Date: Mon, 20 Jan 1997 18:07:12 +1030 (CST) From: Michael Smith <msmith@atrad.adelaide.edu.au> To: uhclem@nemesis.lonestar.org (Frank Durda IV) Cc: freebsd-hackers@FreeBSD.org, uhclem@nemesis.lonestar.org Subject: Re: sup dumping core at end of downloading a set Message-ID: <199701200737.SAA00440@genesis.atrad.adelaide.edu.au> In-Reply-To: <m0vmDIS-000ug2C@nemesis.lonestar.org> from Frank Durda IV at "Jan 20, 97 00:33:00 am"
next in thread | previous in thread | raw e-mail | index | archive | help
Frank Durda IV stands accused of saying:
> The core file was left in /server/stats/bsdcvs/usr/sup (not where
> I invoked sup), and it says:
>
> dbx version 3.11.10
> Type 'help' for help.
> Core file created by program "sup"
>
> signal Segmentation fault at >*[fprintf, 0x3ff800d5ff4] ldq r16, 40(r1)
> (dbx) tstack
> Thread 0x3:
> > 0 fprintf(0xc0080380, 0x140002ca8, 0x1400213a0, 0x11fffdd68, 0x0) [0x3ff800d5ff4]
> 1 finishone(0x1400213a0, 0x11fffdd68, 0x0, 0x5, 0x12000ee40) [0x12000b7fc]
finishone (t,finishfile)
TREE *t;
FILE **finishfile;
{
if ((thisC->Cflags&CFDELETE) == 0 || (t->Tflags&FUPDATE))
if(t->Tflags&FRENAME)
fprintf(*finishfile,"%s %s\n",t->Tname,t->Tnewname);
else
fprintf (*finishfile,"%s\n",t->Tname);
return (SCMOK);
}
(no idea where the extra 3 args to finishfile() come from. dbx looks like
a very inferior debugger 8( )
Still the args to fprintf look fine; it would be handy to know whether
the strings it's getting are properly terminated, and still valid, but I
would expect that they are.
> (dbx) printregs
>
> $vfp= 4831828752 $r0_v0=1
> $r1_t0=3221750656 $r2_t1=19922945
It looks like (*finishfile) is in r1, and it's looking up something in
the FILE structure. The offsets and alignment look OK, and I can't see
how (*finishfile) could have been closed (but that's worth checking)...
> I tried to build cvsup to see if it would work any better, but cvsup won't
> even build on FreeBSD 2.1.5, much less on OSF. The port expects something
> called m3build to be on your system, which is NOT the case on a FreeBSD
> 2.1.5 system installed with the Developer option. Nor is there a word
> in the port files saying what m3build might be. I consider this a bug
> in the port. There are other ports that depend on yet-other ports and have
> the manners to get the pieces they need, or at least have a README
> or something to tell you that you need this or that.
The cvsup port clearly depends on the Modula-3 port :
BUILD_DEPENDS= m3build-4:${PORTSDIR}/lang/modula-3
LIB_DEPENDS= m3\\.4\\.:${PORTSDIR}/lang/modula-3-lib
> Despite that, I believe this might be some Modula-3 thing, but these
> additional requirements make it even less likely that I can get cvsup
> running under OSF. (Attempting to build the Modula-3 port on FreeBSD
> fails because the Modula-3-lib isn't already present. How far does
> this go on?)
Likewise, the modula-3 port clearly depends on the modula-3-lib port :
DEPENDS= ${PORTSDIR}/lang/modula-3-lib
...which contains (in the Makefile) substantial discourse on the
Modula-3 bootstrapping process. Modula-3 is also a DEC project, and
M3 binaries are probably available for OSF.
> So, I would like to stick with sup for this exercise. Does anyone have
> any ideas as to why sup would roll over the way it does? I currently
> don't have the option to run this program under FreeBSD and the OSF
> system has the disk space.
Unfortunately, I don't.
> Thanks for any suggestions.
>
> Frank Durda IV <uhclem@nemesis.lonestar.org>|"The Knights who say "LETNi"
> or uhclem%nemesis@rwsystr.nkn.net | demand... A SEGMENT REGISTER!!!"
> |"A what?"
> or ...letni!rwsys!nemesis!uhclem |"LETNi! LETNi! LETNi!" - 1983
>
>
--
]] Mike Smith, Software Engineer msmith@gsoft.com.au [[
]] Genesis Software genesis@gsoft.com.au [[
]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[
]] realtime instrument control. (ph) +61-8-8267-3493 [[
]] Unix hardware collector. "Where are your PEZ?" The Tick [[
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701200737.SAA00440>