This is the mail archive of the
gdb@sources.redhat.com
mailing list for the GDB project.
Re: recent i386-tdep change breaks build
- To: robertlipe at usa dot net
- Subject: Re: recent i386-tdep change breaks build
- From: Mark Kettenis <kettenis at science dot uva dot nl>
- Date: Wed, 4 Apr 2001 12:28:12 +0200 (MET DST)
- CC: gdb at sources dot redhat dot com
- References: <20010404001258.A7679@rjlinux.sco.com>
Date: Wed, 4 Apr 2001 00:12:58 -0500
From: Robert Lipe <robertlipe@usa.net>
I haven't roamed the CVS logs to be sure Mark's change is the smoking
gun, but this can't be right.
void
i386_extract_return_value (struct type *type, char *regbuf, char *valbuf)
{
int len = TYPE_LENGTH (type);
if (TYPE_CODE (type) == TYPE_CODE_STRUCT
&& TYPE_NFIELDS (type) == 1)
return i386_extract_return_value (TYPE_FIELD_TYPE (type, 0),
regbuf, valbuf);
(and another occurrence of similar return) We're tryint to return a value
from a void function? Is this perhaps another GCC-ism (sigh) that sneaked
in?
Argh, GCC doesn't warn about this. Anyway, this is my mistake. I was
somehow thinking that i386_extract_return_value *did* have a non-void
return type.
I'll fix it when the CVS machine is reachable again.
Thanks for spotting this,
Mark