Improve GLSL integer support

Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1669853002

Committed: https://skia.googlesource.com/skia/+/3a2caf8ecf38124f4ad21a0f6c4dabfcfa17911a

Committed: https://skia.googlesource.com/skia/+/793dc26ca6ba2543bf50e5215b858d1c265af50c

[Chris Dalton, 2016-02-04 20:26:42]

Description was changed from

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
==========

to

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[Chris Dalton, 2016-02-04 20:30:46]

cdalton@nvidia.com changed reviewers:
+ bsalomon@google.com, egdaniel@google.com, ethannicholas@google.com

[Chris Dalton, 2016-02-04 20:30:46]

This is actually a framework change for the instanced stuff as well as a bugfix.

Skia was calling glVertexAttribPointer for integer attribs which isn't
guaranteed to work. This fixes it to use glVertexAttribIPointer instead.

[bsalomon, 2016-02-05 22:19:55]

lgtm

[Chris Dalton, 2016-02-05 22:21:09]

The CQ bit was checked by cdalton@nvidia.com

[commit-bot: I haz the power, 2016-02-05 22:21:22]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1669853002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1669853002/1

[commit-bot: I haz the power, 2016-02-06 00:21:30]

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

[commit-bot: I haz the power, 2016-02-06 00:21:30]

Try jobs failed on following builders:
  Test-Ubuntu-GCC-ShuttleA-GPU-GTX660-x86_64-Release-Trybot on client.skia
(JOB_TIMED_OUT, no build URL)

[Chris Dalton, 2016-02-06 00:22:04]

The CQ bit was checked by cdalton@nvidia.com

[commit-bot: I haz the power, 2016-02-06 00:22:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1669853002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1669853002/1

[commit-bot: I haz the power, 2016-02-06 00:26:37]

Description was changed from

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/3a2caf8ecf38124f4ad21a0f6c4dabfcfa17911a
==========

[commit-bot: I haz the power, 2016-02-06 00:26:38]

Committed patchset #1 (id:1) as
https://skia.googlesource.com/skia/+/3a2caf8ecf38124f4ad21a0f6c4dabfcfa17911a

[dcheng, 2016-02-06 16:41:04]

dcheng@chromium.org changed reviewers:
+ dcheng@chromium.org

[dcheng, 2016-02-06 16:41:04]

This patch is causing errors on the MSAN bots. Example failure:

EnlargedTextureWithAlphaThresholdFilter.GL (run #4):
[ RUN      ] EnlargedTextureWithAlphaThresholdFilter.GL
==28443==WARNING: MemorySanitizer: use-of-uninitialized-value
    #0 0x4568541 in set third_party/skia/src/gpu/gl/GrGLVertexArray.cpp:55:34
    #1 0x4520ade in setupGeometry
third_party/skia/src/gpu/gl/GrGLGpu.cpp:1759:13
    #2 0x452d50d in onDraw third_party/skia/src/gpu/gl/GrGLGpu.cpp:2490:5
    #3 0x4355dde in draw third_party/skia/src/gpu/GrGpu.cpp:383:9
    #4 0x4464172 in onDraw
third_party/skia/src/gpu/batches/GrVertexBatch.cpp:83:13
    #5 0x433bf61 in draw third_party/skia/src/gpu/batches/GrBatch.h:108:43
    #6 0x433bf61 in drawBatches third_party/skia/src/gpu/GrDrawTarget.cpp:214:0
    #7 0x4336837 in flush third_party/skia/src/gpu/GrDrawingManager.cpp:93:9
    #8 0x431c33f in flush third_party/skia/src/gpu/GrContext.cpp:209:9
    #9 0x349bb18 in EndGpuRaster cc/raster/scoped_gpu_raster.cc:38:3
    #10 0x349bb18 in ~ScopedGpuRaster cc/raster/scoped_gpu_raster.cc:21:0
    #11 0x33c7f6d in operator()
buildtools/third_party/libc++/trunk/include/memory:2529:13
    #12 0x33c7f6d in reset base/memory/scoped_ptr.h:177:0
    #13 0x33c7f6d in reset base/memory/scoped_ptr.h:350:0
    #14 0x33c7f6d in operator= base/memory/scoped_ptr.h:344:0
    #15 0x33c7f6d in ~ScopedUseGrContext cc/output/gl_renderer.cc:182:0
    #16 0x338a956 in operator()
buildtools/third_party/libc++/trunk/include/memory:2529:13
    #17 0x338a956 in reset base/memory/scoped_ptr.h:177:0
    #18 0x338a956 in ~scoped_ptr_impl base/memory/scoped_ptr.h:166:0
    #19 0x338a956 in ~scoped_ptr base/memory/scoped_ptr.h:240:0
    #20 0x338a956 in DrawRenderPassQuad cc/output/gl_renderer.cc:998:0
    #21 0x3363eaf in DrawRenderPass cc/output/direct_renderer.cc:516:5
    #22 0x3360e74 in DrawFrame cc/output/direct_renderer.cc:273:7
    #23 0x366bca3 in DrawLayers cc/trees/layer_tree_host_impl.cc:1703:5
    #24 0x398a37c in ?? cc/test/layer_tree_test.cc:247:5
    #25 0x37a7746 in DrawAndSwapInternal cc/trees/proxy_impl.cc:644:5
    #26 0x37a6f2d in ScheduledActionDrawAndSwapIfPossible
cc/trees/proxy_impl.cc:526:10
    #27 0x39a69ee in ScheduledActionDrawAndSwapIfPossible
cc/test/proxy_impl_for_test.cc:47:23
    #28 0x39a69ee in ?? cc/test/proxy_impl_for_test.cc:46:0
    #29 0x352a3d0 in DrawAndSwapIfPossible cc/scheduler/scheduler.cc:668:23
    #30 0x352a3d0 in ProcessScheduledActions cc/scheduler/scheduler.cc:741:0
    #31 0x35276c6 in OnBeginImplFrameDeadline cc/scheduler/scheduler.cc:659:3
    #32 0x3b551f2 in Run base/callback.h:394:12
    #33 0x3b551f2 in RunTask base/debug/task_annotator.cc:51:0
    #34 0x3a393f8 in RunTask base/message_loop/message_loop.cc:486:3
    #35 0x3a3b75c in DeferOrRunPendingTask
base/message_loop/message_loop.cc:495:5
    #36 0x3a3b75c in DoWork base/message_loop/message_loop.cc:607:0
    #37 0x3a479a3 in Run base/message_loop/message_pump_default.cc:33:21
    #38 0x3a71c34 in Run base/run_loop.cc:56:3
    #39 0x3a37894 in ?? base/message_loop/message_loop.cc:293:3
    #40 0x3ac0b3c in ThreadMain base/threading/thread.cc:252:3
    #41 0x3ab4b12 in ThreadFunc base/threading/platform_thread_posix.cc:68:3
    #42 0x7f4682749e99 in start_thread
/build/buildd/eglibc-2.15/nptl/pthread_create.c:308:0
    #43 0x7f468111c38c in ??
/build/buildd/eglibc-2.15/misc/../sysdeps/unix/sysv/linux/x86_64/clone.S:112:0

  Uninitialized value was created by a heap allocation
    #0 0x6a5142 in operator new(unsigned long) ??:0
    #1 0x45025cc in Create third_party/skia/src/gpu/gl/GrGLGpu.cpp:178:16
    #2 0x4359394 in ?? third_party/skia/src/gpu/GrGpuFactory.cpp:36:12
    #3 0x431abfb in init third_party/skia/src/gpu/GrContext.cpp:73:12
    #4 0x431abfb in Create third_party/skia/src/gpu/GrContext.cpp:42:0
    #5 0x431a8fe in Create third_party/skia/src/gpu/GrContext.cpp:35:12
    #6 0x39c9b00 in GrContext cc/test/test_in_process_context_provider.cc:125:32
    #7 0x338a631 in Create cc/output/gl_renderer.cc:175:9
    #8 0x338a631 in DrawRenderPassQuad cc/output/gl_renderer.cc:998:0
    #9 0x3363eaf in DrawRenderPass cc/output/direct_renderer.cc:516:5
    #10 0x3360e74 in DrawFrame cc/output/direct_renderer.cc:273:7
    #11 0x366bca3 in DrawLayers cc/trees/layer_tree_host_impl.cc:1703:5
    #12 0x398a37c in ?? cc/test/layer_tree_test.cc:247:5
    #13 0x37a7746 in DrawAndSwapInternal cc/trees/proxy_impl.cc:644:5
    #14 0x37a6f2d in ScheduledActionDrawAndSwapIfPossible
cc/trees/proxy_impl.cc:526:10
    #15 0x39a69ee in ScheduledActionDrawAndSwapIfPossible
cc/test/proxy_impl_for_test.cc:47:23
    #16 0x39a69ee in ?? cc/test/proxy_impl_for_test.cc:46:0
    #17 0x352a3d0 in DrawAndSwapIfPossible cc/scheduler/scheduler.cc:668:23
    #18 0x352a3d0 in ProcessScheduledActions cc/scheduler/scheduler.cc:741:0
    #19 0x35276c6 in OnBeginImplFrameDeadline cc/scheduler/scheduler.cc:659:3
    #20 0x3b551f2 in Run base/callback.h:394:12
    #21 0x3b551f2 in RunTask base/debug/task_annotator.cc:51:0
    #22 0x3a393f8 in RunTask base/message_loop/message_loop.cc:486:3
    #23 0x3a3b75c in DeferOrRunPendingTask
base/message_loop/message_loop.cc:495:5
    #24 0x3a3b75c in DoWork base/message_loop/message_loop.cc:607:0
    #25 0x3a479a3 in Run base/message_loop/message_pump_default.cc:33:21

(from
https://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20ChromeOS%20...)

[dcheng, 2016-02-06 16:41:19]

A revert of this CL (patchset #1 id:1) has been created in
https://codereview.chromium.org/1674813004/ by dcheng@chromium.org.

The reason for reverting is: MSAN bots are unhappy with this change:
https://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20ChromeOS%20....

[Chris Dalton, 2016-02-07 22:56:03]

Description was changed from

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/3a2caf8ecf38124f4ad21a0f6c4dabfcfa17911a
==========

to

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/3a2caf8ecf38124f4ad21a0f6c4dabfcfa17911a
==========

[Chris Dalton, 2016-02-07 22:57:37]

Ugh. I deleted an extra line. Glad the bots caught it.

Can I rerun the MSAN bots from here?

[dcheng, 2016-02-07 23:08:42]

On 2016/02/07 at 22:57:37, cdalton wrote:
> Ugh. I deleted an extra line. Glad the bots caught it.
> 
> Can I rerun the MSAN bots from here?

I'm not sure if there's an easy way, sorry =/

It might be easiest to just run cc_unittests locally with a MSAN build.

[Chris Dalton, 2016-02-08 00:54:01]

On 2016/02/07 23:08:42, dcheng wrote:
> On 2016/02/07 at 22:57:37, cdalton wrote:
> > Ugh. I deleted an extra line. Glad the bots caught it.
> > 
> > Can I rerun the MSAN bots from here?
> 
> I'm not sure if there's an easy way, sorry =/
> 
> It might be easiest to just run cc_unittests locally with a MSAN build.

Ok, well I reproed the same error with valgrind and verified patchset #2 makes
valgrind happy, so I'm confident the issue is resolved now.

[mtklein, 2016-02-08 01:01:32]

> Can I rerun the MSAN bots from here?

(I'm bringing up MSAN bots for Skia, but it'll be a while yet before we get
things working for GPU bots.  Don't suppose you've got MSAN-instrumented drivers
for us, eh?)

[bsalomon, 2016-02-08 02:41:35]

lgtm if you want to reland.

[Chris Dalton, 2016-02-08 16:46:59]

One question in a comment.

https://codereview.chromium.org/1669853002/diff/20001/src/gpu/gl/GrGLVertexAr...
File src/gpu/gl/GrGLVertexArray.h (left):

https://codereview.chromium.org/1669853002/diff/20001/src/gpu/gl/GrGLVertexAr...
src/gpu/gl/GrGLVertexArray.h:35: {4, GR_GL_INT, false},           //
kInt_GrVertexAttribType
One other thing I forgot to mention, is that kInt used to have a 4 for fCount
and I changed it to a 1 when I moved this block. I think PLS path rendering is
the only thing so far that uses ints. Is there a bot I can run with that to make
sure everything still works as expected?

[ethannicholas, 2016-02-08 17:15:49]

On 2016/02/08 16:46:59, Chris Dalton wrote:
> One question in a comment.
> 
>
https://codereview.chromium.org/1669853002/diff/20001/src/gpu/gl/GrGLVertexAr...
> File src/gpu/gl/GrGLVertexArray.h (left):
> 
>
https://codereview.chromium.org/1669853002/diff/20001/src/gpu/gl/GrGLVertexAr...
> src/gpu/gl/GrGLVertexArray.h:35: {4, GR_GL_INT, false},           //
> kInt_GrVertexAttribType
> One other thing I forgot to mention, is that kInt used to have a 4 for fCount
> and I changed it to a 1 when I moved this block. I think PLS path rendering is
> the only thing so far that uses ints. Is there a bot I can run with that to
make
> sure everything still works as expected?

1 is the correct value here. It won't affect PLS.

[Chris Dalton, 2016-02-08 17:21:02]

On 2016/02/08 17:15:49, ethannicholas wrote:
> On 2016/02/08 16:46:59, Chris Dalton wrote:
> > One question in a comment.
> > 
> >
>
https://codereview.chromium.org/1669853002/diff/20001/src/gpu/gl/GrGLVertexAr...
> > File src/gpu/gl/GrGLVertexArray.h (left):
> > 
> >
>
https://codereview.chromium.org/1669853002/diff/20001/src/gpu/gl/GrGLVertexAr...
> > src/gpu/gl/GrGLVertexArray.h:35: {4, GR_GL_INT, false},           //
> > kInt_GrVertexAttribType
> > One other thing I forgot to mention, is that kInt used to have a 4 for
fCount
> > and I changed it to a 1 when I moved this block. I think PLS path rendering
is
> > the only thing so far that uses ints. Is there a bot I can run with that to
> make
> > sure everything still works as expected?
> 
> 1 is the correct value here. It won't affect PLS.

Ok, thanks.

[Chris Dalton, 2016-02-08 17:21:09]

The CQ bit was checked by cdalton@nvidia.com

[commit-bot: I haz the power, 2016-02-08 17:21:20]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1669853002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1669853002/20001

[commit-bot: I haz the power, 2016-02-08 17:22:05]

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

[commit-bot: I haz the power, 2016-02-08 17:22:06]

Try jobs failed on following builders:
  Build-Mac-Clang-Arm7-Debug-iOS-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Mac-Clang-Arm7...)

[Chris Dalton, 2016-02-08 17:46:17]

The CQ bit was checked by cdalton@nvidia.com

[Chris Dalton, 2016-02-08 17:46:18]

The patchset sent to the CQ was uploaded after l-g-t-m from bsalomon@google.com
Link to the patchset: https://codereview.chromium.org/1669853002/#ps40001
(title: "rebase")

[commit-bot: I haz the power, 2016-02-08 17:46:21]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1669853002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1669853002/40001

[commit-bot: I haz the power, 2016-02-08 18:11:50]

Description was changed from

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/3a2caf8ecf38124f4ad21a0f6c4dabfcfa17911a
==========

to

==========
Improve GLSL integer support

 - Adds shader types for uint.
 - Adds a cap for integer support.
 - Uses glVertexAttribIPointer for integer attribs.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/3a2caf8ecf38124f4ad21a0f6c4dabfcfa17911a

Committed:
https://skia.googlesource.com/skia/+/793dc26ca6ba2543bf50e5215b858d1c265af50c
==========

[commit-bot: I haz the power, 2016-02-08 18:11:50]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/793dc26ca6ba2543bf50e5215b858d1c265af50c