Chromium Main Console

cfredric@chromium.org

[Cookies] Add optional enforcement of validity of webby non-IP hosts

This CL creates new EXCLUDE_INVALID_HOST and WARN_INVALID_HOST exclusion
reason & warning for cookies, a new feature flag to enable URL host
canonicalization (as defined by the cookies spec[1]), and two new
UseCounters that measure how often cookie reads/writes come from
pages/URLs whose hostnames are invalid.

The warning is used when the kCookieRequireCompliantHost feature is
disabled; when the feature is enabled, the exclusion reason is used
instead.

Note that cookie reads/writes that are triggered by pages/URLs with
non-webby schemes (e.g. file://, chrome-extension://) or IP-address
hostnames are not affected.

[1]: https://datatracker.ietf.org/doc/draft-ietf-httpbis-rfc6265bis/

Bug: 580970710
Change-Id: I35923a0e2aeb3cb43757f75e89350389c734a3de
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8709692
Reviewed-by: Maks Orlovich <morlovich@chromium.org>
Commit-Queue: Chris Fredrickson <cfredric@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735152}

satorux@google.com

chromeos: Remove owner no longer on the project from hugepage_text/OWNERS

Remove manojgupta@chromium.org who is no longer on the project.

Bug: 580195798
Change-Id: Iaa1133fdc6f87d4c14c3447b9c84ca42909d6ae4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8689168
Commit-Queue: George Burgess <gbiv@chromium.org>
Auto-Submit: Satoru Takabayashi <satorux@google.com>
Reviewed-by: George Burgess <gbiv@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735151}

blundell@chromium.org

[Blink] Remove non-HDRMetadata Create* overloads from CN2DRP

All production callers of CanvasNon2DResourceProvider::Create(),
CreateForWebGPU(), and CreateForSoftwareCompositor() explicitly pass
const gfx::HDRMetadata&. Remove the 3 wrapper overloads that omit
HDRMetadata and update the single test caller in
OffscreenCanvasPlaceholderTest.

Bug: 352263194
Change-Id: Ic0b7e3d3c4d75159870c355d8e81a3732c94e23d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8714995
Commit-Queue: Colin Blundell <blundell@chromium.org>
Reviewed-by: Vasiliy Telezhnikov <vasilyt@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735150}

mwalachowski@google.com

[Data Controls] Share copy-to-OS-clipboard verdict lookup

CanPopulateFindBarFromSelection(), IsSearchWithAllowed() and
ShouldAllowSearchWith() each resolved the primary main frame endpoint of
a WebContents and then queried ChromeRulesService for the same
copy-to-OS-clipboard verdict.

Move that lookup into a private helper,
GetPageCopyToOSClipboardVerdict(), which returns the endpoint together
with the verdict, and use it in all three functions. There is no
intended behavior change.

IsClipboardCopyAllowedByPolicyForUI() is left as is. It needs the
endpoint for three other checks (ClipboardRestrictionService, the
source-only verdict and content analysis), and the OS clipboard verdict
is only one step in that sequence. Using the helper there did not make
the function shorter or clearer, and the part it shares with the other
functions is already factored out in GetValidURLEndpoint().

Also add a unit test checking that ShouldAllowSearchWith() forwards the
selection size to rule evaluation, since it is the only caller that
passes a size.

Bug: 514618011
Change-Id: Ice49ace8227a0ea1d251b898a99f079f83ba2a0b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8726430
Reviewed-by: Anatoli Hancharou <antoli@google.com>
Reviewed-by: Dominique Fauteux-Chapleau <domfc@chromium.org>
Commit-Queue: Mikołaj Wałachowski <mwalachowski@google.com>
Cr-Commit-Position: refs/heads/main@{#1735149}

amarjotgill@google.com

[iOS] Fix ASan UAF in AgeMismatchSignoutCoordinatorTest

Replace OCMProtocolMock(@protocol(SystemIdentity)) with
[FakeSystemIdentity fakeIdentity1]. Forwarding an unstubbed Objective-C
method returning a non-trivial C++ GaiaId struct leaves stale stack
memory in the return slot, triggering an ASan heap-use-after-free in
FakeSystemIdentityManager::GetCachedAvatarForIdentity.

Bug: 581389154
Change-Id: I62ef888ddfc4645b21745c75da4885bc75926736
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8715889
Commit-Queue: Amarjot Gill <amarjotgill@chromium.org>
Reviewed-by: Boris Sazonov <bsazonov@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735148}

vladyslavvovk@google.com

[iOS] Move manage_sync EG test disable annotations to expectations

Migrates the `FLAKY_` annotations in `ManageSyncSettingsTestCase` into
`ios/testing/test_expectations.txt`, following the team rule:
- flaky -> `[ Pass Failure ]`
- consistently failing -> strict `[ Failure ]` (only with verified data)
- hangs or takes down the run -> `[ Skip ]`

Verification: the class was run 3 times on each of 6 simulator configs
of `flaky-simulator` (iPhone/iPad, iOS 18.5/26.5/27.0). No test failed
consistently across configs, so all 30 remaining `FLAKY_` tests get
`[ Pass Failure ]` with the bug from the removed TODO.

- The in-body iPhone early return in
`testUnsyncedDataDialogShowsInCaseOfUnsyncedReadingListEntry` becomes
an `[ iPhone ] [ Skip ]` entry, keeping the current behavior.
- Behavior change: these tests now run in the regular settings module
instead of only in `ios_chrome_flaky_eg2tests_module`. On the flaky
builder they effectively did not run, because an earlier crashing
test in that module takes down the app for all later suites. This
adds ~20 min of test time per settings-module config; no test hung
in the verification runs.

TAG=agy
CONV=8d8bd63e-8f05-48c1-8d69-c6a4484db52d

Change-Id: I0e3d38f1196f7c7ed219eb9860591ab13dc2118c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8676354
Auto-Submit: Vladyslav Vovk (xWF) <vladyslavvovk@google.com>
Reviewed-by: Olivier Robin <olivierrobin@chromium.org>
Reviewed-by: Arthur Milchior <arthurmilchior@chromium.org>
Commit-Queue: Vladyslav Vovk (xWF) <vladyslavvovk@google.com>
Cr-Commit-Position: refs/heads/main@{#1735147}

chrome-official-brancher@chops-service-accounts.iam.gserviceaccount.com

Updating trunk VERSION from 8143.0 to 8144.0

* This is an automated release commit.
* Do not revert without consulting chrome-pmo@google.com.
NOAUTOREVERT=true

Change-Id: Ida2ffb6e3852bf30526baee888175739c2243823
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8727647
Bot-Commit: chrome-official-brancher@chops-service-accounts.iam.gserviceaccount.com <chrome-official-brancher@chops-service-accounts.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735146}

blundell@chromium.org

[Blink] Drop raster access in CN2DRP::EndExternalWrite()

CanvasNon2DResourceProvider::EndExternalWrite() currently calls
WaitSyncToken(external_write_sync_token) [1] (setting
acquire_sync_token_ and updating the SharedImage destruction sync token)
and then performs a no-op BeginRasterAccess() + EndAccess() round-trip
on RasterInterface to chain a new RasterInterface sync token after
acquire_sync_token_ and pass it to SetReleaseSyncToken().

This round-trip is a holdover from when CanvasResourceSharedImage
generated and verified its release sync token on RasterInterface [2].
Now that the release sync token is set directly via
SetReleaseSyncToken() (and verified via
SharedImageInterface::VerifySyncToken() in CanvasResourceDispatcher
[3]), EndExternalWrite() can pass external_write_sync_token directly to
SetReleaseSyncToken() without bouncing through RasterInterface.

[1] https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/renderer/platform/graphics/canvas_resource.cc;l=322-329;drc=c611b8b9e3e9807ef57a43924674182c260c869c
[2] https://chromium-review.googlesource.com/c/chromium/src/+/6035039
[3] https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/renderer/platform/graphics/canvas_resource_dispatcher.cc;l=155-157;drc=c611b8b9e3e9807ef57a43924674182c260c869c

Bug: 352263194
Change-Id: Ia1f2a6bfe9c587b2528e761d588bd015c5b8d3ba
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8713910
Reviewed-by: Vasiliy Telezhnikov <vasilyt@chromium.org>
Commit-Queue: Colin Blundell <blundell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735145}

blundell@chromium.org

[Blink] Inline CanvasResourceSharedImage::EndExternalWrite()

CanvasResourceSharedImage::EndExternalWrite() has only one caller:
CanvasNon2DResourceProvider::EndExternalWrite(). Inline it into
CanvasNon2DResourceProvider::EndExternalWrite() and remove
CanvasResourceSharedImage::EndExternalWrite().

Bug: 352263194
Change-Id: I8429bc3033ef60552dbf396da2daa6e9ae663dde
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8664664
Reviewed-by: Vasiliy Telezhnikov <vasilyt@chromium.org>
Commit-Queue: Colin Blundell <blundell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735144}

nickdiego@igalia.com

[Bedrock][2/14] Remove //c/t:test_support_ui from chrome/browser

Migrates 25 BUILD.gn files to use direct deps instead of the
//chrome/test:test_support_ui group.

Analysis/Plan: https://notes.nickdiego.dev/chromium/bedrocking-chrome-test

Bug: 353332589
AX-Relnotes: n/a
Change-Id: I9cf349ae88e12508d3178065654bf8ee0b739faa
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8683030
Auto-Submit: Nick Yamane <nickdiego@igalia.com>
Reviewed-by: Colin Blundell <blundell@chromium.org>
Reviewed-by: Nick Yamane <nickdiego@igalia.com>
Commit-Queue: Nick Yamane <nickdiego@igalia.com>
Cr-Commit-Position: refs/heads/main@{#1735143}

ybourouphael@google.com

[WebUI Toolbar] Match native Views icon color for disabled toolbar buttons

Disabled toolbar buttons (<cr-icon-button> and <toolbar-chip-button>)
previously inherited default 0.38 opacity dimming on the active icon
color, whereas native Views ToolbarButton renders disabled buttons with
kColorToolbarButtonIconInactive at full opacity. On macOS inactive
windows (controls-appear-inactive), this also double-faded disabled
buttons by applying 0.38 opacity on top of the inactive color token.

Apply `--color-toolbar-button-icon-inactive` at full opacity when
toolbar buttons or pinned action chips are disabled. Additionally, set
`forced-color-adjust: none` so Windows High Contrast Mode paints
disabled icons with the ColorProvider's inactive color
(kColorNativeGrayText) instead of forcing CanvasText.

Bug: 578457410
Change-Id: I0cc952160aa1abb17e806a69125ac41b61a77bd2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8672534
Reviewed-by: Qingxin Wu <qingxinwu@google.com>
Commit-Queue: Youseff Bourouphel <ybourouphael@google.com>
Cr-Commit-Position: refs/heads/main@{#1735142}

ylahodiuk@google.com

Add PasskeySecurityDomainChecker for Android

Adds PasskeySecurityDomainChecker and
TrustedVaultClient.Backend.getIsSecurityDomainEmpty() to check whether
the passkeys (hw_protected) security domain is empty for an account.

Bug: 540854648
Change-Id: I56ed611647e26e5d083ba0b7d467da58f8d499ff
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8585062
Reviewed-by: Thomas Thrainer <thomasth@google.com>
Commit-Queue: Yurii Lahodiuk <ylahodiuk@google.com>
Cr-Commit-Position: refs/heads/main@{#1735141}

przybytniowska@google.com

[Isolated Mode] Use ScopedIsolatedModeOverride in IsolatedDownloadTest

Enable Isolated Mode in IsolatedDownloadTest with
ScopedIsolatedModeOverrideForTesting instead of the
--force-enterprise-isolated-mode switch. The switch is ignored on
Stable/Beta channels, which required a ScopedChannelOverride workaround
in branded builds; the override is channel-independent.

Bug: 579864999
DD: go/chrome-enterprise-isolated-mode-client-dd
Change-Id: I1053614ace27e38d3452316d31d092f6dfe1882e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8725690
Reviewed-by: Greg Thompson <grt@chromium.org>
Commit-Queue: Julia Przybytniowska <przybytniowska@google.com>
Cr-Commit-Position: refs/heads/main@{#1735140}

ananyaseelam@google.com

[context-input] Default max_total_inputs to `kDefaultMaxTotalInputs`

This CL updates `InputStateModel` to default `max_total_inputs` to
`omnibox::kDefaultMaxTotalInputs` (`10`) when `allowed_input_types` is
non-empty and `rule_set_.has_max_total_inputs()` is false, while keeping
`max_total_inputs` at `0` when no inputs are allowed.

Bug: b:483189410
Change-Id: Ic20bb45509df761f0d02e6329f8da7a322fb5988
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8716909
Reviewed-by: Nihar Majmudar <niharm@google.com>
Commit-Queue: Ananya Seelam <ananyaseelam@google.com>
Cr-Commit-Position: refs/heads/main@{#1735139}

thefrog@chromium.org

Add OWNERS for Safe Browsing URL checker and Chrome subdirectories

Add thefrog@chromium.org and xinghuilu@chromium.org as owners in
components/safe_browsing/core/browser/url_checker/OWNERS, and add OWNERS
files in components/safe_browsing/content/browser/url_checker/ and
chrome/browser/safe_browsing/{db,hashprefix_realtime,realtime}/
referencing their parallel OWNERS files in
components/safe_browsing/core/browser/.

Bug: 578080484
Bypass-Check-License: Moved existing files
Change-Id: I06b4b6fa6c6d1f847ecdceb62fe0ce0cbcbdd853
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8664213
Commit-Queue: thefrog <thefrog@chromium.org>
Reviewed-by: Xinghui Lu <xinghuilu@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735138}

qpubert@google.com

[iOS] Remove testToolbarAboveKeyboardDuringSearch

On iOS 26 (Liquid Glass), UINavigationController replaces UIToolbar with
a private _UIFloatingBarContainerView hosting a SwiftUI
FloatingBarHostingView that ignores
UINavigationController.additionalSafeAreaInsets.bottom and stays behind
the keyboard when the search bar is focused. Since there is no public
API to reposition the native Liquid Glass floating toolbar above the
keyboard, remove
DriveFilePickerTestCase/testToolbarAboveKeyboardDuringSearch and its
test_expectations.txt entry.

Fixed: 581389688
Change-Id: Ie6e54a9d640df337a5cdaf565f4b95c00fe6f346
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8725058
Auto-Submit: Quentin Pubert <qpubert@google.com>
Commit-Queue: Elmehdi Rahmaoui <erahmaoui@google.com>
Commit-Queue: Quentin Pubert <qpubert@google.com>
Reviewed-by: Elmehdi Rahmaoui <erahmaoui@google.com>
Cr-Commit-Position: refs/heads/main@{#1735137}

thefrog@chromium.org

Move chrome/browser/safe_browsing db and realtime files into subdirs

This CL moves the local database, hash-prefix real-time, and URL
real-time factories, helpers, and tests in chrome/browser/safe_browsing/
into db/, hashprefix_realtime/, and realtime/ subdirectories mirroring
components/safe_browsing/core/browser/ so that directory-level OWNERS
can be added.

This CL is intended to be a pure refactor only.

Bug: 578080484
Bypass-Check-License: Moved existing files
Change-Id: Ib177d0528cc6bb350ec39016f44c8737b4384872
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8668624
Reviewed-by: Xinghui Lu <xinghuilu@chromium.org>
Commit-Queue: thefrog <thefrog@chromium.org>
Reviewed-by: Calder Kitagawa <ckitagawa@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735136}

thefrog@chromium.org

Move content/browser URL checker files into url_checker/ subdirectory

This CL moves AsyncCheckTracker, BrowserURLLoaderThrottle,
SafeBrowsingNavigationThrottle, and UrlCheckerHolder in
components/safe_browsing/content/browser/ into
components/safe_browsing/content/browser/url_checker/ so that
directory-level OWNERS can be added.

This CL is intended to be a pure refactor only.

Bug: 578080484
Bypass-Check-License: Moved existing files
Change-Id: Iadcf9f3b225ed83e27793ddf7aa25ae76f1a99dd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8669663
Commit-Queue: thefrog <thefrog@chromium.org>
Reviewed-by: Xinghui Lu <xinghuilu@chromium.org>
Reviewed-by: Calder Kitagawa <ckitagawa@chromium.org>
Reviewed-by: Nate Fischer <ntfschr@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735135}

thefrog@chromium.org

Move core/browser URL checker files into url_checker/ subdirectory

This CL moves SafeBrowsingUrlCheckerImpl, UrlCheckerDelegate, and the
lookup mechanism classes in components/safe_browsing/core/browser/ into
components/safe_browsing/core/browser/url_checker/ so that
directory-level OWNERS can be added.

This CL is intended to be a pure refactor only.

Bug: 578080484
Bypass-Check-License: Moved existing files
Change-Id: Ia10b2f26d9cd1d88bd38354f1158f304c32657a3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8669662
Commit-Queue: thefrog <thefrog@chromium.org>
Reviewed-by: Nate Fischer <ntfschr@chromium.org>
Reviewed-by: Xinghui Lu <xinghuilu@chromium.org>
Reviewed-by: Mustafa Emre Acer <meacer@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735134}

afie@google.com

Enable ClientIdBackupCorrectnessFix in fieldtrial_testing_config.json

Adds the field trial testing configuration entry for ClientIdBackupCorrectnessFix on POSIX platforms (Android, ChromeOS, Linux, Mac).

Bug: 497679558
Change-Id: I3082457f86efefb7a805544ffbcc38ec0d92e707
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8701783
Reviewed-by: Ramon Cano Aparicio <rcanoaparicio@google.com>
Commit-Queue: Andrzej Fiedukowicz <afie@google.com>
Cr-Commit-Position: refs/heads/main@{#1735133}

qpubert@google.com

[iOS] Remove redundant ReaderModeEligibleOnRegularPage unit test

`ReaderModeTabHelperTest.ReaderModeEligibleOnRegularPage` inadvertently
passed an empty `distilled_content` string (`""`) to
`SetReaderModeState` while also calling `ActivateReader`. Because empty
distilled content causes distillation to fail and deactivates Reader
Mode without loading content into the Reader Mode WebState,
`WaitForAvailableReaderModeContentInWebState` timed out on every run (6s
when added in crrev.com/c/6798277, and 16s after crrev.com/c/6829983)
before being disabled in crrev.com/c/8676162.

Rather than fixing `distilled_content`, remove the test altogether as
all of its assertions are already covered by existing tests in
`ReaderModeTabHelperTest`:
- Regular page eligibility and heuristic metrics are verified by
`TriggerHeuristicOnPageLoad` and `EligibilityForEligiblePage`.
- Reader Mode activation and content availability are verified by
`NotifiesObserversOfAvailability`, `TestEligibleContentIsDisplayed`,
and `TestDistillationCompletedAfterTimeout`.

Fixed: 579309262
Change-Id: I17bff111704859dfc0fb3b330e9bddcfa97f33b3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8710975
Reviewed-by: Elmehdi Rahmaoui <erahmaoui@google.com>
Commit-Queue: Elmehdi Rahmaoui <erahmaoui@google.com>
Reviewed-by: Nohemi Fernandez <fernandex@google.com>
Auto-Submit: Quentin Pubert <qpubert@google.com>
Cr-Commit-Position: refs/heads/main@{#1735132}

blundell@chromium.org

[//base] Compare drive letters case-insensitively in FilePath::<=> and hash

On Windows (FILE_PATH_USES_DRIVE_LETTERS), base::FilePath's comparison
and hash operators are inconsistent:
* FilePath::operator== compares a leading [A-Za-z]: drive letter
case-insensitively via EqualDriveLetterCaseInsensitive() [1].
* FilePath::operator<=> (= default) [2] and std::hash<base::FilePath>
[3] compared/hashed raw path_ (value()) directly (case-sensitive on
the drive letter).

This violates both ==/<=> consistency (causing std::set<FilePath> and
std::map<FilePath, ...> to treat "C:\foo" and "c:\foo" as distinct keys
and ordering "D:\foo" before "c:\foo") and the C++ Hash requirement that
a == b implies hash(a) == hash(b).

Normalize a leading [A-Za-z]: drive letter to ASCII lowercase in
FilePath::operator<=> and std::hash<base::FilePath> on
FILE_PATH_USES_DRIVE_LETTERS so that operator==, operator<=>, and
std::hash<base::FilePath> are mutually consistent.

[1]
https://source.chromium.org/chromium/chromium/src/+/main:base/files/file_path.cc;l=307-313;drc=7e66e1e25a381629c798384f97a43224e2e600ef
[2]
https://source.chromium.org/chromium/chromium/src/+/main:base/files/file_path.h;l=216-221;drc=7e66e1e25a381629c798384f97a43224e2e600ef
[3]
https://source.chromium.org/chromium/chromium/src/+/main:base/files/file_path.h;l=560-567;drc=7e66e1e25a381629c798384f97a43224e2e600ef

Bug: 522872468
Change-Id: Ie740ab7c309b794a5b4130ba5a6973610118ab6e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8714480
Reviewed-by: Greg Thompson <grt@chromium.org>
Commit-Queue: Colin Blundell <blundell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735131}

dtapuska@chromium.org

Do not dismiss glic selection UI when WebContents size is unchanged

PrimaryMainFrameWasResized can be invoked when visual properties change
even if the WebContents size remains unchanged (such as updates to
screen info or surface IDs).

This change tracks the WebContents size in GlicSelectionObserver and
ignores resize notifications if the size has not actually changed,
preventing spurious dismissals of the selection UI. Unit and interactive
tests are updated accordingly.

Bug: 572173898
Change-Id: I5576688e6a2638c2b97f128d6d029b2d7f4dcad1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8715630
Reviewed-by: William Liu <liuwilliam@chromium.org>
Reviewed-by: Ian Vollick <vollick@chromium.org>
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735130}

natiahlyi@google.com

[Password Manager] Fix blocklists in HTTP-to-HTTPS migration

FormFetcherImpl previously started HTTP-to-HTTPS migration only when
HTTPS returned zero rows, so a lone HTTPS "Never save" entry blocked
migrating saved HTTP passwords. Start migration whenever HTTPS has no
usable credentials, count the migrator in wait_counter_, and flow
through AddResultsAndFinishIfDone instead of returning early so HTTPS
blocklist entries stay in place for new logins.

Unlike HttpCredentialCleaner's GetAutofillableLogins,
HttpPasswordStoreMigrator's per-origin GetLogins also returns HTTP
blocklist rows, inflating PasswordManager.HttpPasswordMigrationCount2,
re-saving them on every non-HSTS fetch, and hitting a CHECK in
PasswordStore::AddLogins on synced blocklist rows with non-empty
credentials. Filter out blocklisted forms in HttpPasswordStoreMigrator.

HTTP | HTTPS before | HTTPS after (old -> new)
----------+----------------+-----------------------------
pwd | blocklist only | blocklist -> pwd + blocklist
blocklist | empty | blocklist -> empty

Fixed: 373852682, 40927687
Change-Id: I171ef7913811a13ada20c5d648edea70e72b1dad
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5937964
Commit-Queue: Andrii Natiahlyi <natiahlyi@google.com>
Reviewed-by: Viktor Semeniuk <vsemeniuk@google.com>
Cr-Commit-Position: refs/heads/main@{#1735129}

chromium-autoroll@skia-public.iam.gserviceaccount.com

Roll Perfetto Trace Processor Win from 52ac869d7735 to 796a01189b09

https://chromium.googlesource.com/external/github.com/google/perfetto.git/+log/52ac869d7735..796a01189b09

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://autoroll.skia.org/r/perfetto-trace-processor-win-chromium
Please CC chrometto-team+alerts@google.com,chrometto-team@google.com on the revert to ensure that a human
is aware of the problem.

To file a bug in Chromium: https://bugs.chromium.org/p/chromium/issues/entry

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Change-Id: I640d62f4c3e8502e9c21c204d71f2ad8fd6eccca
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8723964
Bot-Commit: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735128}

chromium-autoroll@skia-public.iam.gserviceaccount.com

Roll DevTools Frontend from aa62813f7d10 to 0fc9a516a1e0 (5 revisions)

https://chromium.googlesource.com/devtools/devtools-frontend.git/+log/aa62813f7d10..0fc9a516a1e0

2026-11-04 yangguo@chromium.org [Sources] Use --sys-size-8 for .editable in HeadersView.css
2026-11-04 yangguo@chromium.org [Media] Use size tokens in .player-entry-frame-title
2026-11-04 alinavarkki@google.com [ai_evals] Add encore-tickets LCP render delay AI eval tests
2026-11-04 alinavarkki@google.com [ai_evals] Add vanguard-grid performance AI eval tests
2026-11-04 yangguo@chromium.org [Application] Remove unused button.link:has(devtools-icon) CSS rule

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://autoroll.skia.org/r/devtools-frontend-chromium
Please CC chrome-devtools-staff+oncall-change@google.com,liviurau@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Bug: chromium:578783113,chromium:578789232
Submodule-Gn-Analyze: true
Change-Id: I37a309d16355a739d2ece9c745c6c659f07ad543
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8727288
Bot-Commit: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735127}

awogbemila@chromium.org

[text-decorations] Remove unused DecoratingBox::decorations_

DecoratingBox stores a pointer to an AppliedTextDecorationVector, but
nothing reads it: its AppliedTextDecorations() accessor has no callers.
The field used to back a DCHECK that it matched the style's
AppliedTextDecorations(). That DCHECK was removed in
crrev.com/c/5678276, when the ::first-line case started pushing the
base decorations.

Remove the field, its accessor and the constructor parameter, and stop
passing the decorations when pushing decorating boxes.
TextDecorationInfo looks up decorating boxes by decoration index and
only uses their offset, style and font, so behavior is unchanged.

Bug: 571157138
Change-Id: Ie5b2e55d0475e60ceeba042f0120e65355af4d9c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8716148
Commit-Queue: David A <awogbemila@chromium.org>
Reviewed-by: Kent Tamura <tkent@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735126}

derinel@google.com

[RPS] Add per-actor RemoteActorConfig in components/password_manager

Adds RemoteActorConfig, GetRemoteActorConfig(), and
GetRemoteActorConfigForOrigin() in
components/password_manager/core/browser/remote_actor/ to resolve
per-actor branding and permission parameters (display_name,
learn_more_url, icon_url, oauth_client_id, and task_namespace_id) for
Remote Actor credential sharing:
- Resolves compiled-in defaults for RemoteActorType::kGemini (namespace
30) and RemoteActorType::kSearch (namespace 41), with default actor
display names in components/password_manager_strings.grdp (marked
translateable="false" for now; translations will be enabled later).
- Applies per-actor Finch overrides from actor_configs_json (actor_name,
learn_more_url, icon_url, oauth_client_id, and task_namespace_id) with
HTTPS URL validation and fallback to safe defaults.

Bug: b:580766669
Change-Id: I0d0dda9aeac1c40ba8842a510b348709d3664541
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8714519
Reviewed-by: Maria Kazinova <kazinova@google.com>
Commit-Queue: Adem Derinel <derinel@google.com>
Reviewed-by: Rafał Godlewski <rgod@google.com>
Cr-Commit-Position: refs/heads/main@{#1735125}

rgod@google.com

[iOS] Remove DanglingUntriaged from WebNavigationBrowserAgentTest

Clear the NTP delegate on WebNavigationBrowserAgent in the test fixture
destructor so FakeNTPDelegate is deallocated before TestBrowser closes
its WebStates.

#codehealth

Change-Id: Ied35366dfc36d8127fd36919a2e51116406ff929
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8690310
Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Commit-Queue: Rafał Godlewski <rgod@google.com>
Cr-Commit-Position: refs/heads/main@{#1735124}

nicolasmacbeth@google.com

Revert "[iOS][gardening] Suppress GeminiAppSwitchingTestCase snackbar on Device"

This reverts commit 05fa91ddb25ecf0c1d8bd2cd4597bf1bb2d671bf.

Reason for revert: crrev.com/c/8719968 removes the cause for these
failures

Original change's description:
> [iOS][gardening] Suppress GeminiAppSwitchingTestCase snackbar on Device
>
> Suppress
> testAppSwitchingAISummarizationMismatchedAccountSnackbar on Device
> due to consistent failures on ipad-device:
> - https://ci.chromium.org/b/8666516830537081553
> - https://ci.chromium.org/b/8666497747265006625
>
> No-Try: true
> Bug: 572662973
> Change-Id: Id564ed9cf3b1b08a2b92ad62476a20e868d30e4e
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8688301
> Reviewed-by: Joemer Ramos <joemerramos@google.com>
> Auto-Submit: Kirubel <kaklilu@chromium.org>
> Owners-Override: Kirubel <kaklilu@chromium.org>
> Commit-Queue: Joemer Ramos <joemerramos@google.com>
> Cr-Commit-Position: refs/heads/main@{#1731781}

Bug: 572662973
Fixed: 578284246
Change-Id: I04a0f5571aadd47abc4102229581c5f040ee87d6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8719550
Reviewed-by: Ibrahim Kanouche <kanouche@google.com>
Commit-Queue: Nicolas MacBeth <nicolasmacbeth@google.com>
Cr-Commit-Position: refs/heads/main@{#1735123}

fedegermi@google.com

[iOS] Consume App Group command once in the new startup flow

The new startup flow now removes the App Group command from the shared
defaults when an X-Callback `/app-group-command` task is created, as the
legacy startup flow does. Before, it only read the command time and left
the command in place, so later requests could see it again.

Moreover, histogram "Startup.ApplicationGroupCommandDelay" it's not
recorded anymore in task_request code, that code was moved in a shared
app_group_command_payload.mm logic.

Bug: 493816082
Change-Id: Id90e57737d2c0ec626547053df6096ac19aa91ad
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8704336
Commit-Queue: Federica Germinario <fedegermi@google.com>
Reviewed-by: Sylvain Defresne <sdefresne@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735122}

chromium-autoroll@skia-public.iam.gserviceaccount.com

Roll Perfetto from 796a01189b09 to d5eea1535fd4 (1 revision)

https://chromium.googlesource.com/external/github.com/google/perfetto.git/+log/796a01189b09..d5eea1535fd4

2026-11-04 pablogamito@google.com Expose FlushFlags::CloneTarget in lib C ABI (#8324)

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://autoroll.skia.org/r/perfetto-chromium-autoroll
Please CC chrometto-team+alerts@google.com,kartarsingh@google.com,primiano@chromium.org on the revert to ensure that a human
is aware of the problem.

To file a bug in Chromium: https://bugs.chromium.org/p/chromium/issues/entry

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Cq-Include-Trybots: luci.chromium.try:linux-perfetto-rel
Bug: chromium:578511663
Submodule-Gn-Analyze: true
Change-Id: I4a13cb878f82d4e8e50e22814a591882af637e65
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8725657
Bot-Commit: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735121}

chromium-autoroll@skia-public.iam.gserviceaccount.com

Roll Skia from d65f91e3bd77 to 1c96be51b288 (2 revisions)

https://skia.googlesource.com/skia.git/+log/d65f91e3bd77..1c96be51b288

2026-11-04 skia-autoroll@skia-public.iam.gserviceaccount.com Manual roll ANGLE from 60b8a1c2aa98 to 0a536a02be2d (7 revisions)
2026-11-04 nicolettep@google.com Remove unnecessary host write access flag in Vulkan Ganesh backend

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://autoroll.skia.org/r/skia-autoroll
Please CC jmbetancourt@google.com,skiabot@google.com on the revert to ensure that a human
is aware of the problem.

To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry
To file a bug in Chromium: https://bugs.chromium.org/p/chromium/issues/entry

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:gpu-fyi-cq-android-arm64;luci.chromium.try:linux-blink-rel;luci.chromium.try:linux-chromeos-compile-dbg;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:gpu-fyi-cq-mac-arm64;luci.chromium.try:win_optional_gpu_tests_rel;luci.chromium.try:gpu-fyi-cq-win-arm64
Cq-Do-Not-Cancel-Tryjobs: true
Bug: None
Submodule-Gn-Analyze: true
Change-Id: Ie881a0aba89b2e3aaf824ea693c0f0ba5f0afadd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8727267
Bot-Commit: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735120}

tommasin@chromium.org

Autofill: Reset TTF state before HideTouchToFill call

In TouchToFillAutofillDelegateAndroidImpl::HideTouchToFill,
manager_->client().HideTouchToFillAutofillNotice() is called while
ttf_autofill_state_ is still kShowing. When the Android bridge
synchronously invokes OnDismissed() during hide, OnDismissed()
sees kShowing, transitions to kSuppressing, and calls
TriggerAskForValuesToFill(). Additionally, shown_notice_ was not
cleared on hide or dismiss.

Set ttf_autofill_state_ to kInactive and reset shown_notice_ before
calling HideTouchToFillAutofillNotice(), and add a unit test.

Before: https://issues.chromium.org/579853921#attachment84644072
After: https://issues.chromium.org/579853921#attachment84642965
Screencast: https://issues.chromium.org/579853921#attachment84650537

Bug: 579853921
Change-Id: I31da7259f693251de9cf069a769e49c96b7f10bb
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8687538
Reviewed-by: Timofey Chudakov <tchudakov@google.com>
Commit-Queue: Nicola Tommasi <tommasin@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735119}

chromium-autoroll@skia-public.iam.gserviceaccount.com

Roll Chrome Linux PGO Profile

Roll Chrome Linux PGO profile from chrome-linux-main-1793771794-0d370084997d10187e14bf8e519ee1734cdd57df-bc9c8dd29978a19861ff46463a6c26e0bb1b7309.profdata to chrome-linux-main-1793793385-2043fc54f4439867bc824d93af739750dd0749fe-b8b9e813434efddb2eed020daa9cc44686aeee7a.profdata

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://autoroll.skia.org/r/pgo-linux-chromium
Please CC chrome-brapp-engprod@google.com,pgo-profile-sheriffs@google.com on the revert to ensure that a human
is aware of the problem.

To file a bug in Chromium: https://bugs.chromium.org/p/chromium/issues/entry

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Cq-Include-Trybots: luci.chrome.try:linux-chrome
Merge-Approval-Bypass: Chrome autoroller
Change-Id: Ib004c8104c1e58f31d940bbfe344e02c3c95d51e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8726579
Bot-Commit: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735118}

aattar@google.com

[ResourceBroker] Add routing eligibility metric

Record ResourceBroker.RoutingDecision (with .MainFrame and .Subframe
variants) when creating a network-backed document subresource
URLLoaderFactory while kResourceBroker is enabled.

No factory is interposed yet; this metric measures which factories are
eligible for Resource Broker routing and the first reason others are
excluded (error pages, opaque or non-HTTP(S)/WebUI/extension origins,
Isolated Web Apps, unlocked/PDF/guest/privileged processes, cross-origin
isolated contexts, nonce-partitioned network state, or service worker
control).

Because the RenderFrameHost may still be speculative at creation time,
eligibility is evaluated using only state already set for the committing
document.

Bug: 560232768
Change-Id: I1b58fa1b995ce0607b79c94f8e4338825f7808b9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8082681
Commit-Queue: Alex Attar <aattar@google.com>
Reviewed-by: Caitlin Fischer <caitlinfischer@google.com>
Reviewed-by: Camille Lamy <clamy@chromium.org>
Reviewed-by: Kouhei Ueno <kouhei@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735117}

zequanwu@google.com

[updater] Disable Mach-O fixup chains for x64 debug/ASan UpdaterSetup

After LLVM 88d89008c958 ("[lld/mac] Error out on too large chained
fixups (#233895)"), ld64.lld rejects binaries where a chained fixup's
segment page index reaches UINT16_MAX (65,535) instead of silently
truncating the 16-bit page index and emitting a broken Mach-O binary.

UpdaterSetup embeds updater.zip into __DATA,__updater_zip via
-sectcreate. In static debug and ASan x64 builds (4 KiB pages),
updater.zip exceeds 256 MiB (65,535 * 4 KiB), and because ld64.lld
places thread-local sections at the end of __DATA, __DATA,__thread_data
is pushed past page 65,535.

Pass -Wl,-no_fixup_chains for x64 debug/ASan builds so UpdaterSetup
uses classic LC_DYLD_INFO_ONLY opcodes instead of chained fixups.

Bug: 576214311,581484658
Change-Id: If58e704c7ce449d1031f3f8bb9bfb09d3aee05a2
Cq-Include-Trybots: luci.chromium.try:mac-updater-try-builder-dbg
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8722628
Commit-Queue: Nico Weber <thakis@chromium.org>
Reviewed-by: Hans Wennborg <hans@chromium.org>
Owners-Override: Nico Weber <thakis@chromium.org>
Reviewed-by: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735116}

chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com

Roll clank/internal/apps from c1a92520b93a to 3cc060d6c737 (1 revision)

https://chrome-internal.googlesource.com/clank/internal/apps.git/+log/c1a92520b93a..3cc060d6c737

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://skia-autoroll.corp.goog/r/clank-apps-chromium-autoroll
Please CC chrome-brapp-engprod@google.com,tommasin@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Bug: None
No-Try: true
Change-Id: I03293ecbc69e2a5b0247f80f96caa12910175d3f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8725149
Commit-Queue: chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Bot-Commit: chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735115}

lpromero@google.com

Use FakePageContextWrapper in CSD host tests

Injects PageContextWrapperFactory in the ClientSideDetectionHostIOS
constructor and migrates ClientSideDetectionHostIOSTest to use
FakePageContextWrapper and FakePageContextWrapperFactory instead of
FakeClientSideDetectionPageContextWrapper and
OCMClassMock([PageContextWrapper class]) +alloc stubbing.

Bug: 580897793
Change-Id: Ic4e91471cebeb6d9cc9fd3db06f1fb2f3419c0c4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8702104
Reviewed-by: jdh <jdh@chromium.org>
Auto-Submit: Louis Romero <lpromero@google.com>
Commit-Queue: jdh <jdh@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735114}

lpromero@google.com

Use FakePageContextWrapper in composebox tests

Migrates ComposeboxInputPlateMediator and
ComposeboxInputPlateMediatorTest to use FakePageContextWrapper and
PageContextWrapperFactory instead of a local
FakeComposeboxPageContextWrapper subclass.

Bug: 580897793
Change-Id: I87674636d4c2b9e45ed3cc017f8ad4db623acd25
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8701080
Reviewed-by: Christian Xu <christianxu@chromium.org>
Auto-Submit: Louis Romero <lpromero@google.com>
Commit-Queue: jdh <jdh@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735113}

lpromero@google.com

Add FakePageContextWrapper test support target

Adds FakePageContextWrapper and FakePageContextWrapperFactory in
//ios/chrome/browser/intelligence/proto_wrappers:test_support along with
PageContextWrapperFactory and DefaultPageContextWrapperFactory in
//ios/chrome/browser/intelligence/proto_wrappers so unit tests can
inject and control PageContextWrapper extraction without executing
JavaScript, starting timers, or stubbing +alloc with OCMock.

Migrates GeminiTabHelperTest, GeminiContainerMediatorTest,
ModelLedSuggestionsServiceImplTest, and PageContextUtilTest to use
FakePageContextWrapper and PageContextWrapperFactory injection.

Bug: 580897793
Change-Id: I79011850b879a68a561301c2025d72603241531f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8704262
Commit-Queue: jdh <jdh@chromium.org>
Reviewed-by: Sylvain Defresne <sdefresne@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735112}

aymanm@google.com

Pass -E to diff3 in Cronet Copybara imports

GNU diff3 -m defaults to -A (--show-all), which treats hunks changed
identically in both origin and destination as merge conflicts whenever
another part of the same file differs and bypasses Copybara's whole-file
equality check. Pass a wrapper script via --diff-bin that invokes diff3
with -E (--show-overlap) so non-overlapping reconverged hunks merge
cleanly while real 3-way conflicts still surface.

Bug: b:392575734
Change-Id: I1974411f9ce8b146947b965780d7d26498900c7f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8720248
Auto-Submit: Mohannad Farrag <aymanm@google.com>
Commit-Queue: Mohannad Farrag <aymanm@google.com>
Reviewed-by: Vladyslav Kasprov <vkasprov@google.com>
Commit-Queue: Vladyslav Kasprov <vkasprov@google.com>
Cr-Commit-Position: refs/heads/main@{#1735111}

luci-bisection@appspot.gserviceaccount.com

Revert "Enable startup optimization field trial testing configs"

This reverts commit 11df5a067d538f1ebad492e15862fb15d731e0f7.

Reason for revert:
LUCI Bisection has identified this change as the cause of a test failure. See the analysis: https://ci.chromium.org/ui/p/chromium/bisection/test-analysis/b/5524495298199552

Sample build with failed test: https://ci.chromium.org/b/8666068258673789281
Affected test(s):
[://chrome/test\:sync_integration_tests!gtest::TwoClientThemesSyncTest#CustomThenSyncNative_E2ETest/kSyncTransportOnly](https://ci.chromium.org/ui/test/chromium/:%2F%2Fchrome%2Ftest%5C:sync_integration_tests%21gtest::TwoClientThemesSyncTest%23CustomThenSyncNative_E2ETest%2FkSyncTransportOnly?q=VHash%3A473864b6cd30fe94)

If this is a false positive, please report it at http://b.corp.google.com/createIssue?component=1199205&description=Analysis%3A+https%3A%2F%2Fci.chromium.org%2Fui%2Fp%2Fchromium%2Fbisection%2Ftest-analysis%2Fb%2F5524495298199552&format=PLAIN&priority=P3&title=Wrongly+blamed+https%3A%2F%2Fchromium-review.googlesource.com%2Fc%2Fchromium%2Fsrc%2F%2B%2F8701497&type=BUG

Original change's description:
> Enable startup optimization field trial testing configs
>
> Enable field trial testing configurations for three landed startup
> optimizations that are currently disabled by default:
> - GpuPersistentCacheAsyncInitializeStudy
> (`kGpuPersistentCacheAsyncInitialize`)
> - GtkPrefetchNavButtonIconsStudy (`kGtkPrefetchNavButtonIcons`)
> - OptimizeVariationsIpcStudy (`kOptimizeVariationsIpc`)
>
> Bug: 577796100, 578334326
> Change-Id: I2883e1db61e353ef394aec4ee832fad333fe08bb
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8701497
> Auto-Submit: Eriko Kurimoto <elkurin@chromium.org>
> Reviewed-by: Mingyu Lei <leimy@chromium.org>
> Commit-Queue: Mingyu Lei <leimy@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1734919}
>

Bug: 577796100, 578334326
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Change-Id: If9355f2eeb6511aae95835ee7d616e9c0d0bbfa0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8727367
Pure-Revert: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Noam Rosenthal <nrosenthal@google.com>
Cr-Commit-Position: refs/heads/main@{#1735110}

brunobraga@google.com

[AutofillAi] Remove obsolete first run dialog, strings, and metrics

Following the removal of the Autofill AI Opt-In IPH
(crrev.com/c/8193338), ShowAutofillAiFirstRunDialog and the
feature_first_run helper have no remaining callers.

This CL removes:
- chrome/browser/ui/feature_first_run/ (autofill_ai_first_run_dialog and
feature_first_run_helper, along with its browsertest).
- The Autofill.Ai.OptIn.Funnel histogram, AutofillAiOptInFunnelEvents
enum, and LogOptInFunnelEvent.
- Obsolete strings (IDS_AUTOFILL_AI_FFR_*, IDS_AUTOFILL_AI_OPT_IN_IPH_*,
and IDS_APP_TURN_ON) and the IDR_AUTOFILL_AI_FFR_BANNER_LOTTIE
resource.
- Obsolete UI constants (kFeatureFirstRunDialogContentViewElementId,
kAutofillAiOptInIphElementId, kColorFeatureFirstRun*, and unused
DISTANCE_FEATURE_FIRST_RUN_INFO_BOX_* layout constants).

OBSOLETE_HISTOGRAM[Autofill.Ai.OptIn.Funnel]=Removed along with the
obsolete Autofill AI Opt-In IPH and First Run dialog.

Bug: 440488776
Change-Id: Iedbdda85612c1aded1a7da1782fc4533ecf4ca0c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8715011
Reviewed-by: Jan Keitel <jkeitel@google.com>
Reviewed-by: Rainhard Findling <rainhard@chromium.org>
Commit-Queue: Bruno Braga <brunobraga@google.com>
Cr-Commit-Position: refs/heads/main@{#1735109}

ammarmk@chromium.org

[Passwords] Fix desktop trusted vault save error resolution metrics

When saving a password on desktop is blocked by a trusted vault error:

1. Clicking "Continue" in the save bubble invokes
SaveUpdateBubbleController::OnTrustedVaultUnlockClicked(), which did
not set a dismissal reason. When the recovery tab opened and closed
the bubble, the interaction was recorded as NO_DIRECT_INTERACTION in
dismissal histograms and UKM, incremented the smart-bubble dismissal
count, and sent a "no interaction" vote. Add CLICKED_RESOLVE_ERROR
to UIDismissalReason (mapped to kAccepted in
PasswordFormMetricsRecorder), record it in
OnTrustedVaultUnlockClicked(), and update the unit test that
previously called OnSaveClicked() and masked the issue.
2. PasswordManager.SaveWithTrustedVaultError.Outcome was only logged on
Android. Log one outcome per "Continue" click on desktop in
ManagePasswordsUIController when the pending error-resolution save
concludes (saved after key retrieval, never save for this site, or
abandoned before resolution).

Bug: 579388824
Change-Id: I18424ef6cdf5b6eceab49223ac30eac6f6d8e8a5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8689985
Reviewed-by: Yurii Lahodiuk <ylahodiuk@google.com>
Reviewed-by: Rafał Godlewski <rgod@google.com>
Reviewed-by: Mohamed Amir Yosef <mamir@chromium.org>
Commit-Queue: Ammar Mohammed Kheir <ammarmk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735108}

chromium-autoroll@skia-public.iam.gserviceaccount.com

Roll WebView ARM Orderfile from H7JUNXuUS981avDHq... to KdN4BP4SJyUjCwEfI...

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://autoroll.skia.org/r/orderfile-webview-arm-chromium
Please CC woa-engprod@google.com on the revert to ensure that a human
is aware of the problem.

To file a bug in Chromium: https://bugs.chromium.org/p/chromium/issues/entry

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Merge-Approval-Bypass: Chrome autoroller
Change-Id: Icae94a1d8ecebceaddbc2a95e4ec3c36c96ce81b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8724068
Bot-Commit: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735107}

dpenning@chromium.org

[unified tabstrip] Log TabStrip.TimeToSwitch in unified horizontal

Emit TabStrip.TimeToSwitch in TabStripView::RecordMousePressedInTab when
orientation is horizontal so unified horizontal tabstrip matches legacy
TabStrip::OnMouseEventInTab.

Bug: 523330272
Change-Id: I1937fdcf8854d938d4d29b98a6417dd48e9ff414
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8709317
Reviewed-by: Caroline Rising <corising@chromium.org>
Auto-Submit: David Pennington <dpenning@chromium.org>
Commit-Queue: Caroline Rising <corising@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1735106}

chromium-autoroll@skia-public.iam.gserviceaccount.com

Roll Chrome Mac Arm PGO Profile

Roll Chrome Mac Arm PGO profile from chrome-mac-arm-main-1793786283-04b44570bdfe81098e8b250340e7899e114824c4-11df5a067d538f1ebad492e15862fb15d731e0f7.profdata to chrome-mac-arm-main-1793793385-12b96d7b0e342b8e334d322c0c5dff21e9e432b0-b8b9e813434efddb2eed020daa9cc44686aeee7a.profdata

If this roll has caused a breakage, revert this CL and set the roller
to dry run mode using the controls here:
https://autoroll.skia.org/r/pgo-mac-arm-chromium
Please CC chrome-brapp-engprod@google.com,pgo-profile-sheriffs@google.com on the revert to ensure that a human
is aware of the problem.

To file a bug in Chromium: https://bugs.chromium.org/p/chromium/issues/entry

To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md

Cq-Include-Trybots: luci.chrome.try:mac-chrome
Merge-Approval-Bypass: Chrome autoroller
Change-Id: Ib2c26179f4d762aa121b8eb67fb5c590f459fff3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8726377
Commit-Queue: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Bot-Commit: chromium-autoroll@skia-public.iam.gserviceaccount.com <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1735105}

stefanoduo@google.com

Add a flag to let the built-in resolver succeed with "partial success"

Currently, whenever HostResolverDnsTask is configured both for A and
AAAA transaction, if either address query fails with a
non-ERR_NAME_NOT_RESOLVED error (e.g., ERR_DNS_SERVER_FAILURE,
ERR_DNS_TIMED_OUT, etc), the entire HostResolverDnsTask immediately
aborts. This behavior seems to stem from the assumption that a
HostResolverSystemTask is always present as a fallback. With the
addition of InsecureDnsMode::kEnabledPlatformNoSystem this is no longer
the case. When that mode is selected, HostResolverDnsTask will never
fallback onto HostResolverSystemTask. Failing HostResolverDnsTask, just
because just the A or AAAA query failed, feels suboptimal. It completely
ignores the fact that one of the two queries produced usable results.

This CL introduces kIgnoreIfOtherTransactionResolves, a new
TransactionErrorBehavior. When specified, A succeeding, while AAAA
fails, leads to a HostResolverDnsTask::Results containing only the A
result (or vice-versa). This is what getaddrinfo already does
internally, and this change stops the built-in resolver from having to
fallback onto getaddrinfo to achieve this behavior.

This is achieved by holding:
1. Holding onto single transaction failures, until both A and AAAA have
terminated
2. Once both have terminated, figure out whether we should report a
failure/success/etc

Additionally, this renames the kFallback TransactionErrorBehavior to
kFailTask. Whether failing the task leads to a fallback system task or
not is an implementation detail that should live outside of
HostResolverDnsTask.

Bug: 448975408
Change-Id: Id8b741a612429f489c9277a59d9f8df563294042
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8537215
Auto-Submit: Stefano Duo <stefanoduo@google.com>
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Reviewed-by: Nidhi Jaju <nidhijaju@chromium.org>
Commit-Queue: Stefano Duo <stefanoduo@google.com>
Cr-Commit-Position: refs/heads/main@{#1735104}

hempjudith@google.com

[Permissions] use std::optional<PermissionAction> for chip user decision

PermissionPromptChipModel and the omnibox chip views previously reused
PermissionAction::NUM to represent an unset user decision. Represent
that state with std::optional<PermissionAction> instead so
PermissionAction::NUM can be replaced with kMaxValue in a follow-up.

The views' initial decision is now unset rather than GRANTED, matching
the model; no visible change because callers sync both decision and
prompt style.

Bug: 414527270
Test: unit_tests --gtest_filter=*WebUIPermissionChip*:*WebUILocationBar*
Change-Id: Ie7495c945d08de117e7a5514ff083bdc5f91c49b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/8684805
Reviewed-by: Antonio Sartori <antoniosartori@chromium.org>
Commit-Queue: Judith Hemp <hempjudith@google.com>
Cr-Commit-Position: refs/heads/main@{#1735103}