GitLab

Add an in-process bridge between the gpu, main and drm threads so that
the drm ozone implementation can operate in single-process mode.

BUG=558602

Review URL: https://codereview.chromium.org/1831693002

Cr-Commit-Position: refs/heads/master@{#383856}

It doesn't really make sense to get overflow scrollbars for an iframe (the
real scrollbars are drawn within the frame, and the frame itself cannot
overflow in any useful way).

Rather than address this in the legacy compositing code, this makes the style
system prevent this nonsensical state from ever occurring, by adjusting the
overflow property to visible for all frame elements.

This matches the existing behavior of the non-composited path here: no
scrollbars are drawn in response to the 'overflow' property on a frame.

BUG=384154
TEST=LayoutTests/scrollbars/iframe-overflow-scroll-ignored.html

Review URL: https://codereview.chromium.org/1843723002

Cr-Commit-Position: refs/heads/master@{#383855}

We were calling willAnimaterToOffset with the updated target position before
clearing the animation state when in PostAnimationCleanup. As a result, we
didn't take into account the potentially modified scroll offset by a
programmatic scroll.

BUG=598548

Review URL: https://codereview.chromium.org/1842623003

Cr-Commit-Position: refs/heads/master@{#383854}

This is what I meant to do in https://codereview.chromium.org/1770983002/

BUG=548423

Review URL: https://codereview.chromium.org/1818603002

Cr-Commit-Position: refs/heads/master@{#383853}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/9bcc481fe3e6..2dc8af398989

$ git log 9bcc481fe..2dc8af398 --date=short --no-merges --format='%ad %ae %s'

TBR=catapult-sheriff@chromium.org

Review URL: https://codereview.chromium.org/1836373002

Cr-Commit-Position: refs/heads/master@{#383852}

Revert of Allocator shims working on VS2015. (patchset #11 id:200001 of https://codereview.chromium.org/1414453017/ )

Reason for revert:
It breaks the link on Windows.

FAILED: C:\b\depot_tools\python276_bin\python.exe gyp-win-tool link-with-manifests environment.x86 True views_examples_with_content_exe.exe "C:\b\depot_tools\python276_bin\python.exe gyp-win-tool link-wrapper environment.x86 False link.exe /nologo /OUT:views_examples_with_content_exe.exe @views_examples_with_content_exe.exe.rsp" 1 mt.exe rc.exe "obj\ui\views\examples\views_examples_with_content_exe.views_examples_with_content_exe.exe.intermediate.manifest" obj\ui\views\examples\views_examples_with_content_exe.views_examples_with_content_exe.exe.generated.manifest ..\..\ui\views\examples\views_examples.exe.manifest ..\..\build\win\compatibility.manifest
allocator.lib(allocator.allocator_shim_win.obj) : error LNK2005: __query_new_mode already defined in libucrt.lib(new_mode.obj)

allocator.lib(allocator.allocator_shim_win.obj) : error LNK2005: __set_new_mode already defined in libucrt.lib(new_mode.obj)

allocator.lib(allocator.allocator_shim_win.obj) : error LNK2005: _calloc already defined in libucrt.lib(calloc.obj)

allocator.lib(allocator.allocator_shim_win.obj) : error LNK2005: _free already defined in libucrt.lib(free.obj)

allocator.lib(allocator.allocator_shim_win.obj) : error LNK2005: _malloc already defined in libucrt.lib(malloc.obj)

allocator.lib(allocator.allocator_shim_win.obj) : error LNK2005: _realloc already defined in libucrt.lib(realloc.obj)

views_examples_with_content_exe.exe : fatal error LNK1169: one or more multiply defined symbols found

Original issue's description:
> Allocator shims working on VS2015.
>
> VS2015 is happy for us to simply override the CRT symbols since they
> seem to be defined weakly by the universal CRT.
>
> This shim is far simpler than the previous one, as the libcmt stripping
> technique can be removed.
>
> Allocator shim is also now only be enabled on Release Static builds.
>
> BUG=481611
> TEST=base_unittests --gtest_filter=*Memory* in all configurations
>  (Debug/Release/Static/Component/GN/gyp/x64/x86)
> CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel
>
> Committed: https://crrev.com/e20c2e0ab4e5b796e7a1890fb9048b5eb307d015
> Cr-Commit-Position: refs/heads/master@{#383810}

TBR=thakis@chromium.org,brucedawson@chromium.org,rnk@chromium.org,piman@chromium.org,primiano@chromium.org,wfh@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=481611

Review URL: https://codereview.chromium.org/1839783005

Cr-Commit-Position: refs/heads/master@{#383851}

In the future, content/browser/compositor will likely be
moved to components/display_compositor. In anticipation of
that move, I'm moving files that are content specific to
content/browser/renderer_host.

BUG=598792

Review URL: https://codereview.chromium.org/1841153002

Cr-Commit-Position: refs/heads/master@{#383850}

This is required to transition consumers of PlatformFileForTransit into using
attachment brokering. Alias PlatformFileForTransit to base::SharedMemoryHandle,
since that class has the exact semantics required for brokering a HANDLE.

I tried to avoid unnecessary functionality changes, but couldn't avoid changing
some logic in NaclHostMessageFilter and DesktopSessionProxy. Both classes
receive an IPC message that contain a HANDLE from the sender process, and try to
copy the HANDLE into the current process. By switching to an
attachment-brokerable class, PlatformFileForTransit is automatically brokered,
so the duplication logic is no longer required.

Expect future CLs to convert the remaining consumers of PlatformFileForTransit
into using attachment brokering.

BUG=493414

Review URL: https://codereview.chromium.org/1830853002

Cr-Commit-Position: refs/heads/master@{#383849}

If you edit a DOM item in console it should now properly change view
ports and go into editing mode instead of breaking elements panel and
doing nothing.

BUG=598166
R=lushnikov

Review URL: https://codereview.chromium.org/1844623002

Cr-Commit-Position: refs/heads/master@{#383848}

BUG=597000

Review URL: https://codereview.chromium.org/1837053003

Cr-Commit-Position: refs/heads/master@{#383847}

Three separate crashes were possible when a RenderViewHost is pending
deletion.  We now compute this state on demand rather than storing it
in a boolean, since it can change between the time of swapout and the
next navigation.  We were also failing to check this state in
FrameTree::GetRenderViewHost.

This also clears the pending delete RenderFrameHost if its process
dies, since we won't hear a swap out ACK from it after that.

BUG=581912, 544755, 591478
TEST=See bugs for repro steps.
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review URL: https://codereview.chromium.org/1835833002

Cr-Commit-Position: refs/heads/master@{#383846}

- actually parsing png data
- png dictionary

BUG=584819

Review URL: https://codereview.chromium.org/1839203002

Cr-Commit-Position: refs/heads/master@{#383845}

Adds a few junit tests for the getAllPagesAsync as a template for other
asynchronous functions on OfflinePageBridge.

BUG=598476,595252
R=dewittj@chromium.org

Review URL: https://codereview.chromium.org/1841103002

Cr-Commit-Position: refs/heads/master@{#383844}

https://chromium.googlesource.com/skia.git/+log/cfb6bdf76779..9b2ef62d4708

$ git log cfb6bdf76..9b2ef62d4 --date=short --no-merges --format='%ad %ae %s'
2016-03-29 jcgregorio fiddle: Only output errors to stderr.

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=djsollen@google.com

Review URL: https://codereview.chromium.org/1839183002

Cr-Commit-Position: refs/heads/master@{#383843}

We've switched from using a vector to using a hash map for the cached
items. This allows us to raise the limit of the number of (unlocked)
items we cache. Set this to 1000 (arbitrarily).

R=ericrk, enne
BUG=577421
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1839783004

Cr-Commit-Position: refs/heads/master@{#383842}

https://chromium.googlesource.com/native_client/src/native_client.git/+log/8151fc95fb5c..a36a2c6d3a01

$ git log 8151fc95f..a36a2c6d3 --date=short --no-merges --format='%ad %ae %s'
2016-03-29 smklein PNaCl: Update llvm revision in pnacl/COMPONENT_REVISIONS

CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_chromium_dbg_32_ng,linux_nacl_sdk_build
TBR=mseaborn@chromium.org

Review URL: https://codereview.chromium.org/1838353003

Cr-Commit-Position: refs/heads/master@{#383841}

It generates logs:

V/chromium(22041): [VERBOSE1:blimp_message_pump.cc(59)] Received <BlimpMessage type=COMPOSITOR target_tab_id=0 byte_size=61379>
V/chromium(22041): [VERBOSE1:blimp_connection.cc(66)] Sending <BlimpMessage type=PROTOCOL_CONTROL subtype=CHECKPOINT_ACK checkpoint_id=20 byte_size=15>

BUG=596796

Review URL: https://codereview.chromium.org/1841183002

Cr-Commit-Position: refs/heads/master@{#383840}

SetGeometry when graphics plane resolution changes.

Reasons for this:
* When switching to an app with a different graphics plane
  resolution, the old video plane geometry parameters are now
  invalid since they were relative to the previous graphics
  plane.
* SetGeometry should not always be run when the graphics
  plane resolution changes because some apps (like Backdrop)
  don't have a video element.

BUG=internal b/27554486
TEST=All tests run on dragonfly
     * Switch Youtube -> Backdrop. No SetGeometry calls are made
       after switching to Backdrop.
     * Switch Youtube -> Netflix. SetGeometry calls are made for
       Netflix playback.
     * Revoke resources while playing Youtube video. Then grant
       resources and playback same Youtube video. SetGeometry
       call is made when Youtube video is played again.

Review URL: https://codereview.chromium.org/1838863004

Cr-Commit-Position: refs/heads/master@{#383839}

BUG=596680

Review URL: https://codereview.chromium.org/1836983002

Cr-Commit-Position: refs/heads/master@{#383838}

fast/repaint/repaint-during-scroll-with-zoom.html

TBR=wangxianzhu@chromium.org
BUG=594516

Review URL: https://codereview.chromium.org/1840183002

Cr-Commit-Position: refs/heads/master@{#383837}

In AVDA::QueueInput it was possible that a pending bitstream buffer was
popped from the queue but then a failure caused it to return without
NotifyEndOfBitstreamBuffer. If AVDA was Reset before the error was
propogated to GVD, GVD and AVDA became out of sync.

Now the bitstream buffer notification is done regardless of whether
an error occurred.

BUG=598058

Review URL: https://codereview.chromium.org/1831323002

Cr-Commit-Position: refs/heads/master@{#383836}

BUG=582539

Review URL: https://codereview.chromium.org/1837963003

Cr-Commit-Position: refs/heads/master@{#383835}

We want to connect the source of a resource request with the resource fetch event. Adding the requestId to resource fetch is the first step.

BUG=

Review URL: https://codereview.chromium.org/1815383003

Cr-Commit-Position: refs/heads/master@{#383834}

To support multiple attached devtools, we have to instantiate
individual agents on demand, rather than on creation.
This patch moves regular of agents, with view-specific and worker's left.

BUG=590878

Review URL: https://codereview.chromium.org/1834073002

Cr-Commit-Position: refs/heads/master@{#383833}

Generated extensions code used to rely on linked_ptrs for non-copyable types in
containers. Now, we can make these types movable. Update the following APIs
to use movable types in this CL:
- management.json
- mime_handler_private.idl
- mojo_private.idl
- networking_config.idl
- networking_private.idl
- power.idl
- printer_provider.idl
- printer_provider_internal.idl
- runtime.json

BUG=595949

Review URL: https://codereview.chromium.org/1841543002

Cr-Commit-Position: refs/heads/master@{#383832}

https://boringssl.googlesource.com/boringssl/+log/e29ea166a6ec44bbfce81d1c93f064b9aca0f443..feaa57d13daa0b5bf3c068ce18d24870d50bfae9

BUG=none
TBR=svaldez@chromium.org

Review URL: https://codereview.chromium.org/1844673002

Cr-Commit-Position: refs/heads/master@{#383831}

This CL fixes the following issues in the autoplay experiment:

 - "Any media playback" wasn't checking if the media was playing
   already, so it over-counted quite a bit.

 - "ifmuted" counting was broken, since the autoplay experiment
   didn't call autoplayMediaEncountered in all cases.  For mute,
   this could actually allow the media to start playback without
   preventing additional autoplay metrics for the next playback.

 - made shouldAutoplay not call autoplayMediaEncountered, since
   the experiment might not have overridden the gesture requirement
   yet.  This would generate incorrect metrics in SetReadyState.

 - Adds a metric for "played to completion".

 - Adds metrics for initial playback that capture whether any
   user gesture was seen, and / or if one is active curently.
   This will give us an idea about how often a gesture in
   the inital load() call is responsible for a playback.

 - Added counters for "total media elements seen" and "total video
   elements seen".

BUG=561098

Review URL: https://codereview.chromium.org/1470153004

Cr-Commit-Position: refs/heads/master@{#383830}

Generated extensions code used to rely on linked_ptrs for non-copyable types in
containers. Now, we can make these types movable. Update the following APIs
to use movable types in this CL:
- omnibox.json
- page_action.json
- page_capture.json
- passwords_private.idl
- permissions.json
- platform_keys.idl
- platform_keys_internal.idl
- preferences_private.json
- privacy.json

BUG=595949
TBR=mpearson@chromium.org (trivial c/b/autocomplete/ change)

Review URL: https://codereview.chromium.org/1833773002

Cr-Commit-Position: refs/heads/master@{#383829}

When DevTools draws to capture the screen output for the remote
side, the video draw quad isn't processed as an overlay since
it's not in the root render pass.  However, on android, overlays
aren't optional since Chrome doesn't have the pixels.  So, it
uses the PictureBuffer texture to draw, which obscures the surface
view by removing the transparent box that the overlay put there.

Plus, drawing also happens normally, with overlays (no idea why),
and both results end up on the device display.  The result is that
the user sees a visible flicker as the transparent overlay box
shows up and goes away.  This normal draw has the side-effect of
updating the SurfaceView's contents.

It's unclear to me why this draw happens if we're also going to
display the one for DevTools.  The frame rate is ~halved.

This CL attaches changes the PictureBuffer textures to be 2D, and
attaches a 1x1 transparent textures to them, in the case we're using
SurfaceView.  When DevTools draw uses this texture to draw the video
quad, it draws the transparent box instead of obscuring it.

In the long run, we should probably transition away from SurfaceView
if we ever detect that we're drawing the video quads rather than
processing them for overlays.  This would have the benefit that one
could see the video in DevTools, too.

BUG=592798

Review URL: https://codereview.chromium.org/1785153004

Cr-Commit-Position: refs/heads/master@{#383828}

This adds 1 method, viewRect(), to LayoutViewItem in the new layout API
and calls it from FrameView::collectFrameTimingRequests. There are a
few other references to other methods in the same function which will
require more extensive refactoring, which are not included in this CL.

There are no functional changes.

BUG=499321

Review URL: https://codereview.chromium.org/1842653004

Cr-Commit-Position: refs/heads/master@{#383827}

order in main page.

BUG=None

Review URL: https://codereview.chromium.org/1836343002

Cr-Commit-Position: refs/heads/master@{#383826}

BUG=598736

Review URL: https://codereview.chromium.org/1843663002

Cr-Commit-Position: refs/heads/master@{#383825}

These are deprecated since 10.9. NSUnderlineStyleNone and
NSUnderlineStyleSingle are drop-in replacements with the same enum
value.

BUG=547071

Review URL: https://codereview.chromium.org/1842663003

Cr-Commit-Position: refs/heads/master@{#383824}

Revert of Mac: Decode hardware to 420 instead of 422 (patchset #3 id:40001 of https://codereview.chromium.org/1822173002/ )

Reason for revert:
This likely spiked the crash rates.
BUG=598388

Original issue's description:
> Mac: Decode hardware to 420 instead of 422
>
> If the decoded frame is going to be used as a CALayer overlay, then
> it will never be converted to RGBA. If it is used by OpenGL, then the
> same infrastructure for software frames will be employed to make an
> expensive RGBA copy of the frame.
>
> BUG=594452
>
> Committed: https://crrev.com/24da601d444c8822c5ff0a10751b9384f25dd1b4
> Cr-Commit-Position: refs/heads/master@{#382797}

TBR=sandersd@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=594452

Review URL: https://codereview.chromium.org/1842673003

Cr-Commit-Position: refs/heads/master@{#383823}

The timestamp portion of the EXT_disjoint_timer_query does not have much
support from GPU drivers and emulation of them by Chromium at a level
that is accurate enough for users has been determined to be infeasible.
Therefore timestamps are disabled at the WebGL/client side on all
platforms as time elasped queries should be used by developers instead
as they are more widely supported.

BUG=595172
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel

Review URL: https://codereview.chromium.org/1800383002

Cr-Commit-Position: refs/heads/master@{#383822}

BUG=598393

Review URL: https://codereview.chromium.org/1841613002

Cr-Commit-Position: refs/heads/master@{#383821}

Platform painting should be through skia::ScopedPlatformPaint.
(Only one compositor callsite was calling Begin/End directly.)

Remove skia::EndPlatformPaint() and all device-specific implementations,
which were effectively noops.

Remove skia::BeginPlatformPaint(), and make all device-specific impls
protected.

BUG=598695
R=fmalita@chromium.org
TBR=danakj@chromium.org

Review URL: https://codereview.chromium.org/1839113002

Cr-Commit-Position: refs/heads/master@{#383820}

BUG=453002
TBR=szager@chromium.org

Review URL: https://codereview.chromium.org/1843793003 .

Cr-Commit-Position: refs/heads/master@{#383819}

BUG=

Review URL: https://codereview.chromium.org/1837993002

Cr-Commit-Position: refs/heads/master@{#383818}

Matching rules are fetched when Chrome starts and can take in the order
of 100-200ms. UI navigations that happen before these rules are fetched
will not start the data use tracking. This race condition is fixed by
buffering the navigation events until rule fetch is complete.

BUG=586235

Review URL: https://codereview.chromium.org/1837223002

Cr-Commit-Position: refs/heads/master@{#383817}