Sad: Pcmcia problems on Fujitsu lifebook (TI PCI1225) 

Forum: PCMCIA Installation and Configuration Issues
Date: 2000, Feb 25
From: Wolfgang Wegner Wolfgang

Hello experts, :-) i am having trouble installing pcmcia card services on a Fujitsu/Siemens Lifebook C6325. This is a PII laptop with Intel BX chipset using the TI PCI1225 PCI/CardBus Bridge. The setup is attached below. Some other files can be found on http://berta.e-technik.uni-dortmund.de/m_ww/pcmcia/ I am now running kernel 2.2.14 and pcmcia 3.1.11 I just managed to avoid complete lockups by including only one memory region in config.opts, but the card is still not usable. I had one success after having Windows 98 running with the card and forgot to remove the card. The card services started, but the card could not be configured. After re-insertion of the card, it worked - but i could not reproduce this. Also, in this case "lspci -v" showed different results, but unfortunately i did not save this. The system log of this can be found in the file "log_running" (this was still with old kernel and pcmcia-package). I am not sure if this could be a BIOS problem (look at the memory regions in lspci!). Under Win98, the resources listed for PCMCIA bridge and card are as follows: Socket0: IRQ 11 mem 08000000-08000fff Socket1: IRQ 11 mem 08001000-08001fff mem 08040000-08080fff i/o 1080-10ff Network Card: IRQ 11 mem 0808000-080803ff i/o 1080-10ff mem 08040000-0807ffff (This is why i included this memory region in config.opts) The complete lspci -v output can be found in the file "lspci". lspci -v (only PCI1225 portion): 00:04.0 CardBus bridge: Texas Instruments PCI1225 (rev 01) Subsystem: Texas Instruments: Unknown device ac1c Flags: bus master, medium devsel, latency 168, IRQ 11 Memory at <ignored> (32-bit, non-prefetchable) Bus: primary=00, secondary=20, subordinate=22, sec-latency=32 I/O window 0: 00000000-00000003 I/O window 1: 00000000-00000003 16-bit legacy interface ports at 0007 00:04.1 CardBus bridge: Texas Instruments PCI1225 (rev 01) Subsystem: Texas Instruments: Unknown device ac1c Flags: bus master, medium devsel, latency 168, IRQ 11 Memory at <ignored> (32-bit, non-prefetchable) Bus: primary=00, secondary=23, subordinate=25, sec-latency=32 I/O window 0: 00000000-00000003 I/O window 1: 00000000-00000003 16-bit legacy interface ports at 0007 uname -rv: 2.2.14 #1 Thu Feb 24 17:08:13 CET 2000 config.opts:# # Local PCMCIA Configuration File # # System resources available for PCMCIA devices # include port 0x100-0x4ff, port 0x1000-0x17ff include memory 0x8000000-0x8080fff # include memory 0xc0000-0xfffff # include memory 0xa0000000-0xa0ffffff, memory 0x60000000-0x60ffffff # # Extra port range for IBM Token Ring # include port 0xa00-0xaff # # Resources we should not use, even if they appear to be available # # First built-in serial port exclude irq 4 # Second built-in serial port #exclude irq 3 # First built-in parallel port exclude irq 7 # when including the two lower memory lines, the system locks up completely upon card insertion! (Not very useful, but can be found in "log_lock"). I hope i supplied enough information to find anything out about this strange thing and did not miss anything in the documentation this time. Or perhaps anyone else already went through this and can help me. Best regards, Wolfgang

[ Next-in-Thread ]  [ Next Message ] 

Messages Inline: [ 0 ]  [ 1 ] 

Memory window is too small Re: Sad: Pcmcia problems on Fujitsu lifebook (TI PCI1225) (Wolfgang Wegner) Date: 2000, Feb 25 From: David Hinds <dhinds@pcmcia.sourceforge.org>

With 3.1.3, you're seeing a bug in the CardBus resource allocation code. With 3.1.11, you get an error message that's quite specific: the card is asking for 260K of memory space, and your memory window setting is too small. Enlarge your memory window. I'd try 0x8000000-0x8ffffff. -- Dave

Feedback: Not yet... Re: Memory window is too small (David Hinds) Date: 2000, Feb 28 From: Wolfgang Wegner Wolfgang

Thanks alot for the tip about the memory space. Unfortunately, this does not completely solve the issue. I compared the ioport regions reported by /proc/ioports and those given in /etc/pcmcia/config.opts and split the regions in more smaller regions, excluding the ide ports. This and the larger memory region in fact led to another case of an almost-working pcmcia subsystem. Upon card insertion, the system locked up as usual after the first high beep. No keystroke could be seen on the screen. After removing the card, all pending activities (the second beep, the keystrokes) were issued and the system was running. From now on, the card could be inserted and was usable up to the next reboot. Unfortunately, i could not reproduce this behaviour after reboot... :-( I post the modified /etc/pcmcia/config.opts and the dmesg output of the almost-working case (the two card insertions can be seen. In the first case, the card's eeprom was not seen by the driver). Another thing perhaps worth to be mentioned: Whenever the system locked up and could not be reactivated by removing the card, the system log (/var/log/messages) was filled with a number of binary zero's...?! (That is why i suspected the ide port regions) After changing the ioport regions, this could not be reproduced, so perhaps one more issue is solved. config.opts: # # Local PCMCIA Configuration File # # System resources available for PCMCIA devices # # include port 0x100-0x4ff, port 0x1000-0x17ff include port 0x100-0x16f, port 0x200-0x36f, port 0x1000-0x103f, port 0x1050-0x17 ff include memory 0x8000000-0x8ffffff # include memory 0xc0000-0xfffff # include memory 0xa0000000-0xa0ffffff, memory 0x60000000-0x60ffffff # # Extra port range for IBM Token Ring # include port 0xa00-0xaff # # Resources we should not use, even if they appear to be available # # First built-in serial port exclude irq 4 # Second built-in serial port #exclude irq 3 # First built-in parallel port exclude irq 7 # dmesg output (two subsequent card insertions): Linux PCMCIA Card Services 3.1.11 kernel build: 2.2.14 unknown options: [pci] [cardbus] [apm] [pnp] PCI routing table version 1.0 at 0xfdf60 PnP: PNP BIOS installation structure at 0xc00f7170 PnP: PNP BIOS version 1.0, entry at f0000:a554, dseg at 400 Intel PCIC probe: TI 1225 PCI-to-CardBus at slot 00:04, mem 0x68000000 host opts [0]: [serial pci & irq] [pci irq 11] [lat 64/32] [bus 32/34] host opts [1]: [serial pci & irq] [pci irq 11] [lat 64/32] [bus 35/37] ISA irqs (scanned) = 3,9,10 PCI status changes cs: IO port probe 0x1050-0x17ff: clean. cs: IO port probe 0x1000-0x103f: clean. cs: IO port probe 0x0200-0x036f: clean. cs: IO port probe 0x0100-0x016f: clean. cs: IO port probe 0x0a00-0x0aff: clean.cs: cb_alloc(bus 35): vendor 0x1011, device 0x0019 cs: cb_config(bus 35) fn 0 bar 1: io 0x200-0x27f fn 0 bar 2: mem 0x80c0000-0x80c03ff fn 0 rom: mem 0x8080000-0x80bffff cs: cb_enable(bus 35) bridge io map 0 (flags 0x21): 0x200-0x27f bridge mem map 0 (flags 0x1): 0x8080000-0x80c0fff tulip_reap() tulip_attach(bus 35, function 0) tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.un i-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford) eth0: Digital DS21143 Tulip rev 65 at 0x200,<7>cs: cb_disable(bus 35) cs: cb_release(bus 35) EEPROM not present, 00:4C:69:6E:75:79, IRQ 11. eth0: Missing EEPROM, this interface may not work correctly! cs: cb_free(bus 35) cs: cb_alloc(bus 35): vendor 0x1011, device 0x0019 cs: cb_config(bus 35) fn 0 bar 1: io 0x280-0x2ff fn 0 bar 2: mem 0x80c0000-0x80c03ff fn 0 rom: mem 0x8080000-0x80bffff cs: cb_enable(bus 35) bridge io map 0 (flags 0x21): 0x280-0x2ff bridge mem map 0 (flags 0x1): 0x8080000-0x80c0fff tulip_reap() tulip_attach(bus 35, function 0) eth1: Digital DS21143 Tulip rev 65 at 0x280, 00:E0:98:75:08:67, IRQ 11. eth1: EEPROM default media type Autosense. eth1: Index #0 - Media MII (#11) described by a 21142 MII PHY (3) block. eth1: MII transceiver #0 config 3000 status 7809 advertising 01e1. I really do not know what i am doing wrong here. And i also do not know what is the difference in the rare cases it works. Clueless, Wolfgang

Ok: Seems to be solved...(Fujitsu C6325 now working) Re: Memory window is too small (David Hinds) Date: 2000, Mar 03 From: Wolfgang Wegner Wolfgang

Hi all, The problem seems to be solved now. I think i was a bit blind again... In the rare cases the system "un-freezed" after removing the card, the tulip_cb module was still used (and thus not unloaded), and all subsequent card insertions did not use the ioport 0x200 used upon first card insetion but port 0x280 instead. So i just changed the config to not use the port range from 0x200 but from 0x280 and now all works. I am still not sure what conflict i ran into, but whatever it was, it is in the region 0x200-0x27f. Best regards, Wolfgang

[ Add Message ]  to: "Pcmcia problems on Fujitsu lifebook (TI PCI1225)"

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