Sad: Power settings causing problems with cardbus driver 

Forum: PCMCIA on Toshiba laptops
Date: 2000, May 13
From: Arthur Liu armie

I am having problems with my Toshiba Portege 3020CT. The Portege 3020 has 3 power settings - Full, Low, User Defined. If the machine is booted up and the 3c59x driver is loaded while the power setting is low, it causes a OOPS. Sometimes, it hangs the machine with just the message:   00:00:86:37:43:74, IRQ 3 This is happening with pcmcia-cs 3.1.14 with kernel 2.2.1[45]. Even though it may seem that the card is initialised in the following logs, it is not available. I have attached the relevant bits of the system log. Thanks. Cheers, Arthur ------------------------------------------------------------- May 14 02:08:56 warrior kernel: Linux version 2.2.15 (root@warrior) (gcc version 2.95.2 20000313 (Debian GNU/Linux)) #1 Thu May 4 20:03:49 EST 2000 May 14 02:08:56 warrior kernel: apm: BIOS version 1.2 Flags 0x02 (Driver version 1.12) May 14 02:08:56 warrior kernel: Toshiba System Managment Mode driver v1.6 11/2/2000 May 14 02:08:56 warrior kernel: Linux PCMCIA Card Services 3.1.14 May 14 02:08:56 warrior kernel: kernel build: 2.2.15 unknown May 14 02:08:56 warrior kernel: options: [pci] [cardbus] [apm] May 14 02:08:56 warrior kernel: PCI routing table version 1.0 at 0xf8cd0 May 14 02:08:56 warrior kernel: Intel PCIC probe: May 14 02:08:56 warrior kernel: Toshiba ToPIC95-B rev 07 PCI-to-CardBus at slot 00:13, mem 0x68000000 May 14 02:08:56 warrior kernel: host opts [0]: [slot 0xf0] [ccr 0x10] [cdr 0x86] [rcr 0x02] [no pci irq] [lat 168/176] [bus 20/20] May 14 02:08:56 warrior kernel: host opts [1]: [slot 0xf0] [ccr 0x20] [cdr 0x86] [rcr 0x02] [no pci irq] [lat 168/176] [bus 21/21] May 14 02:08:56 warrior kernel: ISA irqs (default) = 3,4,5,7,9,10,12,15 polling interval = 1000 ms May 14 02:08:56 warrior kernel: cs: cb_alloc(bus 20): vendor 0x10b7, device 0x5157 May 14 02:08:56 warrior kernel: cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff May 14 02:08:56 warrior kernel: cs: IO port probe 0x0800-0x08ff: clean. May 14 02:08:56 warrior kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x330-0x337 0x370-0x37f 0x388-0x38f 0x480-0x48f 0x4d0-0x4d7 May 14 02:08:56 warrior kernel: cs: IO port probe 0x1000-0x17ff: clean. May 14 02:08:56 warrior kernel: cs: IO port probe 0x0a00-0x0aff: clean. May 14 02:08:56 warrior kernel: cs: cb_free(bus 20) May 14 02:08:56 warrior kernel: 3c59x.c:v0.99L 5/28/99 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html May 14 02:08:57 warrior kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000004 May 14 02:08:57 warrior kernel: current->tss.cr3 = 05c45000, %cr3 = 05c45000 May 14 02:08:57 warrior kernel: *pde = 00000000 May 14 02:08:57 warrior kernel: Oops: 0000 May 14 02:08:57 warrior kernel: CPU: 0 May 14 02:08:57 warrior kernel: EIP: 0010:[<c6a3320d>] May 14 02:08:57 warrior kernel: EFLAGS: 00010246 May 14 02:08:57 warrior kernel: eax: c4d35120 ebx: c4d35120 ecx: c5b91cf8 edx: c5fa2b00 May 14 02:08:57 warrior kernel: esi: c6a338c8 edi: 00000000 ebp: c5b91c6c esp: c5b91c5c May 14 02:08:57 warrior kernel: ds: 0018 es: 0018 ss: 0018 May 14 02:08:57 warrior kernel: Process cardmgr (pid: 146, process nr: 11, stackpage=c5b91000) May 14 02:08:57 warrior kernel: Stack: c5b91d01 c6a33070 00000000 c5b91e18 c5b91cac c6a3080c c4d35120 00000015 May 14 02:08:57 warrior kernel: c5b91cac c6a3068a 4050643f c4dc28c0 00000000 00000286 c2946407 c02301f6 May 14 02:08:57 warrior kernel: c5fa2b00 c4dc28c0 c5fa2b00 c4dc28f0 c5b91f8c c6a31157 00000000 c5b91cf8 May 14 02:08:57 warrior kernel: Call Trace: [<c6a33070>] [ds:unregister_pccard_driver+1632/4884] [ds:unregister_pccard_driver+1246/4884] [ds:unregister_pccard_driver+4011/4884] [sock_wmalloc+41/84] [wake_up_process+64/76] [__wake_up+59/68] May 14 02:08:57 warrior kernel: [sock_def_readable+45/56] [sock_def_readable+28/56] [unix_stream_sendmsg+584/620] [get_new_inode+189/308] [sock_sendmsg+115/152] [sys_sendto+229/248] [dput+243/352] [do_no_page+194/212] May 14 02:08:57 warrior kernel: [handle_mm_fault+200/316] [sys_send+36/44] [sys_rt_sigaction+123/208] [sys_ioctl+363/392] [system_call+52/56] May 14 02:08:57 warrior kernel: Code: 83 7f 04 00 74 0d 3b 5f 10 75 08 80 4b 04 08 eb 35 89 f6 8b May 14 02:08:57 warrior kernel: cs: cb_alloc(bus 20): vendor 0x10b7, device 0x5157 May 14 02:08:57 warrior kernel: cs: cb_config(bus 20) May 14 02:08:57 warrior kernel: fn 0 bar 1: io 0x100-0x17f May 14 02:08:57 warrior kernel: fn 0 bar 2: mem 0x60021000-0x6002107f May 14 02:08:57 warrior kernel: fn 0 bar 3: mem 0x60020000-0x6002007f May 14 02:08:57 warrior kernel: fn 0 rom: mem 0x60000000-0x6001ffff May 14 02:08:57 warrior kernel: cs: cb_enable(bus 20) May 14 02:08:57 warrior kernel: bridge io map 0 (flags 0x21): 0x100-0x17f May 14 02:08:57 warrior kernel: bridge mem map 0 (flags 0x1): 0x60000000-0x60021fff May 14 02:08:57 warrior kernel: vortex_reap() May 14 02:08:57 warrior kernel: vortex_attach(bus 20, function 0, device 5157) May 14 02:08:57 warrior kernel: alloc_skb called nonatomically from interrupt c0160d4f May 14 02:08:57 warrior kernel: eth0: 3Com 3CCFE575 Cyclone CardBus at 0x100, 00:00:86:37:43:74, IRQ 3 May 14 02:08:57 warrior kernel: eth0: CardBus functions mapped 60020000->c6a3b000 May 14 02:08:57 warrior kernel: 8K byte-wide RAM 5:3 Rx:Tx split, MII interface. May 14 02:08:57 warrior kernel: MII transceiver found at address 0, status 7809. May 14 02:08:57 warrior kernel: Enabling bus-master transmits and whole-frame receives.

[ Next-in-Thread ]  [ Next Message ] 

Messages Inline: [ 0 ]  [ 1 ] 

A partial fix and something to try Re: Sad: Power settings causing problems with cardbus driver (Arthur Liu) Date: 2000, May 26 From: David Hinds <dhinds@pcmcia.sourceforge.org>

Can you try out the 25-May-00 beta on sourceforge.org in /pcmcia/NEW? I think I've fixed the oops, but probably not the underlying problem (which is that the Toshiba bridge seems to claim the card was ejected then re-inserted). It might be worth playing with some of the timing options for the pcmcia_core module. Like, in /etc/sysconfig/pcmcia, you could try setting: CORE_OPTS="setup_delay=100" The number is in 10ms "ticks", so this is a one second delay. -- Dave

No more oops - card gets initialised 2nd time Re: A partial fix and something to try (David Hinds) Date: 2000, May 28 From: Arthur Liu armie

Thanks Dave, No more oops. The driver doesn't find the card first time though - it takes the second try to find the card. The card is now eth1 instead of eth0. I've attached the log messages again. The setup_delay works... I'll keep that one for now. Thanks for the great work! Cheers, Arthur ---------------------------------------------- May 29 01:40:45 warrior kernel: Linux version 2.2.15 (root@warrior) (gcc version 2.95.2 20000313 (Debian GNU/Linux)) #1 Thu May 4 20:03:49 EST 2000 May 29 01:40:45 warrior kernel: Linux PCMCIA Card Services 3.1.16 May 29 01:40:46 warrior kernel: kernel build: 2.2.15 unknown May 29 01:40:46 warrior kernel: options: [pci] [cardbus] [apm] May 29 01:40:46 warrior kernel: PCI routing table version 1.0 at 0xf8cd0 May 29 01:40:46 warrior kernel: Intel PCIC probe: May 29 01:40:46 warrior kernel: Toshiba ToPIC95-B rev 07 PCI-to-CardBus at slot 00:13, mem 0x68000000 May 29 01:40:46 warrior kernel: host opts [0]: [slot 0xf0] [ccr 0x10] [cdr 0x86] [rcr 0x02] [no pci irq] [lat 168/176] [bus 20/20] May 29 01:40:46 warrior kernel: host opts [1]: [slot 0xf0] [ccr 0x20] [cdr 0x86] [rcr 0x02] [no pci irq] [lat 168/176] [bus 21/21] May 29 01:40:46 warrior kernel: ISA irqs (default) = 3,4,5,7,9,10,12,15 polling interval = 1000 ms May 29 01:40:46 warrior kernel: cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff May 29 01:40:46 warrior kernel: cs: IO port probe 0x0800-0x08ff: clean. May 29 01:40:46 warrior kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x330-0x337 0x370-0x37f 0x388-0x38f 0x480-0x48f 0x4d0-0x4d7 May 29 01:40:46 warrior kernel: cs: IO port probe 0x1000-0x17ff: clean. May 29 01:40:46 warrior kernel: cs: IO port probe 0x0a00-0x0aff: clean. May 29 01:40:46 warrior kernel: cs: cb_alloc(bus 20): vendor 0x10b7, device 0x5157 May 29 01:40:46 warrior kernel: 3c59x.c:v0.99Q 5/16/2000 Donald Becker, becker@scyld.com May 29 01:40:46 warrior kernel: http://www.scyld.com/network/vortex.html May 29 01:40:47 warrior kernel: cs: cb_config(bus 20) May 29 01:40:47 warrior kernel: fn 0 bar 1: io 0x100-0x17f May 29 01:40:47 warrior kernel: fn 0 bar 2: mem 0x60021000-0x6002107f May 29 01:40:47 warrior kernel: fn 0 bar 3: mem 0x60020000-0x6002007f May 29 01:40:47 warrior kernel: fn 0 rom: mem 0x60000000-0x6001ffff May 29 01:40:47 warrior kernel: irq 3 May 29 01:40:47 warrior kernel: cs: cb_enable(bus 20) May 29 01:40:47 warrior kernel: bridge io map 0 (flags 0x21): 0x100-0x17f May 29 01:40:47 warrior kernel: bridge mem map 0 (flags 0x1): 0x60000000-0x60021fff May 29 01:40:47 warrior kernel: vortex_attach(device 14:00.0) May 29 01:40:47 warrior kernel: eth0: 3Com 3CCFE575BT Cyclone CardBus at 0x100, 00:00:86:37:43:74, IRQ 3 May 29 01:40:47 warrior kernel: product code 'NV' rev 07 date 03-10-98 May 29 01:40:47 warrior kernel: eth0: CardBus functions mapped 60020000->c6a3b000. May 29 01:40:47 warrior kernel: 8K byte-wide RAM 5:3 Rx:Tx split, MII interface. May 29 01:40:47 warrior kernel: cs: cb_disable(bus 20) May 29 01:40:47 warrior kernel: cs: cb_release(bus 20) May 29 01:40:47 warrior kernel: ***WARNING*** No MII transceivers found! May 29 01:40:47 warrior kernel: Enabling bus-master transmits and whole-frame receives. May 29 01:40:47 warrior kernel: cs: cb_free(bus 20) May 29 01:40:47 warrior kernel: cs: cb_alloc(bus 20): vendor 0x10b7, device 0x5157 May 29 01:40:47 warrior kernel: eth0: command 0x5800 did not complete! May 29 01:40:47 warrior kernel: eth0: command 0x2800 did not complete! May 29 01:40:49 warrior kernel: cs: cb_config(bus 20) May 29 01:40:49 warrior kernel: fn 0 bar 1: io 0x280-0x2ff May 29 01:40:49 warrior kernel: fn 0 bar 2: mem 0x60021000-0x6002107f May 29 01:40:49 warrior kernel: fn 0 bar 3: mem 0x60020000-0x6002007f May 29 01:40:49 warrior kernel: fn 0 rom: mem 0x60000000-0x6001ffff May 29 01:40:49 warrior kernel: irq 3 May 29 01:40:49 warrior kernel: cs: cb_enable(bus 20) May 29 01:40:49 warrior kernel: bridge io map 0 (flags 0x21): 0x280-0x2ff May 29 01:40:49 warrior kernel: bridge mem map 0 (flags 0x1): 0x60000000-0x60021fff May 29 01:40:49 warrior kernel: vortex_attach(device 14:00.0) May 29 01:40:49 warrior kernel: eth1: 3Com 3CCFE575BT Cyclone CardBus at 0x280, 00:00:86:37:43:74, IRQ 3 May 29 01:40:49 warrior kernel: product code 'NV' rev 07 date 03-10-98 May 29 01:40:49 warrior kernel: eth1: CardBus functions mapped 60020000->c6a3d000. May 29 01:40:49 warrior kernel: 8K byte-wide RAM 5:3 Rx:Tx split, MII interface. May 29 01:40:49 warrior kernel: MII transceiver found at address 0, status 782d. May 29 01:40:49 warrior kernel: Enabling bus-master transmits and whole-frame receives.

[ Add Message ]  to: "Power settings causing problems with cardbus driver"

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