https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123604

--- Comment #8 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
wc -l pr123604.sr15-28{89,90}
  84 pr123604.s15-2889
 123 pr123604.s15-2890
So, at least from line count in assembly, this seems to be a significant
growth.
cprop2 dump is still identical, ce1 contains huge amount of changes.
And, in r15-2889 e.g. the reginfo dump had 127 INSNs/JUMP_INSNs, and in
r16-2890 has 166.

Reply via email to