create shaderproc for nofilter-opaque-dx

create shaderproc for nofilter-opaque-dx


speedup nofilter

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/a40a276bcee2246439dcf816273c1307f5c3c69f

[reed1, 2014-10-23 18:51:53]

reed@google.com changed reviewers:
+ djsollen@google.com, mtklein@google.com

[djsollen, 2014-10-23 19:00:33]

https://codereview.chromium.org/664783004/diff/40001/src/core/SkBitmapProcSta...
File src/core/SkBitmapProcState.cpp (right):

https://codereview.chromium.org/664783004/diff/40001/src/core/SkBitmapProcSta...
src/core/SkBitmapProcState.cpp:593: } else if
(SK_ARM_NEON_WRAP(S32_opaque_D32_nofilter_DX) == fSampleProc32 && clampClamp) {
compile fails because S32_opaque_D32_nofilter_DX_neon doesn't exist

[mtklein, 2014-10-23 19:05:45]

https://codereview.chromium.org/664783004/diff/40001/src/core/SkBitmapProcSta...
File src/core/SkBitmapProcState.cpp (right):

https://codereview.chromium.org/664783004/diff/40001/src/core/SkBitmapProcSta...
src/core/SkBitmapProcState.cpp:1049: if ((unsigned)(fx >> 48) <= maxX &&
(unsigned)((fx + dx * (count - 1)) >> 48) <= maxX) {
Can't hurt to make this a little more readable?

if (maxX >= (unsigned)SkFractionalIntToInt(fx) && 
    maxX >= (unsigned)SkFractionalIntToInt(fx + dx * (count-1))) {
  // No need to clamp anywhere in our x range.
  ...
}

[reed1, 2014-10-23 19:13:56]

ptal

https://codereview.chromium.org/664783004/diff/40001/src/core/SkBitmapProcSta...
File src/core/SkBitmapProcState.cpp (right):

https://codereview.chromium.org/664783004/diff/40001/src/core/SkBitmapProcSta...
src/core/SkBitmapProcState.cpp:593: } else if
(SK_ARM_NEON_WRAP(S32_opaque_D32_nofilter_DX) == fSampleProc32 && clampClamp) {
On 2014/10/23 19:00:33, djsollen wrote:
> compile fails because S32_opaque_D32_nofilter_DX_neon doesn't exist

Done.

https://codereview.chromium.org/664783004/diff/40001/src/core/SkBitmapProcSta...
src/core/SkBitmapProcState.cpp:1049: if ((unsigned)(fx >> 48) <= maxX &&
(unsigned)((fx + dx * (count - 1)) >> 48) <= maxX) {
On 2014/10/23 19:05:45, mtklein wrote:
> Can't hurt to make this a little more readable?
> 
> if (maxX >= (unsigned)SkFractionalIntToInt(fx) && 
>     maxX >= (unsigned)SkFractionalIntToInt(fx + dx * (count-1))) {
>   // No need to clamp anywhere in our x range.
>   ...
> }

Done.

[reed1, 2014-10-23 19:14:01]

The CQ bit was checked by reed@google.com

[commit-bot: I haz the power, 2014-10-23 19:14:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/664783004/120001

[commit-bot: I haz the power, 2014-10-23 19:14:17]

Note for Reviewers:
The CQ is waiting for an approval.
If you believe that the CL is not ready yet, or if you would like to L-G-T-M
with comments then please uncheck the CQ checkbox.

Waiting for LGTM from valid reviewer(s) till 2014-10-24 01:14 UTC

[mtklein, 2014-10-23 19:16:59]

lgtm

[commit-bot: I haz the power, 2014-10-23 19:22:48]

Committed patchset #7 (id:120001) as a40a276bcee2246439dcf816273c1307f5c3c69f

[mtklein, 2014-10-23 19:43:49]

A revert of this CL (patchset #7 id:120001) has been created in
https://codereview.chromium.org/656913005/ by mtklein@google.com.

The reason for reverting is: Many GMs fixed.  Needs rebaseline, perhaps layout
test rebaselines..

[mtklein, 2014-10-23 20:15:47]

On 2014/10/23 19:43:49, mtklein wrote:
> A revert of this CL (patchset #7 id:120001) has been created in
> https://codereview.chromium.org/656913005/ by mailto:mtklein@google.com.
> 
> The reason for reverting is: Many GMs fixed.  Needs rebaseline, perhaps layout
> test rebaselines..

https://codereview.chromium.org/675813002 is the layout test roll to watch.

[mtklein, 2014-10-23 21:25:16]

On 2014/10/23 20:15:47, mtklein wrote:
> On 2014/10/23 19:43:49, mtklein wrote:
> > A revert of this CL (patchset #7 id:120001) has been created in
> > https://codereview.chromium.org/656913005/ by mailto:mtklein@google.com.
> > 
> > The reason for reverting is: Many GMs fixed.  Needs rebaseline, perhaps
layout
> > test rebaselines..
> 
> https://codereview.chromium.org/675813002 is the layout test roll to watch.

linux_blink_* look good.  Going to reland with some GM suppressions.
(Ben tells me we should ignore linux_layout_*.)