Skip to content

GitLab

T
tangled
Switch branch/tag
  1. 28 Jun, 2018 6 commits
    • Xianzhu Wang's avatar
      [CI] Simplify PaintController cache skipping · dc7889d8
      Xianzhu Wang authored 
      The previous cache skipping mechanism supported cache skipping for
some display items of a client while keeping other display items
cacheable. Now simplify by making all display items of a client
non-cacheable once the client creates some display item that skips
cache.

Other changes that are required:
- PaintRecordBuilder needs a transient-mode PaintController (which
  doesn't support caching) to avoid the PaintController from altering
  the cache flags of the painted display items and clients which may
  be cached in other PaintControllers.
- PaintController::AppendDebugDrawingAfterCommit() needs a dedicated
  client to avoid its cache skipping affect the normal caching of
  the client.

Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I0baf248074334198a5b8dce889d566bf75cdbf2b
Reviewed-on: https://chromium-review.googlesource.com/1115619
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: default avatarPhilip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#570957}
      dc7889d8
    • Tiger Oakes's avatar
      supersize html_report: Added check to ensure output directory exists · 7b45c6bd
      Tiger Oakes authored 
      Small convenience feature so we don't need to run mkdir first.
Also backported to old supersize UI.

Change-Id: I96a462ff584992e3b7bff52d43be5706c5f12b31
Reviewed-on: https://chromium-review.googlesource.com/1117713
Commit-Queue: Tiger Oakes <tigero@google.com>
Reviewed-by: default avatarEric Stevenson <estevenson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#570956}
      7b45c6bd
    • Matthew Jones's avatar
      Remove button existence check in top control container · fb3530d0
      Matthew Jones authored 
      The overflow menu is no longer necessarily in the top control
container. This patch removes the assert and the handle to the button
in the control container since it isn't actually used.

Bug: 857124
Change-Id: I6d0d7d913c9743f773a8e1a2b563a38c3d9096d9
Reviewed-on: https://chromium-review.googlesource.com/1117306Reviewed-by: default avatarTed Choc <tedchoc@chromium.org>
Commit-Queue: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#570955}
      fb3530d0
    • Theresa's avatar
      [EoC] A toolbar entry point behind a new flag · c34db2de
      Theresa authored 
      Add an experimental toolbar button that triggers the contextual
suggestions bottom sheet.

Currently the auto-peek behavior for the contextual suggestions bottom
sheet is not disabled when the toolbar button is enabled. This will be
handled in a follow-up CL.

BUG=856403

Change-Id: Ic56aeb44f8ce9dc29606a902553d9d25803df40e
Reviewed-on: https://chromium-review.googlesource.com/1115352
Commit-Queue: Theresa <twellington@chromium.org>
Reviewed-by: default avatarTed Choc <tedchoc@chromium.org>
Reviewed-by: default avatarMatthew Jones <mdjones@chromium.org>
Reviewed-by: default avatarFilip Gorski <fgorski@chromium.org>
Cr-Commit-Position: refs/heads/master@{#570954}
      c34db2de
    • Christian Biesinger's avatar
      [layoutng] Override computeIntrinsicLogicalWidths so we can use the NG code for it · 29b9eafe
      Christian Biesinger authored 
      This is mostly straightforward, except that on Mac, we may have to
compute intrinsic sizes outside of layout. So this patch also adds
special code to avoid layout calls in such cases (for orthogonal
flows). However, this also required me to change some tests to have
the correct "inside of layout" state.

Notes on tests:
  fast/dynamic/012.html
    Had an incorrect rebaseline; rendering matches legacy modulo
    text splitting (positioning is the same)

  css2.1/20110323/table-height-algorithm-023.htm
  css2.1/20110323/table-height-algorithm-024.htm
    Parts of these tests were already using the smaller size due to
    1px text rendering differences; this rebaseline just makes us
    more consistent.

  tables/mozilla/bugs/bug57828.html
    Also a trivial 1px difference

  tables/mozilla/bugs/bug60992.html
    Probably an issue of bug 40634; legacy takes the whitespace before
    <br> (here, a linebreak) into account for calculating the width
    of the cell and NG does not.

  tables/mozilla_expected_failures/bugs/bug1010.html
    Also 1px and 2px text rendering difference

  virtual/layout_ng_experimental/external/wpt/css/css-flexbox/flexbox_stf-table-caption.html
    Fails for an unrelated reason, see newly filed bug 857185

Bug: 635619
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng
Change-Id: Ia11317acd777aec201702fce28ebab74dba99b4f
Reviewed-on: https://chromium-review.googlesource.com/508293
Commit-Queue: Christian Biesinger <cbiesinger@chromium.org>
Reviewed-by: default avatarEmil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#570953}
      29b9eafe
    • Robert Ma's avatar
      Add an expectation for an incoming test to unblock import · 04007607
      Robert Ma authored 
      The test behaves differently on try bots and CQ bots so it can't be
rebaselined automatically.

TBR=raphael.kubo.da.costa@intel.com

No-Try: True
Bug: 857271
Change-Id: I8a1fe9f524cca6cf73370b76bba3d64557ab33f8
Reviewed-on: https://chromium-review.googlesource.com/1117842
Commit-Queue: Robert Ma <robertma@chromium.org>
Reviewed-by: default avatarRobert Ma <robertma@chromium.org>
Cr-Commit-Position: refs/heads/master@{#570952}
      04007607
  3. 27 Jun, 2018 34 commits