Re: Fix PR breakpoints/2080

[Mark Kettenis <mark.kettenis@xs4all.nl>]

> X-Spam-Checker-Version: SpamAssassin 3.1.0 (2005-09-13) on 
> 	elgar.sibelius.xs4all.nl
> X-Spam-Level: 
> X-Spam-Status: No, score=0.0 required=5.0 tests=none autolearn=no 
> 	version=3.1.0
> Date: Tue, 14 Feb 2006 10:52:32 -0500
> From: Daniel Jacobowitz <drow@false.org>
> Cc: gdb-patches@sourceware.org, fnf@specifix.com
> Mail-Followup-To: Mark Kettenis <mark.kettenis@xs4all.nl>,
> 	gdb-patches@sourceware.org, fnf@specifix.com
> Content-Disposition: inline
> X-XS4ALL-DNSBL-Checked: mxdrop9.xs4all.nl checked 66.93.172.17 against DNS blacklists
> X-Virus-Scanned: by XS4ALL Virus Scanner
> X-XS4ALL-Spam-Score: 0 () 
> X-XS4ALL-Spam: NO
> Envelope-To: mark.kettenis@xs4all.nl
> X-UIDL: 1139932355._smtp.mxdrop9.54361,S=1797
> 
> On Tue, Feb 14, 2006 at 04:45:40PM +0100, Mark Kettenis wrote:
> > This fixes the testcase I checked in yesterday, so I committed it.
> 
> FYI, GCC generates this sequence as RTL rather than text, which means
> it can be scheduled and optimized.

Fortunately the scheduler doesn't have a lot of room to do this.  I'm
pretty sure the first three instructions (which are the most important
ones) are pretty much fixed with the current GCC implementation of the
stack alignment.

> Of course in optimized code we don't expect the prologue skipper to
> work very well, and we use dwarf2 to bypass the prologue analyzer,
> so this may not be a big deal.  But I'm somewhat curious to see if
> we could make this extremely sophisticated i386 prologue analyzer
> use Jim's prologue-value.[ch] interface...

Well, writing that prologue analyzer will be hell of a lot more
difficult for CISC than it is for RISC; even with Jim's
prologue-value.[ch].

Mark