chromium/src/build.git
37 hours agodisable more stacktrace usage in debug uclibc builds master
mostynb@opera.com [Fri, 18 Apr 2014 23:40:16 +0000 (23:40 +0000)]
disable more stacktrace usage in debug uclibc builds

Here are a couple more instances of OutputToStream / Print that are
no longer available in uClibc builds.

BUG=361130

Review URL: https://codereview.chromium.org/234803002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264887 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

40 hours agoFinalize apk takes zipalign path as argument instead of hardcoding it
yusufo@chromium.org [Fri, 18 Apr 2014 20:22:43 +0000 (20:22 +0000)]
Finalize apk takes zipalign path as argument instead of hardcoding it

zipalign tool can be in sdk/tools or sdk/build-tools. This change looks
for the right place and uses that instead of having it hardcoded inside
android_sdk_root

BUG=364655

Review URL: https://codereview.chromium.org/238253015

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264842 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

42 hours agoRemove desktop auto-login since we don't plan on shipping it.
rohitrao@chromium.org [Fri, 18 Apr 2014 18:21:25 +0000 (18:21 +0000)]
Remove desktop auto-login since we don't plan on shipping it.

Removes the --enable-autologin switch and associated settings UI.  Autologin is
always enabled for Android builds.  Moves the Android-specific AutoLoginPrompter
class into an android/ directory.

iOS also uses the AutoLoginInfoBarDelegate, so that class is left where it is.

BUG=347214

Review URL: https://codereview.chromium.org/235863025

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264818 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoAdd generate_jni template
cjhopman@chromium.org [Fri, 18 Apr 2014 00:50:10 +0000 (00:50 +0000)]
Add generate_jni template

This adds a template for using the jni_generator. Creates the
base_jni_headers target and updates the base target to depend on it and
actually use the sources that required it.

Adds //third_party/android_tools:cpu_features required by one of the
base files that also required jni.

Disable a bunch of third_party targets for Android builds.

BUG=359249

Review URL: https://codereview.chromium.org/237973002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264695 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days ago[Android] Move screenshot implementation back into AndroidCommands.
jbudorick@chromium.org [Thu, 17 Apr 2014 23:46:31 +0000 (23:46 +0000)]
[Android] Move screenshot implementation back into AndroidCommands.

BUG=267773

Review URL: https://codereview.chromium.org/234803006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264674 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoRemove unused input_shlib_path
cjhopman@chromium.org [Thu, 17 Apr 2014 23:16:50 +0000 (23:16 +0000)]
Remove unused input_shlib_path

This path is now only inferred from test_suite_name.
TBR=darin
NOTRY=true

Review URL: https://codereview.chromium.org/226353002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264666 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoWork on Mac GN build.
brettw@chromium.org [Thu, 17 Apr 2014 20:33:19 +0000 (20:33 +0000)]
Work on Mac GN build.

This fixes a lot of minor mistakes (mostly missing/extra files and flags) for the Mac GN build.

I separated out some clang flags into a config for extra clang warnings. Several of the third party libraries needed to remove this.

Removes the use_nss flag and uses !use_openssl. This is a result of discussion with rsleevi.

Removes extra duplicate net build file from secondary tree.

ui/gesture_events seems to be getting compiled in GN with more strict warnings than in GYP. Rather than fix this, I fixed the warning in the gesture recognizer unit test. It was returning a const copy (the const is pointless when you're copying).

This also removes a bunch of old GYP integration stuff that was left in the GN build.

R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/239543013

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264626 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoIgnore libpng warnings when running aapt crunch.
newt@chromium.org [Thu, 17 Apr 2014 20:28:26 +0000 (20:28 +0000)]
Ignore libpng warnings when running aapt crunch.

process_resources.py currently considers aapt to have failed if it
outputs anything to stderr. Change this to ignore lines printed to
stderr that contain "libpng warning", which is a known non-error
condition.

BUG=364355

Review URL: https://codereview.chromium.org/238043011

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264623 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoRemove all uses of toolkit_use_gtk in the gyp files.
erg@chromium.org [Thu, 17 Apr 2014 19:20:23 +0000 (19:20 +0000)]
Remove all uses of toolkit_use_gtk in the gyp files.

BUG=297026
R=ben@chromium.org, brettw@chromium.org

Review URL: https://codereview.chromium.org/238633004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264609 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoRevert 264559 "Get sample_app running on Android again"
davemoore@chromium.org [Thu, 17 Apr 2014 18:19:07 +0000 (18:19 +0000)]
Revert 264559 "Get sample_app running on Android again"

> Get sample_app running on Android again
>
> R=viettrungluu@chromium.org, viettrungluu
> BUG=
>
> Review URL: https://codereview.chromium.org/240003006

TBR=davemoore@chromium.org

Review URL: https://codereview.chromium.org/238563003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264576 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoGet sample_app running on Android again
davemoore@chromium.org [Thu, 17 Apr 2014 17:26:37 +0000 (17:26 +0000)]
Get sample_app running on Android again

R=viettrungluu@chromium.org, viettrungluu
BUG=

Review URL: https://codereview.chromium.org/240003006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264559 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoRevert accidental dartium code push
jamescook@chromium.org [Thu, 17 Apr 2014 16:50:56 +0000 (16:50 +0000)]
Revert accidental dartium code push

This reverts r264517 to r264538 from terry@google.com

BUG=none
TEST=none
NOTRY=true
NOTREECHECKS=true
TBR=terry@google.com

Review URL: https://codereview.chromium.org/239993009

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264545 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoAdd blink_tests to the dartium_builder target
terry@google.com [Thu, 17 Apr 2014 15:55:25 +0000 (15:55 +0000)]
Add blink_tests to the dartium_builder target

tbr=vsm@google.com

Review URL: https://codereview.chromium.org/132473015

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264531 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoThis fixes debugging chrome apps from the dart editor.
terry@google.com [Thu, 17 Apr 2014 15:55:21 +0000 (15:55 +0000)]
This fixes debugging chrome apps from the dart editor.

R=devoncarew@google.com

Review URL: https://codereview.chromium.org/148523005

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264528 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoUpdate dartium_builder target in gyp file.
terry@google.com [Thu, 17 Apr 2014 15:55:18 +0000 (15:55 +0000)]
Update dartium_builder target in gyp file.

TBR=jacobr@google.com

Review URL: https://codereview.chromium.org/107143004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264527 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agomerged 1908
terry@google.com [Thu, 17 Apr 2014 15:55:16 +0000 (15:55 +0000)]
merged 1908

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264526 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoBranching for 1847 @251904
terry@google.com [Thu, 17 Apr 2014 15:55:13 +0000 (15:55 +0000)]
Branching for 1847 @251904

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264525 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoAdd blink_tests to the dartium_builder target
terry@google.com [Thu, 17 Apr 2014 15:55:06 +0000 (15:55 +0000)]
Add blink_tests to the dartium_builder target

tbr=vsm@google.com

Review URL: https://codereview.chromium.org/132473015

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264521 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoThis fixes debugging chrome apps from the dart editor.
terry@google.com [Thu, 17 Apr 2014 15:55:01 +0000 (15:55 +0000)]
This fixes debugging chrome apps from the dart editor.

R=devoncarew@google.com

Review URL: https://codereview.chromium.org/148523005

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264518 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 days agoUpdate dartium_builder target in gyp file.
terry@google.com [Thu, 17 Apr 2014 15:54:50 +0000 (15:54 +0000)]
Update dartium_builder target in gyp file.

TBR=jacobr@google.com

Review URL: https://codereview.chromium.org/107143004

Conflicts:
build/all.gyp

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264517 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agoDelete unused python_ver gyp variable after PyAuto removal.
thestig@chromium.org [Thu, 17 Apr 2014 06:05:59 +0000 (06:05 +0000)]
Delete unused python_ver gyp variable after PyAuto removal.

BUG=224072

Review URL: https://codereview.chromium.org/238983009

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264442 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agogyp performance: don't invoke python to check dir existance
bratell@opera.com [Wed, 16 Apr 2014 21:52:49 +0000 (21:52 +0000)]
gyp performance: don't invoke python to check dir existance

We spend a few tenths of a second every gyp invocation starting
the dir_exists.py program. Seems a bit unnecessary.

BUG=362075
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/234963003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264316 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agoRename IDS ids to be more consistent in ui_strings.grd.
tnagel@chromium.org [Wed, 16 Apr 2014 21:39:27 +0000 (21:39 +0000)]
Rename IDS ids to be more consistent in ui_strings.grd.

This fixes the awkward sorting in grit_whitelist.txt and had been suggested by
Bartosz some time ago.

BUG=none

Review URL: https://codereview.chromium.org/240113004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264312 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agowin: When use_goma=1, favor fastbuild=1 over win_z7=1.
thakis@chromium.org [Wed, 16 Apr 2014 21:11:46 +0000 (21:11 +0000)]
win: When use_goma=1, favor fastbuild=1 over win_z7=1.

BUG=332697
NOTRY=true

Review URL: https://codereview.chromium.org/239113010

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264304 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agogyp performance: Add a DoMain version of build/compiler_version
bratell@opera.com [Wed, 16 Apr 2014 20:59:15 +0000 (20:59 +0000)]
gyp performance: Add a DoMain version of build/compiler_version

No need to start a full python interpreter for compiler_version.py. This saves about 0.6 CPU seconds.

BUG=362075
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/235083003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264303 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agoFill out the rest of the net targets in the GN build except Android.
brettw@chromium.org [Wed, 16 Apr 2014 20:34:31 +0000 (20:34 +0000)]
Fill out the rest of the net targets in the GN build except Android.

This also fixes some missing base files and libresolv naming.

R=jar@chromium.org

Review URL: https://codereview.chromium.org/236063018

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264296 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agoMove instrumented-libraries-related settings out of common.gypi.
earthdok@chromium.org [Wed, 16 Apr 2014 19:23:26 +0000 (19:23 +0000)]
Move instrumented-libraries-related settings out of common.gypi.

Instead, apply them as direct_dependent_settings to dependents of the
instrumented_libraries target. This is functionally equivalent, but reduces code
duplication, improves locality and also allows us to disable the use of
instrumented libraries on a target-by-target basis.

BUG=313751
TBR=glider@chromium.org
TEST=MSan
NOTRY=true

Review URL: https://codereview.chromium.org/240413004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264281 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agoozone: evdev: Add libgestures bindings for touchpad & mouse support
spang@chromium.org [Wed, 16 Apr 2014 19:04:30 +0000 (19:04 +0000)]
ozone: evdev: Add libgestures bindings for touchpad & mouse support

The gestures library converts kernel input events into gestures. This
currently happens inside X11 in the xf86-input-cmt driver, but without
X11 we have to do use it from EventFactoryEvdev

BUG=351168

Review URL: https://codereview.chromium.org/193813003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264273 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

3 days agoMake NSS work in GN build on Windows.
brettw@chromium.org [Wed, 16 Apr 2014 18:29:50 +0000 (18:29 +0000)]
Make NSS work in GN build on Windows.

This also adds sqlite which is a dependency.

This required some build config changes since nss needs to opt-out of certain default settings.

BUG=
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/235863024

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264255 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agoSplit linux_use_gold_binary into linux_use_bundled_gold and linux_use_bundled_binutils.
sbc@chromium.org [Wed, 16 Apr 2014 09:50:23 +0000 (09:50 +0000)]
Split linux_use_gold_binary into linux_use_bundled_gold and linux_use_bundled_binutils.

linux_use_bundled_binutils adds the binutils to cflags.
linux_use_bundled_gold adds it to the ldflags.  The former
does not work when cross compiling to ARM as the other
binutils do not currently support ARM.  This was causing the
ARM cross compile (for which we do want use the bundled
gold) to break.

I think the word "bundled" works better than "binary" since
it distinguishes from the system gold (which is also a binary).

BUG=363661
TEST=linux_arm_cross_compile trybot

Review URL: https://codereview.chromium.org/239163003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264165 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agowhitespace change to trigger a new build cycle
jochen@chromium.org [Wed, 16 Apr 2014 08:38:59 +0000 (08:38 +0000)]
whitespace change to trigger a new build cycle

BUG=none
TBR=phajdan.jr@chromium.org

Review URL: https://codereview.chromium.org/240153002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264153 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agoenvsetup: Stop adding third_party/android_tools/sdk/tools to PATH.
thakis@chromium.org [Wed, 16 Apr 2014 03:03:48 +0000 (03:03 +0000)]
envsetup: Stop adding third_party/android_tools/sdk/tools to PATH.

If people want to run many binaries from that folder locally, they can add
that folder to their PATH locally.

Depends on:
* https://chrome-internal-review.googlesource.com/160476
* https://codereview.chromium.org/236313004/

BUG=330631
NOTRY=true

Review URL: https://codereview.chromium.org/237953002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@264087 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agoMatch only make in gyp_chromium excludes.
bungeman@chromium.org [Tue, 15 Apr 2014 20:34:51 +0000 (20:34 +0000)]
Match only make in gyp_chromium excludes.

The current test also matches the CMake generator.

R=thakis@chromium.org

Review URL: https://codereview.chromium.org/235943009

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263984 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agoImplement net in GN build.
brettw@chromium.org [Tue, 15 Apr 2014 19:26:44 +0000 (19:26 +0000)]
Implement net in GN build.

This does the net target. I started doing some more that are commented out at the bottom, and there are a few more after that do do. The deps of these targets started to build up to an unreasonable level and this should be a good checkpoint.

Fix grit rule.

Fix SSL dependent configs

Add TLD cleanup

Add gconf and gio targets for Linux.

Add sources filtering for ChromeOS, .rc, and .mm files. Remove built-in code that removes .rc and .mm files (this wasn't quite complete so causes problems).

BUG=
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/236713002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263967 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agoFix provision_devices after the DeviceUtils patch.
jbudorick@chromium.org [Tue, 15 Apr 2014 19:01:12 +0000 (19:01 +0000)]
Fix provision_devices after the DeviceUtils patch.

BUG=363702
NOTRY=true

Review URL: https://codereview.chromium.org/236313010

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263957 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agoSet the default ASan options for executables built with ASan on Linux.
glider@chromium.org [Tue, 15 Apr 2014 18:39:20 +0000 (18:39 +0000)]
Set the default ASan options for executables built with ASan on Linux.

This is a copy of https://codereview.chromium.org/201153007 and https://codereview.chromium.org/213113002
updated to the current trunk with more descriptive comments in sanitizer_options.cc and the additional
legacy_pthread_cond=1 option for the Linux Official builds (legacy_pthread_cond is to be deprecated,
thus we don't enable it on the bots).

This CL introduces a module, base/debug/sanitizer_options.cc, which will override
the defaults for various dynamic tools (only ASan at this moment). For every executable
built with a dynamic tool this module will be linked into that executable, providing
weak functions to be called by the tool.

The existing declaration of __asan_default_options() in chrome/app/chrome_exe_main_gtk.cc
has been moved into sanitizer_options.cc (now every binary built with GOOGLE_CHROME_BUILD=1
will have the same options as google-chrome-asan.
The existing declaration of __asan_default_options() in chrome/nacl/nacl_helper_linux.cc
has been kept as is, but we had to remove -Wl,-u_sanitizer_options_link_helper to avoid picking sanitizer_options.cc.

The default options target is deliberately disabled on 32-bit Chromium OS builds,
where one of the host binaries (mksnapshot.ia32) is 32-bit despite host_arch==x86_64.

GYP changes for OSX and iOS will be committed separately.

TBR=thakis@chromium.org,mseaborn@chromium.org
BUG=302040

Review URL: https://codereview.chromium.org/238123003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263941 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

4 days agoUpdate ARM sysroot image.
sbc@chromium.org [Tue, 15 Apr 2014 17:08:10 +0000 (17:08 +0000)]
Update ARM sysroot image.

This new image include libexif and a few security fixes
from Ubuntu.

Also, cleanup the downloading code so it fails nicely.

R=thestig@chromium.org

Review URL: https://codereview.chromium.org/239113004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263896 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

5 days agoMake resource repacker work on Windows
scottmg@chromium.org [Tue, 15 Apr 2014 12:51:25 +0000 (12:51 +0000)]
Make resource repacker work on Windows

Add pragma compatible with cl, and make parser less strict
about the stuff around whitelisted_resource_NNNN.

R=aurimas@chromium.org,thakis@chromium.org
BUG=338759

Review URL: https://codereview.chromium.org/237663003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263820 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

5 days agoAdd SSL support to the GN build
brettw@chromium.org [Mon, 14 Apr 2014 23:25:13 +0000 (23:25 +0000)]
Add SSL support to the GN build

This moves files from the secondary tree to the main one, and renames the "meta" one to //crypto:ssl. It also adds the crypto unit tests.

R=rsleevi@chromium.org

Review URL: https://codereview.chromium.org/231673006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263744 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

5 days agoRevert 263625 "Set the default ASan options for executables buil..."
fischman@chromium.org [Mon, 14 Apr 2014 15:54:11 +0000 (15:54 +0000)]
Revert 263625 "Set the default ASan options for executables buil..."

Reason for revert: broke all 3 Linux ASan LSan Tests bots:
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/1338
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%282%29/builds/1563
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%283%29/builds/1796

> Set the default ASan options for executables built with ASan on Linux.
>
> This is a copy of https://codereview.chromium.org/201153007 and https://codereview.chromium.org/213113002
> updated to the current trunk with more descriptive comments in sanitizer_options.cc and the additional
> legacy_pthread_cond=1 option for the Linux builds.
>
> This CL introduces a module, base/debug/sanitizer_options.cc, which will override
> the defaults for various dynamic tools (only ASan at this moment). For every executable
> built with a dynamic tool this module will be linked into that executable, providing
> weak functions to be called by the tool.
>
> The existing declaration of __asan_default_options() in chrome/app/chrome_exe_main_gtk.cc
> has been moved into sanitizer_options.cc (now every binary built with GOOGLE_CHROME_BUILD=1
> will have the same options as google-chrome-asan.
> The existing declaration of __asan_default_options() in chrome/nacl/nacl_helper_linux.cc
> has been kept as is, but we had to remove -Wl,-u_sanitizer_options_link_helper to avoid picking sanitizer_options.cc.
>
> The default options target is deliberately disabled on 32-bit Chromium OS builds, where one of the host binaries (mksnapshot.ia32) is 32-bit despite host_arch==x86_64.
>
> GYP changes for OSX and iOS will be committed separately.
>
> BUG=302040,341805
> TBR=thakis@chromium.org,mseaborn@chromium.org
>
> Review URL: https://codereview.chromium.org/230263003

TBR=glider@chromium.org

Review URL: https://codereview.chromium.org/237203003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263634 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

5 days agoSet the default ASan options for executables built with ASan on Linux.
glider@chromium.org [Mon, 14 Apr 2014 13:56:35 +0000 (13:56 +0000)]
Set the default ASan options for executables built with ASan on Linux.

This is a copy of https://codereview.chromium.org/201153007 and https://codereview.chromium.org/213113002
updated to the current trunk with more descriptive comments in sanitizer_options.cc and the additional
legacy_pthread_cond=1 option for the Linux builds.

This CL introduces a module, base/debug/sanitizer_options.cc, which will override
the defaults for various dynamic tools (only ASan at this moment). For every executable
built with a dynamic tool this module will be linked into that executable, providing
weak functions to be called by the tool.

The existing declaration of __asan_default_options() in chrome/app/chrome_exe_main_gtk.cc
has been moved into sanitizer_options.cc (now every binary built with GOOGLE_CHROME_BUILD=1
will have the same options as google-chrome-asan.
The existing declaration of __asan_default_options() in chrome/nacl/nacl_helper_linux.cc
has been kept as is, but we had to remove -Wl,-u_sanitizer_options_link_helper to avoid picking sanitizer_options.cc.

The default options target is deliberately disabled on 32-bit Chromium OS builds, where one of the host binaries (mksnapshot.ia32) is 32-bit despite host_arch==x86_64.

GYP changes for OSX and iOS will be committed separately.

BUG=302040,341805
TBR=thakis@chromium.org,mseaborn@chromium.org

Review URL: https://codereview.chromium.org/230263003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263625 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

6 days agoRemove the no-longer needed src/third_party/gold.
mithro@mithis.com [Sun, 13 Apr 2014 17:34:23 +0000 (17:34 +0000)]
Remove the no-longer needed src/third_party/gold.

BUG=352046,362119

Review URL: https://codereview.chromium.org/232473002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263550 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoMove helper utilities for the DOM Distiller Viewer to core.
nyquist@chromium.org [Sat, 12 Apr 2014 10:45:51 +0000 (10:45 +0000)]
Move helper utilities for the DOM Distiller Viewer to core.

This CL moves the logic for creating HTML content, and retrieving resources
from the content/ subfolder of the component to core/.

It also provides a helper method for calling the right method on the
DomDistillerService based on a URL.

BUG=319881
TBR=jochen@chromium.org
TBR=jshin@chromium.org

Review URL: https://codereview.chromium.org/235833003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263511 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoFix two typos in the documentation of some args of GN.
tfarina@chromium.org [Sat, 12 Apr 2014 06:47:18 +0000 (06:47 +0000)]
Fix two typos in the documentation of some args of GN.

1- s/useing/using
2- s/SHell/Shell

This was found running the following command line and reading the docs.

$ gn args

BUG=None
TBR=brettw@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/236333002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263497 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days ago[Android] Fix UIAutomator, Linker, and downstream tests.
jbudorick@chromium.org [Sat, 12 Apr 2014 05:26:08 +0000 (05:26 +0000)]
[Android] Fix UIAutomator, Linker, and downstream tests.

BUG=267773

Review URL: https://codereview.chromium.org/235943002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263484 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoozone: Initial test blacklist for ChromeOS build
spang@chromium.org [Sat, 12 Apr 2014 05:16:31 +0000 (05:16 +0000)]
ozone: Initial test blacklist for ChromeOS build

We are missing some IME & event functionality. Disable those tests until
we've implemented them, so that we can run more tests.

BUG=361341

Review URL: https://codereview.chromium.org/235673005

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263478 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoRemove stale GPU code on Windows/Mac
piman@chromium.org [Sat, 12 Apr 2014 05:09:08 +0000 (05:09 +0000)]
Remove stale GPU code on Windows/Mac

- on Windows: with Aura + removal of the accelerated surface support, the only
"transport"-type surfaces are TextureImageTransportSurfaces and
IOSurfaceImageTransportSurface. The GpuHostMsg_AcceleratedSurface* messages can
go directly to the UI thread without the need for filtering on the IO thread.
- on both Mac and Windows, we use the compositor thread always, so we don't need
the ViewMsg_SwapBuffers_ACK and associated logic any more.

BUG=None

Review URL: https://codereview.chromium.org/137663006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263475 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days ago[Android] Clean up a few uses of the not-yet-existent DeviceUtils interface.
jbudorick@chromium.org [Sat, 12 Apr 2014 03:16:11 +0000 (03:16 +0000)]
[Android] Clean up a few uses of the not-yet-existent DeviceUtils interface.

BUG=267773

Review URL: https://codereview.chromium.org/234423006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263463 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoRemove a few dead MSVS_VERSION checks.
thakis@chromium.org [Sat, 12 Apr 2014 03:07:46 +0000 (03:07 +0000)]
Remove a few dead MSVS_VERSION checks.

We only support 2013.

TBR=robertshield@chromium.org
BUG=340358

Review URL: https://codereview.chromium.org/233843003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263460 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoCheckpoint work to get GN builds working on Android.
dpranke@chromium.org [Sat, 12 Apr 2014 01:19:16 +0000 (01:19 +0000)]
Checkpoint work to get GN builds working on Android.

This fixes enough issues so that we can compile and link some libraries, but
it is not yet a working build.

Work remaining:
- base needs JNI support ('base_jni_headers', etc.)
- we need to figure out what should link and/or be APKs to run the tests
- we need to adjust GN to only build in the host config, not the target config.

R=brettw@chromium.org, cjhopman@chromium.org
TBR=jam
BUG=360936

Review URL: https://codereview.chromium.org/227673008

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263440 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoMake landmines.py ignore eclipse gyp generator.
newt@chromium.org [Sat, 12 Apr 2014 01:13:21 +0000 (01:13 +0000)]
Make landmines.py ignore eclipse gyp generator.

The eclipse gyp generator is used to generate eclipse project files, not
to actually build anything. Landmines should return early if the gyp
generator is eclipse rather than failing with the error "Unexpected
GYP_GENERATORS (eclipse)"

Review URL: https://codereview.chromium.org/236063002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263438 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoGN Windows build fixes
brettw@chromium.org [Fri, 11 Apr 2014 23:06:17 +0000 (23:06 +0000)]
GN Windows build fixes

This also removes some patterns from the sources assignment filter, and adds scary comments not to make it bigger. The GYP version of this is out of control so I want to set a clear policy of what is included and not.

I removed X-related stuff from the filter (there are only about 50 files around the tree) and added manual rules for the affected files.

BUG=
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/235983002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263394 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoFix get_device_configuration.py/device_utils.py
dtrainor@chromium.org [Fri, 11 Apr 2014 19:49:52 +0000 (19:49 +0000)]
Fix get_device_configuration.py/device_utils.py

- Fix get_device_configuration.py to use build_device.GetAttachedDevices
without the () as it is now a list not a method since () was added at
build_device.GetAttachedDevices
- Fix device_utils.py to check for unicode as well as str since the config
push_libraries was loading had unicode for the device id

NOTRY=true
BUG=267773

Review URL: https://codereview.chromium.org/233203003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263317 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoRevert 263262 "Newly created profiles should have the grey silho..."
viettrungluu@chromium.org [Fri, 11 Apr 2014 18:11:26 +0000 (18:11 +0000)]
Revert 263262 "Newly created profiles should have the grey silho..."

Adds static initializers. See:
http://build.chromium.org/p/chromium/builders/Linux%20x64/builds/63585

# profile_avatar_icon_util.cc profiles::kNoHighResAvatar
# profile_avatar_icon_util.cc profiles::kDefaultAvatarIconResourceFileNames+0xd0

> Newly created profiles should have the grey silhouette as avatar.
> The name should still be automatically generated.
>
> Also, all the avatar icons are becoming squares (31x38 -> 38x38). New resources will be committed and necessary code changes have been made.
>
>
> BUG=337986
> TEST=When creating a new profile from anywhere except the settings. Also, verify avatar icons are still rendered correctly, whether GAIA, the placeholder or a cartoon icon.
> page, the automatically assigned avatar should the grey
> placeholder (silhouette).
>
> Review URL: https://codereview.chromium.org/212603011

TBR=mlerman@chromium.org

Review URL: https://codereview.chromium.org/234583008

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263284 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoAdd support for substring replacement in gypi_to_gn
brettw@chromium.org [Fri, 11 Apr 2014 17:23:40 +0000 (17:23 +0000)]
Add support for substring replacement in gypi_to_gn

Use substring replacement in skia source file list, and update the skia BUILD.gn file accordingly. The .gn file gets a bit simpler now: we can do the substring replacement to make absolute paths rather than doing rebase for all the files.

R=djsollen@google.com, dpranke@chromium.org

Review URL: https://codereview.chromium.org/231983002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263272 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

8 days agoNewly created profiles should have the grey silhouette as avatar.
mlerman@chromium.org [Fri, 11 Apr 2014 16:51:28 +0000 (16:51 +0000)]
Newly created profiles should have the grey silhouette as avatar.
The name should still be automatically generated.

Also, all the avatar icons are becoming squares (31x38 -> 38x38). New resources will be committed and necessary code changes have been made.

BUG=337986
TEST=When creating a new profile from anywhere except the settings. Also, verify avatar icons are still rendered correctly, whether GAIA, the placeholder or a cartoon icon.
page, the automatically assigned avatar should the grey
placeholder (silhouette).

Review URL: https://codereview.chromium.org/212603011

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263262 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoUse -fuse-ld=gold when supported to select gold.
mithro@mithis.com [Fri, 11 Apr 2014 08:05:11 +0000 (08:05 +0000)]
Use -fuse-ld=gold when supported to select gold.

Allow usage of gold on gcc 4.8 systems such as debian.

BUG=352046

Review URL: https://codereview.chromium.org/225393007

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263189 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoUpdating the gold location.
mithro@mithis.com [Fri, 11 Apr 2014 07:56:03 +0000 (07:56 +0000)]
Updating the gold location.

BUG=362119

Review URL: https://codereview.chromium.org/233393002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263185 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoFix screenshot.py
cjhopman@chromium.org [Fri, 11 Apr 2014 03:00:42 +0000 (03:00 +0000)]
Fix screenshot.py

build/android/screenshot.py was creating a DeviceUtils. Then grabbing
it's old_interface (i.e. an AndroidCommands) and calling screenshot on
that. That then called TakeScreenshot from
build/android/pylib/screenshot.py passing itself. But
build/android/pylib/screenshot.py expects a DeviceUtils, not an
AndroidCommands. So wrap the AndroidCommands back into a DeviceUtils
first.

build/android/screenshot.py could probably just call into
build/android/pylib/screenshot.py directly.

TBR=jbudorick
NOTRY=true

Review URL: https://codereview.chromium.org/234533003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263157 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoAndroid: Fix default filename for kTraceStartup file
sievers@google.com [Fri, 11 Apr 2014 01:28:54 +0000 (01:28 +0000)]
Android: Fix default filename for kTraceStartup file

Putting chrometrace.log in the current working directory is not
useful on Android. Generate a filepath on the sdcard instead the
same way we do it for intent triggered tracing.

R=piman@chromium.org, wangxianzhu@chromium.org

Review URL: https://codereview.chromium.org/232053002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263153 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoenvsetup: Unbreak android_gyp for a while.
thakis@chromium.org [Fri, 11 Apr 2014 00:28:35 +0000 (00:28 +0000)]
envsetup: Unbreak android_gyp for a while.

r262292 accidentally broke android_gyp. Unbreak it for a while, by using
an env var with a long name that hopefully nobody will refer to from anywhere
in the 3-4 weeks it's going to be in the tree.

BUG=330631
NOTRY=true

Review URL: https://codereview.chromium.org/234223002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263143 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoAdd GN buildfile for gfx
brettw@chromium.org [Thu, 10 Apr 2014 21:16:59 +0000 (21:16 +0000)]
Add GN buildfile for gfx

Adds libjpeg and some global flags for Ozone from the GYP build.

Improves error messages for mismatched item types I noticed when writing this patch (changes in builder.cc).

BUG=
R=bradnelson@google.com

Review URL: https://codereview.chromium.org/224633002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263086 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoAdd ui/wm to aura_builder target
oshima@chromium.org [Thu, 10 Apr 2014 21:13:48 +0000 (21:13 +0000)]
Add ui/wm to aura_builder target
 This is to mitigate the issue crbug.com/354775 by at least compiling wm_core_unittsts on bots.

BUG=354775
R=jamescook@chromium.org

Review URL: https://codereview.chromium.org/233733002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263083 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days ago[Android] Change object types from AndroidCommands to DeviceUtils in build/android/.
jbudorick@chromium.org [Thu, 10 Apr 2014 19:37:30 +0000 (19:37 +0000)]
[Android] Change object types from AndroidCommands to DeviceUtils in build/android/.

BUG=267773

Review URL: https://codereview.chromium.org/221823011

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263055 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoadd new test targets in chromium_builder_dbg_drmemory_win for DrMemory build
zhaoqin@google.com [Thu, 10 Apr 2014 19:35:34 +0000 (19:35 +0000)]
add new test targets in chromium_builder_dbg_drmemory_win for DrMemory build
- add gpu_unittests

R=bruening@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/233393005

git-svn-id: http://src.chromium.org/svn/trunk/src/build@263053 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoMake sure binutils values are always defined.
mithro@mithis.com [Thu, 10 Apr 2014 14:01:57 +0000 (14:01 +0000)]
Make sure binutils values are always defined.

BUG=352046

Review URL: https://codereview.chromium.org/228763006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262981 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days agoFix variables expansion in build/mac/asan.gyp
glider@chromium.org [Thu, 10 Apr 2014 14:01:18 +0000 (14:01 +0000)]
Fix variables expansion in build/mac/asan.gyp

BUG=344836, 235466
TBR=thakis@chromium.org

Review URL: https://codereview.chromium.org/232723003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262980 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

9 days ago[Android]: Update Arm64 and x86_64 build targets to use the experimental ndk.
rmcilroy@chromium.org [Thu, 10 Apr 2014 13:42:46 +0000 (13:42 +0000)]
[Android]: Update Arm64 and x86_64 build targets to use the experimental ndk.

BUG=354405,346626

Review URL: https://codereview.chromium.org/225093007

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262969 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

10 days agoFix the condition to disable the warning when building with Xcode
sdefresne@chromium.org [Wed, 9 Apr 2014 20:53:00 +0000 (20:53 +0000)]
Fix the condition to disable the warning when building with Xcode

The warning -Wshorten-64-to-32 is automatically enabled by Xcode 5.1
when building for 64-bit architecture. Disable when the generator is
Xcode, not ninja.

BUG=359107

Review URL: https://codereview.chromium.org/231023002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262805 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

10 days agoAdd bookmark undo manager commands to the ios grit whitelist.
erikchen@chromium.org [Wed, 9 Apr 2014 20:41:47 +0000 (20:41 +0000)]
Add bookmark undo manager commands to the ios grit whitelist.

This is in preparation for bringing the bookmark undo manager to ios.

BUG=NONE

Review URL: https://codereview.chromium.org/230003002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262802 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

10 days agoWhen "linux_use_gold_binary" is set, we now use gold from third_party/binutils rather...
mithro@mithis.com [Wed, 9 Apr 2014 20:20:45 +0000 (20:20 +0000)]
When "linux_use_gold_binary" is set, we now use gold from third_party/binutils rather than third_party/gold (allowing us to remove third_party/gold dependency).

As third_party/binutils is 2.24, this also satisfies the binutils version requirements for Debug Fission. Thus, when building with clang on Ubuntu Precise it should now be enabled. This should cause a significant speedup in linking for most chrome developers on Linux.

Requires;
 * https://codereview.chromium.org/209853003/ - Adding binutils as a DEPS to allow DebugFission on Ubuntu Precise when compiling with clang.

BUG=352046

Review URL: https://codereview.chromium.org/196573022

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262794 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

10 days agoAdd Skia to the GN build.
brettw@chromium.org [Wed, 9 Apr 2014 19:55:29 +0000 (19:55 +0000)]
Add Skia to the GN build.

This makes the GN build of Skia compile and adds it to the GN build.
Previously, the configuration of the GN Skia tried to match the confusing array
of skia targets. This new version dispenses with that and just adds all files
(except SSE ones) to one target. I'm not even sure if it's necessary to split
out the SSE ones, but it seems nice since that target will get more complicated
when we add Arm and MIPS support.

I audited the defines again and added some to the main build config that had
been added to the GYP build recently.

This improves some operator error messages that I found confusing when I had errors in my file.

BUG=
R=djsollen@google.com, scottmg@chromium.org

Review URL: https://codereview.chromium.org/214883005

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262779 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoSplit toolchain update and install lookup to make generation faster
scottmg@chromium.org [Wed, 9 Apr 2014 01:56:20 +0000 (01:56 +0000)]
Split toolchain update and install lookup to make generation faster

runhooks does toolchain update, saving relevant information to
build/win_toolchain.json, and during generation this is loaded
and used. This worked almost the same previously, however update
was called at generation time and the .json was deleted after
it was used. This speeds up generation time by a few seconds.

(In preference to https://codereview.chromium.org/228093002/)

R=dpranke@chromium.org
BUG=360878

Review URL: https://codereview.chromium.org/226643011

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262590 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoUse instrumented libc++ by default in TSan builds.
glider@chromium.org [Tue, 8 Apr 2014 23:56:09 +0000 (23:56 +0000)]
Use instrumented libc++ by default in TSan builds.

BUG=318770
TBR=earthdok@chromium.org

Review URL: https://codereview.chromium.org/181013015

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262558 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoTemplate invocation fixes in GN
brettw@chromium.org [Tue, 8 Apr 2014 22:35:18 +0000 (22:35 +0000)]
Template invocation fixes in GN

This adds an error check when invoking templates which caused really confusing messages if the template invocation encountered an error (because we'd continue running).

Hooks up the provider for programatically defined variables in template invocations so those can be used.

Sets the current directory in a template invocation to be that of the invoking file.

No longer define the target-related programatic variables in an import. Using these in an import will give the directory relative to the import, which is probabyl not what you want.

Fix the Windows build by adding a missing library. Add a warning not to add more to the main list (this added .lib is pretty obscure).

BUG=
R=cjhopman@chromium.org

Review URL: https://codereview.chromium.org/226223006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262524 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoUpdate usb charging command to wait for it to succeed.
qsr@chromium.org [Tue, 8 Apr 2014 21:39:21 +0000 (21:39 +0000)]
Update usb charging command to wait for it to succeed.

R=tonyg@chromium.org,pliard@chromium.org

Review URL: https://codereview.chromium.org/228253003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262512 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoFix the Mac toolchain definition in the GN build.
brettw@chromium.org [Tue, 8 Apr 2014 20:01:51 +0000 (20:01 +0000)]
Fix the Mac toolchain definition in the GN build.

I did not update this when I updated the template syntax.

This also fixes a typo in the GCC toolchain definition (it was just in some helpful assertions and didn't affect anything).

BUG=

Review URL: https://codereview.chromium.org/227693007

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262491 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoandroid envsetup: Stop exporting ANDROID_SDK_ROOT.
thakis@chromium.org [Tue, 8 Apr 2014 19:42:03 +0000 (19:42 +0000)]
android envsetup: Stop exporting ANDROID_SDK_ROOT.

The only reference to itwasis from an unused slave script
which I deleted in r262454.

BUG=330631
NOTRY=true

Review URL: https://codereview.chromium.org/226453006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262480 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoRemoves "copy" action for extra_native_libs in java_apk.gypi.
gunsch@chromium.org [Tue, 8 Apr 2014 18:39:54 +0000 (18:39 +0000)]
Removes "copy" action for extra_native_libs in java_apk.gypi.

Since the output lib file is already generated by
strip_native_libraries, the 'copy' action will result in the
unstripped version overwriting the stripped one and getting
packaged in the final APK.

Before change:

$ ninja -C out/Debug -j32 content_shell_apk
[snipped]

$ ls -l out/Debug/apks/ContentShell.apk
-rw-r----- 1 gunsch eng 27M Apr  3 15:09 out/Debug/apks/ContentShell.apk

$ unzip -l out/Debug/apks/ContentShell.apk | grep lib/
   264772  2014-04-03 14:54   lib/armeabi-v7a/gdbserver
 16030344  2014-04-03 15:00   lib/armeabi-v7a/libosmesa.so
 38423836  2014-04-03 15:09   lib/armeabi-v7a/libcontent_shell_content_view.so

$ ls -l out/Debug/lib
total 2.2G
-rwxr-x--- 1 gunsch eng 2.2G Apr  3 15:09 libcontent_shell_content_view.so
-rw-r----- 1 gunsch eng 5.7K Apr  3 15:09 libcontent_shell_content_view.so.TOC
-rwxr-x--- 1 gunsch eng  26K Mar 28 14:25 libmojo_system.so
-rw-r----- 1 gunsch eng  714 Mar 28 14:25 libmojo_system.so.TOC
-rwxr-x--- 2 gunsch eng  16M Apr  3 15:00 libosmesa.so

$ ls -l out/Debug/content_shell_apk/libs/armeabi-v7a/
total 53M
-rwxr-x--- 2 gunsch eng 259K Apr  3 14:54 gdbserver
-rwxr-x--- 1 gunsch eng  37M Apr  3 15:09 libcontent_shell_content_view.so
-rwxr-x--- 2 gunsch eng  16M Apr  3 15:00 libosmesa.so

After change:

$ ls -l out/Debug/apks/ContentShell.apk
-rw-r----- 1 gunsch eng 21M Apr  3 15:31 out/Debug/apks/ContentShell.apk

$ unzip -l out/Debug/apks/ContentShell.apk | grep lib/
   264772  2014-04-03 14:54   lib/armeabi-v7a/gdbserver
  2289816  2014-04-03 15:31   lib/armeabi-v7a/libosmesa.so
 38423836  2014-04-03 15:31   lib/armeabi-v7a/libcontent_shell_content_view.so

$ ls -l out/Debug/lib
total 2.2G
-rwxr-x--- 1 gunsch eng 2.2G Apr  3 15:31 libcontent_shell_content_view.so
-rw-r----- 1 gunsch eng 5.7K Apr  3 15:31 libcontent_shell_content_view.so.TOC
-rwxr-x--- 1 gunsch eng  16M Apr  3 15:30 libosmesa.so

$ ls -l out/Debug/content_shell_apk/libs/armeabi-v7a/
total 40M
-rwxr-x--- 2 gunsch eng 259K Apr  3 14:54 gdbserver
-rwxr-x--- 1 gunsch eng  37M Apr  3 15:31 libcontent_shell_content_view.so
-rwxr-x--- 1 gunsch eng 2.2M Apr  3 15:31 libosmesa.so

BUG=
R=peter@chromium.org

Review URL: https://codereview.chromium.org/224673002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262461 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days ago[Android] Fix gtest test scripts for non-default CHROMIUM_OUT_DIR.
jbudorick@chromium.org [Tue, 8 Apr 2014 17:39:51 +0000 (17:39 +0000)]
[Android] Fix gtest test scripts for non-default CHROMIUM_OUT_DIR.

BUG=347834

Review URL: https://codereview.chromium.org/184733006

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262446 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

11 days agoPull GN @ r262225
brettw@chromium.org [Tue, 8 Apr 2014 16:50:01 +0000 (16:50 +0000)]
Pull GN @ r262225

Implement GN visibility in the build, which this new version implements

BUG=
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/217273004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262438 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

12 days agoLink binaries targeting iOS simulator to the appropriate ASan dynamic runtime.
glider@chromium.org [Tue, 8 Apr 2014 11:11:58 +0000 (11:11 +0000)]
Link binaries targeting iOS simulator to the appropriate ASan dynamic runtime.

BUG=344836,235466
R=thakis@chromium.org

Review URL: https://codereview.chromium.org/218613015

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262372 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

12 days agoandroid envsetup: Stop exporting CHROME_SRC, CURRENT_DIR, SCRIPT_DIR.
thakis@chromium.org [Tue, 8 Apr 2014 03:53:51 +0000 (03:53 +0000)]
android envsetup: Stop exporting CHROME_SRC, CURRENT_DIR, SCRIPT_DIR.

CHROME_SRC is only used inside of envsetup, there's no need to export it.
Depends on https://chrome-internal-review.googlesource.com/159331/

It looks like CURRENT_DIR and SCRIPT_DIR weren't exported intentionally.
Use "local" (which requires putting most of the code in this script in a
function) to make sure they don't make it into the environment that
envsetup is sourced from.

BUG=330631
NOTRY=true
R=yfriedman@chromium.org

Review URL: https://codereview.chromium.org/222933004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262292 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

12 days agoRevert of Revert r245379 Add missing test targets to chrome_official_builder for...
jschuh@chromium.org [Tue, 8 Apr 2014 00:09:30 +0000 (00:09 +0000)]
Revert of Revert r245379 Add missing test targets to chrome_official_builder for Windows (https://codereview.chromium.org/227673007/)

Reason for revert:
The release team owns these tests now.

Original issue's description:
> Revert r245379 Add missing test targets to chrome_official_builder for Windows
>
> These were removed because they break parts of the official build
>
> BUG=334006
> R=scottmg,phajdan.jr@chromium.org
> NOTREECHECKS=true
> NOTRY=true
>
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=262252

TBR=phajdan.jr@chromium.org,scottmg@chromium.org,iannucci@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=334006

Review URL: https://codereview.chromium.org/227703011

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262263 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

12 days agodefine LIBC_GLIBC as 1 for consistency
mostynb@opera.com [Mon, 7 Apr 2014 23:51:45 +0000 (23:51 +0000)]
define LIBC_GLIBC as 1 for consistency

Most of the other defines in build/build_config.h are set to 1, let's
define LIBC_GLIBC as 1 for consistency.

Review URL: https://codereview.chromium.org/227153008

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262255 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

12 days agoRevert r245379 Add missing test targets to chrome_official_builder for Windows
jschuh@chromium.org [Mon, 7 Apr 2014 23:47:08 +0000 (23:47 +0000)]
Revert r245379 Add missing test targets to chrome_official_builder for Windows

These were removed because they break parts of the official build

BUG=334006
R=scottmg,phajdan.jr@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/227673007

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262252 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

12 days ago[Oilpan]: Fix the android builds blink_heap_unittests target.
wibling@chromium.org [Mon, 7 Apr 2014 15:20:18 +0000 (15:20 +0000)]
[Oilpan]: Fix the android builds blink_heap_unittests target.

When moving the blink oilpan heap's infrastructure code I missed updating this path.

R=ager@chromium.org, andrewhayden@chromium.org, erik.corry@gmail.com, haraken@chromium.org, oilpan-reviews@chromium.org, tkent@chromium.org, vegorov@chromium.org, zerny@chromium.org
BUG=360520

Review URL: https://codereview.chromium.org/227373003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262130 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

13 days agoMake it possible to disable fatal linker warnings
phajdan.jr@chromium.org [Mon, 7 Apr 2014 08:17:25 +0000 (08:17 +0000)]
Make it possible to disable fatal linker warnings

This came up in Gentoo Linux bug https://bugs.gentoo.org/show_bug.cgi?id=506268

Fatal linker warnings were originally introduced in https://codereview.chromium.org/196943020

BUG=352985
R=jochen@chromium.org

Review URL: https://codereview.chromium.org/224713007

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262093 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoAdd support for reading .gypi files.
brettw@chromium.org [Sun, 6 Apr 2014 04:35:10 +0000 (04:35 +0000)]
Add support for reading .gypi files.

Adds support for a new value type, a dictionary. To script it looks the same as a scope, but the existing scope didn't have the right memory management semantics for this.

Adds a new input conversion mode to interpret the result of a script as a GN block (basically an eval) and get the result in a dictionary. This allows returning named sets of things from a script. Updates the accessor and defined stuff to work with this new type.

Adds some features to the input file manager so that when we do dynamic reading from a script we keep the text of the read around which means we can give proper errors that refer to the script output, and don't have to do the weird recursive set origin thing to clear out the pointers that became invalid when the InputFile went out of scope.

This allows us to give better error messages when parsing input. The error handling in the input conversion code is changed accordingly.

Checks for unused variables both in code that instantiates a template, and in the template code itself.

Adds a shared python script for outputting values from Python to GN.

Adds a python script to read .gypi files and return the value as a GN scope.

BUG=
R=dpranke@chromium.org, scottmg@chromium.org

Review URL: https://codereview.chromium.org/223783005

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262045 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoRevert 262031 "When "linux_use_gold_binary" is set, we now use g..."
scottmg@chromium.org [Sat, 5 Apr 2014 20:53:01 +0000 (20:53 +0000)]
Revert 262031 "When "linux_use_gold_binary" is set, we now use g..."

Suspected of causing:

http://build.chromium.org/p/chromium.linux/builders/Linux%20Builder%20(dbg)/builds/71338/steps/compile/logs/stdio

[13474/21627] CXX obj/cc/animation/cc_unittests.scrollbar_animation_controller_linear_fade_unittest.o
FAILED: cd ../../remoting; python ../tools/gypv8sh.py ../out/Debug/d8 ../chrome/third_party/mock4js/mock4js.js ../third_party/accessibility-audit/axs_testing.js ../chrome/test/data/webui/test_api.js ../chrome/test/base/js2gtest.js unit "webapp/all_js_load.gtestjs" "chrome/webapp/all_js_load.gtestjs" "../out/Debug/obj/remoting/remoting_unittests.gen/chrome/webapp/all_js_load-gen.cc" "../out/Debug/test_data/chrome/webapp/all_js_load.gtestjs"
../out/Debug/d8: error while loading shared libraries: libicui18n.so: cannot open shared object file: No such file or directory
Traceback (most recent call last):
  File "../tools/gypv8sh.py", line 55, in <module>
    sys.exit(main())
  File "../tools/gypv8sh.py", line 44, in main
    subprocess.check_call(cmd, stdin=subprocess.PIPE, stdout=f)
  File "/usr/lib/python2.7/subprocess.py", line 511, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['../out/Debug/d8', '-e', 'arguments=["../chrome/test/base/js2gtest.js", "webapp/all_js_load.gtestjs", "chrome/webapp/all_js_load.gtestjs", "../out/Debug/obj/remoting/remoting_unittests.gen/chrome/webapp/all_js_load-gen.cc", "unit"]', '../chrome/third_party/mock4js/mock4js.js', '../third_party/accessibility-audit/axs_testing.js', '../chrome/test/data/webui/test_api.js', '../chrome/test/base/js2gtest.js']' returned non-zero exit status 127
FAILED: cd ../../remoting; python ../tools/gypv8sh.py ../out/Debug/d8 ../chrome/third_party/mock4js/mock4js.js ../third_party/accessibility-audit/axs_testing.js ../chrome/test/data/webui/test_api.js ../chrome/test/base/js2gtest.js unit "webapp/browser_globals.gtestjs" "chrome/webapp/browser_globals.gtestjs" "../out/Debug/obj/remoting/remoting_unittests.gen/chrome/webapp/browser_globals-gen.cc" "../out/Debug/test_data/chrome/webapp/browser_globals.gtestjs"
../out/Debug/d8: error while loading shared libraries: libicui18n.so: cannot open shared object file: No such file or directory
Traceback (most recent call last):
  File "../tools/gypv8sh.py", line 55, in <module>
    sys.exit(main())
  File "../tools/gypv8sh.py", line 44, in main
    subprocess.check_call(cmd, stdin=subprocess.PIPE, stdout=f)
  File "/usr/lib/python2.7/subprocess.py", line 511, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['../out/Debug/d8', '-e', 'arguments=["../chrome/test/base/js2gtest.js", "webapp/browser_globals.gtestjs", "chrome/webapp/browser_globals.gtestjs", "../out/Debug/obj/remoting/remoting_unittests.gen/chrome/webapp/browser_globals-gen.cc", "unit"]', '../chrome/third_party/mock4js/mock4js.js', '../third_party/accessibility-audit/axs_testing.js', '../chrome/test/data/webui/test_api.js', '../chrome/test/base/js2gtest.js']' returned non-zero exit status 127
FAILED: cd ../../remoting; python ../tools/gypv8sh.py ../out/Debug/d8 ../chrome/third_party/mock4js/mock4js.js ../third_party/accessibility-audit/axs_testing.js ../chrome/test/data/webui/test_api.js ../chrome/test/base/js2gtest.js unit "webapp/format_iq.gtestjs" "chrome/webapp/format_iq.gtestjs" "../out/Debug/obj/remoting/remoting_unittests.gen/chrome/webapp/format_iq-gen.cc" "../out/Debug/test_data/chrome/webapp/format_iq.gtestjs"
../out/Debug/d8: error while loading shared libraries: libicui18n.so: cannot open shared object file: No such file or directory
Traceback (most recent call last):
  File "../tools/gypv8sh.py", line 55, in <module>
    sys.exit(main())
  File "../tools/gypv8sh.py", line 44, in main
    subprocess.check_call(cmd, stdin=subprocess.PIPE, stdout=f)
  File "/usr/lib/python2.7/subprocess.py", line 511, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['../out/Debug/d8', '-e', 'arguments=["../chrome/test/base/js2gtest.js", "webapp/format_iq.gtestjs", "chrome/webapp/format_iq.gtestjs", "../out/Debug/obj/remoting/remoting_unittests.gen/chrome/webapp/format_iq-gen.cc", "unit"]', '../chrome/third_party/mock4js/mock4js.js', '../third_party/accessibility-audit/axs_testing.js', '../chrome/test/data/webui/test_api.js', '../chrome/test/base/js2gtest.js']' returned non-zero exit status 127
ninja: build stopped: subcommand failed.

> When "linux_use_gold_binary" is set, we now use gold from third_party/binutils rather than third_party/gold (allowing us to remove third_party/gold dependency).
>
> As third_party/binutils is 2.24, this also satisfies the binutils version requirements for Debug Fission. Thus, when building with clang on Ubuntu Precise it should now be enabled. This should cause a significant speedup in linking for most chrome developers on Linux.
>
> Requires;
>  * https://codereview.chromium.org/209853003/ - Adding binutils as a DEPS to allow DebugFission on Ubuntu Precise when compiling with clang.
>
> BUG=352046
>
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=261778
>
> Review URL: https://codereview.chromium.org/196573022

TBR=mithro@mithis.com

Review URL: https://codereview.chromium.org/225093013

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262036 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoRemoves use_mojo
sky@chromium.org [Sat, 5 Apr 2014 20:14:17 +0000 (20:14 +0000)]
Removes use_mojo

Now that we've sorted out the build issues on all platforms it can be
removed.

BUG=353602
TEST=none
R=jam@chromium.org
TBR=tsepez@chromium.org

Review URL: https://codereview.chromium.org/226213002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262032 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoWhen "linux_use_gold_binary" is set, we now use gold from third_party/binutils rather...
mithro@mithis.com [Sat, 5 Apr 2014 20:13:43 +0000 (20:13 +0000)]
When "linux_use_gold_binary" is set, we now use gold from third_party/binutils rather than third_party/gold (allowing us to remove third_party/gold dependency).

As third_party/binutils is 2.24, this also satisfies the binutils version requirements for Debug Fission. Thus, when building with clang on Ubuntu Precise it should now be enabled. This should cause a significant speedup in linking for most chrome developers on Linux.

Requires;
 * https://codereview.chromium.org/209853003/ - Adding binutils as a DEPS to allow DebugFission on Ubuntu Precise when compiling with clang.

BUG=352046

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=261778

Review URL: https://codereview.chromium.org/196573022

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262031 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoRevert "Turn on clang by default on linux."
thakis@chromium.org [Sat, 5 Apr 2014 20:06:28 +0000 (20:06 +0000)]
Revert "Turn on clang by default on linux."

We don't have 32bit clang binaries, but several 32bit builders.
(And since clang now needs a c++11 compiler newer than gcc4.6 to
compile, we can't tell these bots to just build clang themselves
either.)

BUG=360311

Review URL: https://codereview.chromium.org/227133002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262030 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agokick of a round of builds
thakis@chromium.org [Sat, 5 Apr 2014 19:56:38 +0000 (19:56 +0000)]
kick of a round of builds

BUG=none
TBR=awong@chromium.org

Review URL: https://codereview.chromium.org/227143002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262029 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoTurn on clang by default on linux.
thakis@chromium.org [Sat, 5 Apr 2014 18:27:00 +0000 (18:27 +0000)]
Turn on clang by default on linux.

This is mostly to collect performance and size data for now. Unless things
look really good, I'm going to revert this by end-of-day sunday.

BUG=360311
TBR=awong@chromium.org

Review URL: https://codereview.chromium.org/226253008

git-svn-id: http://src.chromium.org/svn/trunk/src/build@262025 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoui/gfx: Add Android support for gfx_unittests target.
tfarina@chromium.org [Sat, 5 Apr 2014 07:45:37 +0000 (07:45 +0000)]
ui/gfx: Add Android support for gfx_unittests target.

By using gtest_target_type we can specialize gfx_unittests for android
and thus get it running on Android.

BUG=331829,299841
TEST=gfx_unittests_apk
R=danakj@chromium.org,bulach@chromium.org

Review URL: https://codereview.chromium.org/225213002

git-svn-id: http://src.chromium.org/svn/trunk/src/build@261986 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoLinux: Do not pass --reduce-memory-overheads to the linker on CrOS.
thestig@chromium.org [Sat, 5 Apr 2014 00:46:14 +0000 (00:46 +0000)]
Linux: Do not pass --reduce-memory-overheads to the linker on CrOS.

BUG=360082
TBR=vapier@chromium.org

Review URL: https://codereview.chromium.org/226013003

git-svn-id: http://src.chromium.org/svn/trunk/src/build@261943 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoSilence git-hooks/pre-commit in the absence of .gitmodules (take 2).
fischman@chromium.org [Sat, 5 Apr 2014 00:29:14 +0000 (00:29 +0000)]
Silence git-hooks/pre-commit in the absence of .gitmodules (take 2).

Other projects use depot_tools (which pulls in this hook) and chromium's build/
(where the hook lives) but don't use git modules (e.g. webrtc).

This first landed as r257470 but was accidentally undone by r259711.

Review URL: https://codereview.chromium.org/225583008

git-svn-id: http://src.chromium.org/svn/trunk/src/build@261931 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks agoRevert of Reapply "Media Galleries API Metadata: Image metadata" (https://codereview...
tommycli@chromium.org [Fri, 4 Apr 2014 22:02:00 +0000 (22:02 +0000)]
Revert of Reapply "Media Galleries API Metadata: Image metadata" (https://codereview.chromium.org/216443005/)

Reason for revert:
Breaking Mac 10.6 release testers. (no idea why, but it is)

Original issue's description:
> Reapply "Media Galleries API Metadata: Image metadata"
>
> Original patch (also PS1) https://codereview.chromium.org/191583002
>
> CQ can't catch the breakages of this patch.
>
> TBR=
> BUG=318450
>
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=261864

TBR=ccameron@chromium.org,thestig@chromium.org,grt@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=318450

Review URL: https://codereview.chromium.org/225023014

git-svn-id: http://src.chromium.org/svn/trunk/src/build@261895 4ff67af0-8c30-449e-8e8b-ad334ec8d88c

2 weeks ago[gin] Add a v8::Platform implementation to allow for v8 posting tasks
jochen@chromium.org [Fri, 4 Apr 2014 21:09:16 +0000 (21:09 +0000)]
[gin] Add a v8::Platform implementation to allow for v8 posting tasks

This currently has no effect unless concurrent sweeping and job based
sweeping is turned on in v8.

BUG=v8:3015
R=abarth@chromium.org,dcarney@chromium.org

Review URL: https://codereview.chromium.org/225413004

git-svn-id: http://src.chromium.org/svn/trunk/src/build@261872 4ff67af0-8c30-449e-8e8b-ad334ec8d88c