GitLab

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/2f912f3..5bb9700

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=rmistry@google.com

Review URL: https://codereview.chromium.org/660733004

Cr-Commit-Position: refs/heads/master@{#299966}

The discussion at [1] gives some background on the root cause of the failure.

[1] https://codereview.chromium.org/610493002/

BUG=323093

Review URL: https://codereview.chromium.org/657253003

Cr-Commit-Position: refs/heads/master@{#299965}

- Keep record of a bad signature in VariationsSeedStore::LoadSeed.
- Register VerifyVariationsSeedSignature in SafeBrowsingService::RegisterAllDelayedAnalysis to send an incident report, if a bad signature is encountered.

BUG=423467

Review URL: https://codereview.chromium.org/646733002

Cr-Commit-Position: refs/heads/master@{#299964}

Not sure what was broken before, but it seems to work now. I also added the
configure flags from debian/rules.

BUG=424313
TBR=glider@chromium.org
NOTRY=true
TEST=MSan

Review URL: https://codereview.chromium.org/660863005

Cr-Commit-Position: refs/heads/master@{#299963}

TBR=keybuk@chromium.org
BUG=417463

Review URL: https://codereview.chromium.org/652663002

Cr-Commit-Position: refs/heads/master@{#299962}

This behavior matches what we do in renderer_main.cc.

R=jamesr@chromium.org

Review URL: https://codereview.chromium.org/658113003

Cr-Commit-Position: refs/heads/master@{#299961}

BUG=338830
TEST=new EME tests pass

Review URL: https://codereview.chromium.org/641443002

Cr-Commit-Position: refs/heads/master@{#299960}

Prior to https://codereview.chromium.org/462813002 the resource
scheduler would exit the critical loading phase as soon as the main
parser parsed the body tag:
- Blocking JS in the head would keep it in the critical phase
- CSS in the head would not (this is the bug that was fixed)
- Non-blocking JS or JS discovered by the preload scanner would NOT
  keep it in the critical phase.

After the fix:
- Blocking JS in the head still kept it in the critical phase
- CSS in the head kept it in the critical phase (fix worked)
- Any JS that was started before the body tag was parsed would keep
  it in the critical phase until that JS finished loading. This inclues
  non-blocking JS (script-injected while in the head) as well as JS
  discovered by the preload scanner.

This change keeps the CSS fix but also restores the JS behavior so
that non-blocking or preload-scanned JS no longer keep the loader
in the critical resources phase.

The fix was really simple since CSS loads at a higher priority
than JS (net::Medium), I just changed the logic to only count
Medium+ as render-blocking (waiting for them to finish regardless of
the state of the body tag) and returned the logic that unblocks
non-render-blocking resources as soon as the body tag was parsed (which
now includes JS). Any blocking JS still keeps the main parser blocked
so only non-blocking and preloaded JS would be pending or loading when
the body tag is parsed.

You can see it working here:
http://www.webpagetest.org/video/compare.php?tests=141016_WH_e488a05d7b224d7fc362dfa91b7c5ec6,141016_H1_2ecb3d397b3c218ff26dc7bdb9cb0858

Request #13 is a non-blocking script that in teh baseline case keeps
images from being loaded but in the fixed case no longer blocks images.

The canonical test page for the CSS bug also shows the critical phase
still being honored when only css is parsed in the head:
http://www.webpagetest.org/video/compare.php?tests=141016_VM_1d25c3ff22f132b4ded97d7d72e5e558,141016_PM_17a2f0282c39f299425ba57c5c5118bf

BUG=423853

Review URL: https://codereview.chromium.org/658163002

Cr-Commit-Position: refs/heads/master@{#299959}

BUG=424093

Review URL: https://codereview.chromium.org/664543002

Cr-Commit-Position: refs/heads/master@{#299958}

This is a mechanical move with opportunistic clang_format
applied for cleanup.

This also introduces a dependency on //google_apis within
//extensions/browser.

BUG=398671

Review URL: https://codereview.chromium.org/654363002

Cr-Commit-Position: refs/heads/master@{#299957}

I think this was missed while refactoring source_control.py in crrev.com/299851

BUG=424296
NOTRY=true

Review URL: https://codereview.chromium.org/647163003

Cr-Commit-Position: refs/heads/master@{#299956}

BUG=423621

Review URL: https://codereview.chromium.org/660873002

Cr-Commit-Position: refs/heads/master@{#299955}

The code to exit was only being hit in some codepaths, but it should be run for every possible way that it could receive a lost context.

BUG=424301

Review URL: https://codereview.chromium.org/657003002

Cr-Commit-Position: refs/heads/master@{#299954}

Bad error messages remind me of working with gyp.

This change just stores the parser error in the InputFileData so that we
can actually report a useful error message from whichever thread first
reports an error.

Review URL: https://codereview.chromium.org/645253002

Cr-Commit-Position: refs/heads/master@{#299953}

BUG=none
TBR=kalman@chromium.org

Review URL: https://codereview.chromium.org/564163002

Cr-Commit-Position: refs/heads/master@{#299952}

This moves ExecProcess to exec_process.h/cc and adds capturing
of stderr on linux.

Adds unittests for ExecProcess.

Windows is not supported.

Review URL: https://codereview.chromium.org/646013003

Cr-Commit-Position: refs/heads/master@{#299951}

By making HidService::Connect return its result asynchronously platform
specific details such as whether device nodes must be opened on a
different thread (due to blocking) or requesting access from the Chrome
OS permission broker can be abstracted away.

BUG=422540

Review URL: https://codereview.chromium.org/660573007

Cr-Commit-Position: refs/heads/master@{#299950}

BUG=423621

Review URL: https://codereview.chromium.org/660663003

Cr-Commit-Position: refs/heads/master@{#299949}

BUG=None
R=oshima@chromium.org
TEST=manually

Review URL: https://codereview.chromium.org/657713002

Cr-Commit-Position: refs/heads/master@{#299948}

Otherwise component build causes linker errors.

BUG=None
R=oshima@chromium.org
TEST=manually

Review URL: https://codereview.chromium.org/656153002

Cr-Commit-Position: refs/heads/master@{#299947}

BUG=423555
TEST=None

Review URL: https://codereview.chromium.org/652613003

Cr-Commit-Position: refs/heads/master@{#299946}

Revert of Ensure that the browser’s copy of page id is in sync with the renderer’s. (patchset #1 id:1 of https://codereview.chromium.org/657233002/)

Reason for revert:
It's crashing... drat.

Original issue's description:
> Ensure that the browser’s copy of page id is in sync with the renderer’s.
>
> This might-maybe-hope-not crash, and so has debugging statements.
>
> BUG=407376
> TEST=no crashing, we hope
>
> Committed: https://crrev.com/b49b70f26e6d0f2f54bb87cf374f30a339c7d7c6
> Cr-Commit-Position: refs/heads/master@{#299822}

TBR=rsesek@chromium.org,creis@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=407376

Review URL: https://codereview.chromium.org/661823002

Cr-Commit-Position: refs/heads/master@{#299945}

According to latest editor's draft http://rawgit.com/w3c/html-api-map/master/index.html
meter tag should have MSAA+IA2 Role as ROLE_SYSTEM_PROGRESSBAR.
And for mac, the AX Mappings should be copy of ARIA role progressbar.

This CL depends on https://codereview.chromium.org/652663003/

This CL also takes care of small fixes related with details tag and div tag.

BUG=421376

Review URL: https://codereview.chromium.org/657513002

Cr-Commit-Position: refs/heads/master@{#299944}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/2606d14..481d9a8

TBR=cbiesinger@chromium.org,dstockwell@chromium.org

Review URL: https://codereview.chromium.org/665453002

Cr-Commit-Position: refs/heads/master@{#299943}

BUG=413863

Review URL: https://codereview.chromium.org/646773005

Cr-Commit-Position: refs/heads/master@{#299942}

Fix indentation issues to allow enabling Checkstyle indentation rule.

BUG=318404

Review URL: https://codereview.chromium.org/644913003

Cr-Commit-Position: refs/heads/master@{#299941}

On suspend, Easy Unlock app gets disabled. When waking up, it takes some
time for the app to reload ( a second or two). During theat time there
is no Easy Unlock icon in the user pod. To improve user experience,
show connecting icon during that time.

BUG=421970
TEST=With Easy Unlock enabled, lock the screen and close the lid (so the
device suspends). When opening the lid spinner icon should be visible
immediately on the user pod.

Review URL: https://codereview.chromium.org/644873002

Cr-Commit-Position: refs/heads/master@{#299940}

There is a race condition in the current state of the code where in cross-process navigation we swap the existing RenderFrameHost with a new RenderFrameHost. If the existing host sends an IPC message to create a new child frame, it arrives on the IO thread, allocates a routing id based of the existing process (p1) and does a PostTask to the UI thread. If there is a CommitPending event either executing on the UI thread or in the task queue before the task posted from the IO thread, it will end up putting the existing RenderFrameHost in swapped out state (or waiting for swapped out). When the task to create a child frame is executed after that, it creates a new RenderFrameHost, but it uses the "current" process (p2), which is different than the process that sent the message (p1). This manifests sometimes as adding duplicate routing ids to RenderProcessHost and is in general really bad bug.

BUG=415059, 423691, 381990

Review URL: https://codereview.chromium.org/642813007

Cr-Commit-Position: refs/heads/master@{#299939}

In order to better integrate the IRT Extension API into NaCl IO, the
first implementation mirrors the other kernel_wrap_* files. All the
unit tests should therefore still be valid. Eventually it may be
easier to just have a separate version of kernel_intercept which
utilizes the IRT extension API directly.

Currently the IRT Extension interface is only supported within newlib,
so I have enabled it for newlib only.

BUG= https://code.google.com/p/nativeclient/issues/detail?id=3918
TEST= trybots

Review URL: https://codereview.chromium.org/608913003

Cr-Commit-Position: refs/heads/master@{#299938}

BUG=None
R=nkostylev@chromium.org
TEST=manually

Review URL: https://codereview.chromium.org/660613002

Cr-Commit-Position: refs/heads/master@{#299937}

BUG=skia:2889

Review URL: https://codereview.chromium.org/662623003

Cr-Commit-Position: refs/heads/master@{#299936}

BUG=423134

Review URL: https://codereview.chromium.org/662503003

Cr-Commit-Position: refs/heads/master@{#299935}

BUG=423134

Review URL: https://codereview.chromium.org/658993002

Cr-Commit-Position: refs/heads/master@{#299934}

Not sure in which condition this can be triggered, my environment is linux x64 with plain "component=shared_library clang=1", the actual linker invoked is "third_party/binutils/Linux_x64/Release/bin/ld.gold"

Here is the error log from linker:
/usr/lib64/gcc/x86_64-unknown-linux-gnu/4.9.1/../../../../include/c++/4.9.1/bits/stl_tree.h:1543: error: undefined reference to 'message_center::CompareTimestampSerial::operator()(message_center::Notification*, message_center::Notification*)'
clang: error: linker command failed with exit code 1 (use -v to see invocation)

BUG=None

Review URL: https://codereview.chromium.org/654093003

Cr-Commit-Position: refs/heads/master@{#299933}

In order to support GL_TEXTURE_RECTANGLE_ARB, cc::GLRenderer needs to
know the size of the mask texture to unnormalize the texture
coordinates.  This has already been done for TileDrawQuad.

As an optimization, rather than adding two more members onto the already
bloated RenderPassDrawQuad, turn mask_uv_rect into mask_uv_scale.  The
position of the mask is now assumed to always be at the origin of the
quad's transform, which it always is in cc.  This is equivalent to
saying that the mask is always attached to the owning layer of a render
surface (since that's what defines the transform origin of the render
surface).

By making this assumption, the old mask_uv_rect can be calculated in
a helper function from just quad->rect and quad->mask_uv_scale.

With this patch, mask_texture_size is currently unused, but its values
are unit tested.

Depends on https://codereview.chromium.org/652393002/

R=danakj@chromium.org,jamesr@chromium.org
BUG=423533

Review URL: https://codereview.chromium.org/659683002

Cr-Commit-Position: refs/heads/master@{#299932}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/7851a56..2f912f3

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=rmistry@google.com

Review URL: https://codereview.chromium.org/640263003

Cr-Commit-Position: refs/heads/master@{#299931}

R=sergeyu@chromium.org
BUG=423134

Review URL: https://codereview.chromium.org/660703003

Cr-Commit-Position: refs/heads/master@{#299930}

BUG=423621

Review URL: https://codereview.chromium.org/657333002

Cr-Commit-Position: refs/heads/master@{#299929}

Other embedded content, such as plug-ins, is also prevented from incurring
auto-log-in requests.

BUG=334230
TEST=You still get the auto-log-in infobar when the requesting site is the
top-level frame, but never otherwise.

Review URL: https://codereview.chromium.org/131483010

Cr-Commit-Position: refs/heads/master@{#299928}

This was blocked on http://crbug.com/423873 which is now fixed.

BUG=414808
R=jam@chromium.org

Review URL: https://codereview.chromium.org/645033003

Cr-Commit-Position: refs/heads/master@{#299927}