Roll old libvpx 70db22..b33794

Roll old libvpx 70db22..b33794

Switch libvpx from static_library to none
https://codereview.chromium.org/1365973002

R=blundell@chromium.org,tomfinegan@chromium.org
BUG=534943
Committed: https://crrev.com/f4e9bd461256bf503772aad27a07df3088f96ae2
Cr-Commit-Position: refs/heads/master@{#351127}

[Johann, 2015-09-24 19:57:15]

Ran 'git cl try -c' - fingers crossed

[Johann, 2015-09-24 20:32:03]

On 2015/09/24 19:57:15, Johann wrote:
> Ran 'git cl try -c' - fingers crossed

Not sure what a runtime deps list file is but I don't have one locally (but I do
have a working clean build and build.ninja file) Taking a look ...

ninja -w dupbuild=err -C /b/build/slave/linux/build/src/out/Release
accessibility_unittests accessibility_unittests_run angle_unittests_run
app_list_unittests app_list_unittests_run app_shell_unittests
app_shell_unittests_run aura_unittests aura_unittests_run base_unittests
base_unittests_run browser_tests browser_tests_run cacheinvalidation_unittests
cacheinvalidation_unittests_run cast_unittests cast_unittests_run cc_unittests
cc_unittests_run chrome chromedriver_unittests chromedriver_unittests_run
chromium_swarm_tests components_browsertests components_browsertests_run
components_unittests components_unittests_run compositor_unittests
compositor_unittests_run content_browsertests content_browsertests_run
content_gl_tests_run content_unittests content_unittests_run crypto_unittests
crypto_unittests_run dbus_unittests device_unittests device_unittests_run
display_unittests display_unittests_run events_unittests events_unittests_run
extensions_browsertests extensions_browsertests_run extensions_unittests
extensions_unittests_run gcm_unit_tests gcm_unit_tests_run gfx_unittests
gfx_unittests_run gl_tests_run gl_unittests gn_unittests gn_unittests_run
google_apis_unittests google_apis_unittests_run gpu_unittests gpu_unittests_run
html_viewer_unittests interactive_ui_tests interactive_ui_tests_run
ipc_mojo_unittests ipc_tests ipc_tests_run jingle_unittests jingle_unittests_run
mandoline:tests media_blink_unittests media_blink_unittests_run media_unittests
media_unittests_run midi_unittests midi_unittests_run mojo_common_unittests
mojo_common_unittests_run mojo_public_application_unittests
mojo_public_bindings_unittests mojo_public_bindings_unittests_run
mojo_public_environment_unittests mojo_public_environment_unittests_run
mojo_public_system_unittests mojo_public_system_unittests_run
mojo_public_utility_unittests mojo_public_utility_unittests_run
mojo_runner_unittests mojo_shell_unittests mojo_surfaces_lib_unittests
mojo_system_unittests mojo_view_manager_lib_unittests nacl_loader_unittests
nacl_loader_unittests_run net_unittests net_unittests_run ppapi_unittests
printing_unittests printing_unittests_run remoting_unittests
remoting_unittests_run resource_provider_unittests sandbox_linux_unittests
sandbox_linux_unittests_run skia_unittests skia_unittests_run sql_unittests
sql_unittests_run sync_integration_tests sync_integration_tests_run
sync_unit_tests sync_unit_tests_run tab_capture_end2end_tests_run
telemetry_gpu_test_run ui_base_unittests ui_base_unittests_run
ui_touch_selection_unittests ui_touch_selection_unittests_run unit_tests
unit_tests_run url_unittests url_unittests_run view_manager_unittests
views_unittests wm_unittests wm_unittests_run -j50
ninja: Entering directory `/b/build/slave/linux/build/src/out/Release'
[1/1] Regenerating ninja files
ninja: error: rebuilding 'build.ninja': subcommand failed
FAILED: /b/build/slave/linux/build/src/buildtools/linux64/gn
--root=/b/build/slave/linux/build/src -q --check
--runtime-deps-list-file=/b/build/slave/linux/build/src/out/Release/runtime_deps
gen //out/Release/
ERROR File for --runtime-deps-list-file doesn't exist.
The file given was "/b/build/slave/linux/build/src/out/Release/runtime_deps"

[Johann, 2015-09-24 22:56:59]

Ehrm, looking closer at this both bots are GN builds:

http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...

FAILED: /b/build/slave/linux/build/src/buildtools/linux64/gn
--root=/b/build/slave/linux/build/src -q --check
--runtime-deps-list-file=/b/build/slave/linux/build/src/out/Release/runtime_deps
gen //out/Release/
ERROR File for --runtime-deps-list-file doesn't exist.
The file given was "/b/build/slave/linux/build/src/out/Release/runtime_deps"
ninja: error: rebuilding 'build.ninja': subcommand failed

http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...

FAILED:
/b/build/slave/linux_chromium_gn_chromeos/build/src/buildtools/linux64/gn
--root=/b/build/slave/linux_chromium_gn_chromeos/build/src -q --check
--runtime-deps-list-file=/b/build/slave/linux_chromium_gn_chromeos/build/src/out/Release/runtime_deps
gen //out/Release/
ERROR File for --runtime-deps-list-file doesn't exist.
The file given was
"/b/build/slave/linux_chromium_gn_chromeos/build/src/out/Release/runtime_deps"

so besides being super weird (didn't touch GN), now I have an idea why my
gyp-based repro attempts aren't getting anywhere and start poking elsewhere

[blundell, 2015-09-25 13:04:45]

+dpranke@

My guess is that the GN compile errors are an unrelated problem that was exposed
by the fact that you did a clobber build.

[Johann, 2015-09-25 19:03:14]

On 2015/09/25 13:04:45, blundell wrote:
> +dpranke@
> 
> My guess is that the GN compile errors are an unrelated problem that was
exposed
> by the fact that you did a clobber build.

https://codereview.chromium.org/1372633003/

Tried another change which simply poked libvpx and provokes the same issue.
Running the bots with -c fails compile with the same message, without -c does
not. "clobber" is roughly deleting out/, right? Wondering why I'm not seeing it
locally (or getting emails about other people seeing it locally)

[Dirk Pranke, 2015-09-25 20:38:13]

Interesting.

Yeah, --clobber isn't going to work on the GN bots at the moment. You've found a
bug.

What's happening is that the bots run two build steps. First, we generate the
build files,
and as part of that, in gn we generate the isolates for swarming; to do that we
write a
file containing the list of targets we want to isolate into the 'runtime_deps'
file in the build directory.

Then, when ninja runs to do the compile, the first thing it does is attempt to
make sure the
.ninja build files are up-to-date; if it thinks they aren't, it attempts to
re-run the GN command
that was used to generate them in the first place.

There's currently a bug in GN where ninja thinks the build files are *always*
out of date at least
once after being generated (crbug.com/535589). 

The problem w/ clobber is that we delete the build directory after running GN
but before running
ninja, so that runtime_deps file is getting deleted.

Why are you running try --clobber ?

[Johann, 2015-09-25 20:41:52]

Ouch, that's interesting

> Why are you running try --clobber ?

blundell@ wanted to make sure the gyp bots were starting from a clean slate.

Sounds like this patch can land then, since it passes without -c

[Johann, 2015-09-25 20:42:02]

The CQ bit was checked by johannkoenig@google.com to run a CQ dry run

[commit-bot: I haz the power, 2015-09-25 20:42:52]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1369553003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1369553003/20001

[commit-bot: I haz the power, 2015-09-25 22:13:16]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-25 22:13:17]

Dry run: Try jobs failed on following builders:
  win_chromium_x64_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[Johann, 2015-09-25 22:15:46]

On 2015/09/25 22:13:17, commit-bot: I haz the power wrote:
> Dry run: Try jobs failed on following builders:
>   win_chromium_x64_rel_ng on tryserver.chromium.win (JOB_FAILED,
>
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

Failure is unrelated.

tomfinegan@, blundell@, ehl gee tea em?

[Dirk Pranke, 2015-09-26 01:21:41]

On 2015/09/25 20:41:52, Johann wrote:
> Ouch, that's interesting
> 
> > Why are you running try --clobber ?
> 
> blundell@ wanted to make sure the gyp bots were starting from a clean slate.
> 
> Sounds like this patch can land then, since it passes without -c

Ah, okay. 

I filed crbug.com/536292 for the GN + clobber issue in case anyone besides me is
interested :).

[blundell, 2015-09-28 07:19:13]

LGTM

Thanks for following up on this!

[Johann, 2015-09-28 17:52:49]

The CQ bit was checked by johannkoenig@google.com

[Tom Finegan, 2015-09-28 17:53:26]

lgtm

[commit-bot: I haz the power, 2015-09-28 17:55:00]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1369553003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1369553003/20001

[commit-bot: I haz the power, 2015-09-28 19:45:10]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-09-28 19:46:28]

Patchset 2 (id:??) landed as
https://crrev.com/f4e9bd461256bf503772aad27a07df3088f96ae2
Cr-Commit-Position: refs/heads/master@{#351127}