skpbench: run for a fixed duration

skpbench: run for a fixed duration

Runs for a fixed amount of time instead of a fixed amount of samples.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2374093002

Committed: https://skia.googlesource.com/skia/+/037adf34f5f00e5806b09b202c96cac5c0a5da0a

[csmartdalton, 2016-09-28 17:40:50]

Description was changed from

==========
skpbench: run for a fixed duration

Runs for a fixed amount of time instead of a fixed amount of samples.

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

to

==========
skpbench: run for a fixed duration

Runs for a fixed amount of time instead of a fixed amount of samples.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2374093002
==========

[csmartdalton, 2016-09-28 17:44:17]

csmartdalton@google.com changed reviewers:
+ bsalomon@google.com

[bsalomon, 2016-09-28 18:51:16]

https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
File tools/skpbench/skpbench.cpp (right):

https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
tools/skpbench/skpbench.cpp:81: GrTAllocator<Sample>* samples) {
Why the change to GrTAllocator?

[csmartdalton, 2016-09-28 19:17:54]

https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
File tools/skpbench/skpbench.cpp (right):

https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
tools/skpbench/skpbench.cpp:81: GrTAllocator<Sample>* samples) {
On 2016/09/28 18:51:16, bsalomon wrote:
> Why the change to GrTAllocator?

Because with "--sampleMs 0" we don't know an upper limit on how many samples
there will be anymore, and the data doesn't need to remain contiguous in the
event that we allocate more space.

It's not a big deal though, is there reason not to use it?

[bsalomon, 2016-09-28 19:20:30]

On 2016/09/28 19:17:54, csmartdalton wrote:
>
https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
> File tools/skpbench/skpbench.cpp (right):
> 
>
https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
> tools/skpbench/skpbench.cpp:81: GrTAllocator<Sample>* samples) {
> On 2016/09/28 18:51:16, bsalomon wrote:
> > Why the change to GrTAllocator?
> 
> Because with "--sampleMs 0" we don't know an upper limit on how many samples
> there will be anymore, and the data doesn't need to remain contiguous in the
> event that we allocate more space.
> 
> It's not a big deal though, is there reason not to use it?

Not particularly... it's not used in many places. If those were removed I'd
propose deleting it. I thought it had already been deleted. :) IIRC, we
currently only use it in places where we need the items stored in it to not be
relocated. But I leave it up to you. lgtm.

[csmartdalton, 2016-09-28 20:37:01]

On 2016/09/28 19:20:30, bsalomon wrote:
> On 2016/09/28 19:17:54, csmartdalton wrote:
> >
>
https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
> > File tools/skpbench/skpbench.cpp (right):
> > 
> >
>
https://codereview.chromium.org/2374093002/diff/20001/tools/skpbench/skpbench...
> > tools/skpbench/skpbench.cpp:81: GrTAllocator<Sample>* samples) {
> > On 2016/09/28 18:51:16, bsalomon wrote:
> > > Why the change to GrTAllocator?
> > 
> > Because with "--sampleMs 0" we don't know an upper limit on how many samples
> > there will be anymore, and the data doesn't need to remain contiguous in the
> > event that we allocate more space.
> > 
> > It's not a big deal though, is there reason not to use it?
> 
> Not particularly... it's not used in many places. If those were removed I'd
> propose deleting it. I thought it had already been deleted. :) IIRC, we
> currently only use it in places where we need the items stored in it to not be
> relocated. But I leave it up to you. lgtm.

I see. It's not worth it if GrTAllocator is on the chopping block. We have an
upper limit anyway as long as sampleMs != 0.

[csmartdalton, 2016-09-28 20:37:04]

The CQ bit was checked by csmartdalton@google.com

[csmartdalton, 2016-09-28 20:37:05]

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/2374093002/#ps40001
(title: "remove GrTAllocator")

[commit-bot: I haz the power, 2016-09-28 20:37:11]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-28 20:56:02]

Description was changed from

==========
skpbench: run for a fixed duration

Runs for a fixed amount of time instead of a fixed amount of samples.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2374093002
==========

to

==========
skpbench: run for a fixed duration

Runs for a fixed amount of time instead of a fixed amount of samples.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2374093002

Committed:
https://skia.googlesource.com/skia/+/037adf34f5f00e5806b09b202c96cac5c0a5da0a
==========

[commit-bot: I haz the power, 2016-09-28 20:56:03]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/037adf34f5f00e5806b09b202c96cac5c0a5da0a