[Profiler] Pay for what you use (v2) #74484
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
Contributor
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 9af2568 (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Please report bugs/suggestions to the (internal) Dr. CI Users group. |
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D34779994 |
robieta
force-pushed
the
export-D34779994
branch
from
1de8e0a to
9df93e2
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D34779994 |
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D34779994 |
robieta
force-pushed
the
export-D34779994
branch
from
9df93e2 to
a99e473
Compare
robieta
force-pushed
the
export-D34779994
branch
from
a99e473 to
610bae5
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D34779994 |
Summary: Pull Request resolved: pytorch#74484 In my first attempt at this in December I stamped out specializations using variadic templates. However I'm able to get comparable performance using simple conditionals since the branch is very predictable and AppendOnlyList::emplace_back is low enough overhead that multiple calls don't cause an issue. This is also a chance to do some BE: rather than force ops and backend events to use the same fields (which in practice means setting a bunch of default values when reporting backend events), I just split them and use a variant. Test Plan: The single threaded benchmark (with no extra options set) improved considerably from ~0.88 us to ~0.62 us. The stress test benchmark improved modestly from ~6.1 us to ~5.8 us. So the bottleneck for multi-threading is somewhere else, but doing less wasted work is still able to move the needle a little bit. Reviewed By: swolchok Differential Revision: D34779994 fbshipit-source-id: 0b0503678929b2d61e9ddb1b27887e64fdcb7fe4
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D34779994 |
robieta
force-pushed
the
export-D34779994
branch
from
610bae5 to
9af2568
Compare
aaronenyeshi
approved these changes
Mar 23, 2022
Member
aaronenyeshi
left a comment
aaronenyeshi
left a comment
There was a problem hiding this comment.
LGTM, reviewed internally! Thanks a lot for this patch!
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 24, 2022
Summary: Pull Request resolved: #74484 In my first attempt at this in December I stamped out specializations using variadic templates. However I'm able to get comparable performance using simple conditionals since the branch is very predictable and AppendOnlyList::emplace_back is low enough overhead that multiple calls don't cause an issue. This is also a chance to do some BE: rather than force ops and backend events to use the same fields (which in practice means setting a bunch of default values when reporting backend events), I just split them and use a variant. Test Plan: The single threaded benchmark (with no extra options set) improved considerably from ~0.88 us to ~0.62 us. The stress test benchmark improved modestly from ~6.1 us to ~5.8 us. So the bottleneck for multi-threading is somewhere else, but doing less wasted work is still able to move the needle a little bit. Reviewed By: swolchok Differential Revision: D34779994 fbshipit-source-id: 392bc7c6f12797fa5e18777063aa21210d9d2067
Contributor
|
Hey @robieta. |
shahofblah
pushed a commit
that referenced
this pull request
Mar 25, 2022
Summary: Pull Request resolved: #74484 In my first attempt at this in December I stamped out specializations using variadic templates. However I'm able to get comparable performance using simple conditionals since the branch is very predictable and AppendOnlyList::emplace_back is low enough overhead that multiple calls don't cause an issue. This is also a chance to do some BE: rather than force ops and backend events to use the same fields (which in practice means setting a bunch of default values when reporting backend events), I just split them and use a variant. Test Plan: The single threaded benchmark (with no extra options set) improved considerably from ~0.88 us to ~0.62 us. The stress test benchmark improved modestly from ~6.1 us to ~5.8 us. So the bottleneck for multi-threading is somewhere else, but doing less wasted work is still able to move the needle a little bit. Reviewed By: swolchok Differential Revision: D34779994 fbshipit-source-id: 392bc7c6f12797fa5e18777063aa21210d9d2067 (cherry picked from commit f0a49ff)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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:
In my first attempt at this in December I stamped out specializations using variadic templates. However I'm able to get comparable performance using simple conditionals since the branch is very predictable and AppendOnlyList::emplace_back is low enough overhead that multiple calls don't cause an issue.
This is also a chance to do some BE: rather than force ops and backend events to use the same fields (which in practice means setting a bunch of default values when reporting backend events), I just split them and use a variant.
Test Plan: The single threaded benchmark (with no extra options set) improved considerably from ~0.88 us to ~0.62 us. The stress test benchmark improved modestly from ~6.1 us to ~5.8 us. So the bottleneck for multi-threading is somewhere else, but doing less wasted work is still able to move the needle a little bit.
Reviewed By: swolchok
Differential Revision: D34779994