Bug#1043281: Aborts with "illegal hardware instruction" on i586 AMD Geode
martin f krafft
Package: sudo
Version: 1.9.13p3-3
Severity: normal
I have two systems running the AMD Geode processor¹ and consequently the non-PAE kernel:
Linux gig 5.10.0-23-686 #1 SMP Debian 5.10.179-2 (2023-07-14) i586 GNU/Linux and
Linux emily 6.1.0-10-686 #1 SMP PREEMPT_DYNAMIC Debian 6.1.38-2 (2023-07-27) i586 GNU/Linux Both have had the bookworm upgrade, but I've not rebooted the former yet, hence the 5.x kernel.
Anywhow, I cannot use sudo on either anymore, because if I do:
emily:~% sudo -l
#(132 pts/0) 08 14:05:20.790
[1] 29199 illegal hardware instruction sudo -l which shows up in dmesg like so:
2023-08-08T14:06:32.666253+02:00 emily kernel: [206827.519200] traps: sudo[29332] trap invalid opcode ip:4ac030 sp:bfb3eedc error:0 in sudo[4ab000+2e000] The system otherwise works fine, it's just sudo that's
not usable.
¹) Here's the CPU data
emily:~% cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 5
model : 10
model name : Geode(TM) Integrated Processor by AMD PCS
stepping : 2
cpu MHz : 498.036
cache size : 128 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
fdiv_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu de pse tsc msr cx8 sep pge cmov clflush mmx mmxext 3dnowext 3dnow cpuid 3dnowprefetch vmmcall
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass
bogomips : 996.07
clflush size : 32
cache_alignment : 32
address sizes : 32 bits physical, 32 bits virtual
power management: .''`. martin f. krafft <madduck@d.o>
: :' : proud Debian developer
`. `'` http://people.debian.org/~madduck
`- Debian - when you have better things to do than fixing systems
Marc Haber
> Anywhow, I cannot use sudo on either anymore, because if I do:
>
> ```
> emily:~% sudo -l
> #(132 pts/0) 08 14:05:20.790
> [1] 29199 illegal hardware instruction sudo -l
If so, this probably boils down on a GCC bug, see
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104713
Greetings
Marc
--
-----------------------------------------------------------------------------
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany | lose things." Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature | How to make an American Quilt | Fax: *49 6224 1600421
martin f krafft
Regarding the following, written by "Marc Haber" on 2023-08-08 at 14:57 Uhr +0200:
Is this probably a duplicate of #1004894?
If so, this probably boils down on a GCC bug, see https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104713
Thanks and sorry for the duplicate. Not sure it is. I did a search, but "opcode" only appeared on my radar later…
.''`. martin f. krafft <madduck@d.o>
: :' : proud Debian developer
`. `'` http://people.debian.org/~madduck
`- Debian - when you have better things to do than fixing systems
"auch der mutigste von uns hat nur selten den mut zu dem,
was er eigentlich weiß."
- friedrich nietzsche
Marc Haber
On Thu, Aug 10, 2023 at 08:36:54AM +1200, martin f krafft wrote:
> Regarding the following, written by "Marc Haber" on 2023-08-08 at 14:57 Uhr +0200:
> > Is this probably a duplicate of #1004894?
> >
> > If so, this probably boils down on a GCC bug, see
> > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104713
>
> Thanks and sorry for the duplicate. Not sure it is. I did a search, but
> "opcode" only appeared on my radar later…
or disabling the hardening option as suggested in 1004894? If either
works, we have a duplicate and sudo sadly cannot do anything about
that.I'd rather not disable hardening for sudo to make it run on obscure
old CPUs until the ctte tells me to.
martin f krafft
Regarding the following, written by "Marc Haber" on 2023-08-14 at 14:03 Uhr +0200:
On Thu, Aug 10, 2023 at 08:36:54AM +1200, martin f krafft wrote: > Regarding the following, written by "Marc Haber" on 2023-08-08 at 14:57 Uhr +0200: > > Is this probably a duplicate of #1004894?
> >
> > If so, this probably boils down on a GCC bug, see > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104713
>
> Thanks and sorry for the duplicate. Not sure it is. I did a search, but > "opcode" only appeared on my radar later…What makes you unsure it is?
Lack of knowledge? It's not important what I am sure and unsure of ;)
I won't have time to recompile any time soon, and in part because I won't be close to those systems in the foreseeable future. They are not important (both are just IPMI jumphosts).
I agree that you shouldn't loosen security on sudo just for a few older systems. It's probably time I replaced them anyway, where are the raspberry pis with two LAN interfaces? ;)
.''`. martin f. krafft <madduck@d.o>
: :' : proud Debian developer
`. `'` http://people.debian.org/~madduck
`- Debian - when you have better things to do than fixing systems
"... alle sätze der logik sagen aber dasselbe. nämlich nichts."
-- wittgenstein
Marc Haber
> I won't have time to recompile any time soon, and in part because I won't be
> close to those systems in the foreseeable future. They are not important
> It's probably time I replaced them anyway, where are the raspberry
> pis with two LAN interfaces? ;)
martin f krafft
Regarding the following, written by "Marc Haber" on 2023-08-15 at 14:11 Uhr +0200:
I won't have time to recompile any time soon, and in part because I won't be > close to those systems in the foreseeable future. They are not important
Would you be willing to try a .deb delivered by me?
Yes of course.
.''`. martin f. krafft <madduck@d.o>
: :' : proud Debian developer
`. `'` http://people.debian.org/~madduck
`- Debian - when you have better things to do than fixing systems
"one should never trust a woman who tells her real age.
if she tells that, she will tell anything."
-- oscar wilde