Chang Heap::FreeQueuedChunks to prepare queued large chunks for StoreBuffer::Filter.

Chang Heap::FreeQueuedChunks to prepare queued large chunks for StoreBuffer::Filter.

We split large chunks into separate smaller kPageSize aligned pieces to ensure that
MemoryChunk::FromAnyPointerAddress would be able to find chunk containing slot even if the chunk was unlinked from the list of large chunks.

R=lrn@chromium.org
BUG=v8:1554
Committed: http://code.google.com/p/v8/source/detail?r=8621

[Lasse Reichstein, 2011-07-13 07:07:13]

LGTM, if you can make the comment.

http://codereview.chromium.org/7356005/diff/1/src/heap.cc
File src/heap.cc (right):

http://codereview.chromium.org/7356005/diff/1/src/heap.cc#newcode5993
src/heap.cc:5993: while (inner <= inner_last) {
It's not obvious why this can't crash, if the end of the large object chunk is
just a few bytes into the inner_last chunk.
Make a comment saying why it won't crash.