Add skeleton for kilobench experiments

Add skeleton for kilobench experiments

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1582903002

Committed: https://skia.googlesource.com/skia/+/99f340680959c07a11b8cd654e501f92b41646a3

[joshua.litt, 2016-01-13 16:57:37]

Description was changed from

==========
Add skeleton for kilobench experiments

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

to

==========
Add skeleton for kilobench experiments

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

[joshualitt, 2016-01-14 15:25:30]

joshualitt@google.com changed reviewers:
+ bsalomon@google.com, djsollen@google.com, joshualitt@google.com

[joshualitt, 2016-01-14 15:25:31]

This is the skeleton for kilobench.  Its pretty rough around the edges, and
currently juts grabs a bunch of skps and loops over them.  I'm not even setting
up a canvas yet, but that will be next.  This should give us something to
iterate on though.

[joshualitt, 2016-01-14 15:39:27]

Description was changed from

==========
Add skeleton for kilobench experiments

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

to

==========
Add skeleton for kilobench experiments

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

[joshualitt, 2016-01-14 15:39:27]

joshualitt@google.com changed reviewers:
+ robertphillips@google.com

[joshualitt, 2016-01-14 15:40:13]

those who ask about kilobench will be sucked into reviewing it

[mtklein, 2016-01-14 15:59:10]

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

[mtklein, 2016-01-14 15:59:11]

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
File tools/kilobench/kilobench.cpp (right):

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:8: #include "Benchmark.h"
If you're only going to bench .skps, you may find this code easier to maintain
if you don't bother with the Benchmark abstraction.

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:34: BenchmarkStream() : fCurrentSKP(0) {
This proved to be a miserable pattern to maintain in nanobench.

If you anticipate this ever benching more than .skps, I think we should get
ahead of that misery now.  If you anticipate only ever benching .skps, you
probably want an array.

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:53:
!bench->isSuitableFor(Benchmark::kGPU_Backend))) {
are there .skps which are not suitable for GPU benching?

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:63: if
(SkCommandLineFlags::ShouldSkip(FLAGS_match, path)) {
If you're sticking to .skps, this is probably the only ShouldSkip() you'll ever
need.

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:101: };
namespaces don't need ; at the end, but it is nice to add something like //
namespace kilobench

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:103: * This is an experimental GPU only
benchmarking program for skps
Up top?

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:107: while (Benchmark* bench = benchStream.next())
{
probably nice to delete these at some point

https://codereview.chromium.org/1582903002/diff/40001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:108: }
aww man, https://www.youtube.com/watch?v=_O1hM-k3aUY

[joshualitt, 2016-01-14 16:16:04]

some feedback incorporated.

Mike, because I'm not sure which direction this project will go, or if we will
even use it, I'd rather not spend any time reducing the flexibility(ie in case
we decide we want microbenchs or 'wrapped' benches like VisualBench).  The
stream is a bit cumbersome, but we can rethink the model once we've proven an
alternative approach to timing is even worth pursuing.

[bsalomon, 2016-01-14 16:45:05]

https://codereview.chromium.org/1582903002/diff/60001/tools/kilobench/kiloben...
File tools/kilobench/kilobench.cpp (right):

https://codereview.chromium.org/1582903002/diff/60001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:111: SkAutoTDelete<Benchmark> bench(b);
Does this work?

while (SkAutoTDelete<Benchmark> bench(benchMarkStream.next()) {

[mtklein, 2016-01-14 16:49:43]

lgtm

https://codereview.chromium.org/1582903002/diff/60001/tools/kilobench/kiloben...
File tools/kilobench/kilobench.cpp (right):

https://codereview.chromium.org/1582903002/diff/60001/tools/kilobench/kiloben...
tools/kilobench/kilobench.cpp:59: bench->unref();
Might want to `delete bench;` here for consistency with the AutoTDelete below. 
This makes it seem like the ref-counting is important.

[joshualitt, 2016-01-14 18:34:35]

The CQ bit was checked by joshualitt@google.com

[joshualitt, 2016-01-14 18:34:35]

The patchset sent to the CQ was uploaded after l-g-t-m from mtklein@google.com
Link to the patchset: https://codereview.chromium.org/1582903002/#ps80001
(title: "feedback inc")

[commit-bot: I haz the power, 2016-01-14 18:34:41]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1582903002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1582903002/80001

[commit-bot: I haz the power, 2016-01-14 18:37:15]

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

[commit-bot: I haz the power, 2016-01-14 18:37:16]

Try jobs failed on following builders:
  Build-Ubuntu-GCC-Arm7-Debug-Android-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-Arm...)
  Build-Ubuntu-GCC-Mips-Debug-Android-Trybot on client.skia.compile (JOB_FAILED,
http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-Mip...)

[joshualitt, 2016-01-14 18:53:37]

The CQ bit was checked by joshualitt@google.com

[joshualitt, 2016-01-14 18:53:38]

The patchset sent to the CQ was uploaded after l-g-t-m from mtklein@google.com
Link to the patchset: https://codereview.chromium.org/1582903002/#ps100001
(title: "android stuff")

[commit-bot: I haz the power, 2016-01-14 18:53:43]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1582903002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1582903002/100001

[joshualitt, 2016-01-14 18:53:45]

The CQ bit was unchecked by joshualitt@google.com

[joshualitt, 2016-01-14 19:02:13]

The CQ bit was checked by joshualitt@google.com

[commit-bot: I haz the power, 2016-01-14 19:02:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1582903002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1582903002/100001

[commit-bot: I haz the power, 2016-01-14 19:05:24]

Description was changed from

==========
Add skeleton for kilobench experiments

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

to

==========
Add skeleton for kilobench experiments

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/+/99f340680959c07a11b8cd654e501f92b41646a3
==========

[commit-bot: I haz the power, 2016-01-14 19:05:25]

Committed patchset #6 (id:100001) as
https://skia.googlesource.com/skia/+/99f340680959c07a11b8cd654e501f92b41646a3