Run pytests
@@ -51,7 +51,7 @@ steps:
grep -q NPKIT_EVENT_EXECUTOR_PUT_PACKET_ENTRY ./npkit_output/npkit_event_trace.json
grep -q NPKIT_EVENT_EXECUTOR_REDUCE_SEND_PACKET_ENTRY ./npkit_output/npkit_event_trace.json
- template: stop.yaml
- template: stop.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}

10
.azure-pipelines/templates/ut.yaml → .azure-pipelines/templates/ut.yml
View File

@@ -10,7 +10,7 @@ parameters:
type: string
steps:
- template: deploy.yaml
- template: deploy.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}
@@ -19,14 +19,14 @@ steps:
deployArgs: 'single-node-test true ${{ parameters.platform }}'
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: UnitTests
displayName: Run mscclpp unit tests
remoteScript: |
./build/bin/unit_tests
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: MpUnitTests
displayName: Run mscclpp multi-process unit tests
@@ -35,14 +35,14 @@ steps:
mpirun --allow-run-as-root -tag-output -np 4 ./build/bin/mp_unit_tests
mpirun --allow-run-as-root -tag-output -np 8 ./build/bin/mp_unit_tests
- template: run-remote-task.yaml
- template: run-remote-task.yml
parameters:
name: PyTests
displayName: Run pytests
remoteScript: |
mpirun --allow-run-as-root -tag-output -x MSCCLPP_HOME=/root/mscclpp -x GPU_MAX_HW_QUEUES=8 -np 8 python3 -m pytest ./python/test/test_mscclpp.py -x
- template: stop.yaml
- template: stop.yml
parameters:
subscription: ${{ parameters.subscription }}
vmssName: ${{ parameters.vmssName }}

12
.azure-pipelines/ut.yml
View File

@@ -43,7 +43,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/ut.yaml
- template: templates/ut.yml
parameters:
subscription: mscclpp-ci
vmssName: mscclpp-ci
@@ -64,7 +64,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/ut-npkit.yaml
- template: templates/ut-npkit.yml
parameters:
subscription: mscclpp-ci
vmssName: mscclpp-ci
@@ -83,7 +83,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/ut.yaml
- template: templates/ut.yml
parameters:
subscription: mscclpp-ci-h100
vmssName: mscclpp-h100-ci
@@ -102,7 +102,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/ut-npkit.yaml
- template: templates/ut-npkit.yml
parameters:
subscription: mscclpp-ci-h100
vmssName: mscclpp-h100-ci
@@ -123,7 +123,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/ut-no-ib-env.yaml
- template: templates/ut-no-ib-env.yml
parameters:
subscription: mscclpp-ci-h100
vmssName: mscclpp-h100-ci
@@ -142,7 +142,7 @@ jobs:
image: $(containerImage)
steps:
- template: templates/ut.yaml
- template: templates/ut.yml
parameters:
subscription: mscclpp-ci-mi300x
vmssName: mscclpp-mi300x-ci

0
.github/workflows/doc-build.yaml → .github/workflows/doc-build.yml vendored
View File

12
README.md
View File

@@ -3,16 +3,16 @@
[![Latest Release](https://img.shields.io/github/release/microsoft/mscclpp.svg)](https://github.com/microsoft/mscclpp/releases/latest)
[![License](https://img.shields.io/github/license/microsoft/mscclpp.svg)](LICENSE)
[![CodeQL](https://github.com/microsoft/mscclpp/actions/workflows/codeql-analysis.yml/badge.svg?branch=main)](https://github.com/microsoft/mscclpp/actions/workflows/codeql-analysis.yml)
[![Docs Build](https://github.com/microsoft/mscclpp/actions/workflows/doc-build.yaml/badge.svg)](https://microsoft.github.io/mscclpp/)
[![Docs Build](https://github.com/microsoft/mscclpp/actions/workflows/doc-build.yml/badge.svg)](https://microsoft.github.io/mscclpp/)
[![codecov](https://codecov.io/gh/microsoft/mscclpp/graph/badge.svg?token=DAV9DGHAY2)](https://codecov.io/gh/microsoft/mscclpp)
| Testing Pipelines | Build Status |
|--------------------------|-------------------|
| Unit Tests (CUDA) | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-ut?branchName=main)](https://msazure.visualstudio.com/One/_build/latest?definitionId=398325&branchName=main) |
| Integration Tests (CUDA) | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-test?branchName=main)](https://msazure.visualstudio.com/One/_build/latest?definitionId=398479&branchName=main) |
| Unit Tests (ROCm) | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-ut-rocm?branchName=main)](https://msazure.visualstudio.com/One/_build/latest?definitionId=399295&branchName=main) |
| NCCL Tests | [![Build Status](https://dev.azure.com/msazure/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-nccl?branchName=main)](https://dev.azure.com/msazure/One/_build/latest?definitionId=320665&branchName=main) |
| RCCL Tests | [![Build Status](https://dev.azure.com/msazure/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-rccl?branchName=main)](https://dev.azure.com/msazure/One/_build/latest?definitionId=448013&branchName=main) |
| Unit Tests (CUDA) | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-ut?branchName=main&jobName=UnitTestH100)](https://msazure.visualstudio.com/One/_build/latest?definitionId=398325&branchName=main) |
| Unit Tests (ROCm) | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-ut?branchName=main&jobName=UnitTestMI300X)](https://msazure.visualstudio.com/One/_build/latest?definitionId=398325&branchName=main) |
| Integration Tests (CUDA) | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-test?branchName=main&jobName=Integration%20test%20H100)](https://msazure.visualstudio.com/One/_build/latest?definitionId=398479&branchName=main) |
| NCCL Tests | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-nccl?repoName=microsoft%2Fmscclpp&branchName=main&jobName=Run%20MSCCLPP%20over%20NCCL%20Test%20(H100))](https://msazure.visualstudio.com/One/_build/latest?definitionId=320665&repoName=microsoft%2Fmscclpp&branchName=main) |
| RCCL Tests | [![Build Status](https://msazure.visualstudio.com/One/_apis/build/status%2FCustom%2FMSCCLPP%2Fmscclpp-rccl?branchName=main&jobName=Run%20MSCCLPP%20over%20RCCL%20Test%20(MI300X))](https://msazure.visualstudio.com/One/_build/latest?definitionId=448013&branchName=main) |
A GPU-driven communication stack for scalable AI applications.