Enable mac_strip_release under ASan on OSX. Remove the .saves files.

Enable mac_strip_release under ASan on OSX. Remove the .saves files.

Unce upon a time the ASan runtime library used to be statically linked into
the executables on OSX.
Because that library provided a number of API functions needed by the shared
libraries, we had custom ASan-specific .saves files that told the `strip`
utility to preserve those functions when stripping the executables.

Then ASan switched to using dynamic runtime library, which instantly broke the
stripping step, because the executables weren't necessarily referencing all the
ASan API functions (issue 242503).
As a result, stripping has been disabled, and we haven't had .dSYM debug info
for ASan builds for more than a year now (issue 148383).

Because the ASan API functions are now undefined in the executables, it's
actually senseless to use the .saves files to preserve those functions.
This CL deletes the .saves files and turns mac_strip_release on for ASan
builds.

BUG=148383, 242503, 170739, 166857
R=mark@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=273199

[Alexander Potapenko, 2014-05-26 14:26:01]

Please take a look.
For some reason the trybot runs still do not have .dSYM files, e.g.
http://build.chromium.org/p/tryserver.chromium/builders/mac_asan_64/builds/71...
contains some unsymbolized reports (search for "AddressSanitizer"), perhaps
that's because the bot has 'fastbuild=1' in GYP_DEFINES (WTF?)

The .dSYM files are being produced locally and I've managed to get them after
SSHing to the trybot and re-running runhooks.

[Alexander Potapenko, 2014-05-26 16:25:47]

FTR https://codereview.chromium.org/299423004/ sets fastbuild=0 for ASan
trybots.

[Mark Mentovai, 2014-05-27 03:33:22]

I came across this again at the end of last week in
https://codereview.chromium.org/295103014/ . Sadly, we’re still hitting bug
162461.

https://codereview.chromium.org/292153006/diff/20001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/292153006/diff/20001/build/common.gypi#newcod...
build/common.gypi:2009: 'mac_strip_release': 1,
Current behavior for mac_strip_release seems to be to only turn it on for
buildtype=Official branding=Chrome builds (line 1716). Is there any reason to
force it on for asan too?

If you removed this condition altogether, does that make things better on the
bot?

[Alexander Potapenko, 2014-05-27 07:22:57]

On 2014/05/27 03:33:22, Mark Mentovai wrote:
> I came across this again at the end of last week in
> https://codereview.chromium.org/295103014/ . Sadly, we’re still hitting bug
> 162461.
We're using ASAN_OPTIONS=replace_intrin=0 to work around this on the bots
(http://dev.chromium.org/developers/testing/addresssanitizer)
The ultimate goal is to make the Mac builds use base/debug/sanitizer_options.cc.

> https://codereview.chromium.org/292153006/diff/20001/build/common.gypi
> File build/common.gypi (right):
> 
>
https://codereview.chromium.org/292153006/diff/20001/build/common.gypi#newcod...
> build/common.gypi:2009: 'mac_strip_release': 1,
> Current behavior for mac_strip_release seems to be to only turn it on for
> buildtype=Official branding=Chrome builds (line 1716). Is there any reason to
> force it on for asan too?
It's used to initialize mac_strip (line 4602), which IIUC is required to run
dsymutil.
I can change that part to unconditionally set mac_strip=1 under ASan.

> If you removed this condition altogether, does that make things better on the
> bot?
Don't think so: https://code.google.com/p/chromium/issues/detail?id=148383#c17
I didn't have .dSYMs before I started setting mac_strip_release.

[Mark Mentovai, 2014-05-27 12:51:00]

LGTM to get this moving. Any difference if you try to set it as
mac_strip_release% with a % at line 2009?

[Alexander Potapenko, 2014-05-27 13:04:27]

On 2014/05/27 12:51:00, Mark Mentovai wrote:
> LGTM to get this moving. Any difference if you try to set it as
> mac_strip_release% with a % at line 2009?

Are you asking about the difference on the trybots?
Right now the trybots are configured to use fastbuild=1, which sets
GCC_GENERATE_DEBUGGING_SYMBOLS=NO (line 2465)
GYP's Xcode emulation invokes dsymutil iff GCC_GENERATE_DEBUGGING_SYMBOLS=YES
(https://code.google.com/p/chromium/codesearch#chromium/src/tools/gyp/pylib/gy...),
so irrespective of any changes in this CL there won't be any .dSYM files on the
trybots right now.

[Mark Mentovai, 2014-05-27 13:06:34]

Oh, I thought you said the trybots were using fastbuild=0.

Still LGTM.

[Alexander Potapenko, 2014-05-27 13:29:44]

The CQ bit was checked by glider@chromium.org

[commit-bot: I haz the power, 2014-05-27 13:29:51]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/glider@chromium.org/292153006/20001

[commit-bot: I haz the power, 2014-05-27 15:21:48]

FYI, CQ is re-trying this CL (attempt #1).
Please consider checking whether the failures are real,
and report flakes to chrome-troopers@google.com.
The failing builders are:
  android_dbg_triggered_tests on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/android_dbg_triggered...)
  chromium_presubmit on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/chromium_presubmit/bu...)

[commit-bot: I haz the power, 2014-05-27 15:24:44]

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

[commit-bot: I haz the power, 2014-05-27 15:24:45]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/chromium_presubmit/bu...)

[Alexander Potapenko, 2014-05-27 15:33:37]

Nico, i can haz owner's lgtm?

[Nico, 2014-05-27 15:37:23]

rslgtm

https://codereview.chromium.org/292153006/diff/20001/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/292153006/diff/20001/build/common.gypi#newcod...
build/common.gypi:2009: 'mac_strip_release': 1,
(I to found this confusing; maybe add a comment explaining why it's here.)

[Alexander Potapenko, 2014-05-28 08:31:03]

The CQ bit was checked by glider@chromium.org

[commit-bot: I haz the power, 2014-05-28 08:31:15]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/glider@chromium.org/292153006/40001

[commit-bot: I haz the power, 2014-05-28 10:32:14]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  ios_rel_device_ninja on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/ios_rel_device_ninja/...)
  linux_chromium_chromeos_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/linux_chromium_chrome...)

[commit-bot: I haz the power, 2014-05-28 13:19:50]

Change committed as 273199