Inline FloatRect::enclosingIntRect

Inline FloatRect::enclosingIntRect

FloatRect::enclosingIntRect is hot on blink_perf.paint's paint offset
microbenchmark. A similar change to inline LayoutRect::enclosingIntRect[1]
was a large perf win [2] and this patch should have similar benefits.
There are relatively few callers so the risk of binary bloat is low.

enclosedIntRect and roundedIntRect are two additional candidates for
inlining but I'd like to separate these changes so we can measure
the impact (perf, binary) independently.

[1] https://crrev.com/af2c8d1e6a537c356b4cf931bb5aca1e5f09f119
[2] https://chromeperf.appspot.com/group_report?rev=436323

BUG=678597
Review-Url: https://codereview.chromium.org/2628903003
Cr-Commit-Position: refs/heads/master@{#443132}
Committed: https://chromium.googlesource.com/chromium/src/+/f76945dbb3afd556549b72544644a2e14fd32271

[pdr., 2017-01-11 21:59:23]

pdr@chromium.org changed reviewers:
+ chrishtr@chromium.org, wangxianzhu@chromium.org

[Xianzhu, 2017-01-11 22:03:34]

lgtm (not an owner).

[chrishtr, 2017-01-11 22:09:06]

The CQ bit was checked by chrishtr@chromium.org

[chrishtr, 2017-01-11 22:09:07]

lgtm

[commit-bot: I haz the power, 2017-01-11 22:10:39]

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

[commit-bot: I haz the power, 2017-01-12 02:57:37]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1484172546742130, "parent_rev":
"9568266bc613e59a97772325b35f6d77af8a0333", "commit_rev":
"f76945dbb3afd556549b72544644a2e14fd32271"}

[commit-bot: I haz the power, 2017-01-12 02:58:10]

Description was changed from

==========
Inline FloatRect::enclosingIntRect

FloatRect::enclosingIntRect is hot on blink_perf.paint's paint offset
microbenchmark. A similar change to inline LayoutRect::enclosingIntRect[1]
was a large perf win [2] and this patch should have similar benefits.
There are relatively few callers so the risk of binary bloat is low.

enclosedIntRect and roundedIntRect are two additional candidates for
inlining but I'd like to separate these changes so we can measure
the impact (perf, binary) independently.

[1] https://crrev.com/af2c8d1e6a537c356b4cf931bb5aca1e5f09f119
[2] https://chromeperf.appspot.com/group_report?rev=436323

BUG=678597
==========

to

==========
Inline FloatRect::enclosingIntRect

FloatRect::enclosingIntRect is hot on blink_perf.paint's paint offset
microbenchmark. A similar change to inline LayoutRect::enclosingIntRect[1]
was a large perf win [2] and this patch should have similar benefits.
There are relatively few callers so the risk of binary bloat is low.

enclosedIntRect and roundedIntRect are two additional candidates for
inlining but I'd like to separate these changes so we can measure
the impact (perf, binary) independently.

[1] https://crrev.com/af2c8d1e6a537c356b4cf931bb5aca1e5f09f119
[2] https://chromeperf.appspot.com/group_report?rev=436323

BUG=678597

Review-Url: https://codereview.chromium.org/2628903003
Cr-Commit-Position: refs/heads/master@{#443132}
Committed:
https://chromium.googlesource.com/chromium/src/+/f76945dbb3afd556549b72544644...
==========

[commit-bot: I haz the power, 2017-01-12 02:58:11]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/f76945dbb3afd556549b72544644...

[pdr., 2017-01-13 19:17:19]

On 2017/01/12 at 02:58:11, commit-bot wrote:
> Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/f76945dbb3afd556549b72544644...

Sadly, this had no effect on measurable performance:
https://chromeperf.appspot.com/group_report?rev=443132 (all false positives)

It did remove enclosingIntRect from my profile but that must be in the noise.
I'm going to leave this patch in for now.

[Xianzhu, 2017-01-13 19:24:51]

On 2017/01/13 19:17:19, pdr. wrote:
> On 2017/01/12 at 02:58:11, commit-bot wrote:
> > Committed patchset #1 (id:1) as
>
https://chromium.googlesource.com/chromium/src/+/f76945dbb3afd556549b72544644...
> 
> Sadly, this had no effect on measurable performance:
> https://chromeperf.appspot.com/group_report?rev=443132 (all false positives)
> 
> It did remove enclosingIntRect from my profile but that must be in the noise.
> I'm going to leave this patch in for now.

There might be improvement, but might be smaller than the threshold to generate
alerts.

[pdr., 2017-01-13 19:29:27]

On 2017/01/13 at 19:24:51, wangxianzhu wrote:
> On 2017/01/13 19:17:19, pdr. wrote:
> > On 2017/01/12 at 02:58:11, commit-bot wrote:
> > > Committed patchset #1 (id:1) as
> >
https://chromium.googlesource.com/chromium/src/+/f76945dbb3afd556549b72544644...
> > 
> > Sadly, this had no effect on measurable performance:
> > https://chromeperf.appspot.com/group_report?rev=443132 (all false positives)
> > 
> > It did remove enclosingIntRect from my profile but that must be in the
noise.
> > I'm going to leave this patch in for now.
> 
> There might be improvement, but might be smaller than the threshold to
generate alerts.

Possibly. Our overall trajectory is looking good:
https://chromeperf.appspot.com/report?sid=3d729b9e37524438c24e621c525378ed556...

Looking forward to seeing what your latest patches do to this line. We may want
to space your patches out to tell the difference independently.