[Webkit-unassigned] [Bug 22736] State-of-art peephole state machine

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Thu Jan 22 18:34:23 PST 2009


https://bugs.webkit.org/show_bug.cgi?id=22736


oliver at apple.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #25843|review?                     |review-
               Flag|                            |




------- Comment #6 from oliver at apple.com  2009-01-22 18:34 PDT -------
(From update of attachment 25843)
While this patch looks like it would be a great addition and improvement over
the existing simplistic optimiser there are a number of issues that we'll need
to address.  The most significant of these is the use of the STL -- basically
WebKit does not use STL, and for the more complex STL structures (such as
vector, etc) we run the risk of mismatching malloc implementations.

My other concern is that the state tables appear to be hand coded, and I would
much rather we had a way to autogenerate them.

Ignoring those issues the code itself looks good (bar some minor style errors)
and probably just needs to be brought up to tip of tree (we do now have length
info for each opcode which should remove/reduce that portion of this patch).

Will CC myself to this bug so i don't miss updates.

Many apologies for the huge delay in review :-(


-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.



More information about the webkit-unassigned mailing list