This is the mail archive of the
mailing list for the GCC project.
Re: A warning patch for config/sparc/*
- To: rth at cygnus dot com
- Subject: Re: A warning patch for config/sparc/*
- From: "Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>
- Date: Sat, 6 Mar 1999 21:54:02 -0500 (EST)
- Cc: egcs-patches at egcs dot cygnus dot com
> From: Richard Henderson <email@example.com>
> On Tue, Mar 02, 1999 at 11:17:23AM -0500, Kaveh R. Ghazi wrote:
> > -static char sccsid = "@(#)gmon.c 5.3 (Berkeley) 5/22/91";
> > +static char sccsid ATTRIBUTE_UNUSED = "@(#)gmon.c 5.3 (Berkeley) 5/22/91";
> If this is retained at all, we should use #ident; that's what it's for.
I'll take it out.
> > - extern etext();
> > + extern int etext();
> > already_setup = 1;
> > - monstartup(0, etext);
> > + monstartup(0, (char *)etext);
> Better to declare as `extern char etext'.
> > #if HOST_BITS_PER_WIDE_INT != 64
> > && ((CONST_DOUBLE_HIGH (operands) == 0
> > && (CONST_DOUBLE_LOW (operands) & 0x80000000) == 0)
> > - || (CONST_DOUBLE_HIGH (operands) == 0xffffffff
> > + || (CONST_DOUBLE_HIGH (operands) == (HOST_WIDE_INT) 0xffffffff
> > && (CONST_DOUBLE_LOW (operands) & 0x80000000) != 0))
> > #endif
> You might also change that to `#if HOST_BITS_PER_WIDE_INT == 32'.
> I don't like presuming 32 and 64 are the only sizes we'll ever see.
> The rest is ok.
It turns out that this patch has a problem on SunOS4 cc which is
solved by http://egcs.cygnus.com/ml/egcs-patches/1999-03/msg00085.html.
Its a short patch and the explanation is contained in that posting, but
when I reread it, it sounded kind of confusing. If I didn't make sense
let me know. :-)
Would you please take a moment to approve that one too before I
checkin the sparc changes?
Kaveh R. Ghazi Engagement Manager / Project Services
firstname.lastname@example.org Qwest Internet Solutions