This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: One more global.c speedup
- To: law at cygnus dot com
- Subject: Re: One more global.c speedup
- From: Joern Rennecke <amylaar at cygnus dot co dot uk>
- Date: Fri, 12 Nov 1999 20:29:45 GMT
- Cc: amylaar at cygnus dot co dot uk, gcc-patches at gcc dot gnu dot org, lucier at math dot purdue dot edu
- Newsgroups: cygnus.egcs.patches
In article <11143.942372278@upchuck> you wrote:
: Thus, it is not necessary to mark conflicts between two pseudos just because
: they are live at the same time -- only need to mark the conflict when one is
: live at the point where the other becomes live. This allows the compiler to
: be more efficient when it builds the conflict matrix.
I think you can make the same argument for hard register - pseudo conflicts.
So record_conflicts can go altogether once the combine_reloads /
find_dummy_reload issues are resolved.