Update skia to test with libpng to 1.6.10

Update skia to test with libpng to 1.6.10

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/f272bb03df9b86e7ea2cf23fb4d5cc56624e0118

[msarett, 2015-08-18 17:21:10]

msarett@google.com changed reviewers:
+ djsollen@google.com, scroggo@google.com

[djsollen, 2015-08-18 17:28:32]

lgtm with comment nit

[msarett, 2015-08-18 17:34:10]

Patchset #3 (id:40001) has been deleted

[msarett, 2015-08-18 17:34:31]

The CQ bit was checked by msarett@google.com

[msarett, 2015-08-18 17:34:31]

The patchset sent to the CQ was uploaded after l-g-t-m from djsollen@google.com
Link to the patchset: https://codereview.chromium.org/1301523005/#ps60001
(title: "Made comment clearer")

[commit-bot: I haz the power, 2015-08-18 17:34:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1301523005/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1301523005/60001

[commit-bot: I haz the power, 2015-08-18 17:46:03]

Committed patchset #3 (id:60001) as
https://skia.googlesource.com/skia/+/f272bb03df9b86e7ea2cf23fb4d5cc56624e0118

[msarett, 2015-08-18 18:36:20]

A revert of this CL (patchset #3 id:60001) has been created in
https://codereview.chromium.org/1298223002/ by msarett@google.com.

The reason for reverting is: DM is failing on gm tests on Android bots.  Cause
is not yet clear..

[scroggo, 2015-08-26 21:53:32]

> Update skia to test with libpng to 1.6.10

Nit: I think this description is a little misleading - this only updates the
version of libpng used by SkImageDecoder on Android, correct? And it's not truly
being updated to 1.6.10; it is being updated to Android's forked version of
1.6.10

On 2015/08/18 18:36:20, msarett wrote:
> A revert of this CL (patchset #3 id:60001) has been created in
> https://codereview.chromium.org/1298223002/ by mailto:msarett@google.com.
> 
> The reason for reverting is: DM is failing on gm tests on Android bots.  Cause
> is not yet clear..

Any update on this? It looks like from the trybots we're failing all (?) subset
decodes, but only on certain devices/builds?

https://codereview.chromium.org/1301523005/diff/80001/dm/DMSrcSink.cpp
File dm/DMSrcSink.cpp (right):

https://codereview.chromium.org/1301523005/diff/80001/dm/DMSrcSink.cpp#newcode84
dm/DMSrcSink.cpp:84: SkDebugf("\nCodec %s\n", fPath.c_str());
I'm guessing we do not want these print statements checked in?

https://codereview.chromium.org/1301523005/diff/80001/platform_tools/android/...
File platform_tools/android/gyp/dependencies.gypi (right):

https://codereview.chromium.org/1301523005/diff/80001/platform_tools/android/...
platform_tools/android/gyp/dependencies.gypi:113: #
skia/third_party/externals/libpng.
nit: Can you explain (briefly) why there is a conflict? (I am guessing this is
temporary, due to the two versions of libpng - the forked version used by
SkImageDecoder and the version used by SkCodec?)

[scroggo, 2015-08-27 20:21:51]

On 2015/08/26 21:53:32, scroggo wrote:
> > Update skia to test with libpng to 1.6.10
> 
> Nit: I think this description is a little misleading - this only updates the
> version of libpng used by SkImageDecoder on Android, correct? And it's not
truly
> being updated to 1.6.10; it is being updated to Android's forked version of
> 1.6.10
> 
> On 2015/08/18 18:36:20, msarett wrote:
> > A revert of this CL (patchset #3 id:60001) has been created in
> > https://codereview.chromium.org/1298223002/ by mailto:msarett@google.com.
> > 
> > The reason for reverting is: DM is failing on gm tests on Android bots. 
Cause
> > is not yet clear..
> 
> Any update on this? It looks like from the trybots we're failing all (?)
subset
> decodes, but only on certain devices/builds?
> 
> https://codereview.chromium.org/1301523005/diff/80001/dm/DMSrcSink.cpp
> File dm/DMSrcSink.cpp (right):
> 
>
https://codereview.chromium.org/1301523005/diff/80001/dm/DMSrcSink.cpp#newcode84
> dm/DMSrcSink.cpp:84: SkDebugf("\nCodec %s\n", fPath.c_str());
> I'm guessing we do not want these print statements checked in?
> 
>
https://codereview.chromium.org/1301523005/diff/80001/platform_tools/android/...
> File platform_tools/android/gyp/dependencies.gypi (right):
> 
>
https://codereview.chromium.org/1301523005/diff/80001/platform_tools/android/...
> platform_tools/android/gyp/dependencies.gypi:113: #
> skia/third_party/externals/libpng.
> nit: Can you explain (briefly) why there is a conflict? (I am guessing this is
> temporary, due to the two versions of libpng - the forked version used by
> SkImageDecoder and the version used by SkCodec?)

Some context: Matt wanted to update our tests to test the version of libpng used
by Android (see b/23265085). He figured out why it had to be reverted - due to
name mangling (and lack thereof), one version of libpng (the one used by
SkImageDecoder) was calling into the *other* version of libpng (the one used by
SkCodec, generally). Matt fixed this in a subsequent patch set (4, I think), but
it was still failing. It is failing because of previously unknown bugs in the
version of libpng used by Android.

Since we're actively working to remove the old version, we decided not to follow
through on this update. Closing this issue.