minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

[CI] Preserve NCCL 2.30.7 after dependency installs (#37508)

Browse Source
This commit is contained in:
Mohammad Miadh Angkad
2026-09-01 19:59:13 -07:00
committed by GitHub
parent 403a15c163
commit d9ae1ebabd
1 changed files with 23 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/ci/cuda/ci_install_dependency.sh
+23 -3
View File
@@ -539,6 +539,8 @@ install_sglang() {
install_nccl() { install_nccl() {
if [ "$CU_MAJOR" = "13" ]; then if [ "$CU_MAJOR" = "13" ]; then
# PyTorch pins 2.29.7, so this override must run after every command
# that resolves Python dependencies (including lmms-eval).
$PIP_CMD install "nvidia-nccl-cu13==2.30.7" \ $PIP_CMD install "nvidia-nccl-cu13==2.30.7" \
--force-reinstall --no-deps $PIP_INSTALL_SUFFIX --force-reinstall --no-deps $PIP_INSTALL_SUFFIX
else else
@@ -827,6 +829,24 @@ verify_imports() {
# One process; torch/cutlass do not import sglang, so the find_spec check # One process; torch/cutlass do not import sglang, so the find_spec check
# still runs ahead of any sglang import. # still runs ahead of any sglang import.
SGLANG_EXPECTED_INIT="${REPO_ROOT}/python/sglang/__init__.py" python3 -c ' SGLANG_EXPECTED_INIT="${REPO_ROOT}/python/sglang/__init__.py" python3 -c '
import ctypes
import importlib.metadata
import os
import sys
if sys.argv[1] == "13":
if importlib.metadata.version("nvidia-nccl-cu13") != "2.30.7":
raise SystemExit("nvidia-nccl-cu13 was changed after the final CI override")
nccl = ctypes.CDLL("libnccl.so.2")
nccl_version = ctypes.c_int()
status = nccl.ncclGetVersion(ctypes.byref(nccl_version))
if status != 0 or nccl_version.value != 23007:
raise SystemExit(
f"expected NCCL runtime 2.30.7, got status={status}, "
f"raw_version={nccl_version.value}"
)
print("NCCL package and runtime versions are 2.30.7")
import torch import torch
print(torch.version.cuda) print(torch.version.cuda)
import deep_ep import deep_ep
@@ -837,7 +857,7 @@ import cutlass.cute
# A shadowed sglang still imports, so without this the failure only surfaces # A shadowed sglang still imports, so without this the failure only surfaces
# as a missing submodule during the test step. find_spec, not import: the # as a missing submodule during the test step. find_spec, not import: the
# finders alone answer this without importing sglang. # finders alone answer this without importing sglang.
import importlib.util, os import importlib.util
want = os.environ["SGLANG_EXPECTED_INIT"] want = os.environ["SGLANG_EXPECTED_INIT"]
spec = importlib.util.find_spec("sglang") spec = importlib.util.find_spec("sglang")
if spec is None: if spec is None:
@@ -860,7 +880,7 @@ for mod in ("server", "grpc", "multimodal"):
except Exception as exc: except Exception as exc:
raise SystemExit(f"{name} is present but does not load: {exc!r}") raise SystemExit(f"{name} is present but does not load: {exc!r}")
print(f"{name} loads") print(f"{name} loads")
' ' "$CU_MAJOR"
mark_step_done "${FUNCNAME[0]}" mark_step_done "${FUNCNAME[0]}"
} }
@@ -886,7 +906,6 @@ main() {
setup_cargo_cache setup_cargo_cache
install_sglang install_sglang
release_cargo_cache_lock release_cargo_cache_lock
install_nccl
# Diffusion B200 CI imports torch inside install_sglang_kernel after removing # Diffusion B200 CI imports torch inside install_sglang_kernel after removing
# stale CUDA 12 NVIDIA wheels, so opt into one early LD_LIBRARY_PATH refresh. # stale CUDA 12 NVIDIA wheels, so opt into one early LD_LIBRARY_PATH refresh.
if [ "${SGLANG_CI_EARLY_LD_LIBRARY_PATH:-0}" = "1" ]; then if [ "${SGLANG_CI_EARLY_LD_LIBRARY_PATH:-0}" = "1" ]; then
@@ -899,6 +918,7 @@ main() {
stabilize_flashinfer_jit_paths stabilize_flashinfer_jit_paths
install_extra_deps install_extra_deps
install_test_tools install_test_tools
install_nccl
prepare_runner prepare_runner
setup_ld_library_path setup_ld_library_path
verify_imports verify_imports