fix bounds for BW lines, AA Hairlines

fix bounds for BW lines, AA Hairlines

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/4c977868bbe100d6d95f9e53cf176d611eceb3dc

[joshualitt, 2015-05-06 20:56:01]

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

[joshualitt, 2015-05-06 20:56:02]

ptal

[robertphillips, 2015-05-07 13:17:17]

https://codereview.chromium.org/1123253003/diff/40001/src/gpu/GrAAHairLinePat...
File src/gpu/GrAAHairLinePathRenderer.cpp (right):

https://codereview.chromium.org/1123253003/diff/40001/src/gpu/GrAAHairLinePat...
src/gpu/GrAAHairLinePathRenderer.cpp:735: // AAHairlines need to be outset a
half pixel in each direction to account for AA
// This is b.c. hairlines are notionally infinitely thin so without expansion
two overlapping lines could be reordered even though they hit the same pixels.
?

https://codereview.chromium.org/1123253003/diff/40001/src/gpu/GrContext.cpp
File src/gpu/GrContext.cpp (right):

https://codereview.chromium.org/1123253003/diff/40001/src/gpu/GrContext.cpp#n...
src/gpu/GrContext.cpp:525: 
rm "then" ?

https://codereview.chromium.org/1123253003/diff/40001/src/gpu/GrContext.cpp#n...
src/gpu/GrContext.cpp:527: // snapping
// This is conservative but, b.c. the hardware can employ somewhat arbitrary
snapping rules to BW lines, it is the best we can do.

?

https://codereview.chromium.org/1123253003/diff/40001/src/gpu/GrContext.cpp#n...
src/gpu/GrContext.cpp:528: fBounds.outset(0.5f, 0.5f);
Shouldn't this only occur for hairlines? For fills we could actually snap
correctly.

[joshualitt, 2015-05-07 13:28:56]

Feedback inc

[robertphillips, 2015-05-07 14:42:02]

https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp
File src/gpu/GrContext.cpp (right):

https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp#n...
src/gpu/GrContext.cpp:525: 
For this case, BW hairline rect draw, I believe we just force the lines to pixel
centers.

[bsalomon, 2015-05-07 14:47:46]

https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp
File src/gpu/GrContext.cpp (right):

https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp#n...
src/gpu/GrContext.cpp:525: 
On 2015/05/07 14:42:02, robertphillips wrote:
> For this case, BW hairline rect draw, I believe we just force the lines to
pixel
> centers.

Correct. This is an unfortunate case of coupling where the user of this batch
did the snapping rather than the batch itself, but the batch has to know about
it.

[joshualitt, 2015-05-12 18:18:43]

On 2015/05/07 14:47:46, bsalomon wrote:
> https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp
> File src/gpu/GrContext.cpp (right):
> 
>
https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp#n...
> src/gpu/GrContext.cpp:525: 
> On 2015/05/07 14:42:02, robertphillips wrote:
> > For this case, BW hairline rect draw, I believe we just force the lines to
> pixel
> > centers.
> 
> Correct. This is an unfortunate case of coupling where the user of this batch
> did the snapping rather than the batch itself, but the batch has to know about
> it.

feedback incorporated, ptal.

[joshualitt, 2015-05-12 20:09:06]

On 2015/05/12 18:18:43, joshualitt wrote:
> On 2015/05/07 14:47:46, bsalomon wrote:
> > https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp
> > File src/gpu/GrContext.cpp (right):
> > 
> >
>
https://codereview.chromium.org/1123253003/diff/60001/src/gpu/GrContext.cpp#n...
> > src/gpu/GrContext.cpp:525: 
> > On 2015/05/07 14:42:02, robertphillips wrote:
> > > For this case, BW hairline rect draw, I believe we just force the lines to
> > pixel
> > > centers.
> > 
> > Correct. This is an unfortunate case of coupling where the user of this
batch
> > did the snapping rather than the batch itself, but the batch has to know
about
> > it.
> 
> feedback incorporated, ptal.

feedback inc

[robertphillips, 2015-05-13 14:43:03]

lgtm + comment suggestion

https://codereview.chromium.org/1123253003/diff/100001/src/gpu/GrContext.cpp
File src/gpu/GrContext.cpp (right):

https://codereview.chromium.org/1123253003/diff/100001/src/gpu/GrContext.cpp#...
src/gpu/GrContext.cpp:668: 
// Non-AA hairlines are snapped to pixel centers to make which pixels are hit
deterministic
?

[joshualitt, 2015-05-13 14:48:47]

The CQ bit was checked by joshualitt@google.com

[joshualitt, 2015-05-13 14:48:47]

The patchset sent to the CQ was uploaded after l-g-t-m from
robertphillips@google.com
Link to the patchset: https://codereview.chromium.org/1123253003/#ps140001
(title: "more")

[commit-bot: I haz the power, 2015-05-13 14:48:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1123253003/140001

[commit-bot: I haz the power, 2015-05-13 15:01:03]

Committed patchset #8 (id:140001) as
https://skia.googlesource.com/skia/+/4c977868bbe100d6d95f9e53cf176d611eceb3dc