https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122954
Jose E. Marchesi <jemarch at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jemarch at gcc dot gnu.org
Resolution|--- |FIXED
Status|UNCONFIRMED |RESOLVED
--- Comment #2 from Jose E. Marchesi <jemarch at gcc dot gnu.org> ---
Hello David.
Thank you very much for checking. I have fixed all the reported problems, but
for two:
- I am keeping the code that triggers the knownConditionTrueFalse in the mode
equivalence code, because the first condition is a heuristic that may at some
point be removed. i.e. STANDARD modes are compared by pointer equality, it
also happens that other modes can be determined to be equal by pointer
equality, but these facts may change independently at some point.
- I am keeping the code that triggers the performance problem, because the ctx
argument is conceptually intended to be copied in a stack fashion. If we add
references to all the functions that happen to not alter ctx (which is most of
them) then this creates a maintenance problem.
Thanks!
