This relands commit e7481c5056.
Which was reverted in commit dc8e02663b.
There was an issue in the original CL where the argument
--test_policy_to_pref_mappings_filter gets popuplated when there are no
policy pref mapping test files. This wasn't noticed because c++ unit
tests and browser tests ignore this flag and work correctly. Android
tests OTOH throw an error because the flag is unknown.
This CL filters policy pref mapping files when building the flag.
Original change's description:
> autotest.py: Support policy to pref mappings tests
>
> Pref mapping test files now work, for example:
> $ tools/autotest.py -C out/Default \
> components/policy/test/data/pref_mapping/NewWindowsInKioskAllowed.json
> $ tools/autotest.py -C out/Default \
> components/policy/test/data/pref_mapping \
> --test_policy_to_pref_mappings_filter=NewWindowsInKioskAllowed
>
> Which are equivalent to:
> $ autoninja -C out/Default browser_tests
> $ out/Default/browser_tests \
> --gtest_filter="*PolicyPrefsTest.PolicyToPrefsMapping" \
> --test_policy_to_pref_mappings_filter="NewWindowsInKioskAllowed"
>
> Bug: b:319254568
> Test: tools/autotest.py -C out/Default components/policy/test/data/pref_mapping
> Test: tools/autotest.py -C out/Default components/policy/test/data/pref_mapping/NewWindowsInKioskAllowed.json
> Change-Id: I8eb36a69abd717bf69c56a7f1aa453209c3da5a2
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5185801
> Reviewed-by: Maksim Ivanov <emaxx@chromium.org>
> Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
> Commit-Queue: Edman Anjos <edman@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1254565}
Bug: b:319254568
Change-Id: I4f3aab804b27ae5939a87804be270b15304d8561
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5257580
Reviewed-by: Maksim Ivanov <emaxx@chromium.org>
Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Commit-Queue: Edman Anjos <edman@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1255138}
Direct access to Layout() is being deprecated. This marks all callsites
for future fixing, and ensures they all go through the common layout
wrapper functions.
Bug: 1521108
Change-Id: I3adff3b9b005329096d2d1fbe960b1bf4b3da076
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5246255
Commit-Queue: Peter Kasting <pkasting@chromium.org>
Reviewed-by: Allen Bauer <kylixrd@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1254666}
This CL is to update the mojo docs to reflect that nullable support for primitive types is currently available. Additionally, it clarifies that arrays, maps, and unions of primitive nullable types are not yet supported.
Bug: 1521618, 657632
Change-Id: I498e07e65aa2dd886abff2340b0e666de9f2861c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5238130
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Kurumi Muto <mkurumi@google.com>
Reviewed-by: Mingyu Lei <leimy@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1253851}
This synchronously invokes Layout(). For now, it does nothing else, but
eventually it can do any pre-layout work necessary, e.g. collecting
metrics or setting state/access control variables.
Ultimately, this should become the only way for any code to reach
Layout().
Bug: 1521108
Change-Id: Idd40832c6e5f2d407e970a089d3c2801423922f6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5243921
Auto-Submit: Peter Kasting <pkasting@chromium.org>
Commit-Queue: Peter Kasting <pkasting@chromium.org>
Reviewed-by: Elaine Chien <elainechien@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1253558}
Currently Chromium is using "expires" option in Use-As-Dictionary
response header for the dictionary lifetime. The option name has been
renamed to "ttl" in the spec discussed in IETF HTTPWG.
And the new change in the spac [1] removed this, and the lifetime should
be calculated from the response's freshness [2].
After this CL, when V2 backend is enabled, Chromium will not check
"expires" option, and it will check the response's freshness.
[1]: https://github.com/httpwg/http-extensions/pull/2709
[2]: https://datatracker.ietf.org/doc/html/rfc9111#name-freshness
Bug: 1413922
Change-Id: Idd2cdeb526099a1f2854ea0094541cedc0a64af8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5227360
Commit-Queue: Tsuyoshi Horo <horo@chromium.org>
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1251343}
A new "Content-Dictionary" response header was introduced by this spec
change [1].
When V2 backend is enabled, if the "Content-Encoding" response header
indicates that a dictionary is used, Chrome will check the
"Content-Dictionary" response header. And if there is no such header or
it doesn't match with the value of "Available-Dictionary" request
header, Chrome will treat the response as an error.
[1]: https://github.com/httpwg/http-extensions/pull/2680
Bug: 1413922
Change-Id: I37d09fad1201f4fb4b059d65c4aaee58cec788ac
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5226641
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Commit-Queue: Tsuyoshi Horo <horo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1251341}
The latest spec discussed in IETF HTTPWG is using `Available-Dictionary`
request header name instead of `Sec-Available-Dictionary`.
Also in the value is using a Structured Field sf-binary sha256 hash
of the dictionary after [1].
So after this CL, when V2 backend is enabled, Chrome will set
`Available-Dictionary` header with a Structured Field sf-binary sha256 hash of the dictionary.
[1]: https://github.com/httpwg/http-extensions/pull/2680
Bug: 1413922
Change-Id: Ibb716ec27bb8f5d58a5266ad6ac3ec75ad643130
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5224886
Reviewed-by: Kenichi Ishibashi <bashi@chromium.org>
Commit-Queue: Tsuyoshi Horo <horo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1251259}
This reverts commit a1d5b0e679.
Reason for revert: Broke Cronet's ARM64 bot (https://ci.chromium.org/ui/p/chromium/builders/ci/android-cronet-arm64-rel/118013/overview)
Original change's description:
> Reland^3 "android: Make arm64 builds HighEnd by default"
>
> This reverts commit 58b4a751db.
>
> Reason for revert: Added pgo to failing bots (separate commits)
>
>
> > > Original change's description:
> > > > android: Make arm64 builds HighEnd by default
> > > >
> > > > The most popular arm64 builds (Trichrome) are shipping with
> > > > is_high_end_android. The Monochrome arm64 builds are shipped for old
> > > > devices, and we expect these to gradually become less popular.
> > > >
> > > > This change sets is_high_end_android to true for arm64 builds by
> > > > default.
> > > >
> > > > This would make local benchmarking slightly less error prone because
> > > > there would be a little less to remember to set in GN args when making a
> > > > build for the most popular official configuration.
> > > >
> > > > Known effects:
> > > >
> > > > 1. Seemingly all official arm64 builders now explicitly set
> > > > is_high_end_android. Hence no effect for them.
> > > >
> > > > 2. Non-official arm64 builders will change to become high-end, even the
> > > > ones that end up running on old devices (example: android-arm64-rel).
> > > > In this patch we make sure that is_high_end_android is ignored for
> > > > non-official builds (disabled RELR relocations for non-official).
> > > > Hence, no effect here as well.
> > > >
> > > > 3. Official builds of Monochrome made by devs would fail to run on
> > > > Android pre-P (because RELR support is lacking). This leads to
> > > > mysterious crashes during something-dlopen in the Android dynamic
> > > > linker. Hopefully nobody does this before we remove support for
> > > > Android O.
> > > >
> > > > Bug: 1515925
> > > > Change-Id: I695c98852bb410894144452211aa06cc32ffb141
> > > > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5169071
> > > > Reviewed-by: Andrew Grieve <agrieve@chromium.org>
> > > > Commit-Queue: Egor Pasko <pasko@chromium.org>
> > > > Cr-Commit-Position: refs/heads/main@{#1247286}
> > >
> > > Bug: 1515925
> > > Change-Id: I4cb32d81b253a24e7c38bbb982fa5bbf0849c236
> > > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5213025
> > > Reviewed-by: Mohamed Heikal <mheikal@chromium.org>
> > > Auto-Submit: Andrew Grieve <agrieve@chromium.org>
> > > Commit-Queue: Andrew Grieve <agrieve@chromium.org>
> > > Cr-Commit-Position: refs/heads/main@{#1249201}
>
> Bug: 1515925
> Change-Id: Ic4cbd4f7e9f0486a3c7922936612f43c98a686a5
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5217121
> Commit-Queue: Andrew Grieve <agrieve@chromium.org>
> Reviewed-by: Mohamed Heikal <mheikal@chromium.org>
> Reviewed-by: Egor Pasko <pasko@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1250269}
Bug: 1515925
Change-Id: I14fa74fd19da70569dc67802edcaff6cc51fd1fa
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5224652
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Commit-Queue: Mohannad Farrag <aymanm@google.com>
Reviewed-by: Egor Pasko <pasko@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1250805}
This reverts commit 58b4a751db.
Reason for revert: Added pgo to failing bots (separate commits)
> > Original change's description:
> > > android: Make arm64 builds HighEnd by default
> > >
> > > The most popular arm64 builds (Trichrome) are shipping with
> > > is_high_end_android. The Monochrome arm64 builds are shipped for old
> > > devices, and we expect these to gradually become less popular.
> > >
> > > This change sets is_high_end_android to true for arm64 builds by
> > > default.
> > >
> > > This would make local benchmarking slightly less error prone because
> > > there would be a little less to remember to set in GN args when making a
> > > build for the most popular official configuration.
> > >
> > > Known effects:
> > >
> > > 1. Seemingly all official arm64 builders now explicitly set
> > > is_high_end_android. Hence no effect for them.
> > >
> > > 2. Non-official arm64 builders will change to become high-end, even the
> > > ones that end up running on old devices (example: android-arm64-rel).
> > > In this patch we make sure that is_high_end_android is ignored for
> > > non-official builds (disabled RELR relocations for non-official).
> > > Hence, no effect here as well.
> > >
> > > 3. Official builds of Monochrome made by devs would fail to run on
> > > Android pre-P (because RELR support is lacking). This leads to
> > > mysterious crashes during something-dlopen in the Android dynamic
> > > linker. Hopefully nobody does this before we remove support for
> > > Android O.
> > >
> > > Bug: 1515925
> > > Change-Id: I695c98852bb410894144452211aa06cc32ffb141
> > > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5169071
> > > Reviewed-by: Andrew Grieve <agrieve@chromium.org>
> > > Commit-Queue: Egor Pasko <pasko@chromium.org>
> > > Cr-Commit-Position: refs/heads/main@{#1247286}
> >
> > Bug: 1515925
> > Change-Id: I4cb32d81b253a24e7c38bbb982fa5bbf0849c236
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5213025
> > Reviewed-by: Mohamed Heikal <mheikal@chromium.org>
> > Auto-Submit: Andrew Grieve <agrieve@chromium.org>
> > Commit-Queue: Andrew Grieve <agrieve@chromium.org>
> > Cr-Commit-Position: refs/heads/main@{#1249201}
Bug: 1515925
Change-Id: Ic4cbd4f7e9f0486a3c7922936612f43c98a686a5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5217121
Commit-Queue: Andrew Grieve <agrieve@chromium.org>
Reviewed-by: Mohamed Heikal <mheikal@chromium.org>
Reviewed-by: Egor Pasko <pasko@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1250269}
This reverts commit 7ce33d1442.
Reason for revert: More bot failures:
* chromium-internal-arm64-builder-rel
* chromium-internal-arm-builder-rel
* arm64-unpublished-builder-rel
* arm64-builder-rel
Original change's description:
> Reland^2 "android: Make arm64 builds HighEnd by default"
>
> This reverts commit df45839bc3.
>
> Reason for reland: android-binary-size-generator now has PGO profiles,
> and this change now updates static initializer logic.
>
> Original change's description:
> > android: Make arm64 builds HighEnd by default
> >
> > The most popular arm64 builds (Trichrome) are shipping with
> > is_high_end_android. The Monochrome arm64 builds are shipped for old
> > devices, and we expect these to gradually become less popular.
> >
> > This change sets is_high_end_android to true for arm64 builds by
> > default.
> >
> > This would make local benchmarking slightly less error prone because
> > there would be a little less to remember to set in GN args when making a
> > build for the most popular official configuration.
> >
> > Known effects:
> >
> > 1. Seemingly all official arm64 builders now explicitly set
> > is_high_end_android. Hence no effect for them.
> >
> > 2. Non-official arm64 builders will change to become high-end, even the
> > ones that end up running on old devices (example: android-arm64-rel).
> > In this patch we make sure that is_high_end_android is ignored for
> > non-official builds (disabled RELR relocations for non-official).
> > Hence, no effect here as well.
> >
> > 3. Official builds of Monochrome made by devs would fail to run on
> > Android pre-P (because RELR support is lacking). This leads to
> > mysterious crashes during something-dlopen in the Android dynamic
> > linker. Hopefully nobody does this before we remove support for
> > Android O.
> >
> > Bug: 1515925
> > Change-Id: I695c98852bb410894144452211aa06cc32ffb141
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5169071
> > Reviewed-by: Andrew Grieve <agrieve@chromium.org>
> > Commit-Queue: Egor Pasko <pasko@chromium.org>
> > Cr-Commit-Position: refs/heads/main@{#1247286}
>
> Bug: 1515925
> Change-Id: I4cb32d81b253a24e7c38bbb982fa5bbf0849c236
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5213025
> Reviewed-by: Mohamed Heikal <mheikal@chromium.org>
> Auto-Submit: Andrew Grieve <agrieve@chromium.org>
> Commit-Queue: Andrew Grieve <agrieve@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1249201}
Bug: 1515925
Change-Id: Ie26da9ac5d6b82b089623f86efc3633af6d0b77c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5217468
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Auto-Submit: Andrew Grieve <agrieve@chromium.org>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#1249367}
