Nvclock crash ("*** stack smashing detected ***")

[Anton Kraus]

Binary package hint: nvclock

I'm running Jaunty with nvclock at version 0.8b3-1ubuntu1.

Whenever I try to use it with a parameter (e.g.: -i, -T or -s), it crashes with the following error message:

*** stack smashing detected ***: nvclock terminated
======= Backtrace: =========
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x48)[0xb7eacda8]
/lib/tls/i686/cmov/libc.so.6(__fortify_fail+0x0)[0xb7eacd60]
nvclock[0x805884c]
nvclock[0x8059c0c]
nvclock[0x805a1dd]
nvclock[0x805a044]
nvclock[0x804e020]
nvclock[0x804db09]
nvclock[0x80574ce]
nvclock[0x804a34b]
/lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5)[0xb7dc5775]
nvclock[0x80491e1]
======= Memory map: ========
08048000-08062000 r-xp 00000000 08:02 411292 /usr/bin/nvclock
08062000-08063000 rw-p 0001a000 08:02 411292 /usr/bin/nvclock
09220000-09241000 rw-p 09220000 00:00 0 [heap]
b7d18000-b7d25000 r-xp 00000000 08:02 24250 /lib/libgcc_s.so.1
b7d25000-b7d26000 r--p 0000c000 08:02 24250 /lib/libgcc_s.so.1
b7d26000-b7d27000 rw-p 0000d000 08:02 24250 /lib/libgcc_s.so.1
b7d36000-b7d46000 r--s 00000000 08:03 1794367 /home/toe/.nvclock/bios0.rom
b7d46000-b7d56000 rw-s de300000 00:0e 7894 /dev/nvidia0
b7d56000-b7d86000 rw-s de000000 00:0e 7894 /dev/nvidia0
b7d86000-b7d88000 rw-p b7d86000 00:00 0
b7d88000-b7d8c000 r-xp 00000000 08:02 413903 /usr/lib/libXdmcp.so.6.0.0
b7d8c000-b7d8d000 rw-p 00003000 08:02 413903 /usr/lib/libXdmcp.so.6.0.0
b7d8d000-b7d8f000 r-xp 00000000 08:02 25196 /lib/tls/i686/cmov/libdl-2.9.so
b7d8f000-b7d90000 r--p 00001000 08:02 25196 /lib/tls/i686/cmov/libdl-2.9.so
b7d90000-b7d91000 rw-p 00002000 08:02 25196 /lib/tls/i686/cmov/libdl-2.9.so
b7d91000-b7da9000 r-xp 00000000 08:02 413453 /usr/lib/libxcb.so.1.1.0
b7da9000-b7daa000 r--p 00017000 08:02 413453 /usr/lib/libxcb.so.1.1.0
b7daa000-b7dab000 rw-p 00018000 08:02 413453 /usr/lib/libxcb.so.1.1.0
b7dab000-b7dad000 r-xp 00000000 08:02 413515 /usr/lib/libXau.so.6.0.0
b7dad000-b7dae000 r--p 00001000 08:02 413515 /usr/lib/libXau.so.6.0.0
b7dae000-b7daf000 rw-p 00002000 08:02 413515 /usr/lib/libXau.so.6.0.0
b7daf000-b7f0b000 r-xp 00000000 08:02 24889 /lib/tls/i686/cmov/libc-2.9.so
b7f0b000-b7f0c000 ---p 0015c000 08:02 24889 /lib/tls/i686/cmov/libc-2.9.so
b7f0c000-b7f0e000 r--p 0015c000 08:02 24889 /lib/tls/i686/cmov/libc-2.9.so
b7f0e000-b7f0f000 rw-p 0015e000 08:02 24889 /lib/tls/i686/cmov/libc-2.9.so
b7f0f000-b7f13000 rw-p b7f0f000 00:00 0
b7f13000-b7ffd000 r-xp 00000000 08:02 411405 /usr/lib/libX11.so.6.2.0
b7ffd000-b7ffe000 ---p 000ea000 08:02 411405 /usr/lib/libX11.so.6.2.0
b7ffe000-b7fff000 r--p 000ea000 08:02 411405 /usr/lib/libX11.so.6.2.0
b7fff000-b8001000 rw-p 000eb000 08:02 411405 /usr/lib/libX11.so.6.2.0
b8001000-b8002000 rw-p b8001000 00:00 0
b8002000-b800f000 r-xp 00000000 08:02 413570 /usr/lib/libXext.so.6.4.0
b800f000-b8011000 rw-p 0000c000 08:02 413570 /usr/lib/libXext.so.6.4.0
b801a000-b801c000 rw-s de680000 00:0e 7894 /dev/nvidia0
b801c000-b801e000 rw-s de601000 00:0e 7894 /dev/nvidia0
b801e000-b801f000 rw-s de100000 00:0e 7894 /dev/nvidia0
b801f000-b8020000 rw-s de101000 00:0e 7894 /dev/nvidia0
b8020000-b8022000 rw-p b8020000 00:00 0
b8022000-b8023000 r-xp b8022000 00:00 0 [vdso]
b8023000-b803f000 r-xp 00000000 08:02 24349 /lib/ld-2.9.so
b803f000-b8040000 r--p 0001b000 08:02 24349 /lib/ld-2.9.so
b8040000-b8041000 rw-p 0001c000 08:02 24349 /lib/ld-2.9.so
bf92c000-bf941000 rw-p bffeb000 00:00 0 [stack]
Aborted

Running nvclock without a parameter works as expected and brings up the list of possible commands.

My graphics card is a Geforce 6600 LE. Its lspci output:

01:00.0 VGA compatible controller: nVidia Corporation NV43 [GeForce 6600 LE] (rev a2)
 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
 Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 248 (1250ns min, 250ns max)
 Interrupt: pin A routed to IRQ 16
 Region 0: Memory at de000000 (32-bit, non-prefetchable) [size=16M]
 Region 1: Memory at e0000000 (32-bit, prefetchable) [size=256M]
 Region 2: Memory at dd000000 (32-bit, non-prefetchable) [size=16M]
 [virtual] Expansion ROM at dffe0000 [disabled] [size=128K]
 Capabilities: [60] Power Management version 2
  Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
  Status: D0 PME-Enable- DSel=0 DScale=0 PME-
 Capabilities: [44] AGP version 3.0
  Status: RQ=32 Iso- ArqSz=0 Cal=0 SBA+ ITACoh- GART64- HTrans- 64bit- FW+ AGP3- Rate=x1,x2,x4
  Command: RQ=32 ArqSz=0 Cal=0 SBA+ AGP+ GART64- 64bit- FW+ Rate=x4
 Kernel driver in use: nvidia
 Kernel modules: nvidia, nvidiafb

Back in Intrepid everything worked fine.

[Anton Kraus (done)]

I just discovered that bug 72117, which was marked as "Fix released" two years ago seems to be a duplicate of this.

[Obrouni (nick-fisk)]

I also get this problem:-

Jaunty (2.6.28-8-generic)

[Geoff123 (gsking1)]

Same Problem Here.
This is a critical program for me since I need it to control the nvidia fan speed which does not work for my card (6600 GT).
This has worked fine in last several versions of Ubuntu.

A newer version 0.8b4 from http://www.linuxhardware.org/nvclock/ compiles and works fine. So maybe his can be fixed by getting the new version; there is also a request for that in bug #326365

However, 0.8b3 crashes both from the ubuntu packages and when I compile myself.

Linux version 2.6.28-8-generic

Description: Ubuntu jaunty (development branch)
Release: 9.04

nvclock:
  Installed: 0.8b3-1ubuntu1
  Candidate: 0.8b3-1ubuntu1
  Version table:
 *** 0.8b3-1ubuntu1 0
        500 http://us.archive.ubuntu.com jaunty/universe Packages
        100 /var/lib/dpkg/status

[Anton Kraus (done)]

The newest nvclock version in the Jaunty repositories (0.8b4-1ubuntu1) fixed this bug for me. :)

[Steve Langasek (vorlon)]

Thanks, closing the bug per the latest follow-up.

[Roger Binns (ubuntu-rogerbinns)]

0.8b4-1ubunt2 still crashes on AMD64

[Roger Binns (ubuntu-rogerbinns)]

On AMD64 (latest Jaunty, all updates applied) it still crashes 0.8b4-1ubuntu2 although it is a little later in running. For example nvclock -i (get info) shows a few things then voltages:

$ nvclock -i
[ nice information elided ]
Voltage level 2: 1.05V, VID: 2
Voltage level 3: 1.10V, VID: 3

*** stack smashing detected ***: nvclock terminated
======= Backtrace: =========
/lib/libc.so.6(__fortify_fail+0x37)[0x3e078ff2c7]
/lib/libc.so.6(__fortify_fail+0x0)[0x3e078ff290]
nvclock[0x4059b3]
/lib/libc.so.6(__libc_start_main+0xe6)[0x3e0781e5a6]
nvclock[0x401889]
======= Memory map: ========
00400000-00421000 r-xp 00000000 08:02 79881523 /usr/bin/nvclock
00620000-00621000 r--p 00020000 08:02 79881523 /usr/bin/nvclock
00621000-00622000 rw-p 00021000 08:02 79881523 /usr/bin/nvclock
00622000-00623000 rw-p 00622000 00:00 0
0063e000-0065f000 rw-p 0063e000 00:00 0 [heap]
.... system shared libraries from this point .....

Other commands seem to work without the crashes, but I won't try most of them since I don't want to possibly damage my hardware.

[Roger Binns (ubuntu-rogerbinns)]

Debug symbols aren't available so I can't get anything more helpful. gdb doesn't show anything useful except the call to abort. valgrind doesn't complain about anything at all.

[TSanchess (tsanchess)]

I also get this problem on freshly installed Jaunty with nvclock (0.8b4-1ubuntu2)
2.6.28-11-generic(i686)

[Martin P (martin-pittamitz)]

I am experiencing the same problem, 2.6.28-11-generic (x86_64), nvclock 0.8b4-1ubuntu2.

[gnufreex (gnufreex)]

I to have that problem. Jaunty 64-bit, GF 8800GT, newest nvclock version (and only) in Ubuntu repositories. Is there a workaround yet?

[gnufreex (gnufreex)]

nvclock actually does its job. I only need nvclock to adjust GPU fan rpm, and that part (thanks God) work fine. Command "nvclock -f --fanspeed XX" (where XX is number of %) does not give any error and does its job. But it would be nice if that "stack smacking" issue get fixed...

[aftertaf (david-wooffindin)]

Would like to open bug again for karmic koala.
0.8 beta4 in shell says cant find /dev/nvidia0, and both _qt and _gtk version core dump/seg fault.

this is since update to karmic, using -185 nvidia drivers from repositories.

lspci:
01:00.0 VGA compatible controller: nVidia Corporation NV40 [GeForce 6800] (rev a1)

[gnufreex (gnufreex)]

I updated to Karmic Koala, and I think there is no any bug anymore. When I type 'nvclock --info', there is no that "stack smacking" anymore. 'nvclock -f --fanspeed XX' works fine as well. Well done fixing that bug.

@aftertaf

What qt and gtk+ versions? I use only terminal version. Did not know that threre is any graphical one. :D