The diff jobs currently take a Firefox package and compares against the
one from another build. When that fails, it also makes the diff of the
generated files, which can be useful when there are differences, but
can also be useful on its own. Also, because sometimes, like right now,
there are differences in generated files that have no impact on Firefox
itself, when differences do show up for Firefox, the differences in
generated files are added noise that sheriffs can't work around.
Differential Revision: https://phabricator.services.mozilla.com/D67113
--HG--
extra : moz-landing-system : lando
The fork server is not launched as expected with the
preference. This patch fix the issue.
Differential Revision: https://phabricator.services.mozilla.com/D65930
--HG--
extra : moz-landing-system : lando
The produce the same outcome as originally done in bug 1487797. However, we only do so on the last block ever retrieved.
Ideally, we should calculate the start time based on the number of frames previously returned. It would avoid such inconsistency and while we had one file presenting such symptom, I'm not sure we can always rely for the frame_or_sample_num value to be valid in reference to the first frame block size.
Depends on D66989
Differential Revision: https://phabricator.services.mozilla.com/D66990
--HG--
extra : moz-landing-system : lando
It didn't take into account that there could be a mixture of variable block size and fixed block size and could use the start time of an invalid block to determine the end of the previous frame as the code relied on Frame::FindNext() to have always returning a valid frame; but this isn't always the case.
We also want to keep the scope of each classes separate, so that it is up to the FrameParser to correct a Frame should it be invalid or not set properly.
Differential Revision: https://phabricator.services.mozilla.com/D66989
--HG--
extra : moz-landing-system : lando
`clone_from_bag` lets Rust code create a thread-safe `HashPropertyBag`
from any kind of `nsIPropertyBag` by cloning its contents. This is
important when you want to use a property bag that's passed to an XPIDL
method off the main thread. An `nsIPropertyBag` passed from JS isn't
thread-safe, since it's cycle-collected, and its properties are
`XPCVariant`s. But a Rust `HashPropertyBag` is, because it's backed by
a non-cycle collected `nsHashPropertyBag`, and the only supported
properties are thread-safe storage variants.
It also marks `HashPropertyBag` as `Send + Sync`, given that the only
way to construct one is through `HashPropertyBag::new`, and
`VariantType` is only implemented for storage variants.
Finally, it adds more type tags for string (narrow and wide) variants,
and changes the `VariantType for nsCString` getter to use
`GetAsAUTF8String`, since that's what mozStorage implements for its
text variant, not `GetAsACString`.
Differential Revision: https://phabricator.services.mozilla.com/D67075
--HG--
extra : moz-landing-system : lando
OSKeyStore doesn't need its own thread and can use the background thread pool instead.
Differential Revision: https://phabricator.services.mozilla.com/D66692
--HG--
extra : moz-landing-system : lando
When the signature verification fails after retry we don't want to apply the changes, and if possible we want to leave the local data as it was before sync.
Differential Revision: https://phabricator.services.mozilla.com/D67139
--HG--
extra : moz-landing-system : lando
Automatic update from web-platform-tests
Fix ci_resources_unittest (#22249)
Firefox has changed an error message recently, and the test is so
verbose that any change in output will cause failures.
--
wpt-commits: e19e45639560ff2a89a826c21bbe26d2bbee29df
wpt-pr: 22249
Automatic update from web-platform-tests
Add 'reftest-wait' class with timeout to reference pages for certain wpt
The reference pages for the 'embedded_style_media_queries*' tests
previously didn't wait to take a screenshot (as opposed to their
counterparts which wait 100ms to take a screenshot), causing them to be
a bit flaky. This CL fixes that.
Change-Id: Ie8dd12b3b4dd7f810d72b3ba94ae2e5cb3644fe0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2103142
Commit-Queue: Fredrik Söderquist <fs@opera.com>
Reviewed-by: Fredrik Söderquist <fs@opera.com>
Cr-Commit-Position: refs/heads/master@{#750509}
--
wpt-commits: 5f0d13915e4f46c7d25958f1a59155c200421a15
wpt-pr: 22251
Automatic update from web-platform-tests
[WPT/mixed-content] Add tests for SharedWorkers with no opt-in headers
Manual change: wpt/mixed-content/spec.src.json.
Previously, it omitted `http-rp/unset` tests because it expected
redundant tests `meta/unset`. However, in certain worker-related
tests, there are no `meta` tests, and thus `http-rp/unset` is
not redundant.
This CL enables generating such non-redundant `http-rp/unset` tests,
namely `sharedworker-{classic,module}.http-rp/unset`,
i.e. SharedWorkers served from HTTPS origin without no
opt-in CSP headers.
All other changes are mechanical.
Bug: 1061679, 1056500
Change-Id: Ib8c323cf06e1c6e0a0bdeae713bd13848335b95a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2103967
Commit-Queue: Hiroshige Hayashizaki <hiroshige@chromium.org>
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#750486}
--
wpt-commits: 370b686e596beb32e32a742e6c921bfdd84c35f0
wpt-pr: 22272
Automatic update from web-platform-tests
[WPT/common/security-features] Faster generator by less json.dumps()
Previously, dump_test_parameters() (which uses json.dumps())
was used to calculate the keys of `exclusion_dict`,
so that the keys of `exclusion_dict` include
all the fields of the selections.
However, this was slow.
This CL instead uses `excluded_selection_pattern % selection`
where excluded_selection_pattern incudes "`%(key)s`" for
all the keys of the selections, except for "expansion".
(The exclusion of "expansion" is probably good, as
differences in "expansion" fields shouldn't prevent
excluded test matching)
This CL improves the generator's performance about 5x.
This CL doesn't change the generated results or test behavior.
Bug: 906850
Change-Id: I67f2cf6e960d7867a9b409cb84bf8249ac1912c7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2100562
Commit-Queue: Hiroshige Hayashizaki <hiroshige@chromium.org>
Reviewed-by: Hiroki Nakagawa <nhiroki@chromium.org>
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#750484}
--
wpt-commits: f64deda054dda4d72ad4d91f480ec584be6fb7e1
wpt-pr: 22271
Automatic update from web-platform-tests
[WPT/common/security-features] Remove spec_json.js from repository
`spec_json.js` are not used in WPT production runs.
This CL removes them, to omit `spec_json.js`-related diffs
in upcoming CLs.
This CL doesn't change the generated results or test behavior.
Bug: 906850
Change-Id: I1001286f2680294f1d8e747f23a25040f5bfd253
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2103728
Commit-Queue: Hiroshige Hayashizaki <hiroshige@chromium.org>
Reviewed-by: Hiroki Nakagawa <nhiroki@chromium.org>
Reviewed-by: Eriko Kurimoto <elkurin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#750483}
--
wpt-commits: 4568db58b5528eeb9edf931ab34d83fd6883600f
wpt-pr: 22265
Brian Birtles
Mike Hommey
Masayuki Nakano
Thinker Li
Jean-Yves Avenard
Bryce Seager van Dyk
Logan Smyth
James Teh
Cosmin Sabou
Andy Grover
Chun-Min Chang
Paul Bone
Michael
Lina Cambridge
Kartikaya Gupta
Dana Keeler
Ciure Andrei
Agi Sferro
alwu
Shane Caraveo
Olli Pettay
Mathieu Leplatre
Kershaw Chang
Jonathan Kew
moz-wptsync-bot
Robert Ma
Will Cassella
Hiroshige Hayashizaki