Add a Subzero variant of Chrome PNaCl error handling test.

Add a Subzero variant of Chrome PNaCl error handling test.

Make sure that subzero can also handle various pexes with a
bad header. Test with a tiny fake .pexe (already had
that test case) and a larger realistic nonfinalized pexe.

The nonfinalized pexe requires streaming multiple chunks
and so is able to tickle an error-reporting race condition
(use-after-free) + possible deadlock. I tried it on subzero
from last week and was able to tickle the crash.
It should no longer tickle that bug this week unless
there's a regression.

The tiny version runs pretty quickly, but otherwise had
not been crashing before. It merely returned an imprecise
error message ("Some error occurred"), which should
now be fixed too.

BUG= https://code.google.com/p/nativeclient/issues/detail?id=4163

CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:linux_chromium_dbg_32_ng

Committed: https://crrev.com/89cda3776ecc975213ac4b626845deea417f977f
Cr-Commit-Position: refs/heads/master@{#333769}

[jvoung (off chromium), 2015-06-08 18:09:28]

jvoung@chromium.org changed reviewers:
+ dschuff@chromium.org, kschimpf@google.com, ncbray@chromium.org,
stichnot@chromium.org

[Nick Bray (chromium), 2015-06-09 17:25:00]

LGTM w/ one question.

https://codereview.chromium.org/1157213005/diff/20001/chrome/test/data/nacl/p...
File chrome/test/data/nacl/pnacl_error_handling/pnacl_nonfinal_pexe_O0.nmf
(right):

https://codereview.chromium.org/1157213005/diff/20001/chrome/test/data/nacl/p...
chrome/test/data/nacl/pnacl_error_handling/pnacl_nonfinal_pexe_O0.nmf:6: "url":
"pnacl_errors_newlib_pnacl.pexe.debug"
In what conditions is this file created?  Will it exist on release-mode
builders?

[jvoung (off chromium), 2015-06-09 18:03:11]

https://codereview.chromium.org/1157213005/diff/20001/chrome/test/data/nacl/p...
File chrome/test/data/nacl/pnacl_error_handling/pnacl_nonfinal_pexe_O0.nmf
(right):

https://codereview.chromium.org/1157213005/diff/20001/chrome/test/data/nacl/p...
chrome/test/data/nacl/pnacl_error_handling/pnacl_nonfinal_pexe_O0.nmf:6: "url":
"pnacl_errors_newlib_pnacl.pexe.debug"
On 2015/06/09 17:25:00, Nick Bray (chromium) wrote:
> In what conditions is this file created?  Will it exist on release-mode
> builders?

It's currently always created by build_nexe.py. However, just in case that's
optimized to leave out the intermediate file, I modified the .gyp to pass along
the "extra_args" for nonstable-pnacl which should make the ".pexe" not
finalized. Then we don't need to rely on the .debug file existing as a
side-effect of the build_nexe build process.

[jvoung (off chromium), 2015-06-09 21:25:50]

The CQ bit was checked by jvoung@chromium.org

[jvoung (off chromium), 2015-06-09 21:25:51]

The patchset sent to the CQ was uploaded after l-g-t-m from ncbray@chromium.org
Link to the patchset: https://codereview.chromium.org/1157213005/#ps40001
(title: "make more explicit")

[commit-bot: I haz the power, 2015-06-09 21:27:41]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1157213005/40001

[commit-bot: I haz the power, 2015-06-09 23:31:05]

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

[commit-bot: I haz the power, 2015-06-09 23:31:06]

Try jobs failed on following builders:
  android_chromium_gn_compile_dbg on tryserver.chromium.linux (JOB_TIMED_OUT, no
build URL)

[jvoung (off chromium), 2015-06-10 18:13:32]

The CQ bit was checked by jvoung@chromium.org

[commit-bot: I haz the power, 2015-06-10 18:13:54]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1157213005/40001

[commit-bot: I haz the power, 2015-06-10 18:59:18]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-06-10 19:00:24]

Patchset 3 (id:??) landed as
https://crrev.com/89cda3776ecc975213ac4b626845deea417f977f
Cr-Commit-Position: refs/heads/master@{#333769}