DM: also run benches once.

DM: also run benches once.

Also:
  - make GrMemoryPoolBenches threadsafe
  - some tweaks to various DM code
  - rename GM::shortName() to getName() to match benches and tests

On my desktop, (289 GMs, 617 benches) x 4 configs, 227 tests takes 46s in Debug, 14s in Release.  (Still minutes faster than running tests && bench && gm.)  GPU singlethreading is definitely the limiting factor again; going to reexamine whether that's helpful to thread it again.

BUG=skia:

Committed: http://code.google.com/p/skia/source/detail?r=13603

[mtklein, 2014-02-26 20:13:33]

Mike, I believe this is what you were asking for last week.

[reed1, 2014-02-26 20:30:58]

I presume the benchiness can be opt'd in or out from the command-line?

lgtm

[bsalomon, 2014-02-26 20:37:45]

As I understand it, this won't run GMs-as-benches. We now let GMs know whether
they're being run in bench mode or not so they can have different behavior.
Currently, only one GM observes the mode and it tests a subset of code paths in
bench mode. So it's not a problem now but it could become one.

[mtklein, 2014-02-26 20:44:29]

On 2014/02/26 20:30:58, reed1 wrote:
> I presume the benchiness can be opt'd in or out from the command-line?
> 
> lgtm
Yes, opt out is --nobenches, like --notests and --nogms.

On 2014/02/26 20:37:45, bsalomon wrote:
> As I understand it, this won't run GMs-as-benches. We now let GMs know whether
> they're being run in bench mode or not so they can have different behavior.
> Currently, only one GM observes the mode and it tests a subset of code paths
in
> bench mode. So it's not a problem now but it could become one.
Right, this runs only GMs-as-GMs and benches-as-benches.  I hadn't realized the
GMs-as-benches knew they were in a different mode.  It'd be easy to add
GMs-as-benches here.  Let me try to do that now so that the purpose doesn't slip
our minds before we do it.

[mtklein, 2014-02-26 20:56:03]

On 2014/02/26 20:44:29, mtklein wrote:
> On 2014/02/26 20:30:58, reed1 wrote:
> > I presume the benchiness can be opt'd in or out from the command-line?
> > 
> > lgtm
> Yes, opt out is --nobenches, like --notests and --nogms.
> 
> On 2014/02/26 20:37:45, bsalomon wrote:
> > As I understand it, this won't run GMs-as-benches. We now let GMs know
whether
> > they're being run in bench mode or not so they can have different behavior.
> > Currently, only one GM observes the mode and it tests a subset of code paths
> in
> > bench mode. So it's not a problem now but it could become one.
> Right, this runs only GMs-as-GMs and benches-as-benches.  I hadn't realized
the
> GMs-as-benches knew they were in a different mode.  It'd be easy to add
> GMs-as-benches here.  Let me try to do that now so that the purpose doesn't
slip
> our minds before we do it.

Actually, I take it back.  I'd want to think a bit about how to do this.  I've
added a caveat note instead.

[bsalomon, 2014-02-26 21:06:00]

On 2014/02/26 20:56:03, mtklein wrote:
> On 2014/02/26 20:44:29, mtklein wrote:
> > On 2014/02/26 20:30:58, reed1 wrote:
> > > I presume the benchiness can be opt'd in or out from the command-line?
> > > 
> > > lgtm
> > Yes, opt out is --nobenches, like --notests and --nogms.
> > 
> > On 2014/02/26 20:37:45, bsalomon wrote:
> > > As I understand it, this won't run GMs-as-benches. We now let GMs know
> whether
> > > they're being run in bench mode or not so they can have different
behavior.
> > > Currently, only one GM observes the mode and it tests a subset of code
paths
> > in
> > > bench mode. So it's not a problem now but it could become one.
> > Right, this runs only GMs-as-GMs and benches-as-benches.  I hadn't realized
> the
> > GMs-as-benches knew they were in a different mode.  It'd be easy to add
> > GMs-as-benches here.  Let me try to do that now so that the purpose doesn't
> slip
> > our minds before we do it.
> 
> Actually, I take it back.  I'd want to think a bit about how to do this.  I've
> added a caveat note instead.

lgtm

[mtklein, 2014-02-26 21:44:58]

The CQ bit was checked by mtklein@google.com

[commit-bot: I haz the power, 2014-02-26 21:45:07]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/mtklein@chromium.org/178473006/70001

[commit-bot: I haz the power, 2014-02-26 23:02:07]

Change committed as 13603