Sad: 3.1.15 fatal crash (Toshiba laptop w Xircom CBEII-10/100) 

Forum: PCMCIA on Toshiba laptops
Date: 2000, Jun 02
From: Fabien Tassin fta

Context: laptop Toshiba Satellite Pro 4270 running kernel 2.2.15, pcmcia 3.1.15 with a Xircom CBEII-10/100 Cardbus card. I was running succesfully a pre 3.1.15 version (~ May 15) of the PCMCIA drivers. It was usable but I have to make some manipulations to get it started after each reboot (else, the interface was up but no way to pass a single packet). I've decided to upgrade to 3.1.15 today but now, the system crashes systematically while loading the module i82365.o (no way to recover, not even with magic sysReq). I've also tried pre-3.1.16 (01-Jun-00) but it is the same thing. Strangely, I'm able to start the pcmcia service while in "single user" mode.. but I'm unable to pass a single packet :( Unfortunatly, I've overwrited my pre-3.1.15 copy (with final 3.1.15) and it's no longer available on the ftp site so I'm not able to downgrade :(( Here is the lspci output (bridge, cardbus briges, xircom card [from the single user boot]) : 00:05.3 Bridge: Intel Corporation 82371AB PIIX4 ACPI (rev 03) 00:0b.0 CardBus bridge: Toshiba America Info Systems ToPIC95 PCI to Cardbus Bridge with ZV Support (rev 20) 00:0b.1 CardBus bridge: Toshiba America Info Systems ToPIC95 PCI to Cardbus Bridge with ZV Support (rev 20) 10:00.0 Ethernet controller: Xircom Cardbus Ethernet 10/100 (rev 03). When it is working, here is the output of 'insmod i82365.o' : Intel PCIC probe: Toshiba ToPIC100 rev 20 PCI-to-CardBus at slot 00:0b, mem 0x68000000 host opts [0]: [slot 0xf0] [ccr Ox10] [cdr 0x86] [rcr 0xc000000] [pci irq 11] [lat 64/176] [bus 20/20] host opts [1]: [slot 0xf0] [ccr Ox20] [cdr 0x86] [rcr 0xc000000] [pci irq 11] [lat 64/176] [bus 21/21] ISA irqs (default) = 4,5,7,9,10,12 PCI status changes Else, no output [ even with PCMCIA_DEBUG=1 ] As it crashes even when the slots are all empty, I don't think this is a Xircom problem. Any idea ?

[ Next-in-Thread ]  [ Next Message ] 

Messages Inline: [ 0 ]  [ 1 ] 

More messages might help Re: Sad: 3.1.15 fatal crash (Toshiba laptop w Xircom CBEII-10/100) (Fabien Tassin) Date: 2000, Jun 03 From: David Hinds <dhinds@pcmcia.sourceforge.org>

> > When it is working, here is the output of 'insmod i82365.o' : So are these messages from the pre-3.1.15 drivers, or the 3.1.15 drivers booted in single user mode? (it would be useful to compare those two cases) Looking at the changes in 3.1.15, I'm not seeing any good candidates for where the bug might be. I assume you did not change anything else around the time you tried to upgrade to 3.1.15? (no kernel changes? no changes in how other devices are configured?) The PCMCIA-HOWTO has some hints for debugging system lockups that might be helpful. There are also some suggestions in the chapter on diagnosing configuration problems. -- Dave

More: Re: More messages might help Re: More messages might help (David Hinds) Date: 2000, Jun 03 From: Fabien Tassin fta

> > When it is working, here is the output of 'insmod i82365.o' : > So are these messages from the pre-3.1.15 drivers, or the > 3.1.15 drivers booted in single user mode? 3.1.15 in single user mode. > (it would be useful to compare those two cases) as previously mentionned, I no longer have the pre-3.1.15 source tree :( > Looking at the changes in 3.1.15, I'm not seeing any good > candidates for where the bug might be. I assume you did > not change anything else around the time you tried to > upgrade to 3.1.15? (no kernel > changes? > no changes in how other devices are configured?) I've investigated a bit to understand why it is working in single user mode and found that the problem occurs when ALSA drivers are loaded before PCMCIA. I have a sound card that requires the latest CVS ALSA source tree in order to work (a YAMAHA YMF-744B card supported by the new ymfpci.o module). BTW, I had no problem with PCMCIA pre-3.1.15 and ALSA CVS-2000-05-31. It started to crash with PCMCIA 3.1.15 (or pre-3.1.16) and ALSA CVS-2000-06-02. I've upgraded ALSA first and AFAIR it was still working.. I'll try to look more deeply into this later.. OTOH, I'm not able to pass packets through my Xircom card until I enable the promiscuous-mode. With pre-3.1.15, a ifdown/ifup was usually enough.

Something to try Re: More: Re: More messages might help (Fabien Tassin) Date: 2000, Jun 09 From: David Hinds <dhinds@pcmcia.sourceforge.org>

Try with the 09-Jun-00 beta on sourceforge.org in /pcmcia/NEW. I changed the i82365 module to skip the interrupt probe for ToPIC bridges. I think the ALSA driver upgrade is what is making the problem visible. -- Dave

[ Add Message ]  to: "3.1.15 fatal crash (Toshiba laptop w Xircom CBEII-10/100)"

[ Members ]  [ Subscribe ]  [ Admin Mode ] 
[ Show Frames ]  [ Help for HyperNews at pcmcia-cs.sourceforge.net 1.10 ]