Failure to apply trivial peephole optimizations

[Richard Henderson]

On Sat, Dec 14, 2002 at 01:42:15AM -0800, Martin Buchholz wrote:
> A trivial peephole optimizer with a two instruction window can convert
> the sequence
> 
> 	subl	$40, %esp
> 	movl	%ebp, %esp
> to 
> 	movl	%ebp, %esp

Peephole optimization is not the issue.  Policy with regards to
stack frame manipulation is.  What we want to do is enable
something akin to -momit-leaf-frame-pointer by default, but at
present this causes regressions in the gdb testsuite, so we can't.


r~