Recalc sibling styles for forward positional rules on mutations only.

Recalc sibling styles for forward positional rules on mutations only.

Forward positional rules only affect selector matching of siblings when
nodes are added to or removed from the DOM. However, selectors with indirect
adjacent combinators may affect sibling evaluation without node addition or
removal. For instance if a class selector left of an indirect adjacent
combinator is set on an element.

Since the indirect adjacent combinator was considered as a forward
positional rule, we were recalculating sibling styles in the presence of an
nth-selector. We now split indirect adjacent selector matching as a separate
flag to be able to optimize recalc in the presence if nth-selectors and
skip sibling recalc for forward positional rules if no mutations happened.

ElementRareData and ShadowRootRareData do not increase in size since they
were already padded in the relevant parts.

R=ojan
BUG=160212
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=169471

[rune, 2014-03-18 08:45:39]

This needs another rebase now after https://codereview.chromium.org/202633006/

[esprehn, 2014-03-18 09:31:36]

lgtm, one nit to fix before landing. This is a great patch (once you rebase it)
:)

https://codereview.chromium.org/198783003/diff/20001/LayoutTests/fast/css/nth...
File LayoutTests/fast/css/nth-child-no-mutations.html (right):

https://codereview.chromium.org/198783003/diff/20001/LayoutTests/fast/css/nth...
LayoutTests/fast/css/nth-child-no-mutations.html:21:
internals.updateStyleAndReturnAffectedElementCount();
Just do document.body.offsetTop if you're going to ignore the return value.

[rune, 2014-03-18 11:01:47]

https://codereview.chromium.org/198783003/diff/20001/LayoutTests/fast/css/nth...
File LayoutTests/fast/css/nth-child-no-mutations.html (right):

https://codereview.chromium.org/198783003/diff/20001/LayoutTests/fast/css/nth...
LayoutTests/fast/css/nth-child-no-mutations.html:21:
internals.updateStyleAndReturnAffectedElementCount();
On 2014/03/18 09:31:37, esprehn wrote:
> Just do document.body.offsetTop if you're going to ignore the return value.

Done.

[rune, 2014-03-18 11:02:14]

The CQ bit was checked by rune@opera.com

[commit-bot: I haz the power, 2014-03-18 11:02:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rune@opera.com/198783003/60001

[commit-bot: I haz the power, 2014-03-18 11:09:01]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-03-18 11:09:03]

Try jobs failed on following builders:
  tryserver.blink on mac_blink_compile_dbg

[rune, 2014-03-18 19:56:02]

The CQ bit was checked by rune@opera.com

[commit-bot: I haz the power, 2014-03-18 19:56:06]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rune@opera.com/198783003/60001

[commit-bot: I haz the power, 2014-03-18 20:29:48]

Change committed as 169471