Re: evbmips-mips64el 9.99.17 userland segfaults

["John D. Baker" <jdbaker%consolidated.net@localhost>]

On Sat, 19 Oct 2019, Michael wrote:

> On Fri, 18 Oct 2019 22:46:06 -0500 (CDT)
> "John D. Baker" <jdbaker%consolidated.net@localhost> wrote:
> 
> > After a long hiatus, I finally got around to booting (tftp, NFS root)
> > my Lemote YeeLoong with 9.99.17 from a couple of days ago.
> > 
> > Userland programs are very likely to segfault.  The various tools
> > involved in 'etcupdate' procedures (particularly 'md5') would segfault
> > frequently during the process.
> > 
> > Anyone tried -current on mips64 (with n32 userland) lately?  See anything
> > similar?
> 
> Hmm, that's odd. I'll dust off my gdium and have a look. It's been a
> while since I messed with that. May have something to do with the gcc
> update.

Since the 'etcupdate' attempt had as its current directory the
RELEASEDIR/binary/sets directory, I found zero-length core files for
the various userland programs that were being used (and segfaulting):

cat.core
cmp.core
cp.core
dirname.core
grep.core
ident.core
md5.core
mkdir.core
rm.core
sed.core
sh.core
sort.core
stat.core
sysctl.core
uname.core

Thanks for looking at this.

-- 
|/"\ John D. Baker, KN5UKS               NetBSD     Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net  OpenBSD            FreeBSD
| X  No HTML/proprietary data in email.   BSD just sits there and works!
|/ \ GPGkeyID:  D703 4A7E 479F 63F8 D3F4  BD99 9572 8F23 E4AD 1645