Skip to content

fix: compile flags for trtllm fmha_v2 #2175

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

Merged
yzh119 merged 2 commits into from
Dec 5, 2025
Merged

fix: compile flags for trtllm fmha_v2 #2175

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
bbfd407
ampere and hopper flags and cleanup
jimmyzho Dec 4, 2025
1e7fe36
only sm120
jimmyzho Dec 4, 2025
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
Empty file modified .pre-commit-config.yaml
100644 → 100755
View file
Open in desktop
Empty file.
4 changes: 2 additions & 2 deletions flashinfer/jit/attention/fmha_v2/generator_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3711,10 +3711,10 @@ def generate_files(specs_names):
]
if "CUDA_PATH" in os.environ:
cmd[0] = os.environ["CUDA_PATH"] + "/bin/" + cmd[0]
print('Running command "{}" to build "bin/print_traits.exe":'.format(" ".join(cmd)))
# print('Running command "{}" to build "bin/print_traits.exe":'.format(" ".join(cmd)))

@yzh119 yzh119 Dec 4, 2025

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are these changes relevant to the PR?

@jimmyzho jimmyzho Dec 4, 2025

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

not really, just commenting these out from the original trtllm script to clean up the stdout

process = subprocess.Popen(cmd, stdin=subprocess.PIPE, stdout=subprocess.PIPE)
output, error = process.communicate()
print('Running "bin/print_traits.exe":')
# print('Running "bin/print_traits.exe":')
Comment on lines +3714 to +3717

@gemini-code-assist gemini-code-assist Bot Dec 4, 2025

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

Instead of commenting out these print statements, consider using the logging module. This allows for more flexible control over verbosity (e.g., via log levels like INFO or DEBUG) and is a better practice for maintainability. The information about the commands being run is valuable for debugging the kernel generation process.

process = subprocess.Popen(
"bin/print_traits.exe", stdin=subprocess.PIPE, stdout=subprocess.PIPE
)
Expand Down
3 changes: 2 additions & 1 deletion flashinfer/jit/attention/modules.py
100644 → 100755
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1901,9 +1901,10 @@ def gen_trtllm_fmha_v2_module() -> JitSpec:
source_paths = kernel_paths + [binding_source_path]

nvcc_flags = current_compilation_context.get_nvcc_flags_list(
supported_major_versions=[10, 11, 12]
supported_major_versions=[8, 9, 10, 11, 12]
Comment thread
yzh119 marked this conversation as resolved.
Outdated
)
nvcc_flags.append(f"-I{jit_env.FLASHINFER_CSRC_DIR / 'fmha_v2'}")
nvcc_flags.append("-Wno-deprecated-gpu-targets")

return gen_jit_spec(
uri,
Expand Down