SDK building script now uses package_version to extract.

NaCl: Update revision in DEPS, r13018 -> r13062.

Chromium NaCl scripts now uses package_version to extract.

Now that we have split up the toolchain into multiple tars, we should
use the package versioning script to manage which tars to extract.
Unfortunately, currently the SDK uses different directory names compared
to what the NaCl toolchain directories normally use, so we also must map
and move the toolchain directories to the SDK ones.

TEST= trybots

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=265303

[David Yen, 2014-04-11 23:13:05]

As discussed, here are the changes necessary for build_sdk to use
package_version.py once the DEPS roll occurs from the NaCl side to the chromium
side.

As Derek suggested, I will also use this CL to test the changes before we
actually do the DEPS roll just to keep track of the testing to be sure
everything works.

[David Yen, 2014-04-11 23:33:25]

I've added Brad as a reviewer for the changes involved in replacing
download_toolchains.py.

Noel should review the build_sdk.py file.

[noelallen1, 2014-04-12 22:52:15]

one nit, otherwise build_sdk.py LGTM

test= is really for manual testing instructions, it's assumed you will use the
trybots.  In this case:

test=on linux: build_sdk.py --enable-bionic

https://codereview.chromium.org/235603004/diff/20001/native_client_sdk/src/bu...
File native_client_sdk/src/build_tools/build_sdk.py (right):

https://codereview.chromium.org/235603004/diff/20001/native_client_sdk/src/bu...
native_client_sdk/src/build_tools/build_sdk.py:57: CYGTAR =
os.path.join(BUILD_DIR, 'cygtar.py')
Is CYGTAR still used?

[David Yen, 2014-04-14 16:09:59]

https://codereview.chromium.org/235603004/diff/20001/native_client_sdk/src/bu...
File native_client_sdk/src/build_tools/build_sdk.py (right):

https://codereview.chromium.org/235603004/diff/20001/native_client_sdk/src/bu...
native_client_sdk/src/build_tools/build_sdk.py:57: CYGTAR =
os.path.join(BUILD_DIR, 'cygtar.py')
On 2014/04/12 22:52:16, noelallen1 wrote:
> Is CYGTAR still used?

It looks like it is used for tarring up the pepper bundle and the NaCl ports
bundle.

[noelallen1, 2014-04-14 19:42:20]

Still LGTM, I went ahead and reviewed the other files.

https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
File build/download_nacl_toolchains.py (right):

https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
build/download_nacl_toolchains.py:30: # TODO (robertm): Finish getting PNaCl
ready for prime time.
Can this TODO go away now?

[David Yen, 2014-04-14 20:20:55]

https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
File build/download_nacl_toolchains.py (right):

https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
build/download_nacl_toolchains.py:30: # TODO (robertm): Finish getting PNaCl
ready for prime time.
On 2014/04/14 19:42:21, noelallen1 wrote:
> Can this TODO go away now?

Done.

[David Yen, 2014-04-15 22:34:27]

On 2014/04/14 20:20:55, David Yen wrote:
>
https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
> File build/download_nacl_toolchains.py (right):
> 
>
https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
> build/download_nacl_toolchains.py:30: # TODO (robertm): Finish getting PNaCl
> ready for prime time.
> On 2014/04/14 19:42:21, noelallen1 wrote:
> > Can this TODO go away now?
> 
> Done.

I've added a deps roll to 13034 in this CL since the NaCl -> Chromium DEPS roll
relies on this CL.

[Mark Seaborn, 2014-04-18 20:09:23]

Are you going to try committing this, then?  Does anything block updating
nacl_revision in DEPS at the moment?

[David Yen, 2014-04-18 20:15:04]

On 2014/04/18 20:09:23, Mark Seaborn wrote:
> Are you going to try committing this, then?  Does anything block updating
> nacl_revision in DEPS at the moment?

None of the windows builds work with the latest PNaCl toolchain. Derek thinks it
could be the new mingw version so we are currently testing if the cygwin version
will work.

[David Yen, 2014-04-19 04:48:03]

The CQ bit was checked by dyen@chromium.org

[commit-bot: I haz the power, 2014-04-19 04:48:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dyen@chromium.org/235603004/320001

[commit-bot: I haz the power, 2014-04-19 06:20:56]

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

[commit-bot: I haz the power, 2014-04-19 06:20:56]

Retried try job too often on chromium_presubmit for step(s) presubmit
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

[David Yen, 2014-04-21 15:03:35]

On 2014/04/19 06:20:56, I haz the power (commit-bot) wrote:
> Retried try job too often on chromium_presubmit for step(s) presubmit
>
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=chromium_p...

It looks like I need a LGTM from Brad, PTAL.

[David Yen, 2014-04-21 16:51:21]

Brad is out sick today so I need someone else to look over my changes. They are
pretty small, just a change from using download_toolchains to package_version.

Jan, can you look over pnacl_support_extension.gyp?
Mark, can you look over buildbot_chrome_nacl_stage.py?

[jvoung (off chromium), 2014-04-21 18:08:52]

pnacl_support_extension LGTM

https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
File build/download_nacl_toolchains.py (right):

https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
build/download_nacl_toolchains.py:30: # TODO (robertm): Finish getting PNaCl
ready for prime time.
On 2014/04/14 20:20:56, David Yen wrote:
> On 2014/04/14 19:42:21, noelallen1 wrote:
> > Can this TODO go away now?
> 
> Done.

Does anything actually use --optional-pnacl anymore?

The DEPS hook used to:
https://chromiumcodereview.appspot.com/10914169/diff/9008/DEPS

Maybe the code checking for --optional-pnacl can be removed later too.

[David Yen, 2014-04-21 18:18:56]

On 2014/04/21 18:08:52, jvoung (cr) wrote:
> pnacl_support_extension LGTM
> 
>
https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
> File build/download_nacl_toolchains.py (right):
> 
>
https://codereview.chromium.org/235603004/diff/20001/build/download_nacl_tool...
> build/download_nacl_toolchains.py:30: # TODO (robertm): Finish getting PNaCl
> ready for prime time.
> On 2014/04/14 20:20:56, David Yen wrote:
> > On 2014/04/14 19:42:21, noelallen1 wrote:
> > > Can this TODO go away now?
> > 
> > Done.
> 
> Does anything actually use --optional-pnacl anymore?
> 
> The DEPS hook used to:
> https://chromiumcodereview.appspot.com/10914169/diff/9008/DEPS
> 
> Maybe the code checking for --optional-pnacl can be removed later too.

You are probably right that it is no longer used. I'll remove it later when I
remove extraction of tars during run hooks.

[Mark Seaborn, 2014-04-22 15:50:52]

On 2014/04/21 16:51:21, David Yen wrote:
> Mark, can you look over buildbot_chrome_nacl_stage.py?

LGTM for that.

[David Yen, 2014-04-22 16:05:35]

The CQ bit was checked by dyen@chromium.org

[commit-bot: I haz the power, 2014-04-22 16:06:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dyen@chromium.org/235603004/340001

[commit-bot: I haz the power, 2014-04-22 18:28:09]

Change committed as 265303

[Vadim Sh., 2014-04-23 01:07:03]

On 2014/04/22 18:28:09, I haz the power (commit-bot) wrote:
> Change committed as 265303

browser_tests on Linux are failing with "Plugin did not load. 'NaCl module load
failed: could not initialize module.":
http://build.chromium.org/p/chromium.linux/builders/Linux%20Tests%20%28dbg%29...

Not sure it's related to that CL though...

[hidehiko, 2014-04-23 07:04:11]

A revert of this CL has been created in
https://codereview.chromium.org/247143007/ by hidehiko@chromium.org.

The reason for reverting is: This update breaks PPAPI*NonSfi tests on
linux-32bits bot. See also crbug.com/365817.