Default to cores/2 threads in JNI layer #6042
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/6042
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit c8c1650 with merge base 866b40c ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Oct 9, 2024
Summary: Default to using cores/2 threadpool threads. The long-term plan is to improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default. Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using all cores can be 10x slower than using cores/2. This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware. Differential Revision: D64107326
GregoryComer
force-pushed
the
export-D64107326
branch
from
20ebaf3 to
4b5639d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Oct 9, 2024
Summary: Default to using cores/2 threadpool threads. The long-term plan is to improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default. Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using all cores can be 10x slower than using cores/2. This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware. Differential Revision: D64107326
GregoryComer
force-pushed
the
export-D64107326
branch
from
4b5639d to
c43f3fc
Compare
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Oct 9, 2024
Summary: Default to using cores/2 threadpool threads. The long-term plan is to improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default. Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using all cores can be 10x slower than using cores/2. This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware. Differential Revision: D64107326
GregoryComer
force-pushed
the
export-D64107326
branch
from
c43f3fc to
2204a7e
Compare
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
kirklandsign
approved these changes
Oct 9, 2024
|
Please fix lintrunner |
GregoryComer
force-pushed
the
export-D64107326
branch
from
2204a7e to
b1cc8f7
Compare
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Oct 9, 2024
Summary: Default to using cores/2 threadpool threads. The long-term plan is to improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default. Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using all cores can be 10x slower than using cores/2. This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware. Differential Revision: D64107326
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Oct 9, 2024
Summary: Default to using cores/2 threadpool threads. The long-term plan is to improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default. Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using all cores can be 10x slower than using cores/2. This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware. Differential Revision: D64107326
GregoryComer
force-pushed
the
export-D64107326
branch
from
b1cc8f7 to
064c095
Compare
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Oct 9, 2024
Summary: Default to using cores/2 threadpool threads. The long-term plan is to improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default. Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using all cores can be 10x slower than using cores/2. This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware. Reviewed By: kirklandsign Differential Revision: D64107326
GregoryComer
force-pushed
the
export-D64107326
branch
from
064c095 to
bf88166
Compare
Summary: Default to using cores/2 threadpool threads. The long-term plan is to improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default. Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using all cores can be 10x slower than using cores/2. This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware. Reviewed By: kirklandsign Differential Revision: D64107326
GregoryComer
force-pushed
the
export-D64107326
branch
from
bf88166 to
c8c1650
Compare
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D64107326 |
|
This pull request has been merged in 27330f2. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Default to using cores/2 threadpool threads. The long-term plan is to
improve performant core detection in CPUInfo, but for now we can use cores/2 as a sane default.
Based on testing, this is almost universally faster than using all cores, as efficiency cores can be quite slow. In extreme cases, using
all cores can be 10x slower than using cores/2.
This also matches Lite Interpreter's default behavior when it doesn't have a more precise heuristic for the target hardware.
Differential Revision: D64107326