Include close errnos in crash dumps.

Include close errnos in crash dumps.

We may be crashing here because close is failing with EBADF (signalling other
code having problems with fds) or because networked filesystems are buggy and
failing with some other errno. Alias a copy of errno on the stack so we may
distinguish the two.

BUG=603354
Committed: https://crrev.com/3b6205579e92ceb47a07040a1c79d87923de7512
Cr-Commit-Position: refs/heads/master@{#401427}

[davidben, 2016-06-22 19:34:32]

davidben@chromium.org changed reviewers:
+ rsesek@chromium.org

[davidben, 2016-06-22 19:34:33]

rsesek: What do you think? Is this worthwhile? It seems we don't know yet
whether this is due to EBADF (we should go hunting for bad closes) or some weird
thing where network filesystems cause close to fail (maybe we should only crash
on EBADF and ignore the others???).

[Robert Sesek, 2016-06-22 19:53:01]

Yeah, this seems reasonable.

https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc
File base/files/scoped_file.cc (right):

https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc#n...
base/files/scoped_file.cc:40: PCHECK(0 == ret);
While you're here, PCHECK_EQ would be better.

[davidben, 2016-06-22 19:58:28]

https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc
File base/files/scoped_file.cc (right):

https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc#n...
base/files/scoped_file.cc:40: PCHECK(0 == ret);
On 2016/06/22 19:53:01, Robert Sesek wrote:
> While you're here, PCHECK_EQ would be better.

Done.

[Robert Sesek, 2016-06-22 20:21:58]

LGTM

On 2016/06/22 19:58:28, davidben wrote:
> https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc
> File base/files/scoped_file.cc (right):
> 
>
https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc#n...
> base/files/scoped_file.cc:40: PCHECK(0 == ret);
> On 2016/06/22 19:53:01, Robert Sesek wrote:
> > While you're here, PCHECK_EQ would be better.
> 
> Done.

It looks like I invented PCHECK_EQ in my mind... sorry about that :(.

[davidben, 2016-06-22 20:39:04]

On 2016/06/22 20:21:58, Robert Sesek wrote:
> LGTM
> 
> On 2016/06/22 19:58:28, davidben wrote:
> > https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc
> > File base/files/scoped_file.cc (right):
> > 
> >
>
https://codereview.chromium.org/2090703002/diff/1/base/files/scoped_file.cc#n...
> > base/files/scoped_file.cc:40: PCHECK(0 == ret);
> > On 2016/06/22 19:53:01, Robert Sesek wrote:
> > > While you're here, PCHECK_EQ would be better.
> > 
> > Done.
> 
> It looks like I invented PCHECK_EQ in my mind... sorry about that :(.

Haha. My bad for not checking. :-)

[davidben, 2016-06-22 20:39:22]

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

[davidben, 2016-06-22 20:39:22]

+thakis for OWNERS.

[Nico, 2016-06-22 20:44:43]

lgtm

[davidben, 2016-06-22 21:46:08]

The CQ bit was checked by davidben@chromium.org

[davidben, 2016-06-22 21:46:09]

The patchset sent to the CQ was uploaded after l-g-t-m from rsesek@chromium.org
Link to the patchset: https://codereview.chromium.org/2090703002/#ps40001
(title: "no such thing as PCHECK_EQ")

[commit-bot: I haz the power, 2016-06-22 21:47:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/2090703002/40001

[commit-bot: I haz the power, 2016-06-22 22:07:16]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-06-22 22:09:52]

Description was changed from

==========
Include close errnos in crash dumps.

We may be crashing here because close is failing with EBADF (signalling other
code having problems with fds) or because networked filesystems are buggy and
failing with some other errno. Alias a copy of errno on the stack so we may
distinguish the two.

BUG=603354
==========

to

==========
Include close errnos in crash dumps.

We may be crashing here because close is failing with EBADF (signalling other
code having problems with fds) or because networked filesystems are buggy and
failing with some other errno. Alias a copy of errno on the stack so we may
distinguish the two.

BUG=603354

Committed: https://crrev.com/3b6205579e92ceb47a07040a1c79d87923de7512
Cr-Commit-Position: refs/heads/master@{#401427}
==========

[commit-bot: I haz the power, 2016-06-22 22:09:53]

Patchset 3 (id:??) landed as
https://crrev.com/3b6205579e92ceb47a07040a1c79d87923de7512
Cr-Commit-Position: refs/heads/master@{#401427}