Skip to content

Add --cuda-event-trace=false to nsys command #180

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
hemildesai merged 2 commits into from
Apr 8, 2025
Merged

Add --cuda-event-trace=false to nsys command #180

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
88e2fa5
Set cuda event trace to false in nsys
guyueh1 Apr 8, 2025
71948c2
Formatting
guyueh1 Apr 8, 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
16 changes: 10 additions & 6 deletions nemo_run/core/execution/launcher.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,15 @@ class Launcher(ConfigurableMixin):
nsys_profile: bool = False
nsys_folder: str = "nsys_profile"
nsys_trace: list[str] = field(default_factory=lambda: ["nvtx", "cuda"])
nsys_extra_args: list[str] = field(
default_factory=lambda: [
"--force-overwrite=true",
"--capture-range=cudaProfilerApi",
"--capture-range-end=stop",
"--cuda-graph-trace=node",
"--cuda-event-trace=false",
]
)

def get_nsys_prefix(self, profile_dir: str) -> Optional[list[str]]:
"""Make a command prefix for nsys profiling"""
Expand All @@ -27,12 +36,7 @@ def get_nsys_prefix(self, profile_dir: str) -> Optional[list[str]]:
",".join(self.nsys_trace),
"-o",
f"{profile_out_path}/profile_%p",
"--force-overwrite",
"true",
"--capture-range=cudaProfilerApi",
"--capture-range-end=stop",
"--cuda-graph-trace=node",
]
] + self.nsys_extra_args
return args

def transform(self, cmd: list[str]) -> Optional[Script]: ...
Expand Down
2 changes: 1 addition & 1 deletion nemo_run/package_info.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@
# limitations under the License.
from packaging.version import Version

__version__ = '0.5.0rc0.dev0'
__version__ = "0.5.0rc0.dev0"

MAJOR = Version(__version__).major
MINOR = Version(__version__).minor
Expand Down
4 changes: 2 additions & 2 deletions test/core/execution/test_slurm_templates.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -470,11 +470,11 @@ def test_dummy_batch_request_nsys(
"nvtx,cuda",
"-o",
"/nemo_run/nsys_profile/profile_%p",
"--force-overwrite",
"true",
"--force-overwrite=true",
"--capture-range=cudaProfilerApi",
"--capture-range-end=stop",
"--cuda-graph-trace=node",
"--cuda-event-trace=false",
]

def test_dummy_batch_request_warn(
Expand Down
Loading