Oilpan: remove invalid asserts from ListHashSet tracing.

Oilpan: remove invalid asserts from ListHashSet tracing.

When conservatively tracing ListHashSets we can end up tracing the
ListHashSet itself where the ListHashSet backing has been marked
and is on the marking stack for later tracing. Therefore, the
asserts that the backing has already been fully traced if the
backing is marked when we trace the ListHashSet is invalid.

R=erik.corry@gmail.com, oilpan-reviews@chromium.org
BUG=375669
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=176110

[Erik Corry, 2014-06-13 11:47:33]

lgtm

[Mads Ager (chromium), 2014-06-13 11:49:12]

The CQ bit was checked by ager@chromium.org

[commit-bot: I haz the power, 2014-06-13 11:49:44]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ager@chromium.org/336693007/1

[sof, 2014-06-13 12:16:32]

That that marking&trace order invalidates the assert makes sense.

You could instead assert that m_head and m_tail have equal mark bits (modulo
being non-null), I suppose.

[sof, 2014-06-13 12:19:43]

On 2014/06/13 12:16:32, sof wrote:
> That that marking&trace order invalidates the assert makes sense.
> 
> You could instead assert that m_head and m_tail have equal mark bits (modulo
> being non-null), I suppose.

Hmm, no :) You could end up with a mark stack that contained one or the other by
the time you invoke the above trace().

[Mads Ager (chromium), 2014-06-13 12:20:38]

On 2014/06/13 12:19:43, sof wrote:
> On 2014/06/13 12:16:32, sof wrote:
> > That that marking&trace order invalidates the assert makes sense.
> > 
> > You could instead assert that m_head and m_tail have equal mark bits (modulo
> > being non-null), I suppose.
> 
> Hmm, no :) You could end up with a mark stack that contained one or the other
by
> the time you invoke the above trace().

Yeah, not sure we can actually add much of a meaningful assert about the marking
state of head and tail here. :)

[commit-bot: I haz the power, 2014-06-13 13:03:31]

Change committed as 176110

[haraken, 2014-06-13 13:19:05]

LGTM