2006-02-09 02:37:32

by Anthony Tippett

[permalink] [raw]
Subject: kernel BUG at mm/swap.c:49

Can someone direct to whom I should send this kernel BUG or what else I
should send as debugging information.

It looks like it might be an issue with Xorg and fglrx (ati drivers)

Feb 8 16:50:42 act kernel: kernel BUG at mm/swap.c:49!
Feb 8 16:50:42 act kernel: invalid operand: 0000 [#1]
Feb 8 16:50:42 act kernel: Modules linked in: fglrx binfmt_misc nfsd
exportfs lockd nfs_acl sunrpc parport_pc lp parport autofs4 ipv6 deflate
zlib_deflate twofish serpent aes blowfish des sha256 sha1 crypto_null
af_key snd_bt87x emu10k1_gp gameport snd_emu10k1_synth snd_emux_synth
snd_seq_virmidi snd_seq_midi_emul snd_seq_dummy snd_seq_oss snd_seq_midi
snd_seq_midi_event snd_seq snd_emu10k1 snd_rawmidi snd_seq_device
snd_util_mem snd_hwdep ohci1394 ieee1394 snd_intel8x0
snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer
snd soundcore snd_page_alloc forcedeth ehci_hcd ohci_hcd ide_cd cdrom
ide_disk ide_generic usb_storage nvidia_agp agpgart usbhid usbcore
tvaudio tuner msp3400 bttv video_buf firmware_class v4l2_common
btcx_risc tveeprom videodev i2c_algo_bit i2c_core amd74xx ide_core
joydev ext3 jbd mbcache sd_mod sata_sil libata scsi_mod shpchp
pci_hotplug evdev mousedev
Feb 8 16:50:42 act kernel: CPU: 0
Feb 8 16:50:42 act kernel: EIP: 0060:[put_page+78/105] Tainted:
P B VLI
Feb 8 16:50:42 act kernel: EFLAGS: 00213256 (2.6.15-1-k7)
Feb 8 16:50:42 act kernel: EIP is at put_page+0x4e/0x69
Feb 8 16:50:42 act kernel: eax: 00000000 ebx: c12e6560 ecx:
c12e6560 edx: c12e6560
Feb 8 16:50:42 act kernel: esi: d658d5a0 edi: 00000020 ebp:
b6968000 esp: da65be94
Feb 8 16:50:42 act kernel: ds: 007b es: 007b ss: 0068
Feb 8 16:50:42 act kernel: Process Xorg (pid: 4479, threadinfo=da65a000
task=daf54050)
Feb 8 16:50:42 act kernel: Stack: c013b59f c12e6560 c12e6560 dea10a20
ffffffff 00000000 b6968000 dda2fb68
Feb 8 16:50:42 act kernel: b6969000 dda2fb68 c013b6e9 c032443c
d942f6a4 dda2fb68 b6968000 b6969000
Feb 8 16:50:42 act kernel: da65bf14 00000000 dda2fb68 b6968000
d942f6a4 b6969000 da65bf48 c013b79f
Feb 8 16:50:42 act kernel: Call Trace:
Feb 8 16:50:42 act kernel: [zap_pte_range+358/476]
zap_pte_range+0x166/0x1dc
Feb 8 16:50:42 act kernel: [unmap_page_range+212/227]
unmap_page_range+0xd4/0xe3
Feb 8 16:50:42 act kernel: [unmap_vmas+167/339] unmap_vmas+0xa7/0x153
Feb 8 16:50:42 act kernel: [unmap_region+114/209] unmap_region+0x72/0xd1
Feb 8 16:50:42 act kernel: [do_munmap+201/232] do_munmap+0xc9/0xe8
Feb 8 16:50:42 act kernel: [sys_munmap+57/84] sys_munmap+0x39/0x54
Feb 8 16:50:42 act kernel: [sysenter_past_esp+84/117]
sysenter_past_esp+0x54/0x75
Feb 8 16:50:42 act kernel: Code: 00 46 a1 27 c0 83 42 04 ff 0f 98 c0 84
c0 74 35 8b 4a 30 89 54 24 04 ff e1 8b 02 89 d1 f6 c4 40 74 03 8b 4a 0c
8b 41 04 40 75 08 <0f> 0b 31 00 46 a1 27 c0 83 42 04 ff 0f 98 c0 84 c0
74 07 89 d0


2006-02-09 09:35:46

by Marc Burkhardt

[permalink] [raw]
Subject: Re: kernel BUG at mm/swap.c:49

* Anthony Tippett <[email protected]> [2006-02-08 19:24:47 -0800]:

> Can someone direct to whom I should send this kernel BUG or what else I
> should send as debugging information.
>
> It looks like it might be an issue with Xorg and fglrx (ati drivers)
>
> Feb 8 16:50:42 act kernel: kernel BUG at mm/swap.c:49!
> Feb 8 16:50:42 act kernel: invalid operand: 0000 [#1]
> Feb 8 16:50:42 act kernel: Modules linked in: fglrx binfmt_misc nfsd
> exportfs lockd nfs_acl sunrpc parport_pc lp parport autofs4 ipv6 deflate
> zlib_deflate twofish serpent aes blowfish des sha256 sha1 crypto_null
> af_key snd_bt87x emu10k1_gp gameport snd_emu10k1_synth snd_emux_synth
> snd_seq_virmidi snd_seq_midi_emul snd_seq_dummy snd_seq_oss snd_seq_midi
> snd_seq_midi_event snd_seq snd_emu10k1 snd_rawmidi snd_seq_device
> snd_util_mem snd_hwdep ohci1394 ieee1394 snd_intel8x0
> snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer
> snd soundcore snd_page_alloc forcedeth ehci_hcd ohci_hcd ide_cd cdrom
> ide_disk ide_generic usb_storage nvidia_agp agpgart usbhid usbcore
> tvaudio tuner msp3400 bttv video_buf firmware_class v4l2_common
> btcx_risc tveeprom videodev i2c_algo_bit i2c_core amd74xx ide_core
> joydev ext3 jbd mbcache sd_mod sata_sil libata scsi_mod shpchp
> pci_hotplug evdev mousedev

Uhm ... for what reason do you use fglrx and nvidia_agp at the same time?

I guess - as usual - you have to get on to your hardware vendor when using
closed source drivers.

Marc

2006-02-09 20:44:42

by Jim Crilly

[permalink] [raw]
Subject: Re: kernel BUG at mm/swap.c:49

On 02/09/06 10:35:49AM +0100, Marc Koschewski wrote:
> * Anthony Tippett <[email protected]> [2006-02-08 19:24:47 -0800]:
>
> > Can someone direct to whom I should send this kernel BUG or what else I
> > should send as debugging information.
> >
> > It looks like it might be an issue with Xorg and fglrx (ati drivers)
> >
> > Feb 8 16:50:42 act kernel: kernel BUG at mm/swap.c:49!
> > Feb 8 16:50:42 act kernel: invalid operand: 0000 [#1]
> > Feb 8 16:50:42 act kernel: Modules linked in: fglrx binfmt_misc nfsd
> > exportfs lockd nfs_acl sunrpc parport_pc lp parport autofs4 ipv6 deflate
> > zlib_deflate twofish serpent aes blowfish des sha256 sha1 crypto_null
> > af_key snd_bt87x emu10k1_gp gameport snd_emu10k1_synth snd_emux_synth
> > snd_seq_virmidi snd_seq_midi_emul snd_seq_dummy snd_seq_oss snd_seq_midi
> > snd_seq_midi_event snd_seq snd_emu10k1 snd_rawmidi snd_seq_device
> > snd_util_mem snd_hwdep ohci1394 ieee1394 snd_intel8x0
> > snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer
> > snd soundcore snd_page_alloc forcedeth ehci_hcd ohci_hcd ide_cd cdrom
> > ide_disk ide_generic usb_storage nvidia_agp agpgart usbhid usbcore
> > tvaudio tuner msp3400 bttv video_buf firmware_class v4l2_common
> > btcx_risc tveeprom videodev i2c_algo_bit i2c_core amd74xx ide_core
> > joydev ext3 jbd mbcache sd_mod sata_sil libata scsi_mod shpchp
> > pci_hotplug evdev mousedev
>
> Uhm ... for what reason do you use fglrx and nvidia_agp at the same time?
>
> I guess - as usual - you have to get on to your hardware vendor when using
> closed source drivers.
>
> Marc

I could be wrong, but I'm pretty sure nvidia_agp is the OSS nVidia AGPGART
driver, so I assume he has an nVidia-based motherboard and an ATI video
card.

Jim.

2006-02-09 21:02:24

by Marc Burkhardt

[permalink] [raw]
Subject: Re: kernel BUG at mm/swap.c:49

* Jim Crilly <[email protected]> [2006-02-09 15:43:18 -0500]:

> On 02/09/06 10:35:49AM +0100, Marc Koschewski wrote:
> > * Anthony Tippett <[email protected]> [2006-02-08 19:24:47 -0800]:
> >
> > > Can someone direct to whom I should send this kernel BUG or what else I
> > > should send as debugging information.
> > >
> > > It looks like it might be an issue with Xorg and fglrx (ati drivers)
> > >
> > > Feb 8 16:50:42 act kernel: kernel BUG at mm/swap.c:49!
> > > Feb 8 16:50:42 act kernel: invalid operand: 0000 [#1]
> > > Feb 8 16:50:42 act kernel: Modules linked in: fglrx binfmt_misc nfsd
> > > exportfs lockd nfs_acl sunrpc parport_pc lp parport autofs4 ipv6 deflate
> > > zlib_deflate twofish serpent aes blowfish des sha256 sha1 crypto_null
> > > af_key snd_bt87x emu10k1_gp gameport snd_emu10k1_synth snd_emux_synth
> > > snd_seq_virmidi snd_seq_midi_emul snd_seq_dummy snd_seq_oss snd_seq_midi
> > > snd_seq_midi_event snd_seq snd_emu10k1 snd_rawmidi snd_seq_device
> > > snd_util_mem snd_hwdep ohci1394 ieee1394 snd_intel8x0
> > > snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer
> > > snd soundcore snd_page_alloc forcedeth ehci_hcd ohci_hcd ide_cd cdrom
> > > ide_disk ide_generic usb_storage nvidia_agp agpgart usbhid usbcore
> > > tvaudio tuner msp3400 bttv video_buf firmware_class v4l2_common
> > > btcx_risc tveeprom videodev i2c_algo_bit i2c_core amd74xx ide_core
> > > joydev ext3 jbd mbcache sd_mod sata_sil libata scsi_mod shpchp
> > > pci_hotplug evdev mousedev
> >
> > Uhm ... for what reason do you use fglrx and nvidia_agp at the same time?
> >
> > I guess - as usual - you have to get on to your hardware vendor when using
> > closed source drivers.
> >
> > Marc
>
> I could be wrong, but I'm pretty sure nvidia_agp is the OSS nVidia AGPGART
> driver, so I assume he has an nVidia-based motherboard and an ATI video
> card.

OK, that could be the reason. I never had anything else than Intel stuff in my
machines. Except my graphics. Dunno why. Maybe destiny. So far I never had
probs with the chips. And ... anything works perfectly with my beloved Linux! :)

Marc

2006-02-10 12:34:46

by Alistair John Strachan

[permalink] [raw]
Subject: Re: kernel BUG at mm/swap.c:49

On Thursday 09 February 2006 03:24, Anthony Tippett wrote:
> Can someone direct to whom I should send this kernel BUG or what else I
> should send as debugging information.
>
> It looks like it might be an issue with Xorg and fglrx (ati drivers)

Could be, no way to tell from the trace. Try to reproduce without the driver.

Also you don't way what architecture this is, or what kernel version you're
using. Assuming it's not 2.6.15, please upgrade and repeat the test. If you
still have a problem then please reply to this thread.

> Feb 8 16:50:42 act kernel: kernel BUG at mm/swap.c:49!
> Feb 8 16:50:42 act kernel: invalid operand: 0000 [#1]
> Feb 8 16:50:42 act kernel: Modules linked in: fglrx binfmt_misc nfsd
> exportfs lockd nfs_acl sunrpc parport_pc lp parport autofs4 ipv6 deflate
> zlib_deflate twofish serpent aes blowfish des sha256 sha1 crypto_null
> af_key snd_bt87x emu10k1_gp gameport snd_emu10k1_synth snd_emux_synth
> snd_seq_virmidi snd_seq_midi_emul snd_seq_dummy snd_seq_oss snd_seq_midi
> snd_seq_midi_event snd_seq snd_emu10k1 snd_rawmidi snd_seq_device
> snd_util_mem snd_hwdep ohci1394 ieee1394 snd_intel8x0
> snd_ac97_codec snd_ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_timer
> snd soundcore snd_page_alloc forcedeth ehci_hcd ohci_hcd ide_cd cdrom
> ide_disk ide_generic usb_storage nvidia_agp agpgart usbhid usbcore
> tvaudio tuner msp3400 bttv video_buf firmware_class v4l2_common
> btcx_risc tveeprom videodev i2c_algo_bit i2c_core amd74xx ide_core
> joydev ext3 jbd mbcache sd_mod sata_sil libata scsi_mod shpchp
> pci_hotplug evdev mousedev
> Feb 8 16:50:42 act kernel: CPU: 0
> Feb 8 16:50:42 act kernel: EIP: 0060:[put_page+78/105] Tainted:
> P B VLI
> Feb 8 16:50:42 act kernel: EFLAGS: 00213256 (2.6.15-1-k7)
> Feb 8 16:50:42 act kernel: EIP is at put_page+0x4e/0x69
> Feb 8 16:50:42 act kernel: eax: 00000000 ebx: c12e6560 ecx:
> c12e6560 edx: c12e6560
> Feb 8 16:50:42 act kernel: esi: d658d5a0 edi: 00000020 ebp:
> b6968000 esp: da65be94
> Feb 8 16:50:42 act kernel: ds: 007b es: 007b ss: 0068
> Feb 8 16:50:42 act kernel: Process Xorg (pid: 4479, threadinfo=da65a000
> task=daf54050)
> Feb 8 16:50:42 act kernel: Stack: c013b59f c12e6560 c12e6560 dea10a20
> ffffffff 00000000 b6968000 dda2fb68
> Feb 8 16:50:42 act kernel: b6969000 dda2fb68 c013b6e9 c032443c
> d942f6a4 dda2fb68 b6968000 b6969000
> Feb 8 16:50:42 act kernel: da65bf14 00000000 dda2fb68 b6968000
> d942f6a4 b6969000 da65bf48 c013b79f
> Feb 8 16:50:42 act kernel: Call Trace:
> Feb 8 16:50:42 act kernel: [zap_pte_range+358/476]
> zap_pte_range+0x166/0x1dc
> Feb 8 16:50:42 act kernel: [unmap_page_range+212/227]
> unmap_page_range+0xd4/0xe3
> Feb 8 16:50:42 act kernel: [unmap_vmas+167/339] unmap_vmas+0xa7/0x153
> Feb 8 16:50:42 act kernel: [unmap_region+114/209] unmap_region+0x72/0xd1
> Feb 8 16:50:42 act kernel: [do_munmap+201/232] do_munmap+0xc9/0xe8
> Feb 8 16:50:42 act kernel: [sys_munmap+57/84] sys_munmap+0x39/0x54
> Feb 8 16:50:42 act kernel: [sysenter_past_esp+84/117]
> sysenter_past_esp+0x54/0x75
> Feb 8 16:50:42 act kernel: Code: 00 46 a1 27 c0 83 42 04 ff 0f 98 c0 84
> c0 74 35 8b 4a 30 89 54 24 04 ff e1 8b 02 89 d1 f6 c4 40 74 03 8b 4a 0c
> 8b 41 04 40 75 08 <0f> 0b 31 00 46 a1 27 c0 83 42 04 ff 0f 98 c0 84 c0
> 74 07 89 d0

--
Cheers,
Alistair.

'No sense being pessimistic, it probably wouldn't work anyway.'
Third year Computer Science undergraduate.
1F2 55 South Clerk Street, Edinburgh, UK.

2006-02-10 14:20:21

by Hugh Dickins

[permalink] [raw]
Subject: Re: kernel BUG at mm/swap.c:49

On Fri, 10 Feb 2006, Alistair John Strachan wrote:
> On Thursday 09 February 2006 03:24, Anthony Tippett wrote:
> >
> > It looks like it might be an issue with Xorg and fglrx (ati drivers)
>
> Could be, no way to tell from the trace. Try to reproduce without the driver.
>
> > Feb 8 16:50:42 act kernel: kernel BUG at mm/swap.c:49!

It's just fglrx wrapper not uptodate with 2.6.15,
I sent him the wrapper patch offline.

Hugh