2.6.12-rc4, -mm: bad ide-cs problems
Hi!
I see some problems in pcmcia subsystem in 2.6.12-rc4:
On compaq nx5000, inserting CF ide card is recognized as anonymous memory. On compaq evo, it is recognized okay and mounts/works. Unfortunately when I unplug the card, I get an oops:
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000010
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: printing eip:
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: *pde = 00000000
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: Oops: 0000 [#1]
. -mm kernel actually works better on nx5000; it behaves similary to -rc4 on evo; unfortunately it produces similar oops on card unplug.
Pavel
Quoted 15 lines · Pavel Machek
Hi!
I see some problems in pcmcia subsystem in 2.6.12-rc4:
On compaq nx5000, inserting CF ide card is recognized as anonymous memory. On compaq evo, it is recognized okay and mounts/works. Unfortunately when I unplug the card, I get an oops:
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000010
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: printing eip:
Quoted 8 lines
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: *pde = 00000000
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: Oops: 0000 [#1]
. -mm kernel actually works better on nx5000; it behaves similary to -rc4 on evo; unfortunately it produces similar oops on card unplug.
This is full oops from -rc4 on compaq evo:
hde: cache flushes not supported
hde: hde1
ide-cs: hde: Vcc = 3.3, Vpp = 0.0
hde: hde1
hde: hde1
Unable to handle kernel NULL pointer dereference at virtual address
00000010
printing eip:c033d618
*pde = 00000000
Oops: 0000 [#1] PREEMPT Modules linked in: CPU: 0 EIP: 0060:[<c033d618>] Not tainted VLI EFLAGS: 00010292 (2.6.12-rc4)
EIP is at ide_drive_remove+0x8/0x10
eax: c07825f8 ebx: c0782704 ecx: c07826f0 edx: 00000000
esi: c07826e0 edi: c065fed8 ebp: 00000001 esp: dfce5e84
ds: 007b es: 007b ss: 0068
Process pccardd (pid: 932, threadinfo=dfce4000 task=dfc84a40)
Stack: c02dc4b4 c07826e0 c065fa00 c0782568 c02dc724 c07826e0 c0782a80
c02db5e2
c07826e0 df373a80 c02db628 c07825f8 c033bcfb c0782568 00000002
df373b80
df373b80 c0670680 c0670728 c034f63b df758200 00000000 c034f67a
c0386108
Call Trace:
[<c02dc4b4>] device_release_driver+0x74/0x80
[<c02dc724>] bus_remove_device+0x64/0xa0
[<c02db5e2>] device_del+0x52/0x90
[<c02db628>] device_unregister+0x8/0x10
[<c033bcfb>] ide_unregister+0x3ab/0x4a0
[<c034f63b>] ide_release+0x5b/0x60
[<c034f67a>] ide_event+0x3a/0xc0
[<c0386108>] send_event_callback+0x28/0x30
[<c02dc0e0>] __bus_for_each_dev+0x50/0x80
[<c02dc1b8>] bus_for_each_dev+0x28/0x50
[<c03860e0>] send_event_callback+0x0/0x30
[<c038615f>] send_event+0x4f/0x70
[<c03860e0>] send_event_callback+0x0/0x30
[<c038621e>] ds_event+0x9e/0xb0
[<c0380375>] send_event+0x95/0x120
[<c0380418>] socket_shutdown+0x8/0x30
[<c0380958>] socket_remove+0x8/0x70
[<c0380a18>] socket_detect_change+0x58/0x70
[<c0380bdb>] pccardd+0x1ab/0x200
[<c0119810>] default_wake_function+0x0/0x10
[<c0102fd2>] ret_from_fork+0x6/0x14
[<c0119810>] default_wake_function+0x0/0x10
[<c0380a30>] pccardd+0x0/0x200
[<c010133d>] kernel_thread_helper+0x5/0x18
Code: c1 dd ff 8b 43 08 a8 08 75 07 31 c0 e9 6d ff ff ff e8 dd e7 1c00 eb f2 e8 d6 e7 1c 00 eb a7 8d 74 26 00 2d e8 00 00 00 8b 50 1c <ff> 52 10 31 c0 c3 89 f6 55 57 89 c7 b8 01 00 00 00 56 53 83 ec
Quoted 50 lines · Pavel Machek
Quoted 15 lines · Pavel Machek
Hi!
I see some problems in pcmcia subsystem in 2.6.12-rc4:
On compaq nx5000, inserting CF ide card is recognized as anonymous memory. On compaq evo, it is recognized okay and mounts/works. Unfortunately when I unplug the card, I get an oops:
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000010
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: printing eip:
Quoted 8 lines
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: *pde = 00000000
Message from syslogd@Elf at Wed May 25 13:25:25 2005 ... Elf kernel: Oops: 0000 [#1]
. -mm kernel actually works better on nx5000; it behaves similary to -rc4 on evo; unfortunately it produces similar oops on card unplug.
This is full oops from -rc4 on compaq evo:
hde: cache flushes not supported
hde: hde1
ide-cs: hde: Vcc = 3.3, Vpp = 0.0
hde: hde1
hde: hde1Unable to handle kernel NULL pointer dereference at virtual address00000010
printing eip:c033d618
*pde = 00000000Oops: 0000 [#1] PREEMPT Modules linked in: CPU: 0 EIP: 0060:[<c033d618>] Not tainted VLI EFLAGS: 00010292 (2.6.12-rc4)EIP is at
ide_drive_remove+0x8/0x10
eax:c07825f8ebx:c0782704ecx:c07826f0edx: 00000000
esi:c07826e0edi:c065fed8ebp: 00000001 esp:dfce5e84
ds: 007b es: 007b ss: 0068
Process pccardd (pid: 932, threadinfo=dfce4000 task=dfc84a40)
Quoted 8 lines
Stack: c02dc4b4 c07826e0 c065fa00 c0782568 c02dc724 c07826e0 c0782a80
c02db5e2
c07826e0 df373a80 c02db628 c07825f8 c033bcfb c0782568 00000002
df373b80
df373b80 c0670680 c0670728 c034f63b df758200 00000000 c034f67a
c0386108
Call Trace:
[<c02dc4b4>] device_release_driver+0x74/0x80Pavel,
I had a similar problem which I fixed amd posted to lkml a while ago though I seems to got ignored by the ide maintainer. Please see if this helps in your case:
http://marc.theaimsgroup.com/?l=linux-kernel&m=111409743421578&w=2
Signature · 1 line
Andreas Steinmetz SPAMmers use [email protected]
Hi!
Quoted 30 lines
Quoted 21 lines
hde: cache flushes not supported
hde: hde1
ide-cs: hde: Vcc = 3.3, Vpp = 0.0
hde: hde1
hde: hde1Unable to handle kernel NULL pointer dereference at virtual address00000010
printing eip:c033d618
*pde = 00000000Oops: 0000 [#1] PREEMPT Modules linked in: CPU: 0 EIP: 0060:[<c033d618>] Not tainted VLI EFLAGS: 00010292 (2.6.12-rc4)EIP is at
ide_drive_remove+0x8/0x10
eax:c07825f8ebx:c0782704ecx:c07826f0edx: 00000000
esi:c07826e0edi:c065fed8ebp: 00000001 esp:dfce5e84
ds: 007b es: 007b ss: 0068
Process pccardd (pid: 932, threadinfo=dfce4000 task=dfc84a40)Quoted 8 lines
Stack: c02dc4b4 c07826e0 c065fa00 c0782568 c02dc724 c07826e0 c0782a80 c02db5e2 c07826e0 df373a80 c02db628 c07825f8 c033bcfb c0782568 00000002 df373b80 df373b80 c0670680 c0670728 c034f63b df758200 00000000 c034f67a c0386108 Call Trace: [<c02dc4b4>] device_release_driver+0x74/0x80
Quoted 6 lines
Pavel,
I had a similar problem which I fixed amd posted to lkml a while ago though I seems to got ignored by the ide maintainer. Please see if this helps in your case:http://marc.theaimsgroup.com/?l=linux-kernel&m=111409743421578&w=2
Yes, thanks, that works. Andrew, is there reason not to push that patch into -mm/linus?
Pavel
Quoted 37 lines · Pavel Machek
Hi!
Quoted 35 lines
Quoted 21 lines
hde: cache flushes not supported
hde: hde1
ide-cs: hde: Vcc = 3.3, Vpp = 0.0
hde: hde1
hde: hde1Unable to handle kernel NULL pointer dereference at virtual address00000010
printing eip:c033d618
*pde = 00000000Oops: 0000 [#1] PREEMPT Modules linked in: CPU: 0 EIP: 0060:[<c033d618>] Not tainted VLI EFLAGS: 00010292 (2.6.12-rc4)EIP is at
ide_drive_remove+0x8/0x10
eax:c07825f8ebx:c0782704ecx:c07826f0edx: 00000000
esi:c07826e0edi:c065fed8ebp: 00000001 esp:dfce5e84
ds: 007b es: 007b ss: 0068
Process pccardd (pid: 932, threadinfo=dfce4000 task=dfc84a40)Quoted 8 lines
Stack: c02dc4b4 c07826e0 c065fa00 c0782568 c02dc724 c07826e0 c0782a80 c02db5e2 c07826e0 df373a80 c02db628 c07825f8 c033bcfb c0782568 00000002 df373b80 df373b80 c0670680 c0670728 c034f63b df758200 00000000 c034f67a c0386108 Call Trace: [<c02dc4b4>] device_release_driver+0x74/0x80Pavel,
I had a similar problem which I fixed amd posted to lkml a while ago though I seems to got ignored by the ide maintainer. Please see if this helps in your case:
Quoted 4 lines
Yes, thanks, that works. Andrew, is there reason not to push that patch into -mm/linus?
Full fix is in -mm (but still needs few changes).