Skip to content

GitLab

T
tangled
Switch branch/tag
  1. 21 Oct, 2018 6 commits
    • Ehsan Karamad's avatar
      Implement feature-policy: 'layout-animations' · 9024e95a
      Ehsan Karamad authored 
      This CL converts the current "animations" feature policy to
"layout-animations" feature policy: the former policy is about blocking
non-composited animations whereas the new policy only blocks a handful
of CSS animations which are known to (potentially) cause re-layout and
are also widely used. The current set of CSS properties considered in
the new policy is: {bottom, height, left, right,top, width}.

Bug: 874218
Change-Id: I97f08ce6b2b902c7057ea821416a9c145d26f068
Reviewed-on: https://chromium-review.googlesource.com/c/1235055
Commit-Queue: Ehsan Karamad <ekaramad@chromium.org>
Reviewed-by: default avatarJeremy Roman <jbroman@chromium.org>
Reviewed-by: default avatarRobert Flack <flackr@chromium.org>
Reviewed-by: default avatarKen Buchanan <kenrb@chromium.org>
Reviewed-by: default avatarIan Clelland <iclelland@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601412}
      9024e95a
    • Ehsan Karamad's avatar
      Reland "Implement fallback content for RemoteFrameOwner" · 192a8da2
      Ehsan Karamad authored 
      This is a reland of 02cb80e6

The issues seems to have been the test case {'a', 'b', 'b'} where
apparently the failed navigation still fires a load event which might
arrive sooner than fallback update. The reland modifies the test file
to suppress sending 'OBJECT_LOAD' event from the test page.

Original change's description:
> Implement fallback content for RemoteFrameOwner
>
> RemoteFrameOwner does not implement the logic for fallback content. This
> means if a cross-origin navigation fails with some error, the owner
> element (in this context, <object>) in the parent process does not get
> notified and will not use its fallback content (instead the frame might
>  show an error page).
>
> When the <object> has fallback content, it should always use that over
> the frame's error message. To support this matter, this CL implements
> fallback methods in RemoteFrameOwner. Essentially,
>
>   * When <object> creates a local frame, the corresponding frame tree
> node will be marked as the type that "can" render fallback content. This
> will propagte everywhere using FrameReplicationState.
>
>   * When the provisional loading of a frame fails, RemoteFrameOwner will
> notify the browser through the current proxy for navigation. The browser
> then uses the parent frame to notify the renderer process that the owner
> element should render its own fallback content.
>
>   * If the <object> does not specify fallback, the navigation to error
> page commits and the error page is shown.
>
>   * When the <object> renders its own content, the remote frame stays
> alive. This is a bug but not a new one; It will be fixed in future CLs.
>
> Bug: 853140
> Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng
> Change-Id: Icad3934ccfc1823c0cdecd8e1223e6370ea4b3bb
> Reviewed-on: https://chromium-review.googlesource.com/c/1105635
> Commit-Queue: Ehsan Karamad <ekaramad@chromium.org>
> Reviewed-by: Daniel Cheng <dcheng@chromium.org>
> Reviewed-by: Ehsan Karamad <ekaramad@chromium.org>
> Reviewed-by: Wei Li <weili@chromium.org>
> Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#601194}

TBR=dcheng@chromium.org,alexmos@chromium.org,weili@chromium.org

Bug: 853140
Change-Id: Ib1594aa642ea238f745fd29d191818fddc503afc
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng
Reviewed-on: https://chromium-review.googlesource.com/c/1292899Reviewed-by: default avatarEhsan Karamad <ekaramad@chromium.org>
Commit-Queue: Ehsan Karamad <ekaramad@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601411}
      192a8da2
    • chromium-internal-autoroll's avatar
      Roll src-internal 91b0f1063bdf..410c469a02ae (4 commits) · ff9175cd
      chromium-internal-autoroll authored 
      https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/91b0f1063bdf..410c469a02ae


Created with:
  gclient setdep -r src-internal@410c469a02ae

The AutoRoll server is located here: https://autoroll-internal.skia.org/r/src-internal-chromium-autoroll

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


TBR=mmoss@chromium.org

Change-Id: I082bcad16b0770c4256cd13c0415236859dc78f9
Reviewed-on: https://chromium-review.googlesource.com/c/1292799Reviewed-by: default avatarchromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Commit-Queue: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#601410}
      ff9175cd
    • Christopher Cameron's avatar
      BrowserCompositorMac: Remove detached compositor state · a71197c1
      Christopher Cameron authored 
      With MacViews, it is now very rare to actually need a
RenderWidgetHostViewMac to have its own compositor. The only
circumstances where this happens are
- content shell
- popup windows (the date/time picker)
- tab capture of backgrounded tabs

Remove the concept of having a "detached" compositor, and simplify
the transitions between states. This change will introduce flashes to
paths that rely on the RenderWidgetHostViewMac having its own
compositor (which are now vanishingly rare).

Change-Id: I0feeec67a42f05de05aa4d72fe9af93f033d9d64
Reviewed-on: https://chromium-review.googlesource.com/c/1292377
Commit-Queue: ccameron <ccameron@chromium.org>
Commit-Queue: Fady Samuel <fsamuel@chromium.org>
Reviewed-by: default avatarFady Samuel <fsamuel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601409}
      a71197c1
    • Saman Sami's avatar
      Fix blank page after switching out of simplified view · 4580e632
      Saman Sami authored 
      Sometimes EmbedSurface is called right before WasShown when the tab
is hidden. When frame evictor is notified of this new surface, it
might immediately evict it because it doesn't know the tab will be
visible soon. This will result in an evicted surface getting embedded
which will show blank. Do two things:
- Only call SwappedFrame when the tab is visible so don't have the
problem mention above. Generally there is no point in calling
SwappedFrame when it's invisible because even though a new id
is allocated, there will not be any CompositorFrames submitted to this
surface until the tab is shown, so there is no point in notifying the
frame evictor.
- Notify RHWVAndroid via WasEvicted when the surface is evicted so that
it can allocate a new id for the next time it's shown.

Bug: 896633,893731
Change-Id: I6fbc339f606d70e65ecc79b0f74bcd44133bd4fc
Reviewed-on: https://chromium-review.googlesource.com/c/1289953
Commit-Queue: Saman Sami <samans@chromium.org>
Reviewed-by: default avatarKhushal <khushalsagar@chromium.org>
Reviewed-by: default avatarFady Samuel <fsamuel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601408}
      4580e632
    • chrome-release-bot's avatar
      Updating trunk VERSION from 3587.0 to 3588.0 · 9b88efb6
      chrome-release-bot authored 
      # This is an automated release commit.
# Do not revert without consulting chrome-pmo@google.com.
NOAUTOREVERT=true
TBR=mmoss@chromium.org

Change-Id: I4859bb33a11bff11c65a9568155e3fb114039193
Reviewed-on: https://chromium-review.googlesource.com/c/1292381Reviewed-by: default avatarchrome-release-bot@chromium.org <chrome-release-bot@chromium.org>
Cr-Commit-Position: refs/heads/master@{#601407}
      9b88efb6
  3. 20 Oct, 2018 34 commits