Make SkFixedRound/Ceil/FloorToFixed as inline func

Make SkFixedRound/Ceil/FloorToFixed as inline func

The macros that we were using will return unsigned int32 instead of
signed int32 because of the last 0xFFFF0000 mask. That may bring
problems if we right shift that result.

Special thanks to mtklein@google.com for helping me find out this issue.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2433233003

Committed: https://skia.googlesource.com/skia/+/3f490cc642202c8ab3d51a8ffbd1782c63b9e6a0

[liyuqian, 2016-10-20 17:26:42]

Description was changed from

==========
Make SkFixedRound/Ceil/FloorToFixed as inline func

The macros that we were using will return unsigned int32 instead of
signed int32 because of the last 0xFFFF0000 mask. That may bring
problems if we right shift that result.

Special thanks to mtklein@google.com for helping me find out this issue.

BUG=skia:
==========

to

==========
Make SkFixedRound/Ceil/FloorToFixed as inline func

The macros that we were using will return unsigned int32 instead of
signed int32 because of the last 0xFFFF0000 mask. That may bring
problems if we right shift that result.

Special thanks to mtklein@google.com for helping me find out this issue.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2433233003
==========

[liyuqian, 2016-10-20 17:27:47]

Description was changed from

==========
Make SkFixedRound/Ceil/FloorToFixed as inline func

The macros that we were using will return unsigned int32 instead of
signed int32 because of the last 0xFFFF0000 mask. That may bring
problems if we right shift that result.

Special thanks to mtklein@google.com for helping me find out this issue.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2433233003
==========

to

==========
Make SkFixedRound/Ceil/FloorToFixed as inline func

The macros that we were using will return unsigned int32 instead of
signed int32 because of the last 0xFFFF0000 mask. That may bring
problems if we right shift that result.

Special thanks to mtklein@google.com for helping me find out this issue.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2433233003
==========

[liyuqian, 2016-10-20 17:27:47]

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

[reed1, 2016-10-20 17:39:53]

easy to write a unit-test that would have failed before? We have a bunch around
SkFixed afaik already...

https://codereview.chromium.org/2433233003/diff/1/include/private/SkFixed.h
File include/private/SkFixed.h (right):

https://codereview.chromium.org/2433233003/diff/1/include/private/SkFixed.h#n...
include/private/SkFixed.h:76: return (((x) + SK_Fixed1 - 1) & 0xFFFF0000);
nit: more () than needed in each impl

[liyuqian, 2016-10-20 17:55:14]

On 2016/10/20 17:39:53, reed1 wrote:
> easy to write a unit-test that would have failed before? We have a bunch
around
> SkFixed afaik already...
> 
> https://codereview.chromium.org/2433233003/diff/1/include/private/SkFixed.h
> File include/private/SkFixed.h (right):
> 
>
https://codereview.chromium.org/2433233003/diff/1/include/private/SkFixed.h#n...
> include/private/SkFixed.h:76: return (((x) + SK_Fixed1 - 1) & 0xFFFF0000);
> nit: more () than needed in each impl

Unit tests added :)

[liyuqian, 2016-10-20 17:55:36]

Nit fixed and unit tests added :)

https://codereview.chromium.org/2433233003/diff/1/include/private/SkFixed.h
File include/private/SkFixed.h (right):

https://codereview.chromium.org/2433233003/diff/1/include/private/SkFixed.h#n...
include/private/SkFixed.h:76: return (((x) + SK_Fixed1 - 1) & 0xFFFF0000);
On 2016/10/20 17:39:53, reed1 wrote:
> nit: more () than needed in each impl

Done.

[reed1, 2016-10-20 17:56:35]

lgtm

https://codereview.chromium.org/2433233003/diff/20001/include/private/SkFixed.h
File include/private/SkFixed.h (right):

https://codereview.chromium.org/2433233003/diff/20001/include/private/SkFixed...
include/private/SkFixed.h:79: return (x & 0xFFFF0000);
can remove one more () pair :)

[liyuqian, 2016-10-20 18:00:48]

The CQ bit was checked by liyuqian@google.com

[liyuqian, 2016-10-20 18:00:48]

The patchset sent to the CQ was uploaded after l-g-t-m from reed@google.com
Link to the patchset: https://codereview.chromium.org/2433233003/#ps40001
(title: "Remove one more bracket")

[commit-bot: I haz the power, 2016-10-20 18:00:54]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[mtklein_C, 2016-10-20 18:06:53]

mtklein@chromium.org changed reviewers:
+ mtklein@chromium.org

[mtklein_C, 2016-10-20 18:06:54]

lgtm

[commit-bot: I haz the power, 2016-10-20 18:23:11]

Description was changed from

==========
Make SkFixedRound/Ceil/FloorToFixed as inline func

The macros that we were using will return unsigned int32 instead of
signed int32 because of the last 0xFFFF0000 mask. That may bring
problems if we right shift that result.

Special thanks to mtklein@google.com for helping me find out this issue.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2433233003
==========

to

==========
Make SkFixedRound/Ceil/FloorToFixed as inline func

The macros that we were using will return unsigned int32 instead of
signed int32 because of the last 0xFFFF0000 mask. That may bring
problems if we right shift that result.

Special thanks to mtklein@google.com for helping me find out this issue.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2433233003

Committed:
https://skia.googlesource.com/skia/+/3f490cc642202c8ab3d51a8ffbd1782c63b9e6a0
==========

[commit-bot: I haz the power, 2016-10-20 18:23:12]

Committed patchset #3 (id:40001) as
https://skia.googlesource.com/skia/+/3f490cc642202c8ab3d51a8ffbd1782c63b9e6a0