Use error-checking mutexes in Debug builds.

Use error-checking mutexes in Debug builds.

Upside:   our deadlock detection is now thread-safe.
Downside: static mutexes on Macs won't have deadlock detection until we drop 10.6.

BUG=skia:3294
Committed: https://skia.googlesource.com/skia/+/8b50ac242bc3a41cf81d5d9bbc24d49b9f85d834

[mtklein, 2015-01-09 13:59:30]

The CQ bit was checked by mtklein@google.com

[commit-bot: I haz the power, 2015-01-09 14:00:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/816833005/80001

[commit-bot: I haz the power, 2015-01-09 14:00:22]

Note for Reviewers:
The CQ is waiting for an approval.
If you believe that the CL is not ready yet, or if you would like to L-G-T-M
with comments then please uncheck the CQ checkbox.

Waiting for LGTM from valid reviewer(s) till 2015-01-09 20:00 UTC

[mtklein_C, 2015-01-09 14:12:00]

mtklein@chromium.org changed reviewers:
+ jyasskin@chromium.org
- mtklein@google.com

[Jeffrey Yasskin, 2015-01-09 16:31:55]

lgtm

https://codereview.chromium.org/816833005/diff/80001/src/ports/SkMutex_pthread.h
File src/ports/SkMutex_pthread.h (right):

https://codereview.chromium.org/816833005/diff/80001/src/ports/SkMutex_pthrea...
src/ports/SkMutex_pthread.h:33: SkASSERT(0 == rc);
It would be nice for debugging if there were some way to print 'rc' when this
fails, but without a DCHECK_EQ() macro in Skia that takes significantly more
code, so I don't insist.

https://codereview.chromium.org/816833005/diff/80001/src/ports/SkMutex_pthrea...
src/ports/SkMutex_pthread.h:77: #if defined(SK_DEBUG) &&
defined(PTHREAD_ERRORCHECK_MUTEX_INITIALIZER)
Very nice. I had no idea PTHREAD_ERRORCHECK_MUTEX_INITIALIZER existed. :)

[mtklein, 2015-01-09 16:39:24]

mtklein@google.com changed reviewers:
+ mtklein@google.com

[mtklein, 2015-01-09 16:39:25]

https://codereview.chromium.org/816833005/diff/80001/src/ports/SkMutex_pthread.h
File src/ports/SkMutex_pthread.h (right):

https://codereview.chromium.org/816833005/diff/80001/src/ports/SkMutex_pthrea...
src/ports/SkMutex_pthread.h:33: SkASSERT(0 == rc);
On 2015/01/09 16:31:55, Jeffrey Yasskin wrote:
> It would be nice for debugging if there were some way to print 'rc' when this
> fails, but without a DCHECK_EQ() macro in Skia that takes significantly more
> code, so I don't insist.

Agreed, though in practice these rarely trigger.  I'll certainly add a little
macro to do it once I see one trigger.  Incidentally, we've also been thinking
about just depending on or mimicking glog...

https://codereview.chromium.org/816833005/diff/80001/src/ports/SkMutex_pthrea...
src/ports/SkMutex_pthread.h:77: #if defined(SK_DEBUG) &&
defined(PTHREAD_ERRORCHECK_MUTEX_INITIALIZER)
On 2015/01/09 16:31:55, Jeffrey Yasskin wrote:
> Very nice. I had no idea PTHREAD_ERRORCHECK_MUTEX_INITIALIZER existed. :)

Yeah, me neither.  I accidentally stumbled upon it when Spotlight pulled up
Bionic's pthread.h as its top hit today.

[mtklein, 2015-01-09 17:00:13]

The CQ bit was unchecked by mtklein@google.com

[mtklein, 2015-01-09 17:00:19]

The CQ bit was checked by mtklein@google.com

[commit-bot: I haz the power, 2015-01-09 17:00:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/816833005/80001

[commit-bot: I haz the power, 2015-01-09 17:00:22]

Note for Reviewers:
The CQ is waiting for an approval.
If you believe that the CL is not ready yet, or if you would like to L-G-T-M
with comments then please uncheck the CQ checkbox.

Waiting for LGTM from valid reviewer(s) till 2015-01-09 23:00 UTC

[mtklein, 2015-01-09 17:01:01]

lgtm

[borenet, 2015-01-09 17:02:24]

LGTM?

[commit-bot: I haz the power, 2015-01-09 17:02:47]

Committed patchset #5 (id:80001) as
https://skia.googlesource.com/skia/+/8b50ac242bc3a41cf81d5d9bbc24d49b9f85d834