feat(expr-eval): Fix flaky adaptiveCpuSamplingPerFunctionRates test (#17002)#17002
Closed
Rajeev975 wants to merge 1 commit intofacebookincubator:mainfrom
Closed
feat(expr-eval): Fix flaky adaptiveCpuSamplingPerFunctionRates test (#17002)#17002Rajeev975 wants to merge 1 commit intofacebookincubator:mainfrom
Rajeev975 wants to merge 1 commit intofacebookincubator:mainfrom
Conversation
✅ Deploy Preview for meta-velox canceled.
|
meta-cla
bot
added
the
CLA Signed
|
@Rajeev975 has exported this pull request. If you are a Meta employee, you can view the originating Diff in D99126870. |
Build Impact AnalysisSelective Build Targets (building these covers all 1 affected)Total affected: 1/557 targets Affected targets (1)Directly changed (1)
Fast path • Graph from main@b65b5c1c56942f7ef34c3e34fb641e50229266cf |
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 1, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
meta-codesync
bot
changed the title
Rajeev975
force-pushed
the
export-D99126870
branch
from
e8d5afb to
5124ff2
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 1, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
5124ff2 to
367f6e2
Compare
Contributor
Author
|
Will add another stress test to verify that adaptiveCpuSamplingPerFunctionRates is not flaky |
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 1, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
367f6e2 to
37b132e
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 1, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
37b132e to
39024e1
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 1, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
2 times, most recently
from
0f2b7b5 to
d76b1c3
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 1, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
d76b1c3 to
aa1d16a
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…ion evaluation (facebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 This is similar to facebookincubator#16646 Previous landing was reverted because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
aa1d16a to
eaa17c9
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…acebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 Previous diff (facebookincubator#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
eaa17c9 to
0b40baa
Compare
meta-codesync
bot
changed the title
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…acebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 Previous diff (facebookincubator#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
0b40baa to
e6e04d0
Compare
pedroerp
approved these changes
Apr 3, 2026
Contributor
pedroerp
left a comment
pedroerp
left a comment
There was a problem hiding this comment.
Thanks @Rajeev975 . Please revert the documentation comment slashes change first.
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…acebookincubator#17002) Summary: Previous diff (facebookincubator#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Reviewed By: pedroerp Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
e6e04d0 to
c7ce597
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…acebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 Previous diff (facebookincubator#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Reviewed By: pedroerp Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
c7ce597 to
41eb95e
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…acebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 Previous diff (facebookincubator#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Reviewed By: pedroerp Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
2 times, most recently
from
edf9acb to
df52e01
Compare
Rajeev975
added a commit
to Rajeev975/velox
that referenced
this pull request
Apr 3, 2026
…acebookincubator#17002) Summary: Previous diff (facebookincubator#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Reviewed By: pedroerp Differential Revision: D99126870
…acebookincubator#17002) Summary: Pull Request resolved: facebookincubator#17002 Previous diff (facebookincubator#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering. Reviewed By: pedroerp Differential Revision: D99126870
Rajeev975
force-pushed
the
export-D99126870
branch
from
df52e01 to
c53dfda
Compare
|
This pull request has been merged in 7ea5609. |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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:
Previous diff (#16646) had a flaky test because the test asserted that slow_add must be in kAlwaysTrack state and plus must be in kSampling state. Replaced absolute state assertions with a relative comparison: slow_add sampling rate must be ≤ plus sampling rate. This is robust because both functions share the same timerOverheadNanos_ per ExprSet, so measurement noise affects both equally and cannot flip the relative ordering.
Reviewed By: pedroerp
Differential Revision: D99126870