Refactor position computation to enable device space "nudge"

Refactor position computation to enable device space "nudge"

To match raster's handling of BW geometry we want to be able to perform a device space "nudge" on all geometry. This CL sets us up to be able to do that in GrGLVertexBuilder::transformToNormalizedDeviceSpace.

BUG=423834
TBR=bsalomon@google.com

Committed: https://skia.googlesource.com/skia/+/46d36f0e7b709a077c647841eee23bd3efdc4117

[robertphillips, 2015-01-15 19:25:30]

robertphillips@google.com changed reviewers:
+ joshualitt@chromium.org, joshualitt@google.com

[joshualitt, 2015-01-15 19:53:58]

On 2015/01/15 19:46:24, robertphillips wrote:

nits

[joshualitt, 2015-01-15 19:54:04]

https://codereview.chromium.org/854013002/diff/20001/src/gpu/GrGeometryProces...
File src/gpu/GrGeometryProcessor.cpp (right):

https://codereview.chromium.org/854013002/diff/20001/src/gpu/GrGeometryProces...
src/gpu/GrGeometryProcessor.cpp:266: gpArgs->fPositionVar.set(kVec2f_GrSLType,
GrShaderVar::kNone_TypeModifier, "pos2");
Maybe make a new setter with type and name?

https://codereview.chromium.org/854013002/diff/20001/src/gpu/gl/builders/GrGL...
File src/gpu/gl/builders/GrGLVertexShaderBuilder.cpp (right):

https://codereview.chromium.org/854013002/diff/20001/src/gpu/gl/builders/GrGL...
src/gpu/gl/builders/GrGLVertexShaderBuilder.cpp:51:
this->codeAppendf("gl_Position = vec4(dot(%s.xz, %s.xy)/%s.z, dot(%s.yz,
%s.zw)/%s.z, 0, 1.0f);",
linewrap?  I guess its a bit tough

[robertphillips, 2015-01-15 20:55:33]

https://codereview.chromium.org/854013002/diff/20001/src/gpu/GrGeometryProces...
File src/gpu/GrGeometryProcessor.cpp (right):

https://codereview.chromium.org/854013002/diff/20001/src/gpu/GrGeometryProces...
src/gpu/GrGeometryProcessor.cpp:266: gpArgs->fPositionVar.set(kVec2f_GrSLType,
GrShaderVar::kNone_TypeModifier, "pos2");
On 2015/01/15 19:54:04, joshualitt wrote:
> Maybe make a new setter with type and name?

Done (maybe). WDYT?

https://codereview.chromium.org/854013002/diff/20001/src/gpu/gl/builders/GrGL...
File src/gpu/gl/builders/GrGLVertexShaderBuilder.cpp (right):

https://codereview.chromium.org/854013002/diff/20001/src/gpu/gl/builders/GrGL...
src/gpu/gl/builders/GrGLVertexShaderBuilder.cpp:51:
this->codeAppendf("gl_Position = vec4(dot(%s.xz, %s.xy)/%s.z, dot(%s.yz,
%s.zw)/%s.z, 0, 1.0f);",
I've shortened a bit but breaking it into multiple lines is going to be ugly.

[joshualitt, 2015-01-15 21:08:00]

On 2015/01/15 20:55:33, robertphillips wrote:
>
https://codereview.chromium.org/854013002/diff/20001/src/gpu/GrGeometryProces...
> File src/gpu/GrGeometryProcessor.cpp (right):
> 
>
https://codereview.chromium.org/854013002/diff/20001/src/gpu/GrGeometryProces...
> src/gpu/GrGeometryProcessor.cpp:266: gpArgs->fPositionVar.set(kVec2f_GrSLType,
> GrShaderVar::kNone_TypeModifier, "pos2");
> On 2015/01/15 19:54:04, joshualitt wrote:
> > Maybe make a new setter with type and name?
> 
> Done (maybe). WDYT?
> 
>
https://codereview.chromium.org/854013002/diff/20001/src/gpu/gl/builders/GrGL...
> File src/gpu/gl/builders/GrGLVertexShaderBuilder.cpp (right):
> 
>
https://codereview.chromium.org/854013002/diff/20001/src/gpu/gl/builders/GrGL...
> src/gpu/gl/builders/GrGLVertexShaderBuilder.cpp:51:
> this->codeAppendf("gl_Position = vec4(dot(%s.xz, %s.xy)/%s.z, dot(%s.yz,
> %s.zw)/%s.z, 0, 1.0f);",
> I've shortened a bit but breaking it into multiple lines is going to be ugly.

lol lgtm

[robertphillips, 2015-01-18 16:04:59]

The CQ bit was checked by robertphillips@google.com

[commit-bot: I haz the power, 2015-01-18 16:06:00]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/854013002/60001

[commit-bot: I haz the power, 2015-01-18 16:06:05]

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

[commit-bot: I haz the power, 2015-01-18 16:06:06]

Presubmit check for 854013002-60001 failed and returned exit status 1.

Running presubmit commit checks ...

** Presubmit ERRORS **
Since the CL is editing public API, you must have an LGTM from one of:
('reed@chromium.org', 'reed@google.com', 'bsalomon@chromium.org',
'bsalomon@google.com', 'djsollen@chromium.org', 'djsollen@google.com')

Was the presubmit check useful? If not, run "git cl presubmit -v"
to figure out which PRESUBMIT.py was run, then run git blame
on the file to figure out who to ask for help.

[robertphillips, 2015-01-18 16:07:32]

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

[robertphillips, 2015-01-18 16:07:38]

The CQ bit was checked by robertphillips@google.com

[commit-bot: I haz the power, 2015-01-18 16:08:00]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/854013002/60001

[commit-bot: I haz the power, 2015-01-18 16:14:20]

Committed patchset #4 (id:60001) as
https://skia.googlesource.com/skia/+/46d36f0e7b709a077c647841eee23bd3efdc4117

[bsalomon, 2015-01-20 15:19:51]

https://codereview.chromium.org/854013002/diff/60001/src/gpu/GrGeometryProces...
File src/gpu/GrGeometryProcessor.cpp (right):

https://codereview.chromium.org/854013002/diff/60001/src/gpu/GrGeometryProces...
src/gpu/GrGeometryProcessor.cpp:221: vb->codeAppendf("%s = (%s * %s).xy;",
v.vsOut(), uniName, posVar.c_str());
Is this right? It seems like we know that the vm is likely persp since the GP
emitted a vec3 pos but that the coordxform is non-persp. Don't we have to
homogenize either before or after the xform matrix multiply? Or would we never
get here because varyingType would already be vec3 because we earlier checked
the vm for persp?