Install DeepEP from release wheels (#33932)

This commit is contained in:
Baizhou Zhang
2026-08-07 15:38:44 -07:00
committed by GitHub
parent 115cd7bde1
commit eb3cc879e0
31 changed files with 141 additions and 353 deletions
+3 -51
View File
@@ -59,8 +59,6 @@ env:
SGLANG_CUDA_COREDUMP: "1"
SGLANG_JIT_DEEPGEMM_FAST_WARMUP: true
SKIP_PR_TEST_HEALTH_CHECK: ${{ (inputs.skip_pr_test_health_check == true || inputs.test_parallel_dispatch == true || inputs.run_all_tests == true) && 'true' || 'false' }}
# TEMP: force DeepEP rebuilds during the initial Torch 2.13 rollout so every CI runner replaces its Torch 2.11 build; remove after the runner fleet has cycled.
FORCE_REBUILD_DEEPEP: '1'
# Schedule / main-branch dispatch / workflow_call from main use refs/heads/main; PR events use refs/pull/*/merge
PR_TEST_BYPASS_MAINTENANCE_ON_MAIN: ${{ github.ref == 'refs/heads/main' && 'true' || 'false' }}
USE_VENV: false
@@ -400,6 +398,8 @@ jobs:
caller_inputs: ${{ toJson(inputs) }}
partitions: ${{ needs.check-changes.outputs.partitions }}
run_timeout_minutes: '30'
warmup_deep_gemm_models: 'lmsys/sglang-ci-dsv3-test:4'
warmup_server_models: 'lmsys/sglang-ci-dsv3-test:4'
rust_ext_artifact: ${{ needs.rust-ext-build.outputs.artifact_name }}
secrets: inherit
@@ -414,6 +414,7 @@ jobs:
caller_inputs: ${{ toJson(inputs) }}
partitions: ${{ needs.check-changes.outputs.partitions }}
run_timeout_minutes: '30'
timeout_per_file: '1800'
# Per-model TP must match the test's launch in test/registered/ -- see
# FALLBACK_ARGS in scripts/ci/cuda/warmup_deep_gemm.py for extra dp/ep
# flags. Only models that actually invoke DeepGEMM kernels at runtime
@@ -438,52 +439,6 @@ jobs:
rust_ext_artifact: ${{ needs.rust-ext-build.outputs.artifact_name }}
secrets: inherit
base-c-test-deepep-4-gpu-h100:
needs: [check-changes, call-gate, wait-for-base-b, sgl-kernel-build-wheels, rust-ext-build]
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/_pr-test-stage.yml
with:
self_name: base-c-test-deepep-4-gpu-h100
runner_config: deepep-4-gpu-h100
check_changes: ${{ toJson(needs.check-changes.outputs) }}
caller_inputs: ${{ toJson(inputs) }}
partitions: ${{ needs.check-changes.outputs.partitions }}
run_timeout_minutes: '30'
warmup_deep_gemm_models: 'lmsys/sglang-ci-dsv3-test:4'
warmup_server_models: 'lmsys/sglang-ci-dsv3-test:4'
rust_ext_artifact: ${{ needs.rust-ext-build.outputs.artifact_name }}
secrets: inherit
base-c-test-deepep-4-gpu-b200:
needs: [check-changes, call-gate, wait-for-base-b, sgl-kernel-build-wheels, rust-ext-build]
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/_pr-test-stage.yml
with:
self_name: base-c-test-deepep-4-gpu-b200
runner_config: deepep-4-gpu-b200
check_changes: ${{ toJson(needs.check-changes.outputs) }}
caller_inputs: ${{ toJson(inputs) }}
partitions: ${{ needs.check-changes.outputs.partitions }}
run_timeout_minutes: '30'
timeout_per_file: '1800'
rust_ext_artifact: ${{ needs.rust-ext-build.outputs.artifact_name }}
secrets: inherit
base-c-test-deepep-8-gpu-h200:
needs: [check-changes, call-gate, wait-for-base-b, sgl-kernel-build-wheels, rust-ext-build]
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/_pr-test-stage.yml
with:
self_name: base-c-test-deepep-8-gpu-h200
runner_config: deepep-8-gpu-h200
check_changes: ${{ toJson(needs.check-changes.outputs) }}
caller_inputs: ${{ toJson(inputs) }}
partitions: ${{ needs.check-changes.outputs.partitions }}
run_timeout_minutes: '30'
timeout_per_file: '1800'
rust_ext_artifact: ${{ needs.rust-ext-build.outputs.artifact_name }}
secrets: inherit
base-c-test-4-gpu-b200:
needs: [check-changes, call-gate, wait-for-base-b, sgl-kernel-build-wheels, rust-ext-build]
if: ${{ !failure() && !cancelled() }}
@@ -561,9 +516,6 @@ jobs:
base-c-test-4-gpu-h100,
base-c-test-8-gpu-h20,
base-c-test-8-gpu-h200,
base-c-test-deepep-4-gpu-h100,
base-c-test-deepep-4-gpu-b200,
base-c-test-deepep-8-gpu-h200,
base-c-test-4-gpu-b200,
base-c-test-4-gpu-gb300,
base-c-test-8-gpu-b300,