Skip to content

llm release with logging #16496

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

Open
JacobSzwejbka wants to merge 1 commit into
base: main
Choose a base branch
from
Open

llm release with logging #16496

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
34 changes: 34 additions & 0 deletions CMakePresets.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -145,6 +145,17 @@
"CMAKE_INSTALL_PREFIX": "${sourceDir}/cmake-out"
}
},
{
"name": "llm-release-logging",
Copy link
Contributor

@Gasoonjia Gasoonjia Jan 7, 2026

Choose a reason for hiding this comment

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

can we just concentrate all things inside debug mode? Now we have too much different modes

Copy link
Contributor Author

@JacobSzwejbka JacobSzwejbka Jan 7, 2026 •
edited
Loading

Choose a reason for hiding this comment

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

Right now we have logging coupled with optimizations (debug mode). Since we are for some reason using ET logs to report useful step information to the user in the examples binaries it would be useful to have these be independent decisions. Debug mode is at least twice as slow as release for cpu portable ops

"displayName": "LLM release build with logging enabled",
"inherits": [
"llm-release"
],
"cacheVariables": {
"EXECUTORCH_ENABLE_LOGGING": "ON",
"ET_MIN_LOG_LEVEL": "Info"
}
},
{
"name": "llm-release-cuda",
"displayName": "LLM release build with CUDA",
Expand Down Expand Up @@ -294,6 +305,15 @@
],
"jobs": 0
},
{
"name": "llm-release-logging-install",
"displayName": "Build and install LLM extension release artifacts (with logging)",
"configurePreset": "llm-release-logging",
"targets": [
"install"
],
"jobs": 0
},
{
"name": "llm-debug-install",
"displayName": "Build and install LLM extension debug artifacts",
Expand Down Expand Up @@ -365,6 +385,20 @@
}
]
},
{
"name": "llm-release-logging",
Copy link
Contributor

@Gasoonjia Gasoonjia Jan 7, 2026

Choose a reason for hiding this comment

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

what's the difference between this and the llm-release-logging above?

Copy link
Contributor Author

@JacobSzwejbka JacobSzwejbka Jan 7, 2026

Choose a reason for hiding this comment

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

The other stuff lets you do

cmake --workflow --preset llm-release-logging

vs

cmake --preset llm-release-logging
cmake --build --preset llm-release-logging-install

"displayName": "Configure, build and install ExecuTorch LLM extension with logging enabled",
"steps": [
{
"type": "configure",
"name": "llm-release-logging"
},
{
"type": "build",
"name": "llm-release-logging-install"
}
]
},
{
"name": "llm-debug",
"displayName": "Configure, build and install ExecuTorch LLM extension with default CPU backend (Debug)",
Expand Down
Loading