Don't merge the mappings if their exec bit are different.

Don't merge the mappings if their exec bit are different.

BUG=585534
R=primiano@chromium.org

Committed: https://chromium.googlesource.com/breakpad/breakpad/+/8915f7be39448d9257b6da3ad0233944d1d9a92a

[michaelbai, 2016-02-29 23:01:03]

Description was changed from

==========
Don't merge the mappings if their exec bit are different.

BUG=585534
==========

to

==========
Don't merge the mappings if their exec bit are different.

BUG=585534
==========

[michaelbai, 2016-02-29 23:01:03]

michaelbai@chromium.org changed reviewers:
+ primiano@chromium.org, rmcilroy@chromium.org

[michaelbai, 2016-02-29 23:01:37]

PTAL

[Primiano Tucci (use gerrit), 2016-02-29 23:39:46]

primiano@chromium.org changed reviewers:
+ ted.mielczarek@gmail.com

[Primiano Tucci (use gerrit), 2016-02-29 23:39:46]

+ted.mielczarek who introduced the original code.

I wonder if this breaks somehow the crazylinker + relro packing. Did you test
it?

[michaelbai, 2016-03-01 04:25:24]

On 2016/02/29 23:39:46, Primiano (throttled til Mar 4) wrote:
> +ted.mielczarek who introduced the original code.
> 
> I wonder if this breaks somehow the crazylinker + relro packing. Did you test
> it?

I don't know how to verify it, but from what I tested, the mapping size of
ChromeModern's libchrome.so and RELRO:libchrome.so weren't changed. Does some
part of library mapping not have exec bit for the case of crazylinker + relro?

[Primiano Tucci (use gerrit), 2016-03-01 17:26:07]

On 2016/03/01 04:25:24, michaelbai wrote:
> On 2016/02/29 23:39:46, Primiano (throttled til Mar 4) wrote:
> > +ted.mielczarek who introduced the original code.
> > 
> > I wonder if this breaks somehow the crazylinker + relro packing. Did you
test
> > it?
> 
> I don't know how to verify it, but from what I tested, the mapping size of
> ChromeModern's libchrome.so and RELRO:libchrome.so weren't changed. Does some
> part of library mapping not have exec bit for the case of crazylinker + relro?

Build chrome_apk in official build mode, make it crash
(chrome://inducebrowsercrashforrealz) and see if the stack produced by
microdump_stackwalk is sane.

[Primiano Tucci (use gerrit), 2016-03-03 18:26:03]

Had a chat offline and Michael suggests he has tested this and the crash reports
w/ crazylinker are still sane.

LGTM

[Primiano Tucci (use gerrit), 2016-03-03 18:27:25]

On 2016/03/03 18:26:03, Primiano (throttled til Mar 4) wrote:
> Had a chat offline and Michael suggests he has tested this and the crash
reports
> w/ crazylinker are still sane.
> 
> LGTM

Ah please make sure that the breakpad tests (make check) are still green before
landing.

[michaelbai, 2016-03-07 19:04:08]

On 2016/03/03 18:27:25, Primiano (throttled til Mar 4) wrote:
> On 2016/03/03 18:26:03, Primiano (throttled til Mar 4) wrote:
> > Had a chat offline and Michael suggests he has tested this and the crash
> reports
> > w/ crazylinker are still sane.
> > 
> > LGTM
> 
> Ah please make sure that the breakpad tests (make check) are still green
before
> landing.

All test past

[michaelbai, 2016-03-07 19:05:22]

On 2016/03/07 19:04:08, michaelbai wrote:
> On 2016/03/03 18:27:25, Primiano (throttled til Mar 4) wrote:
> > On 2016/03/03 18:26:03, Primiano (throttled til Mar 4) wrote:
> > > Had a chat offline and Michael suggests he has tested this and the crash
> > reports
> > > w/ crazylinker are still sane.
> > > 
> > > LGTM
> > 
> > Ah please make sure that the breakpad tests (make check) are still green
> before
> > landing.
> 
> All test past

I meant all test passed.

[michaelbai, 2016-03-07 19:17:33]

On 2016/03/07 19:05:22, michaelbai wrote:
> On 2016/03/07 19:04:08, michaelbai wrote:
> > On 2016/03/03 18:27:25, Primiano (throttled til Mar 4) wrote:
> > > On 2016/03/03 18:26:03, Primiano (throttled til Mar 4) wrote:
> > > > Had a chat offline and Michael suggests he has tested this and the crash
> > > reports
> > > > w/ crazylinker are still sane.
> > > > 
> > > > LGTM
> > > 
> > > Ah please make sure that the breakpad tests (make check) are still green
> > before
> > > landing.
> > 
> > All test past
> 
> I meant all test passed.

I can't 'git cl land', could you help to land it?

[Primiano Tucci (use gerrit), 2016-03-08 08:46:04]

Description was changed from

==========
Don't merge the mappings if their exec bit are different.

BUG=585534
==========

to

==========
Don't merge the mappings if their exec bit are different.

BUG=585534
R=primiano@chromium.org

Committed:
https://chromium.googlesource.com/breakpad/breakpad/+/8915f7be39448d9257b6da3...
==========

[Primiano Tucci (use gerrit), 2016-03-08 08:46:06]

Committed patchset #1 (id:1) manually as
8915f7be39448d9257b6da3ad0233944d1d9a92a (presubmit successful).

[Primiano Tucci (use gerrit), 2016-03-08 08:46:29]

On 2016/03/08 08:46:06, Primiano (throttled til Mar 4) wrote:
> Committed patchset #1 (id:1) manually as
> 8915f7be39448d9257b6da3ad0233944d1d9a92a (presubmit successful).

Done. Remember to roll breakpad in chromium for this to have effect.