This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Stack-based pseudos


I was thinking about stack-based pseudos in stack-reorg.c last night, and
realized they might be useful for solving other problems as well.

The previous problem I reported with function prologue/epilogue
scheduling:

http://gcc.gnu.org/ml/gcc-bugs/2003-02/msg01290.html
http://gcc.gnu.org/ml/gcc-bugs/2003-02/msg01322.html

...might be solvable by having a new barrier type which prevents movement
of stack-based pseudos across the barrier.

In addition, it seems applicable to;

http://gcc.gnu.org/ml/gcc-bugs/2003-03/msg00004.html

...to improve alias analysis on stack-based arrays.

Am I expecting too much, or are these solvable with stack-based pseudos?

Toshi



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]