8.39% twitter fcp (Linux) regression on Wed May 18 2022
Categories
(Core :: DOM: Core & HTML, defect)
Tracking
()
Tracking | Status | |
---|---|---|
firefox-esr91 | --- | unaffected |
firefox100 | --- | unaffected |
firefox101 | --- | unaffected |
firefox102 | --- | fixed |
firefox103 | --- | fixed |
People
(Reporter: afinder, Assigned: emilio)
References
(Regression)
Details
(Keywords: perf, perf-alert, regression)
Attachments
(5 files)
13.61 KB,
text/html
|
Details | |
48 bytes,
text/x-phabricator-request
|
pascalc
:
approval-mozilla-beta+
|
Details | Review |
48 bytes,
text/x-phabricator-request
|
pascalc
:
approval-mozilla-beta+
|
Details | Review |
48 bytes,
text/x-phabricator-request
|
Details | Review | |
48 bytes,
text/x-phabricator-request
|
Details | Review |
Perfherder has detected a browsertime performance regression from push 208368baa3eef0d10fea2952ba53f8703e4488d3. As author of one of the patches included in that push, we need your help to address this regression.
Regressions:
Ratio | Test | Platform | Options | Absolute values (old vs new) |
---|---|---|---|---|
8% | twitter fcp | linux1804-64-shippable-qr | fission warm webrender | 186.02 -> 201.62 |
Details of the alert can be found in the alert summary, including links to graphs and comparisons for each of the affected tests. Please follow our guide to handling regression bugs and let us know your plans within 3 business days, or the offending patch(es) will be backed out in accordance with our regression policy.
If you need the profiling jobs you can trigger them yourself from treeherder job view or ask a sheriff to do that for you.
For more information on performance sheriffing please see our FAQ.
Comment 1•2 years ago
|
||
Set release status flags based on info from the regressing bug 1768198
Assignee | ||
Comment 2•2 years ago
|
||
The twitter page has:
<meta http-equiv="origin-trial" content="Ap6SMBNB0lQoXpXl4I9vyTJqJ7Y0X9tPd6Q6rN697iHdubQQxBcWHy21N3N7uEz7Ba5UKMbN+eLvDczBSbi27AsAAABfeyJvcmlnaW4iOiJodHRwczovL3R3aXR0ZXIuY29tOjQ0MyIsImZlYXR1cmUiOiJCYWRnaW5nIiwiZXhwaXJ5IjoxNTY0NTgyNzY2LCJpc1N1YmRvbWFpbiI6dHJ1ZX0=">
<meta http-equiv="origin-trial" content="Apir4chqTX+4eFxKD+ErQlKRB/VtZ/dvnLfd9Y9Nenl5r1xJcf81alryTHYQiuUlz9Q49MqGXqyaiSmqWzHUqQwAAABneyJvcmlnaW4iOiJodHRwczovL3R3aXR0ZXIuY29tOjQ0MyIsImZlYXR1cmUiOiJDb250YWN0c01hbmFnZXIiLCJleHBpcnkiOjE1NzUwMzUyODMsImlzU3ViZG9tYWluIjp0cnVlfQ==">
<meta http-equiv="origin-trial" content="AleGS26SZL7UA8Fe1DbvXzoay74bPTvrfKKGimIu1RI8vA+RtXOSVlizUkz2zU/fQoFoOTgCiCciP6pM5teaeQgAAABjeyJvcmlnaW4iOiJodHRwczovL3R3aXR0ZXIuY29tOjQ0MyIsImZlYXR1cmUiOiJTbXNSZWNlaXZlciIsImV4cGlyeSI6MTU3OTAyMDkyMSwiaXNTdWJkb21haW4iOnRydWV9">
Assignee | ||
Comment 3•2 years ago
|
||
Assignee | ||
Comment 4•2 years ago
|
||
This takes half the time of the signature validation process, and we
trust the key we're providing, so we shouldn't need to do this.
Plus, PK11_VerifyWithMechanism verifies the key again
(see bug 1770921).
Updated•2 years ago
|
Assignee | ||
Comment 5•2 years ago
|
||
As verifying the signature can be slow-ish (filed bug 1770921) for this.
This makes the overhead of origin trial tokens negligible in the
attached test-case, as opposed to taking ~60ms.
Depends on D147171
Assignee | ||
Comment 7•2 years ago
|
||
As per recommendation in the other revision.
Assignee | ||
Comment 8•2 years ago
|
||
This avoids the key copy, and should prevent further verification when
using the returned key.
Comment 10•2 years ago
|
||
Comment 11•2 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/319e6905f56c
https://hg.mozilla.org/mozilla-central/rev/54ab422e8b53
https://hg.mozilla.org/mozilla-central/rev/2a21aa12623f
https://hg.mozilla.org/mozilla-central/rev/1ab65da024c4
Comment 12•2 years ago
|
||
The patch landed in nightly and beta is affected.
:emilio, is this bug important enough to require an uplift?
If not please set status_beta
to wontfix
.
For more information, please visit auto_nag documentation.
Assignee | ||
Comment 13•2 years ago
|
||
Comment on attachment 9277931 [details]
Bug 1770885 - Don't verify the origin-trials public key. r=smaug
Beta/Release Uplift Approval Request
- User impact if declined: Perf regression on sites that use Chrome origin trials.
- Is this code covered by automated tests?: Yes
- Has the fix been verified in Nightly?: Yes
- Needs manual test from QE?: No
- If yes, steps to reproduce:
- List of other uplifts needed: none
- Risk to taking this patch: Low
- Why is the change risky/not risky? (and alternatives if risky): Just reorders and avoids some operations to be faster. Other fixes are probably not so critical to uplift.
- String changes made/needed: none
- Is Android affected?: Yes
Assignee | ||
Updated•2 years ago
|
Comment 14•2 years ago
|
||
Comment on attachment 9277931 [details]
Bug 1770885 - Don't verify the origin-trials public key. r=smaug
Approved for 102 beta 5, thanks.
Updated•2 years ago
|
Comment 15•2 years ago
|
||
bugherder uplift |
Comment 16•2 years ago
|
||
(In reply to Pulsebot from comment #10)
Pushed by ealvarez@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/2a21aa12623f
Cache existing public key rather than recreating it over and over. r=keeler
https://hg.mozilla.org/integration/autoland/rev/1ab65da024c4
Avoid key copy + re-verification in CreateECPublicKey. r=keeler
== Change summary for alert #34332 (as of Tue, 07 Jun 2022 06:50:39 GMT) ==
Improvements:
Ratio | Test | Platform | Options | Absolute values (old vs new) |
---|---|---|---|---|
13% | twitch fcp | linux1804-64-shippable-qr | fission warm webrender | 58.90 -> 51.04 |
6% | twitter fcp | linux1804-64-shippable-qr | fission warm webrender | 198.08 -> 186.00 |
5% | twitter fcp | windows10-64-shippable-qr | fission warm webrender | 101.85 -> 96.79 |
4% | twitter fcp | windows10-64-shippable-qr | cold fission webrender | 196.21 -> 188.67 |
For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=34332
Description
•