chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

gold has a bug where it ignores section alignment when merging functions.
This means that member functions and regular functions with the same
contents can be merged even if the regular function has an odd address.
In the Itanium ABI, member function pointers with an odd address denote
virtual member function points, so if a non-virtual member function
gets folded into an identical non-member function that happens to be
at an odd address, the generated code will conclude at runtime that
the call is meant to be virtual, leading to a crash.

This only happens with -Os, at -O2 and -O0 functions are always
aligned at 16-byte boundaries.  This only happens with clang because
we currently never pass --icf to gold in android builds with gcc.

The gold bug is fixed upstream and is in the progress of being merged
into the NDK's gold, but it's not there yet.  Until it is, disable
full ICF on android (where we use -Os).

BUG=663886
Committed: https://crrev.com/aaa006d4a8d979e96d2ac248306240a60590b0bc
Cr-Commit-Position: refs/heads/master@{#433797}

[Nico, 2016-11-21 21:46:24]

The CQ bit was checked by thakis@chromium.org to run a CQ dry run

[Nico, 2016-11-21 21:47:10]

thakis@chromium.org changed reviewers:
+ rnk@chromium.org

[Nico, 2016-11-21 21:47:10]

With this, the clang-built libchrome.so is 3MB larger than the gcc-built one on
x64 :-(

[commit-bot: I haz the power, 2016-11-21 21:47:22]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Nico, 2016-11-21 21:48:38]

The CQ bit was checked by thakis@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-11-21 21:49:10]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Nico, 2016-11-21 22:05:58]

Description was changed from

==========
chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

gold has a bug where it ignores section alignment when merging functions.
This means that member functions and regular functions with the same
contents can be merged even if the regular function has an odd address.
In the Itanium ABI, member function pointers with an odd address denote
virtual member function points, so if a non-virtual member function
gets folded into an identical non-member function that happens to be
at an odd address, the generated code will conclude at runtime that
the call is meant to be virtual, leading to a crash.

This only happens with -Os, at -O2 and -O0 functions are always
aligned at 16-byte boundaries.

The gold bug is fixed upstream and is in the progress of being merged
into the NDK's gold, but it's not there yet.  Until it is, disable
full ICF on android (where we use -Os).

BUG=663886
==========

to

==========
chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

gold has a bug where it ignores section alignment when merging functions.
This means that member functions and regular functions with the same
contents can be merged even if the regular function has an odd address.
In the Itanium ABI, member function pointers with an odd address denote
virtual member function points, so if a non-virtual member function
gets folded into an identical non-member function that happens to be
at an odd address, the generated code will conclude at runtime that
the call is meant to be virtual, leading to a crash.

This only happens with -Os, at -O2 and -O0 functions are always
aligned at 16-byte boundaries.  This only happens with clang because
we currently never pass --icf to gold in android builds with gcc.

The gold bug is fixed upstream and is in the progress of being merged
into the NDK's gold, but it's not there yet.  Until it is, disable
full ICF on android (where we use -Os).

BUG=663886
==========

[Reid Kleckner, 2016-11-21 22:07:02]

Bummer, lgtm.

Good to know Sri fixed the gold bug, though. He sits one cube away, I'll thank
him. :)

[Nico, 2016-11-21 22:15:42]

From my reading of the bug, Gergely Nagy fixed the bug and Sri merged the fix.
But yeah, I'll go say hello the next time I'm in MTV :-)

[Nico, 2016-11-21 22:15:46]

The CQ bit was unchecked by thakis@chromium.org

[Nico, 2016-11-21 22:15:52]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2016-11-21 22:16:12]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-22 00:04:23]

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

[commit-bot: I haz the power, 2016-11-22 00:04:23]

Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[Nico, 2016-11-22 04:41:27]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2016-11-22 04:41:47]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-22 05:55:25]

CQ is committing da patch.

Bot data: {"patchset_id": 20001, "attempt_start_ts": 1479789687958770,
"parent_rev": "1c0e9c09682403eb59439b9eab28ceae53219aa5", "commit_rev":
"e0ae7aa6e7a054ee274a4102f9df2d64dfb62b20"}

[commit-bot: I haz the power, 2016-11-22 05:55:54]

Description was changed from

==========
chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

gold has a bug where it ignores section alignment when merging functions.
This means that member functions and regular functions with the same
contents can be merged even if the regular function has an odd address.
In the Itanium ABI, member function pointers with an odd address denote
virtual member function points, so if a non-virtual member function
gets folded into an identical non-member function that happens to be
at an odd address, the generated code will conclude at runtime that
the call is meant to be virtual, leading to a crash.

This only happens with -Os, at -O2 and -O0 functions are always
aligned at 16-byte boundaries.  This only happens with clang because
we currently never pass --icf to gold in android builds with gcc.

The gold bug is fixed upstream and is in the progress of being merged
into the NDK's gold, but it's not there yet.  Until it is, disable
full ICF on android (where we use -Os).

BUG=663886
==========

to

==========
chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

gold has a bug where it ignores section alignment when merging functions.
This means that member functions and regular functions with the same
contents can be merged even if the regular function has an odd address.
In the Itanium ABI, member function pointers with an odd address denote
virtual member function points, so if a non-virtual member function
gets folded into an identical non-member function that happens to be
at an odd address, the generated code will conclude at runtime that
the call is meant to be virtual, leading to a crash.

This only happens with -Os, at -O2 and -O0 functions are always
aligned at 16-byte boundaries.  This only happens with clang because
we currently never pass --icf to gold in android builds with gcc.

The gold bug is fixed upstream and is in the progress of being merged
into the NDK's gold, but it's not there yet.  Until it is, disable
full ICF on android (where we use -Os).

BUG=663886
==========

[commit-bot: I haz the power, 2016-11-22 05:55:55]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-11-22 05:57:38]

Description was changed from

==========
chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

gold has a bug where it ignores section alignment when merging functions.
This means that member functions and regular functions with the same
contents can be merged even if the regular function has an odd address.
In the Itanium ABI, member function pointers with an odd address denote
virtual member function points, so if a non-virtual member function
gets folded into an identical non-member function that happens to be
at an odd address, the generated code will conclude at runtime that
the call is meant to be virtual, leading to a crash.

This only happens with -Os, at -O2 and -O0 functions are always
aligned at 16-byte boundaries.  This only happens with clang because
we currently never pass --icf to gold in android builds with gcc.

The gold bug is fixed upstream and is in the progress of being merged
into the NDK's gold, but it's not there yet.  Until it is, disable
full ICF on android (where we use -Os).

BUG=663886
==========

to

==========
chrome/android/intel/clang: Use --icf=safe instead of --icf=all.

gold has a bug where it ignores section alignment when merging functions.
This means that member functions and regular functions with the same
contents can be merged even if the regular function has an odd address.
In the Itanium ABI, member function pointers with an odd address denote
virtual member function points, so if a non-virtual member function
gets folded into an identical non-member function that happens to be
at an odd address, the generated code will conclude at runtime that
the call is meant to be virtual, leading to a crash.

This only happens with -Os, at -O2 and -O0 functions are always
aligned at 16-byte boundaries.  This only happens with clang because
we currently never pass --icf to gold in android builds with gcc.

The gold bug is fixed upstream and is in the progress of being merged
into the NDK's gold, but it's not there yet.  Until it is, disable
full ICF on android (where we use -Os).

BUG=663886

Committed: https://crrev.com/aaa006d4a8d979e96d2ac248306240a60590b0bc
Cr-Commit-Position: refs/heads/master@{#433797}
==========

[commit-bot: I haz the power, 2016-11-22 05:57:39]

Patchset 2 (id:??) landed as
https://crrev.com/aaa006d4a8d979e96d2ac248306240a60590b0bc
Cr-Commit-Position: refs/heads/master@{#433797}