PING^n fwprop and pr/19653

Paolo Bonzini
Wed Apr 12 09:17:00 GMT 2006

> I investigated the problem with complex-6.c and the point of divergence
> is in .136r.lreg where we choose different register preferences
> from before your change.
Can you check if reverting Dale's patch fixes the failures?  I'm more 
and more inclined to do so since (as expected) the reload changes had no 
effect on everything but i386, while the regclass changes are triggering 
bugs or other crashes on powerpc, sh64, and now x86_64.  Which is not 
appropriate for stage3, I think.


More information about the Gcc-patches mailing list