biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[CI] Breakup h200 tests (#30499)

Browse Source 
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
This commit is contained in:
Lucas Wilkinson
2026-01-15 23:23:22 -07:00
committed by GitHub
parent 4ae77dfd42
commit 14ce524249
5 changed files with 390 additions and 296 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/env_override.py
View File

@@ -95,7 +95,7 @@ def memory_plan_reuse_patched(self):
# ===================================================
# This change monkeypatches get_graph_partition_signature in pytorch 2.9.0 to
# fix inductor partition + attention-nvfp4 quant fusion, tested in
# `tests/compile/distributed/test_fusions_e2e.py::test_attn_quant`.
# `tests/compile/test_fusion_attn.py::test_attn_quant`.
# For more context, see https://github.com/pytorch/pytorch/pull/165815.