Patch ping ^ 2

Zdenek Dvorak rakdver@atrey.karlin.mff.cuni.cz
Mon Oct 18 08:37:00 GMT 2004


Hello,

I have thought about the situation a bit more.  My opinion now is:

1) Unless your patch to make voperands represent output dependences again
   (http://gcc.gnu.org/ml/gcc-patches/2004-10/msg01437.html) gets into
   mainline, the rewrite of store motion
   (http://gcc.gnu.org/ml/gcc-patches/2004-09/msg01120.html) is the best
   choice.

2) If your patch gets to mainline, it will be still necessary to rewrite
   store motion to fix its current compile time problems.  It will
   however be useful to use SSA form for voperands then.  This however
   will be just a fairly small change over
   http://gcc.gnu.org/ml/gcc-patches/2004-09/msg01120.html, since the
   only thing that will need to be changed is the way how the
   information about aliasing is determined.

Any of these options is fine with me, although 2) will be easier to
update when your changes to aliasing info representation are included
(it should just work without any further changes, in fact, but of course
one cannot be completely sure).

Note that there are no real dependences between the two patches, i.e.
they can be reviewed independently.

Zdenek



More information about the Gcc-patches mailing list