Re: [FIXED 2.5.69] Re: [2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

Re: [FIXED 2.5.69] Re: [2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

[Matthew Harrell]

: It was the fact that when I booted my system with my 3com cardbus NIC inserted 
: in the slot it would oops, giving me a non-working pcmcia/cardbus system.
: 
: Booting without the card worked fine, and inserting the card afterwards also 
: worked great.
: 
: thread is here:
: 
: http://marc.theaimsgroup.com/?l=linux-kernel&m=105107713129722&w=2

Oh, mine's different.  Total lockup with no messages.  Starts a few seconds 
after the yenta module is loaded.  I think it occurs during the yenta probe
but I haven't been able to prove that yet.  

: Ah. I don't use ACPI, because well... it caused me more pain than it helped me 
: in the past, in both OS' on my system (being Linux and win2k)

Yeah, it's a pain but this laptop no longer has APM so I need some way to read
my batteries and soft power the system off.  But, if it weren't for the 
batteries I wouldn't bother because it is a real pain to deal with

-- 
  Matthew Harrell                          I love defenseless animals,
  Bit Twiddlers, Inc.                       especially in a good gravy.
  mharrell@bittwiddlers.com     

[2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

[DevilKin]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

Kernels 2.5.6[78] hang when the pcmcia yenta_socket is initialised.

The last message I get is this:

Apr 23 06:54:22 laptop kernel: Linux Kernel Card Services 3.1.22
Apr 23 06:54:22 laptop kernel:   options:  [pci] [cardbus] [pm]
Apr 23 06:54:22 laptop kernel: Intel PCIC probe: not found.
Apr 23 06:54:22 laptop kernel: PCI: Found IRQ 11 for device 00:03.0
Apr 23 06:54:22 laptop kernel: PCI: Sharing IRQ 11 with 00:03.1
Apr 23 06:54:22 laptop kernel: PCI: Sharing IRQ 11 with 00:07.2
Apr 23 06:54:22 laptop kernel: Yenta IRQ list 0698, PCI irq11
Apr 23 06:54:22 laptop kernel: Socket status: 30000020
Apr 23 06:54:22 laptop kernel: PCI: Found IRQ 11 for device 00:03.1
Apr 23 06:54:22 laptop kernel: PCI: Sharing IRQ 11 with 00:03.0
Apr 23 06:54:22 laptop kernel: PCI: Sharing IRQ 11 with 00:07.2
Apr 23 06:54:22 laptop kernel: Yenta IRQ list 0698, PCI irq11
Apr 23 06:54:23 laptop kernel: Socket status: 30000010


After this, the system becomes unresponsive. Reboot is possible using the 
SysRq keys.

System information:

System:
Dell Latitude CPx J 650MHz - Pentium 3 650

Distribution:
Debian Unstable (Sid)

Hardware:
00:03.0 CardBus bridge: Texas Instruments PCI1225 (rev 01)
        Subsystem: Dell Computer Corporation: Unknown device 00bb
        Flags: bus master, medium devsel, latency 168, IRQ 11
        Memory at 10000000 (32-bit, non-prefetchable) [size=4K]
        Bus: primary=00, secondary=02, subordinate=05, sec-latency=176
        Memory window 0: 10400000-107ff000 (prefetchable)
        Memory window 1: 10800000-10bff000
        I/O window 0: 00001000-000010ff
        I/O window 1: 00001400-000014ff
        16-bit legacy interface ports at 0001

00:03.1 CardBus bridge: Texas Instruments PCI1225 (rev 01)
        Subsystem: Dell Computer Corporation: Unknown device 00bb
        Flags: bus master, medium devsel, latency 168, IRQ 11
        Memory at 10001000 (32-bit, non-prefetchable) [size=4K]
        Bus: primary=00, secondary=06, subordinate=09, sec-latency=176
        Memory window 0: 10c00000-10fff000 (prefetchable)
        Memory window 1: 11000000-113ff000
        I/O window 0: 00001800-000018ff
        I/O window 1: 00001c00-00001cff
        16-bit legacy interface ports at 0001

To reproduce: boot with any 2.5.6[78] kernel with pcmcia as modules, and have 
it loaded through pcmcia-cs at bootup

It works fine on kernel 2.5.66.

DK
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+pijspuyeqyCEh60RAl/7AJwOq6hubHFvffLFVO6RJ+yttCGoZgCfU4zS
fOYYQnc7w12OKOe50rx8WuE=
=Wuuf
-----END PGP SIGNATURE-----

Re: [2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

[DevilKin-LKML]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wednesday 23 April 2003 21:43, Russell King wrote:
> On Wed, Apr 23, 2003 at 08:50:39PM +0200, DevilKin-LKML wrote:
> > Unable to handle kernel paging request at virtual address d10545c0
> >  printing eip:
> > c01b0368
> > *pde = 0fb66067
> > *pte = 00000000
> > Oops: 0000 [#1]
> > CPU:    0
> > EIP:    0060:[<c01b0368>]    Not tainted
> > EFLAGS: 00010286
> > EIP is at pci_bus_match+0x18/0xb0
> > eax: 00000000   ebx: cf2a8800   ecx: d10545c0   edx: 00000000
> > esi: cf2a884c   edi: ffffffed   ebp: c12d284c   esp: ce591e88
> > ds: 007b   es: 007b   ss: 0068
> > Process pcmcia/0 (pid: 388, threadinfo=ce590000 task=ce6f7300)
> > Stack: d109edc8 c01d1abf cf2a884c d109edc8 d109edf8 cf2a884c c02d7ff8
> > c01d1b5f cf2a884c d109edc8 cf2a884c c02d7fa0 cf2a8888 c01d1d24 cf2a884c
> > c0282ddf c02de000 cf2a884c 00000000 cf2a8888 c01d0ef0 cf2a884c cf2a8800
> > cffd08b4 Call Trace:
> >  [<d109edc8>] m3_pci_driver+0x28/0xa0 [maestro3]
> >  [<c01d1abf>] bus_match+0x2f/0x80
> >  [<d109edc8>] m3_pci_driver+0x28/0xa0 [maestro3]
> >  [<d109edf8>] m3_pci_driver+0x58/0xa0 [maestro3]
> >  [<c01d1b5f>] device_attach+0x4f/0x90
> >  [<d109edc8>] m3_pci_driver+0x28/0xa0 [maestro3]
>
> I don't think this is PCMCIA related - something else is going on here.
>
> My guess is we're trying to locate a driver for the card, and we get to
> maestro3.  This works as expected, but the next driver on the chain
> seems to be a module which was may have been removed but which left
> its pci_driver structure behind (at 0xd10545c0.)

I have a Maestro3 in the laptop (which is also why the module is loaded), but 
I never unload it once it's loaded at bootup...

Anything else I can try?

Jan
- -- 
To be sure of hitting the target, shoot first and, whatever you hit,
call it the target.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+p5TlpuyeqyCEh60RAuAuAJ9WByWWm+pwMNT3HrarHxiW/6HfKACeMOPZ
2d2+dnUbdZhDG2A3vfVNHoI=
=rw/c
-----END PGP SIGNATURE-----

Re: [2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

[Russell King]

On Thu, Apr 24, 2003 at 09:40:18AM +0200, DevilKin-LKML wrote:
> I have a Maestro3 in the laptop (which is also why the module is loaded), but 
> I never unload it once it's loaded at bootup...
> 
> Anything else I can try?

Maybe looking in /sysfs/bus/pci/drivers before inserting the card
(this may cause an oops as well - if so, it confirms my suspicion.)

-- 
Russell King (rmk@arm.linux.org.uk)                The developer of ARM Linux
             http://www.arm.linux.org.uk/personal/aboutme.html

[FIXED 2.5.69] Re: [2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

[DevilKin]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Just reporting that this now works fine under 2.5.69. Whatever the problem, 
it's definitely gone now.

DK
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+tiX5puyeqyCEh60RAvk1AJ9Qvygxk+70PFSKSSLKTgUi2mWmJwCaAjf0
5AR6Cz0LF6fC9g6Nd5t2eP8=
=o4WM
-----END PGP SIGNATURE-----

Re: [FIXED 2.5.69] Re: [2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

[Matthew Harrell]

Was this one you reported - I didn't see this one.  

I still get a hang on any 2.5.50+ kernel when I load yenta and have ACPI
turned on.  It locks solid about three seconds after loading the module


> 
> Just reporting that this now works fine under 2.5.69. Whatever the problem, 
> it's definitely gone now.
> 

-- 
  Matthew Harrell                          Preserve wildlife --
  Bit Twiddlers, Inc.                       pickle a squirrel today!
  mharrell@bittwiddlers.com     

Re: [FIXED 2.5.69] Re: [2.5.67 - 2.5.68] Hangs on pcmcia yenta_socket initialisation

[DevilKin]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wednesday 07 May 2003 01:48, Matthew Harrell wrote:
> Was this one you reported - I didn't see this one.

It was the fact that when I booted my system with my 3com cardbus NIC inserted 
in the slot it would oops, giving me a non-working pcmcia/cardbus system.

Booting without the card worked fine, and inserting the card afterwards also 
worked great.

thread is here:

http://marc.theaimsgroup.com/?l=linux-kernel&m=105107713129722&w=2

>
> I still get a hang on any 2.5.50+ kernel when I load yenta and have ACPI
> turned on.  It locks solid about three seconds after loading the module
>

Ah. I don't use ACPI, because well... it caused me more pain than it helped me 
in the past, in both OS' on my system (being Linux and win2k)

Jan
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE+uJdQpuyeqyCEh60RAv/oAKCAUsots5uqvrL0O3dsrBGVXJ9icwCfQS/c
H6BRfG+uFkYWWsFQL8SJsSs=
=cGpw
-----END PGP SIGNATURE-----