[webkit-reviews] review denied: [Bug 132319] CSS JIT: optimize direct / indirect adjacent's traversal backtracking : [Attachment 230473] Patch

[bugzilla-daemon at webkit.org]

Benjamin Poulain <benjamin at webkit.org> has denied Yusuke Suzuki
<utatane.tea at gmail.com>'s request for review:
Bug 132319: CSS JIT: optimize direct / indirect adjacent's traversal
backtracking
https://bugs.webkit.org/show_bug.cgi?id=132319

Attachment 230473: Patch
https://bugs.webkit.org/attachment.cgi?id=230473&action=review

------- Additional Comments from Benjamin Poulain <benjamin at webkit.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=230473&action=review


The patch looks great.

The comments you have added on bugzilla are very useful. You should put those
explanation in the changelog too.

r- because I think the test needs to be extended. The patch looks correct to me
otherwise.

> Source/WebCore/cssjit/SelectorCompiler.cpp:610
> +static inline BacktrackingAction solveTraversalBacktrackingAction(const
SelectorFragment& fragment, bool hasDescendantRelationOnTheRight)

This is only for adjacent backtracking action, let's name this
solveAdjacentTraversalBacktrackingAction() for clarity.

> LayoutTests/fast/selectors/backtracking-adjacent.html:6
> +/* default */

You can remove this comment.

> LayoutTests/fast/selectors/backtracking-adjacent.html:113
> +description('The backtracking direct adjacent combinator with descendant
tail cases');

Hum, I am confused here, none of the cases tested needs a descendant tail.

I think what would make a good test here is cases of adjacent chain inside a
descendant chain where a descendant tail would have been needed before your
backtracking optimizations.