This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: PREFERRED_STACK_BOUNDARY clarification desired
- To: law at cygnus dot com
- Subject: Re: PREFERRED_STACK_BOUNDARY clarification desired
- From: Joern Rennecke <amylaar at cygnus dot co dot uk>
- Date: Thu, 4 Feb 1999 04:29:24 +0000 (GMT)
- Cc: john at feith dot com, egcs at egcs dot cygnus dot com
> I don't think any existing optimization pass tries to optimize stack pointer
> updates. I think part of the reason is we don't have REG_INC notes on insns
> which modify the stack pointer via autoinc addressing.
We got -fdefer-pop, but that's more of a code generation strategy.
Now, if we hacked regmove to run after reload and then operate on hard
registers....