Importing blink mojo types into other directories that aren't blink causes imports of internal blink header files (WTF::String/Vector/HashMap). This can then be abused by the caller to create structures that it shouldn't have access to. This is preliminary cleanup for adding an enforcement to the blink mojo header files.
Change-Id: I0aa1e30f07f595023191af1c96cdf58a258dc118
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5953880
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1372961}
This CL creates a mojo struct that represents a source location. The goal is to have only two states when we manage source locations: 1. source location exists 2. no source location at all. Having a struct for source location allows us to make a source location itself nullable (avoiding having optional fields inside) and limit the possible cases to only the two.
Here are the specific changes in this CL:
- Created a new mojom file and defined a struct for source location there.
- BFCache-related structs now use the source location struct in common.
Bug: 324201881
Change-Id: I4061ad708010f895416e5881d46c16ba8a3251d6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5276765
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Fergal Daly <fergal@chromium.org>
Commit-Queue: Kurumi Muto <mkurumi@google.com>
Cr-Commit-Position: refs/heads/main@{#1263204}
This CL is to change the mojo struct `BackForwardCacheNotRestoredReasons` so it can pass source location information (i.e. url, line number, column number) to the renderer side. In a follow-up, we will send the actual source locations of reasons to the renderer's web local frame, so the NotRestoredReasons API can provide source locations.
Before this CL,
- In the mojo struct `BackForwardCacheNotRestoredReasons`, BFCache blocking reasons were represented as a vector of string without source location.
- The message is sent to the renderer side as mojo `CommitNavigationParams`'s not_restored_reasons: https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/public/mojom/navigation/navigation_params.mojom;l=595?q=CommitNavigationParams%20-f:out&ss=chromium%2Fchromium%2Fsrc
In this CL,
- The mojo struct has fields for source location, therefore `CommitNavigationParams`'s not_restored_reasons can now contain source location information.
- But, actual source locations are not sent yet.
In follow-up CLs,
- Actual source locations gathered in back_forward_cache_impl.cc will be sent to a web local frame in the renderer.
- The NotRestoredReasons API will provide source locations.
Change-Id: I0c63b19ea3076ffa5aeb734ad87388a285961a3b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5250654
Commit-Queue: Kurumi Muto <mkurumi@google.com>
Reviewed-by: Kouhei Ueno <kouhei@chromium.org>
Reviewed-by: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1257775}
This CL updates NotRestoredReasons API's implementation to conform to the spec draft[1].
Namely, this CL
(1) removes "preventedBFCache" enum,
(2) adds "masked" to the outermost main frame when cross-origin iframes are blocking, and
(3) creates NotRestoredReasonDetails class on the blink side to make the reason into a struct like {"reason": "error-document"}.
As for (2), when `kAllowCrossOriginSubframes` flag is on, we add "masked" to the randomly selected cross-origin iframe.
Also with (2), `reasons` field is no longer specific to same-origin frames, so I'm moving the field out of `SameOriginDetails`.
We also need to add "masked" when UA internal reasons such as memory pressure block, and that's going to be addressed in a follow-up.
[1]: https://github.com/whatwg/html/pull/9360/files
Bug: 1472637
Change-Id: I89702696a128eea31b2dd2af0ef8c9874e7213ca
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5086563
Reviewed-by: Takashi Toyoshima <toyoshim@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1249965}
This CL starts saving a map of [feature, vector of BlockingDetailsPtr] as part of CanStoreDocumentResult.
This way the information is now passed to DevTools (we already pass CanStoreDocumentResult to DevTools).
Bug:1366675
Change-Id: I3269b0d20abef3241cdc5f1a1d5ea41b613babc8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4332776
Reviewed-by: Fergal Daly <fergal@chromium.org>
Commit-Queue: Yuzu Saijo <yuzus@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1176443}
This is a reland of commit a7387520f7
Calls to parent classes' SetUpCommandLine were missing.
Original change's description:
> Add BackForwardCacheUnloadAllowed flag.
>
> This flag controls the interaction between back/forward cache and
> unload. When it's enabled, pages with unload handlers may enter the
> cache.
>
> It is enabled by default on Android.
>
> Add/update tests to cover all scenarios.
>
> Change-Id: Ibc45486cbff6f07aa2533414f59ec8e1220772d2
> Bug: 1419856
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4293576
> Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
> Commit-Queue: Fergal Daly <fergal@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1110753}
Bug: 1419856
Change-Id: Id1bf47532a44393ba559e98c34e05afea83b2d6e
Cq-Include-Trybots: luci.chromium.try:android-bfcache-rel,linux-bfcache-rel
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4297556
Commit-Queue: Rakina Zata Amni <rakina@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Auto-Submit: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1110850}
This flag controls the interaction between back/forward cache and
unload. When it's enabled, pages with unload handlers may enter the
cache.
It is enabled by default on Android.
Add/update tests to cover all scenarios.
Change-Id: Ibc45486cbff6f07aa2533414f59ec8e1220772d2
Bug: 1419856
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4293576
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1110753}
in the frame is invalid
IsDOMContentLoaded can be completed when the url is valid.
If a subframe with an invalid url exists in a page, bfcache does not work because IsDOMContentLoaded of the frame is always false.
This required test updates:
- ChildFrameCaptureContentFirst disable BFCache because the test requires that the frame be evicted
- the data= version of embedded-not-found-expected.txt now passes
This adds WebContentsWaiter a base class for making test-waiters that wait for a WebContentObserver event to occur.
R=toyoshim@chromium.org
Bug: 1366707
Change-Id: I371cd9ad55f342d9c71ba24eaf5b8b5981d818ba
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3912822
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1107116}
The network limit is sometimes exceeded, e.g.
BackForwardCacheBrowserTestWithMediaSession.CacheWhenMediaSessionServiceIsNotUsed
plays a large video file and it might exceed the limit.
FIX: Set the limit to INT_MAX by default and only set a low limit in
tests that test the limit.
Also set the grace_period_to_finish_loading_in_seconds as it's on the same feature and it makes sense to do it (even though nothing was flaky due to this).
Bug: 1338348
Change-Id: If952526139c3179307b2743f24747e023a9522cf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3956324
Reviewed-by: Yuzu Saijo <yuzus@chromium.org>
Commit-Queue: Fergal Daly <fergal@chromium.org>
Auto-Submit: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1060852}
A base::Feature should not be copyable; a feature is intended to have a
single global instance, as it caches internal mutable state.
Feature vectors in conjunction with `base::test::ScopedFeatureList` are
by far the most common use of the base::Feature copy constructor. This
is a mostly automated replacement of `std::vector<base::Feature>` with
`std::vector<base::test::FeatureRef>` to remove that dependency.
This CL was uploaded by git cl split.
R=nasko@chromium.org
Bug: 1370572
Change-Id: Iecfb813c08c475e042234aab91577cff999b4101
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3933905
Commit-Queue: Daniel Cheng <dcheng@chromium.org>
Auto-Submit: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Nasko Oskov <nasko@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1055143}
A base::Feature should not be copyable; a feature is intended to have a
single global instance, as it caches internal mutable state.
The feature copying here ends up being problematic in conjunction with
the migration to use base::test::FeatureRef for ScopedFeatureList: a
FeatureRef ends up taking a reference to a stack temporary, causing a
use-after-return.
The fix is to just pass base::test::FeatureRef all the way through; once
all feature copies are eliminated, the copy constructor will be deleted,
making future bugs of this sort harder to write.
This also cleans up some of the map-related handling: in particular,
using unnamed namespaces in headers is discouraged, since each
translation unit that includes the header ends up with its own private
copy of the functions in the unnamed namespace(s). Since this code is
not performance critical, just use std::map for simplicity.
Bug: 1370572
Change-Id: Ic1c11571c62e1c8124d6a214fc3c4a4c78968e21
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3933984
Commit-Queue: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Fergal Daly <fergal@chromium.org>
Commit-Queue: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1055118}
This is a reland of commit 43acbdc6b7
The previous commit didn't remove the flag from `testing/buildbot`,
which caused failure in `blink_web_tests`. See https://ci.chromium.org/ui/p/chromium/builders/ci/linux-bfcache-rel/36826/overview
In this patch, the `enable-back-forward-cache-same-site` flag is
removed:
https://chromium-review.googlesource.com/c/chromium/src/+/3865414/1..2
Original change's description:
> Clean up `enable_same_site` and `skip_same_site_if_unload_exists`
> feature and the corresponding logic since we may not need to control
> them any more.
>
> As we will assume the same-site BFCache to be always enabled
> instead of reading from the feature param, this CL removes the
> following functions and replace them with corresponding general
> BFCache checks instead:
> - `content::BackForwardCache::
> IsSameSiteBackForwardCacheFeatureEnabled()` is replaced by
> `content::BackForwardCache::IsBackForwardCacheFeatureEnabled()`.
> - `content::IsSameSiteBackForwardCacheEnabled()` is replaced by
> `content::IsBackForwardCacheEnabled()`.
>
> Similarly, as we will not blocking same site BFCache even if unload
> handler exists, the `skip_same_site_if_unload_exists` flag will be
> processed as `false` all the time. The following function is removed:
> - `content::ShouldSkipSameSiteBackForwardCacheForPageWithUnload()`.
>
> This CL also updates all the browser tests and unit tests to stop
> setting up testing context with variants in `enable_same_site`
> feature, as well as to clear all the tests against the cases when
> same-site BFCache is disabled. The testing class dedicated for the
> `skip_same_site_if_unload_exists` is also removed.
>
>
> Bug: 1353982
> Change-Id: Ib299db733fc9af35e5d35b63b1b166e1880502c9
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3858766
> Reviewed-by: Stephen Nusko <nuskos@chromium.org>
> Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
> Reviewed-by: David Trainor <dtrainor@chromium.org>
> Reviewed-by: Michael Bai <michaelbai@chromium.org>
> Commit-Queue: Mingyu Lei <leimy@chromium.org>
> Reviewed-by: Arthur Sonzogni <arthursonzogni@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1041643}
Bug: 1353982
Change-Id: Ibb5c67e85248e5f5a05c9602f22dc0266c5f7844
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3865414
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Reviewed-by: David Trainor <dtrainor@chromium.org>
Reviewed-by: Taylor Bergquist <tbergquist@chromium.org>
Commit-Queue: Mingyu Lei <leimy@chromium.org>
Reviewed-by: Stephen Nusko <nuskos@chromium.org>
Reviewed-by: Arthur Sonzogni <arthursonzogni@chromium.org>
Reviewed-by: Hans Wennborg <hans@chromium.org>
Reviewed-by: Michael Bai <michaelbai@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1044366}
This reverts commit 43acbdc6b7.
Reason for revert: blink_web_tests failing with 'AssertionError raised: enable-back-forward-cache-same-site is not defined in FlagSpecificConfig' on Ubuntu, see https://ci.chromium.org/ui/p/chromium/builders/ci/linux-bfcache-rel/36826/overview
Original change's description:
> Clean up `enable_same_site` and `skip_same_site_if_unload_exists`
> feature and the corresponding logic since we may not need to control
> them any more.
>
> As we will assume the same-site BFCache to be always enabled
> instead of reading from the feature param, this CL removes the
> following functions and replace them with corresponding general
> BFCache checks instead:
> - `content::BackForwardCache::
> IsSameSiteBackForwardCacheFeatureEnabled()` is replaced by
> `content::BackForwardCache::IsBackForwardCacheFeatureEnabled()`.
> - `content::IsSameSiteBackForwardCacheEnabled()` is replaced by
> `content::IsBackForwardCacheEnabled()`.
>
> Similarly, as we will not blocking same site BFCache even if unload
> handler exists, the `skip_same_site_if_unload_exists` flag will be
> processed as `false` all the time. The following function is removed:
> - `content::ShouldSkipSameSiteBackForwardCacheForPageWithUnload()`.
>
> This CL also updates all the browser tests and unit tests to stop
> setting up testing context with variants in `enable_same_site`
> feature, as well as to clear all the tests against the cases when
> same-site BFCache is disabled. The testing class dedicated for the
> `skip_same_site_if_unload_exists` is also removed.
>
>
> Bug: 1353982
> Change-Id: Ib299db733fc9af35e5d35b63b1b166e1880502c9
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3858766
> Reviewed-by: Stephen Nusko <nuskos@chromium.org>
> Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
> Reviewed-by: David Trainor <dtrainor@chromium.org>
> Reviewed-by: Michael Bai <michaelbai@chromium.org>
> Commit-Queue: Mingyu Lei <leimy@chromium.org>
> Reviewed-by: Arthur Sonzogni <arthursonzogni@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#1041643}
Bug: 1353982
Change-Id: I0e4cf162bbc8ca0029ac91f3e4957550f7bebb01
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3868134
Owners-Override: Taylor Bergquist <tbergquist@chromium.org>
Auto-Submit: Taylor Bergquist <tbergquist@chromium.org>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Taylor Bergquist <tbergquist@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1041777}
feature and the corresponding logic since we may not need to control
them any more.
As we will assume the same-site BFCache to be always enabled
instead of reading from the feature param, this CL removes the
following functions and replace them with corresponding general
BFCache checks instead:
- `content::BackForwardCache::
IsSameSiteBackForwardCacheFeatureEnabled()` is replaced by
`content::BackForwardCache::IsBackForwardCacheFeatureEnabled()`.
- `content::IsSameSiteBackForwardCacheEnabled()` is replaced by
`content::IsBackForwardCacheEnabled()`.
Similarly, as we will not blocking same site BFCache even if unload
handler exists, the `skip_same_site_if_unload_exists` flag will be
processed as `false` all the time. The following function is removed:
- `content::ShouldSkipSameSiteBackForwardCacheForPageWithUnload()`.
This CL also updates all the browser tests and unit tests to stop
setting up testing context with variants in `enable_same_site`
feature, as well as to clear all the tests against the cases when
same-site BFCache is disabled. The testing class dedicated for the
`skip_same_site_if_unload_exists` is also removed.
Bug: 1353982
Change-Id: Ib299db733fc9af35e5d35b63b1b166e1880502c9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3858766
Reviewed-by: Stephen Nusko <nuskos@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Reviewed-by: David Trainor <dtrainor@chromium.org>
Reviewed-by: Michael Bai <michaelbai@chromium.org>
Commit-Queue: Mingyu Lei <leimy@chromium.org>
Reviewed-by: Arthur Sonzogni <arthursonzogni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1041643}
- Make impl provides `histogram_tester_`.
- Update the signature of `ukm_recorder()`.
- Move `AddSampleToBuckets()` to local.
- Add more documentation about how this class actually works.
Bug: 1352894
Change-Id: I1835a33c06318493e498ff5e2be91f00eee21e84
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3832154
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Ming-Ying Chung <mych@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1035380}
This CL does not change any test behaviors nor modify any methods. Rather it helps reduce the size of the grand class `BackForwardCacheBrowserTest` by extracting histogram & ukm related Expect*() methods into a base class `BackForwardCacheTestMatcher`, such that other non-bfcached tests can also reuse the same logic. It also removes a dead method `ExpectTotalCount()`.
The next possible work is to not expose the private methods, and remove the complex implicit dependencies between those methods.
Change-Id: I9329819870e94716d43df4d609c16c17ca58a391
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3828394
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Ming-Ying Chung <mych@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1035000}
Previously we used dedicated worker but that is about to be unblocked.
This switches to using plugin, which will never be unblocked and also
encapsulates the choice of feature so we have only one place to update
in future.
It also
- adds a main-frame test which seems to have been missing
- adds documentation on when to use this vs UseDummyStickyBackForwardCacheDisablingFeatureForTesting
Change-Id: I769e676de5f9a58d79bbe1bbb16f95d7503b1eeb
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3747520
Commit-Queue: Fergal Daly <fergal@chromium.org>
Auto-Submit: Fergal Daly <fergal@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1021173}
It was returning a base::ListValue, which is deprecated.
Also update consumers as necessary (none of which really care much
about the ListValue vs Value difference).
Bug: 1338341
Change-Id: Icaa4f734c87177f130ed8c0a2266a940b727fd75
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3727194
Reviewed-by: Avi Drissman <avi@chromium.org>
Commit-Queue: Matt Menke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1018740}
This allows modules/levels to be added to vlog at runtime.
Use this to enable back_forward_cache_impl's vlogging in BackForwardCacheTest so that failing tests are accompanied by detailed logs.
This ends up updating the protections on g_vlog_info to prevent threads from clobbering the value when they initialize.
Change-Id: Id9a4eef496f109d2fdd07486960f76a722638de2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3648806
Commit-Queue: Fergal Daly <fergal@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1013834}
This patch must be a no-op with all the common build flags.
Add the "DanglingUntriaged" raw_ptr annotation. It indicates a
raw_ptr becomes dangling, and it should be triaged/fixed.
This annotates the one firing during content_browsertests.
Multiple follow-up will complete the list, up until being able to
enable dangling pointer detection on a bot.
Stats:
- 175 DanglingUntriaged in code
- 69 DanglingUntriaged in tests
Bug=1291138
Change-Id: I00771bd46403f90297aa5b972fdd1ddc23b18e07
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3687960
Commit-Queue: Arthur Sonzogni <arthursonzogni@chromium.org>
Reviewed-by: Bartek Nowierski <bartekn@chromium.org>
Owners-Override: Bartek Nowierski <bartekn@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1013525}
Pages with unload handlers were put into bfcache when unload strategy was
set to kAlways, but in Finch we always set this to kNo.
This CL blocks pages with unload handlers by default, and only allow them
on Android.
Bug: 1312486, 1316120
Change-Id: I80237668984ba79c90807176db67584c00df9732
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3590233
Reviewed-by: Fergal Daly <fergal@chromium.org>
Commit-Queue: Yuzu Saijo <yuzus@chromium.org>
Cr-Commit-Position: refs/heads/main@{#994048}
This CL uses the new class to iterate thru the frame tree to check
bfcache eligibility, in order to correctly capture eviction reasons.
Now NotRestoredReason tree is only created through the builder.
This also adds test cases to make sure eviction reasons are recorded.
Bug: 1297914
Change-Id: I727da918ff4c71152554d0af598fa25fd2457b42
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3467300
Reviewed-by: Fergal Daly <fergal@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Yuzu Saijo <yuzus@chromium.org>
Cr-Commit-Position: refs/heads/main@{#984701}
In the bug, a crash occurs because we try to cache an interstitial. We
catch some error documents via status codes etc but interstitials do
not consistently set those. Checking IsErrorDocument() is more reliable.
Bug: 1274308,1287996
Change-Id: Ifec662c169c77e33ca5dc4d56b0e42c8d71f1d97
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3319862
Commit-Queue: Fergal Daly <fergal@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Reviewed-by: Andrey Kosyakov <caseq@chromium.org>
Reviewed-by: Alexander Timin <altimin@chromium.org>
Cr-Commit-Position: refs/heads/main@{#981026}
This CL enables the experiment to only consider per-process network
request buffer limit by default.
The experiment has been running on stable 50% and the results look good,
and now we are proceeding to advance it to stable 100% via finch.
But before that we need to enable it on TOT.
Also this CL removes browser tests that use per-request limit, which will
be no longer used after enabling the feature.
Bug: 1243600
Change-Id: Id65d6554a1c8ec9f087916dc03cc8a4fc3529e85
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3337833
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Fergal Daly <fergal@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Yuzu Saijo <yuzus@chromium.org>
Cr-Commit-Position: refs/heads/main@{#952362}
This makes tests a bit clearer and more concise but I have also seen 2
tests in the last 2 days that incorrectly omitted the WaitForLoadStop.
This updates all of BFCache's tests to use these. It also changes all
EXPECTs to ASSERTs because continuing after the history traversal fails
is almost certainly going to generate noise, not signal.
Change-Id: I61d6f0797f4e6bc52f927bc83cccc2361fd045f6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3288729
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#943378}
Bluetooth is only used by one test but ends up getting compiled in for
all groups of tests. This will reduce include size and compile time a
little for those.
Bug: 1267091
Change-Id: I8f96c4b1632559983cc41614cd99f29daa742854
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3269792
Auto-Submit: Fergal Daly <fergal@chromium.org>
Reviewed-by: Rakina Zata Amni <rakina@chromium.org>
Commit-Queue: Fergal Daly <fergal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#940194}
