Remove backend factories

Remove backend factories

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/eb2a6761654307e8aeeeaabdd63c6bf9ab0411e9

[joshualitt, 2014-12-03 21:48:02]

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

[joshualitt, 2014-12-03 21:48:03]

Now, this is a big CL, but its actually pretty mindless.  I simply deleted the
factories and moved the factory calls to the processors.

There are a bunch of cleanups I want to do later, like deleting GrGLProcessor,
and getting rid of GenKey on the GL*Processor base classes, etc, but this CL is
already waaayy too big so I didn't do that here.

[bsalomon, 2014-12-04 14:31:28]

https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrFragmentPr...
File include/gpu/GrFragmentProcessor.h (right):

https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrFragmentPr...
include/gpu/GrFragmentProcessor.h:35: for the given GrProcessor; caller is
responsible for deleting
GrProcessor->GrFragmentProcessor

https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrProcessor.h
File include/gpu/GrProcessor.h (right):

https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrProcessor....
include/gpu/GrProcessor.h:22: * Used by processors to build their keys. It
incorporates each per-processor key into a larger shader
linewrap here

https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrProcessor....
include/gpu/GrProcessor.h:105: virtual uint32_t classID() const = 0;
needs to be virtual? Could we have a member initialized by the subclass? I think
we call this a fair amount.

MySubClass::MySubClass() {
   static uint32_t kClassID = GenClassID();
   fClassID = kClassID;
}

Or we could try to be a bit helpful:

template <typename PROC_SUBCLASS> uint32_t GrProcessor::initClassID() {
    static uint32_t kClassID = GenClassID(); 
    fClassID = kClassID;
}

The base class could init fClassID to kIllegal and then getClassID() could
assert that fClassID != kIllegal.

[joshualitt, 2014-12-04 15:19:47]

On 2014/12/04 14:31:28, bsalomon wrote:
>
https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrFragmentPr...
> File include/gpu/GrFragmentProcessor.h (right):
> 
>
https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrFragmentPr...
> include/gpu/GrFragmentProcessor.h:35: for the given GrProcessor; caller is
> responsible for deleting
> GrProcessor->GrFragmentProcessor
> 
> https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrProcessor.h
> File include/gpu/GrProcessor.h (right):
> 
>
https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrProcessor....
> include/gpu/GrProcessor.h:22: * Used by processors to build their keys. It
> incorporates each per-processor key into a larger shader
> linewrap here
> 
>
https://codereview.chromium.org/778453002/diff/40001/include/gpu/GrProcessor....
> include/gpu/GrProcessor.h:105: virtual uint32_t classID() const = 0;
> needs to be virtual? Could we have a member initialized by the subclass? I
think
> we call this a fair amount.
> 
> MySubClass::MySubClass() {
>    static uint32_t kClassID = GenClassID();
>    fClassID = kClassID;
> }
> 
> Or we could try to be a bit helpful:
> 
> template <typename PROC_SUBCLASS> uint32_t GrProcessor::initClassID() {
>     static uint32_t kClassID = GenClassID(); 
>     fClassID = kClassID;
> }
> 
> The base class could init fClassID to kIllegal and then getClassID() could
> assert that fClassID != kIllegal.

Feedback incorporated.  The only advantage to the virtual is its compile time
checked, but it is obviously slower.

[bsalomon, 2014-12-04 15:27:33]

https://codereview.chromium.org/778453002/diff/60001/include/gpu/GrProcessor.h
File include/gpu/GrProcessor.h (right):

https://codereview.chromium.org/778453002/diff/60001/include/gpu/GrProcessor....
include/gpu/GrProcessor.h:127: static uint32_t GenClassID() {
Can we move this and fCurrProcessorClassID to private? Prolly kIllegal as well
right?

Also, statics get a 'g' prefix instead of 'f'

otherwise, lgtm.

[joshualitt, 2014-12-04 18:48:33]

The CQ bit was checked by joshualitt@google.com

[commit-bot: I haz the power, 2014-12-04 18:49:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/778453002/100001

[commit-bot: I haz the power, 2014-12-04 18:53:00]

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

[commit-bot: I haz the power, 2014-12-04 18:53:01]

Try jobs failed on following builders:
  Build-Ubuntu13.10-Clang-x86_64-Debug-Trybot on client.skia.compile
(http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu13.10-Cl...)

[joshualitt, 2014-12-04 19:25:02]

The CQ bit was checked by joshualitt@google.com

[commit-bot: I haz the power, 2014-12-04 19:26:09]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/778453002/140001

[commit-bot: I haz the power, 2014-12-04 19:35:38]

Committed patchset #8 (id:140001) as
https://skia.googlesource.com/skia/+/eb2a6761654307e8aeeeaabdd63c6bf9ab0411e9