Create a baseline RenderBlock::lineHeight() performance

Create a baseline RenderBlock::lineHeight() performance

This test calls RenderBlock::lineHeight() ~425,000 times on each run.
I hope to remove the caching of RenderStyle::computedLineHeight() by
|m_lineHeight| in RenderBlock so this will baseline will ensure it
doesn't regress performance.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175356

[rhogan, 2014-04-28 20:09:21]

The follow up to this will be : https://codereview.chromium.org/260073005/. Look
there for more context on the motivation behind this test.

[Julien - ping for review, 2014-04-28 20:11:37]

lgtm

https://codereview.chromium.org/256593009/diff/1/PerformanceTests/Layout/line...
File PerformanceTests/Layout/line-layout-line-height.html (right):

https://codereview.chromium.org/256593009/diff/1/PerformanceTests/Layout/line...
PerformanceTests/Layout/line-layout-line-height.html:4: <title>Line height
calculation performance test</title>
Can't we dump this into the test or would this impact its results?

https://codereview.chromium.org/256593009/diff/1/PerformanceTests/Layout/line...
PerformanceTests/Layout/line-layout-line-height.html:14: var rows = 10000;
s/rows/lines/?

https://codereview.chromium.org/256593009/diff/1/PerformanceTests/Layout/line...
PerformanceTests/Layout/line-layout-line-height.html:22: style.display =
"block";
FWIW you are also checking the attachment / detachment by including these
display changes in the test.

https://codereview.chromium.org/256593009/diff/1/PerformanceTests/Layout/line...
PerformanceTests/Layout/line-layout-line-height.html:32:
PerfTestRunner.measureTime({ run: test });
I think it's preferred to measure runs per seconds as it is unbounded and it
wouldn't run into clock accuracy (if you go below 10ms, it starts to get hairy).

[commit-bot: I haz the power, 2014-04-28 20:11:55]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/robhogan@gmail.com/256593009/1

[Julien - ping for review, 2014-04-28 20:12:53]

Oups, click epic failure. I didn't really mean to LGTM as this is more Levi's
turf. The idea behind this testing is definitely cool and has my entire support
though!

[Julien - ping for review, 2014-04-28 20:12:59]

The CQ bit was unchecked by jchaffraix@chromium.org

[leviw_travelin_and_unemployed, 2014-04-28 20:42:18]

LGTM with Julien's nits addressed.

[leviw_travelin_and_unemployed, 2014-05-01 08:11:35]

For the record, regardless of what you found in your checks, I support more
blink perf tests :)

[rhogan, 2014-06-03 10:30:24]

The CQ bit was checked by robhogan@gmail.com

[commit-bot: I haz the power, 2014-06-03 10:30:57]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/robhogan@gmail.com/256593009/1

[commit-bot: I haz the power, 2014-06-03 11:30:13]

Change committed as 175356