Skip to content

[None][infra] Disable fmha warmup to reduce non-perf test idle time #16416

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
pengbowang-nv wants to merge 1 commit into
base: main
Choose a base branch
from
+8 −0
Draft

[None][infra] Disable fmha warmup to reduce non-perf test idle time #16416

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions jenkins/L0_Test.groovy
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1135,6 +1135,10 @@ def getPytestBaseCommandLine(
extraInternalEnv += " CBTS_COVERAGE_CONFIG=${coverageConfigFile}"
extraInternalEnv += " CBTS_MARKER_FILE=${outputPath}/cbts_current_test.txt"
}
// Skip TRTLLM-Gen FMHA JIT warmup for non-performance tests.
if (!stageName.contains("-Perf-") && !stageName.contains("-PerfSanity-")) {
extraInternalEnv += " TLLM_ENABLE_TRTLLM_GEN_FMHA_JIT_WARMUP=0"
}

// Container port allocation environment variables for avoiding port conflicts
def portEnvVars = ""
Expand Down
4 changes: 4 additions & 0 deletions tensorrt_llm/_torch/pyexecutor/model_engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1280,6 +1280,10 @@ def _run_attention_warmup(self,
if not issubclass(self.attn_backend.Metadata, TrtllmAttentionMetadata):
return

if os.environ.get("TLLM_ENABLE_TRTLLM_GEN_FMHA_JIT_WARMUP", "1") != "1":
logger.info("Skipping TRTLLM-Gen FMHA JIT warmup")
return

@contextlib.contextmanager
def trtllm_gen_fmha_jit_warmup():
previous = self._trtllm_gen_jit_warmup
Expand Down
Loading