GitLab

There was a mismatch between core_services_initialization.cc, the
BUILD.gn file, and the package listing.

BUG=484234

Review URL: https://codereview.chromium.org/1141593004

Cr-Commit-Position: refs/heads/master@{#330181}

BUG=457697

Review URL: https://codereview.chromium.org/1146473002

Cr-Commit-Position: refs/heads/master@{#330180}

This CL lands server change 91813814 by birenroy.

BUG=488484

Review URL: https://codereview.chromium.org/1134603005

Cr-Commit-Position: refs/heads/master@{#330179}

Adds an error log entry to chrome://media-internals when MSE's
MP4 stream parser detects an unfragmented MP4. This error log should
help make it more clear why the parse fails in this unfortunately common
case.

BUG=487410
R=dalecurtis,chcunningham
TEST=Manually verified the log entry occurs for the repro in bug 487410.

Review URL: https://codereview.chromium.org/1147453002

Cr-Commit-Position: refs/heads/master@{#330178}

This CL lands server change 91620063 by elving.

BUG=488484

Review URL: https://codereview.chromium.org/1123153003

Cr-Commit-Position: refs/heads/master@{#330177}

This CL lands server change 93659491 by mpw.

BUG=488484

Review URL: https://codereview.chromium.org/1139953003

Cr-Commit-Position: refs/heads/master@{#330176}

At the same time, add a "disable" flag as an escape hatch for when
things go bad.

BUG=467832

Review URL: https://codereview.chromium.org/1124083004

Cr-Commit-Position: refs/heads/master@{#330175}

Summary of changes available at:
https://boringssl.googlesource.com/boringssl/+log/771a138..a07c0fc

This pulls in the SSL_get_current_cipher fix.

BUG=none

Review URL: https://codereview.chromium.org/1126233010

Cr-Commit-Position: refs/heads/master@{#330174}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/7e26b7f..f3c3784

TBR=timvolodine@chromium.org,kbr@chromium.org

Review URL: https://codereview.chromium.org/1140253002

Cr-Commit-Position: refs/heads/master@{#330173}

It doesn't seem like the suggestions page has been used in a while.

There's a bunch of code that says:

  if (IsSuggestionsPageEnabled()) {
    ...
  }

But the implementation of the feature switching method is:

  bool IsSuggestionsPageEnabled() {
    return false;  // Maybe we'll use this someday.
  }

We shouldn't bloat the binary for a "maybe we'll use it later".

BUG=329637
R=estade@chromium.org

Review URL: https://codereview.chromium.org/1141843004

Cr-Commit-Position: refs/heads/master@{#330172}

Make session restore forced tab load delay Finch configurable, and separate first tab loads from subsequent.

BUG=472772

Review URL: https://codereview.chromium.org/1130673003

Cr-Commit-Position: refs/heads/master@{#330171}

Make one part handle updating status (now private, and called
automatically when content or viewport params change), and one part
called on commit or renderer change.

BUG=464892

Review URL: https://codereview.chromium.org/1134123005

Cr-Commit-Position: refs/heads/master@{#330170}

BUG=484699

Review URL: https://codereview.chromium.org/1128173003

Cr-Commit-Position: refs/heads/master@{#330169}

This CL supports HTTPConnectionDelegate.OnReceivedRequest() and adds corresponding tests.

The following work will be in future CLs:
- support WebSocket.
- more tests.

BUG=478249
TEST=Newly-added tests.

Review URL: https://codereview.chromium.org/1128863004

Cr-Commit-Position: refs/heads/master@{#330168}

This CL fixes
WebLayerImplFixedBoundsTest.CompareToWebLayerImplComplex and
WebLayerImplFixedBoundsTest.CompareToWebLayerImplSimple tests.

BUG=440441

R=enne, danakj

Review URL: https://codereview.chromium.org/1130123005

Cr-Commit-Position: refs/heads/master@{#330167}

BUG=None

Review URL: https://codereview.chromium.org/1141233002

Cr-Commit-Position: refs/heads/master@{#330166}

Tested on Linux with the following command lines:

$ gn gen out-gn/Debug
$ gn check out-gn/Debug //gpu:*

BUG=367595, 376000
TEST=see above
R=sievers@chromium.org,dpranke@chromium.org
TBR=brettw@chromium.org # for .gn

Review URL: https://codereview.chromium.org/1136283008

Cr-Commit-Position: refs/heads/master@{#330165}

This is necessary because it is one of the dependencies of the
'blink_tests' group that should be built.

BUG=483171
R=avi@chromium.org,dpranke@chromium.org

CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:android_chromium_gn_compile_dbg,android_chromium_gn_compile_rel;tryserver.chromium.win:win8_chromium_gn_rel,win8_chromium_gn_dbg;tryserver.chromium.mac:mac_chromium_gn_rel,mac_chromium_gn_dbg

Review URL: https://codereview.chromium.org/1138653004

Cr-Commit-Position: refs/heads/master@{#330164}

The component was experimental and uses org.apache.http library,
which is now deprecated in Android.

BUG=488202
TBR=erikwright@chromium.org

Review URL: https://codereview.chromium.org/1142463003

Cr-Commit-Position: refs/heads/master@{#330163}

Adds support to the PacketDroppingTestWriter for dropping the first N
packets of a connection.  Will be used for testing stateless rejects,
when the initial stateless reject (SREJ) from the server back to the
client gets lost.

Test only and not even used in this CL.

Merge internal change: 93446002
https://codereview.chromium.org/1136553007/

Minor cleanup of obsolete comment and code.  No functional change.

Merge internal change: 93441848
https://codereview.chromium.org/1139153003/

Remove the no longer used kPACE connection option.  No functional change.

Merge internal change: 93437185
https://codereview.chromium.org/1135253004/

Rename local variable only - consumed_window to available_window.

QuicFlowController::MaybeSendWindowUpdate() - rename counterintuitive
consumed_window.

Merge internal change: 93431527
https://codereview.chromium.org/1145433003/

Protected by FLAGS_enable_quic_stateless_reject_support.  Add support
for stateless rejects to the QuicDispatcher code.

QuicServerSession - Added extra setter, which the dispatcher uses to
indicate that session should issue stateless rejects (if supported by
the peer).

QuicDispatcher - Will use stateless rejects if the session map is over
a threshold, configurable via flag.  If a stateless reject has been
issued, the dispatcher will close the session silently.  It will also
add the connection-id to the time-wait list, indicating that further
packets for the connection-id should be black-holed.

Changes to the end_to_end_tests will be made in a separate CL.

Merge internal change: 93414515
https://codereview.chromium.org/1128103007/

Added support for stateless time-wait entries.  Will be used in a
subsequent CL.

Time-wait list entries can now be marked as stateless, which means
that the corresponding connection was closed via stateless reject.
Since all state for that connection id is dropped after the stateless
reject, the time-wait list manager will not respond to incoming
packets for the connection ID with public resets.  Instead, it will
black-hole the packets.

Added corresponding tests.

Merge internal change: 93412303
https://codereview.chromium.org/1135113003/

Remove FLAGS_quic_use_optimized_packet_reader now that it's been deployed.

Merge internal change: 93331362
https://codereview.chromium.org/1128933010/

Fixing a bug in the QUIC code where we could create new streams
server-side even once the connection was killed off.

Not creating QUIC streams after disconnect.  Not flag protected.

Merge internal change: 92727137
https://codereview.chromium.org/1142543002/

Flag-protected. Add stateless reject support to crypto streams.

Client crypto stream:

The client now sends all connection options on all handshake messages.
(Previously, no options were sent on inchoate hellos).  This is so
that the client can indicate support for stateless rejects during the
initial handshake.

If the crypto client stream receives a stateless reject from the
server, it will continue validating and cacheing the proof.  It will
abandon the connection with a special error code immediately
thereafter.  It will NOT send close messages after a reject, since
there is no state at the server to close.  (We may have to change this
when we do stateless by default).

Server crypto stream:

The server now processes connection options speculatively on all
handshake messages.  This is so that the server can detect that the
client supports stateless rejects prior to sending the reject itself.
It will only consider the options to be "negotiated", however, prior
to sending an SHLO.

The crypto server stream now also maintains an additional boolean,
indicating whether it should use stateless rejects when rejecting
messages on the stream ("use_stateless_rejects").  If rejecting, it
will only emit a stateless reject if both using stateless rejects and
the client supports it.

Merge internal change: 92677792
https://codereview.chromium.org/1139183002/

Crypto-client config no longer gives stateless reject error.

QuicCryptoClientConfig::ProcessRejection no longer produces an error
when processing a stateless reject.  The client can always inspect the
message tag directly to determine that the reject is stateless.

flag-protected.

Merge internal change: 92637704
https://codereview.chromium.org/1145453002/

Remove FLAGS_quic_limit_max_cwnd_to_receive_buffer.

Merge internal change: 92419361
https://codereview.chromium.org/1141743004/

R=rch@chromium.org

Review URL: https://codereview.chromium.org/1138443003

Cr-Commit-Position: refs/heads/master@{#330162}

This CL lands server change 90723544 by yinjie.

BUG=488484

Review URL: https://codereview.chromium.org/1146533003

Cr-Commit-Position: refs/heads/master@{#330161}

R=vabr@chromium.org
BUG=486739

Review URL: https://codereview.chromium.org/1140253003

Cr-Commit-Position: refs/heads/master@{#330160}

- This is in preparation for split apks, where we'll need to call
apkbuilder multiple times.
- This also makes including a classes.dex file optional

BUG=484797

Review URL: https://codereview.chromium.org/1138953003

Cr-Commit-Position: refs/heads/master@{#330159}

Convert code to use enum DomCode rather than strings.

BUG=444048
R=wez@chromium.org,bbudge@chromium.org,bauerb@chromium.org,jcivelli@chromium.org,sadrul@chromium.org

Presubmit has failed to acknowledge the "LGTM" at https://codereview.chromium.org/1124963003/#msg29 so I'm adding this TBR as a workaround:

TBR=jcivelli

Review URL: https://codereview.chromium.org/1124963003

Cr-Commit-Position: refs/heads/master@{#330158}

Reset unreachable data reduction proxy message when data savings is
switched off. Clear counts of successful requests through proxy and
net errors.

Reset bad proxy list so that the data reduction proxies are retried when Data
Saver is turned back on.

BUG=436326,438694

Review URL: https://codereview.chromium.org/1006583002

Cr-Commit-Position: refs/heads/master@{#330157}

(Previously, this was used only with the "Chrome" branding).

BUG= 488572

R=gunsch@chromium.org

Review URL: https://codereview.chromium.org/1130893008

Cr-Commit-Position: refs/heads/master@{#330156}

Recent work on didStopLoading makes it possible for us to use
WaitForLoadStop to wait for all child frames to be loaded when
navigating a subframe.

BUG=

Review URL: https://codereview.chromium.org/1123233005

Cr-Commit-Position: refs/heads/master@{#330155}

WebM initData is a key ID, and Android only supports 128-bit key IDs, so fail if a larger |init_data| is
passed to CreateSessionAndGenerateRequest().

BUG=456397
TEST=manual test on Android

Review URL: https://codereview.chromium.org/1113883003

Cr-Commit-Position: refs/heads/master@{#330154}

The histograms removed in this CL are not declared in histograms.xml
and so are not collected at all - removing what is essentially dead
code.

R=sky@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/1132333004

Cr-Commit-Position: refs/heads/master@{#330153}

One of the intermediate steps in the build process for
creating Android APKs depends on the output from another
intermediate step, but that dependency was not *also* declared
in the 'deps' target in the GN build file.

As a result, 'gn refs' (and hence 'mb analyze') would not calculate
the right set of dependencies and we would incorrectly conclude that
no compiles were necessary in cases where they really were needed.

More details in the bug.

R=cjhopman@chromium.org, brettw@chromium.org
BUG=487897

Review URL: https://codereview.chromium.org/1129773004

Cr-Commit-Position: refs/heads/master@{#330152}

BUG=467779

Review URL: https://codereview.chromium.org/1136663006

Cr-Commit-Position: refs/heads/master@{#330151}

PpapiDecryptor::Create() must not proceed with decryptor initialization
if the CDM wrapper could not be created.
PpapiDecryptor::PpapiDecryptor() asserts the CDM wrapper is valid and
InitializeCdm() uses it unconditionally.

BUG=488363
TEST=EME tests pass

Review URL: https://codereview.chromium.org/1128693004

Cr-Commit-Position: refs/heads/master@{#330150}

This also changes IsSmartLockedBrandingEnabled() to only return true if the user is syncing without using a custom passphrase.

BUG=486739

Review URL: https://codereview.chromium.org/1138523005

Cr-Commit-Position: refs/heads/master@{#330149}

BUG=486482

Review URL: https://codereview.chromium.org/1129063010

Cr-Commit-Position: refs/heads/master@{#330148}

Revert of Non-SFI mode: Build nacl_helper_nonsfi_unittests (patchset #4 id:80001 of https://codereview.chromium.org/1137553003/)

Reason for revert:
This breaks the asan build. See comments on the original CL.

Original issue's description:
> Non-SFI mode: Build nacl_helper_nonsfi_unittests
>
> This CL starts to build nacl_helper_nonsfi_unittests.
> The binary is not yet enabled to run on bots. It will be done in a following CL.
>
> TEST=Ran locally. Ran build bots. Ran build bots with editing configuration to include nacl_helper_nonsfi_unittests.
> BUG=358465
>
> Committed: https://crrev.com/72c418b09bbf90505f54df2b76285650acd97960
> Cr-Commit-Position: refs/heads/master@{#330069}

TBR=mseaborn@chromium.org,mdempsky@chromium.org,thakis@chromium.org,jln@chromium.org,hidehiko@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=358465

Review URL: https://codereview.chromium.org/1145593002

Cr-Commit-Position: refs/heads/master@{#330147}

Workarounds to support grabbing the DRM devices as master after GPU crashes
made the GPU lock unnecessary.

BUG=none

Review URL: https://codereview.chromium.org/1141243002

Cr-Commit-Position: refs/heads/master@{#330146}

BUG=441605
TBR=davidben,mmenke

Review URL: https://codereview.chromium.org/1143633003

Cr-Commit-Position: refs/heads/master@{#330145}

This service was used as part of a previous Sync end-to-end testing
effort and is no longer used.

BUG=NONE

Review URL: https://codereview.chromium.org/1138113003

Cr-Commit-Position: refs/heads/master@{#330144}

Refcount an extension so that it if it happens to be reloaded while the dialog
is showing, we don't crash.

BUG=487161
TBR=sky@chromium.org (small extension_uninstall_dialog_view.cc changes, reviewed by kalman)

Review URL: https://codereview.chromium.org/1140053004

Cr-Commit-Position: refs/heads/master@{#330143}

ozone: Clarify All SurfaceFactoryOzone APIs except for CreateNativePixmap() must be used on the main thread in Gpu process

CreateNativePixmap() can be used in IO thread because IO thread creates GpuMemoryBuffer for low latency.

BUG=472610

Review URL: https://codereview.chromium.org/1068723005

Cr-Commit-Position: refs/heads/master@{#330142}