Use this page to discuss SCSI host adapter issues. Before posting, be sure you've checked any relevant sections of the PCMCIA-HOWTO, such as: Particularly useful bits of information to post include your laptop's make and model; the type(s) of card(s) and SCSI device(s) you are using; and, of course, any relevant system log messages.

Messages Inline: 0 1

Sad Problem w/aha152x_cs: "soft interrupt lost"(kernel 2.2.13 only!)

Date: 1999, Dec 06
From: Serge Winitzki winitzki

Hi,

I'm having a problem starting my Adaptec SlimSCSI 1460. It used to work fine with 2.0.34 and 2.0.36 kernels, pcmcia-3.0.x but I can't make it work with my new 2.2.13 kernel and pcmcia-3.x.x at all.

Here are messages from successful boot (2.0.36):
Linux PCMCIA Card Services 3.0.5
  kernel build: 2.0.36 unknown
  options: [pci] [cardbus]
Intel PCIC probe:
  Cirrus PD6729 PCI at port 0x3000 ofs 0x00, 2 sockets
    host opts [0]: [ring] [1/5/0] [1/20/0]
    host opts [1]: [ring] [1/5/0] [1/20/0]
    ISA irqs (default) = 3,4,5,7,9,10,11,12 status change on irq 11
cs: IO port probe 0x1000-0x17ff: excluding 0x1068-0x106f 0x1220-0x1227 0x1340
-0x1347 0x1388-0x138f 0x13f0-0x13f7 0x1468-0x146f 0x17f0-0x17f7
cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x227 0x340-0x347 0x388-0x38
f 0x408-0x40f 0x468-0x46f 0x480-0x48f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0x0d0000-0x0dffff: clean.
aha152x: processing commandline: ok
aha152x: BIOS test: passed, detected 1 controller(s)
aha152x0: vital data: PORTBASE=0x140, IRQ=4, SCSI ID=7, reconnect=enabled, pa
rity=enabled, synchronous=disabled, delay=100, extended translation=disabled
aha152x: trying software interrupt, ok.
scsi0 : Adaptec 152x SCSI driver; $Revision: 1.18 $
scsi : 1 host.

---
Now, if I boot 2.2.13 on the same root partition with pcmcia 3.1.5, I get the following errors:

Linux PCMCIA Card Services 3.1.5
  kernel build: 2.2.13 #9 Mon Dec 6 01:19:38 EST 1999
  options: [pci] [cardbus] [apm]
Intel PCIC probe:
  Cirrus PD6729 PCI-to-PCMCIA at bus 0 slot 24, port 0x3000, 2 sockets
    host opts [0]: [ring] [1/5/0] [1/20/0]
    host opts [1]: [ring] [1/5/0] [1/20/0]
    ISA irqs (default) = 3,4,5,7,9,10,11,12 polling interval = 1000 ms
cs: IO port probe 0x1000-0x17ff: excluding 0x1068-0x106f 0x1220-0x1227 0x1340
-0x1347 0x1388-0x138f 0x13f0-0x13ff 0x1468-0x146f 0x17f0-0x17ff
cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x227 0x340-0x347 0x388-0x38
f 0x408-0x40f 0x468-0x46f 0x480-0x48f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0x0d0000-0x0dffff: clean.
aha152x: processing commandline:
aha152x: PCMCIA setup
aha152x: invalid line (controller=1)
ok
aha152x: BIOS test: passed, detected 1 controller(s)
aha152x0: vital data: PORTBASE=0x1740, IRQ=3, SCSI ID=7, reconnect=enabled, parity=enabled, synchronous=disabled, delay=100, extended translation=disabled
aha152x: trying software interrupt, lost.
aha152x: IRQ 3 possibly wrong. Please verify.
aha152x_cs: no SCSI devices found
scsi : 0 hosts.

---
I have both irq 3 and 4 available (I disabled both serial ports). irq 9 and 10 are used by the sound card on my laptop (no-name pentium 150). What bothers me is that I don't seem to be able to make it work after I upgraded the kernel.
First, I can't find the "invalid line controller=1" anywhere in my options. And it's odd that this error appears with pcmcia 3.1.5 but not with pcmcia 3.0.x when booting with the same (I use all defaults) pcmcia options.
Second, I tried disabling various interrupts and putting irq_list=9,10,11 into the options, but the result is the same ("lost software interrupt").
Third, notice that there is now a different "PORTBASE" address (0x1740 instead of 0x140) detected.

Finally I tried compiling pcmcia-3.0.14 with linux 2.2.13 and here's what I get:

Linux PCMCIA Card Services 3.0.14
  kernel build: 2.2.13 #10 Mon Dec 6 01:33:03 EST 1999
  options: [pci] [apm]
Intel PCIC probe:
  Cirrus PD6729 PCI-to-PCMCIA at bus 0 slot 24, port 0x3000, 2 sockets
    host opts [0]: [ring] [1/5/0] [1/20/0]
    host opts [1]: [ring] [1/5/0] [1/20/0]
    ISA irqs (default) = 3,4,5,7,9,10,11,12 polling interval = 1000 ms
cs: IO port probe 0x1000-0x17ff: excluding 0x1068-0x106f 0x1220-0x122f 0x1330
-0x1337 0x1340-0x1347 0x1388-0x138f 0x13f0-0x13f7 0x1468-0x146f 0x1530-0x1537
 0x17f0-0x17f7
cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x330-0x337 0x340-0x347 0x388-0x38f 0x408-0x40f 0x468-0x46f 0x480-0x48f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0x0d0000-0x0dffff: clean.
aha152x: processing commandline: ok
aha152x: BIOS test: passed, detected 1 controller(s)
aha152x0: vital data: PORTBASE=0x140, IRQ=5, SCSI ID=7, reconnect=enabled, parity=enabled, synchronous=disabled, delay=100, extended translation=disabled
aha152x: trying software interrupt, lost.
aha152x: IRQ 5 possibly wrong. Please verify.
aha152x_cs: no SCSI devices found
scsi : 0 hosts.

---

So the "commandline" is ok now; I put "irq_list=5,11" and also tried just "11" but the result is the same.
Notice that the "PORTBASE" is back to 0x140.

I'll appreciate any help.

Regards,
Serge

None Too many things changing at once...

Re: Sad Problem w/aha152x_cs: "soft interrupt lost"(kernel 2.2.13 only!) (Serge Winitzki)
Date: 1999, Dec 07
From: David Hinds <dhinds@pcmcia.sourceforge.org>

Ok...

according to your log, 3.0.5 worked with irq 4 and io 0x140.

3.1.5 did not work, with irq 3 and io 0x1740.

3.0.14 did not work, with irq 5 and io 0x140.

It doesn't really matter what interrupts appear to be available.  Some
interrupts may not even be physically connected to your PCMCIA bridge. 
Also, if a device is "disabled", that doesn't necessarily mean that
its interrupt is available for use by another arbitrary device.

What I would do is to try to recreate the 3.0.5 situation.  Exclude
irq 3 in your config.opts file, but don't exclude irq 4.  And exclude
the 0x1000-0x17ff io port range.

-- Dave

Ok thank you, it works now

Re: None Too many things changing at once... (David Hinds)
Date: 1999, Dec 07
From: Serge Winitzki winitzki

Thanks, I was able to make it work.

I excluded the upper port range 0x1000-0x17ff, included irq4 and specified "irq_list=4,5,12" and the card worked after a reboot. (Was a reboot really necessary?
Now the log is the same as the old pcmcia-3.0.5 log. Thanks a lot!
Regards,
Serge

Question Bus Toaster Freezes my box.

Date: 1999, Dec 08
From: Francisco Escarpa PacoEscarpa

I'm using SuSE Linux with 2.2.13 kernel, and pcmcia_cs 3.1.4.

As soon I insert the card in the running system, it reboots when the driver begins to run; If I insert the card and then boot the machine, it ( usualy ) works, and recognize the host, - I give a serie of messages about fast PIO error, and the devices ( Minolta scanner and Iomega Zip drive ) are not recognized.

I use the following /etc/rc.d/pcmcia options:

    ...
    # Should be either i82365 or tcic
    PCIC=i82365
    # Put socket driver timing parameters here
    PCIC_OPTS="poll_interval=100"
    # Put pcmcia_core options here
    CORE_OPTS="pc_debug=4"
    # Put cardmgr options here
    CARDMGR_OPTS=
    ...

and /etc/pcmcia/config device "sym53c500_cs" class"scsi"module"scsi/scsi_mod","scsi/sd_mod","scsi/sg","sym53c500_cs"

If somebody is using this card and / or scanner, I will be very pleased to receive ideas, comments, ...

Thanks in advance,

Paco.

None More information would be helpful

Re: Question Bus Toaster Freezes my box. (Francisco Escarpa)
Date: 1999, Dec 13
From: David Hinds <dhinds@pcmcia.sourceforge.org>

The actual message log would be helpful (including the messages from
when the i82365 driver loads, describing your PCMCIA bridge details).

-- Dave

None More information ...

Re: None More information would be helpful (David Hinds)
Date: 1999, Dec 14
From: Francisco Escarpa PacoEscarpa

Sorry; long message.

I am using two additional cards ( modem and ethernet ) that are recognized without problem (I work everyday with them ).

Debug level=4 That is the log from a unsuccesfully session on start.

Dec  2 23:25:56 dione syslogd 1.3-3: restart.
Dec  2 23:25:57 dione kernel: klogd 1.3-3, log source = /proc/kmsg started.
Dec  2 23:25:57 dione kernel: Inspecting /boot/System.map-2.2.13
Dec  2 23:25:57 dione kernel: Loaded 6071 symbols from /boot/System.map-2.2.13.
Dec  2 23:25:57 dione kernel: Symbols match kernel version 2.2.13.
Dec  2 23:25:57 dione kernel: Loaded 12 symbols from 5 modules.
Dec  2 23:25:57 dione kernel: Serial driver version 4.27 with SHARE_IRQ enabled
Dec  2 23:25:57 dione kernel: Linux PCMCIA Card Services 3.1.4
Dec  2 23:25:57 dione kernel:   kernel build: 2.2.13 #2 Fri Nov 19 22:54:56 CET 1999
Dec  2 23:25:57 dione kernel:   options:  [pci] [apm] [pnp]
Dec  2 23:25:57 dione kernel: cs.c 1.235 1999/11/11 17:52:05 (David Hinds)
Dec  2 23:25:57 dione kernel: PnP: PNP BIOS installation structure at 0xc00ff010
Dec  2 23:25:57 dione kernel: PnP: PNP BIOS version 1.0, entry at ec000:1dbb, dseg at ec000
Dec  2 23:25:57 dione kernel: cs: CardServices(GetCardServicesInfo, 0xc1cd5f2c, 0xc2821000)
Dec  2 23:25:57 dione kernel: i82365.c 1.265 1999/11/10 18:36:21 (David Hinds)
Dec  2 23:25:57 dione kernel: Intel PCIC probe: 
Dec  2 23:25:57 dione kernel:   Cirrus PD6832 PCI-to-CardBus at bus 0 slot 25, mem 0x08000000, 2 sockets
Dec  2 23:25:57 dione kernel:     host opts [0]: [ring] [0/7/4] [3/17/4] [no pci irq] [lat 128/176] [bus 32/34]
Dec  2 23:25:57 dione kernel:     host opts [1]: [ring] [0/7/4] [3/17/4] [no pci irq] [lat 128/176] [bus 35/37]
Dec  2 23:25:57 dione kernel:   testing ISA irq 3
Dec  2 23:25:57 dione kernel: -> hit on irq 3
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 4
Dec  2 23:25:57 dione kernel: -> hit on irq 4
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 5
Dec  2 23:25:57 dione kernel: -> hit on irq 5
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 7
Dec  2 23:25:57 dione kernel: -> hit on irq 7
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 9
Dec  2 23:25:57 dione kernel:     hits = 0
Dec  2 23:25:57 dione kernel:   testing ISA irq 10
Dec  2 23:25:57 dione kernel:     hits = 0
Dec  2 23:25:57 dione kernel:   testing ISA irq 11
Dec  2 23:25:57 dione kernel: -> hit on irq 11
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 12
Dec  2 23:25:57 dione kernel:     hits = 0
Dec  2 23:25:57 dione kernel:   testing ISA irq 14
Dec  2 23:25:57 dione kernel:   testing ISA irq 3
Dec  2 23:25:57 dione kernel: -> hit on irq 3
Dec  2 23:25:57 dione kernel: -> hit on irq 3
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 4
Dec  2 23:25:57 dione kernel: -> hit on irq 4
Dec  2 23:25:57 dione kernel: -> hit on irq 4
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 5
Dec  2 23:25:57 dione kernel: -> hit on irq 5
Dec  2 23:25:57 dione kernel: -> hit on irq 5
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 7
Dec  2 23:25:57 dione kernel: -> hit on irq 7
Dec  2 23:25:57 dione kernel: -> hit on irq 7
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:   testing ISA irq 11
Dec  2 23:25:57 dione kernel: -> hit on irq 11
Dec  2 23:25:57 dione kernel: -> hit on irq 11
Dec  2 23:25:57 dione kernel:     hits = 1
Dec  2 23:25:57 dione kernel:     ISA irqs (scanned) = 3,4,5,7,11 polling interval = 1000 ms
Dec  2 23:25:57 dione kernel: cs: register_ss_entry(2, 0xc2827250)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 4, 0xc191e808)
Dec  2 23:25:57 dione kernel: i82365: SetSocket(0, flags 0x000, Vcc 0, Vpp 0, io_irq 0, csc_mask 0x80)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 6, 0xc1cd5ec4)
Dec  2 23:25:57 dione kernel: i82365: SetIOMap(0, 0, 0x00, 0 ns, 0x0000-0x0001)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 6, 0xc1cd5ec4)
Dec  2 23:25:57 dione kernel: i82365: SetIOMap(0, 1, 0x00, 0 ns, 0x0000-0x0001)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(0, 0, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(0, 1, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(0, 2, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(0, 3, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(0, 4, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 1, 0xc191e814)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 11, 0xc1880e00)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 4, 0xc191e008)
Dec  2 23:25:57 dione kernel: i82365: SetSocket(1, flags 0x000, Vcc 0, Vpp 0, io_irq 0, csc_mask 0x80)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 6, 0xc1cd5ec4)
Dec  2 23:25:57 dione kernel: i82365: SetIOMap(1, 0, 0x00, 0 ns, 0x0000-0x0001)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 6, 0xc1cd5ec4)
Dec  2 23:25:57 dione kernel: i82365: SetIOMap(1, 1, 0x00, 0 ns, 0x0000-0x0001)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(1, 0, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(1, 1, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 8, 0xc1cd5ecc)
Dec  2 23:25:59 dione named[142]: starting
Dec  2 23:25:59 dione named[142]: cache zone "" (IN) loaded (serial 0)
Dec  2 23:25:59 dione named[142]: master zone "escarpa.es" (IN) loaded (serial 1)
Dec  2 23:26:00 dione named[142]: master zone "ofitec.torre" (IN) loaded (serial 1)
Dec  2 23:26:00 dione named[142]: master zone "0.168.192.IN-ADDR.ARPA" (IN) loaded (serial 1)
Dec  2 23:26:00 dione named[142]: master zone "0.0.127.IN-ADDR.ARPA" (IN) loaded (serial 1)
Dec  2 23:26:00 dione named[142]: listening on [127.0.0.1].53 (lo)
Dec  2 23:26:00 dione named[142]: listening on [192.168.0.2].53 (dummy0)
Dec  2 23:26:00 dione named[142]: Forwarding source address is [0.0.0.0].1024
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(1, 2, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(1, 3, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 8, 0xc1cd5ecc)
Dec  2 23:25:57 dione kernel: i82365: SetMemMap(1, 4, 0x00, 0 ns, 0x00000-0x00000, 0x00000)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 1, 0xc191e014)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 11, 0xc184a160)
Dec  2 23:25:57 dione kernel: ds.c 1.100 1999/11/08 20:47:02 (David Hinds)
Dec  2 23:25:57 dione kernel: cs: CardServices(GetCardServicesInfo, 0xc1cd5efc, 0xc282e000)
Dec  2 23:25:57 dione kernel: cs: CardServices(BindDevice, 0xc1cd5ef4, 0xc282e000)
Dec  2 23:25:57 dione kernel: cs: bind_device(): client 0xc1737360, sock 0, dev Driver Services
Dec  2 23:25:57 dione kernel: cs: CardServices(RegisterClient, 0xc184a480, 0xc1cd5f10)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 0, 0xc1cd5ea0)
Dec  2 23:25:57 dione kernel: pcic_ioctl(0, 2, 0xc1cd5e9c)
Dec  2 23:25:57 dione kernel: i82365: GetStatus(0) = 0x0041
Dec  2 23:25:57 dione kernel: cs: register_client(): client 0xc1737360, sock 0, dev Driver Services
Dec  2 23:25:57 dione kernel: cs: CardServices(BindDevice, 0xc1cd5ef4, 0xc282e000)
Dec  2 23:25:57 dione kernel: cs: bind_device(): client 0xc1737480, sock 1, dev Driver Services
Dec  2 23:25:57 dione kernel: cs: CardServices(RegisterClient, 0xc184a4b4, 0xc1cd5f10)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 0, 0xc1cd5ea0)
Dec  2 23:25:57 dione kernel: pcic_ioctl(1, 2, 0xc1cd5e9c)
Dec  2 23:25:57 dione kernel: i82365: GetStatus(1) = 0x0041
Dec  2 23:25:57 dione kernel: cs: register_client(): client 0xc1737480, sock 1, dev Driver Services
Dec  2 23:25:57 dione kernel: ds_open(socket 0)
Dec  2 23:25:57 dione kernel: ds_open(socket 1)
Dec  2 23:25:57 dione kernel: ds_open(socket 2)
Dec  2 23:25:57 dione kernel: ds_open(socket 2)
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0x80146401, 0xbffffb40)
Dec  2 23:25:57 dione kernel: cs: CardServices(GetCardServicesInfo, 0xc0275cf8, 0xc1fb65c0)
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067db0)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: cs: IO port probe 0x1000-0x17ff: clean.
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067d90)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x100-0x107 0x200-0x207 0x240-0x24f 0x2d0-0x2d7 0x300-0x307 0x388-0x38f
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067dd0)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067e10)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067df0)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067e30)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067e50)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: ds_ioctl(socket 0, 0xc0146402, 0x8067e70)
Dec  2 23:25:57 dione kernel: cs: CardServices(AdjustResourceInfo, 0xc1737360, 0xc0275cf8)
Dec  2 23:25:57 dione kernel: ds_poll(socket 0)
Dec  2 23:25:57 dione kernel: ds_poll(socket 1)
Dec  2 23:25:57 dione kernel: ds_poll(socket 0)
Dec  2 23:25:57 dione kernel: ds_poll(socket 1)
Dec  2 23:26:03 dione kernel: Installing knfsd (copyright (C) 1996 okir@monad.swb.de).
Dec  2 23:26:03 dione kernel: nfsd_init: initialized fhcache, entries=256
Dec  2 23:30:53 dione kernel: i82365: socket 1 event 0x80
Dec  2 23:30:53 dione kernel: pcic_ioctl(1, 2, 0xc01dbee8)
Dec  2 23:30:53 dione kernel: i82365: GetStatus(1) = 0x01c1
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 2, 0xc01dbf2c)
Dec  2 23:30:54 dione kernel: i82365: GetStatus(1) = 0x01c1
Dec  2 23:30:54 dione kernel: cs: setup_socket(1): applying power
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 4, 0xc191e008)
Dec  2 23:30:54 dione kernel: i82365: SetSocket(1, flags 0x000, Vcc 50, Vpp 50, io_irq 0, csc_mask 0x80)
Dec  2 23:30:54 dione kernel: cs: resetting socket 1
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 4, 0xc191e008)
Dec  2 23:30:54 dione kernel: i82365: SetSocket(1, flags 0x240, Vcc 50, Vpp 50, io_irq 0, csc_mask 0x80)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 4, 0xc191e008)
Dec  2 23:30:54 dione kernel: i82365: SetSocket(1, flags 0x200, Vcc 50, Vpp 50, io_irq 0, csc_mask 0x80)
Dec  2 23:30:54 dione kernel: e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e00000-0xa0e00fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e01000-0xa0e01fff, 0x00000)
Dec  2 23:30:54 dione kernel: cs: CardServices(GetFirstTuple, 0xc1737480, 0xc0275624)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione cardmgr[80]: initializing socket 1
Dec  2 23:30:54 dione cardmgr[80]: socket 1: New Media Bus Toaster SCSI
Dec  2 23:30:55 dione cardmgr[80]: executing: 'insmod /lib/modules/2.2.13/scsi/scsi_mod.o'
Dec  2 23:30:55 dione cardmgr[80]: executing: 'insmod /lib/modules/2.2.13/scsi/sd_mod.o'
Dec  2 23:30:55 dione cardmgr[80]: executing: 'insmod /lib/modules/2.2.13/scsi/sg.o'
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e01000-0xa0e01fff, 0x00000)
Dec  2 23:30:54 dione kernel: pcic_ioctl(1, 8, 0xc191e06c)
Dec  2 23:30:54 dione kernel: i82365: SetMemMap(1, 0, 0x21, 300 ns, 0xa0e01000-0xa0e01fff, 0x00000)

and the machine hungs ...

Thanks in advance ...

Paco.

None These are not the messages you're looking for

Re: None More information ... (Francisco Escarpa)
Date: 1999, Dec 15
From: David Hinds <dhinds@pcmcia.sourceforge.org>

The extra debugging information is not helpful in this case.

I'd re-build with debugging disabled.  Then follow the instructions in
the HOWTO, for debugging a system lockup: the last kernel messages are
probably lost because the lockup happens before they can be written to
disk.  Redirect them to the console, and you'll probably see something
more useful.

My guess is that you're experiencing an IO port conflict.  If you can
squeeze out a couple more messages, they'll tell you what IO resources
the card is trying to use.

-- Dave

More A couple more messages :-).

Re: None These are not the messages you're looking for (David Hinds)
Date: 1999, Dec 20
From: Francisco Escarpa PacoEscarpa

Using the tips at PCMCIA HOWTO ( console 8 and so .. ), I get the following:

cs:memory probe 0xa0000000 - 0xa0ffffff clean SYM53C500: invalid IRQ3 specified SCSI0: SYM53C500 at 0x130, IRQ 0 , fast PIO mode SCSI: 1 host

Here, usually the computer hungs for 3-4 seconds and reboots.

The IRQ is free ( IRQ 5 has the same behaviour; they are used troubleless for other PC cards ).

Sometimes ( usualy the first/second try ), I get the following:

Dec 16 21:25:05 dione kernel: SYM53C500: Warning: PIO error!
Dec 16 21:25:05 dione kernel: SYM53C500: Warning: PIO error!
Dec 16 21:25:05 dione kernel: scsi0 channel 0 : resetting for second half of retries.
Dec 16 21:25:05 dione kernel: SCSI bus is being reset for host 0 channel 0.
Dec 16 21:25:05 dione kernel: scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 0, lun 0 0x03 00 00 00 10 00 
... ... ( for each pid up to 6 ) ...
Dec 16 22:00:06 dione kernel: scsi : 0 hosts.
Dec 16 22:00:06 dione cardmgr[88]: shutting down socket 1
Dec 16 22:00:06 dione cardmgr[88]: executing: './scsi stop n/a'
Dec 16 22:00:07 dione cardmgr[88]: + open() failed: No such file or directory
Dec 16 22:00:07 dione cardmgr[88]: + /dev/n/a: No such file or directory
Dec 16 22:00:07 dione cardmgr[88]: stop cmd exited with status 1
Dec 16 22:00:07 dione cardmgr[88]: executing: 'rmmod sym53c500_cs'
Dec 16 22:00:07 dione cardmgr[88]: executing: 'rmmod sg'

If I activate de debugging option in the SYM53C500, I get a cascade of ( until now ) unreadable messages, and the screen border changes to blue ( is a debugging feature).

Thanks again, and again :-)

Question Apa1480 cardbus + Fujitsu M2513A MO drive problem

Date: 1999, Dec 12
From: Hin-Tak

MO drive doesn't want to mount disk when connected to
laptop via Apa1480 (formatted by a pentium desktop with AHA2940 card, same drive). Time out and then complains of too many filesystem mounted, wrong fstype etc.

Kernel message: (scsi:0:4:0) Parity error during Data-In phase

pcmcia package on laptop(Sony Viao F350): 3.0.14 as
shipped with Slackware 7; kernel 2.2.13 recompiled with
gcc 2.95.2. The Apa1480 card on the laptop works perfectly 
with an OEM Teac CD-R55S (cd recorder) with Schilly's cdrecord. Scsi core and the 4 drivers are all compiled into the kernel. I also have a cardbus 3com ethernet card which works happily. I don't usually use both simultaneously, so probably irrelevant.

Desktop has kernel 2.2.10 compiled with egcs 1.1.2 before the upgrade to slackware 7. 

Separately, Pcmcia APA1480 SCSI card works (laptop + cd writer) and the Fujitsu M2513A MO drive works (desktop + AHA2940). But put together, they don't. 

Is this the problem mentioned in the "KNOWN BUGS" of pcmcia 3.1.15 about the apa1480_cs being a bit quirky about exotic scsi hardware? Would upgrading to 3.1.16 help?

Except for the "Parity error during Data-In phase" message, 
syslog shows the usual thing: card detected, drive detected, etc. 

None Maybe a kernel version issue

Re: Question Apa1480 cardbus + Fujitsu M2513A MO drive problem
Date: 1999, Dec 13
From: David Hinds <dhinds@pcmcia.sourceforge.org>

That's odd, because the AHA2940 is using the same Linux SCSI driver as
the APA1480 card.

Upgrading PCMCIA is unlikely to help.  The base PCMCIA stuff is
working fine, since your other SCSI device is working.  The only
variable that looks suspicious would be the kernel version (2.2.10
versus 2.2.13): I know that there have been changes in the aic7xxx
driver between those releases.  PCMCIA just pulls the aic7xxx driver
out of your kernel tree when you compile it.  Can you try 2.2.10 on
the laptop?

-- Dave

Question bus toaster problem with aha152x

Date: 1999, Dec 14
From: Ted Herman <herman@cs.uiowa.edu>

Hi,

I have a bus toaster SCSI adaptor that works fine on an older (circa 1994) laptop, using the old toaster_cs module for its support. Now I am trying a new laptop, the Casio Fiva, and using Slackware 4 for the software. The bus toaster card doesn't work in this situation. I am wondering from the following messages whether the problem is software or could it be that the Fiva uses nonstandard hardware? The Fiva does properly work with my New Media Ethernet pcmcia card, so I know that at least one pcmcia module works OK.

-- Ted.

Log follows.

Linux version 2.2.6 (root@zap) (gcc version 2.7.2.3) #52 Tue Apr 27 17:00:17 CDT 1999
...
... (first try)
...
Linux PCMCIA Card Services 3.0.9
  kernel build: 2.2.6 #20 Sat Apr 17 23:17:12 CDT 1999
  options:  [pci] [cardbus]
Intel PCIC probe:
  Ricoh RL5C475 PCI-to-CardBus at bus 0 slot 1, mem 0x68000000, 1 socket
    host opts [0]: [isa irq] [io 3/6/1] [mem 3/6/1] [no pci irq] [lat 168/176] [bus 32/34]
    ISA irqs (default) = 3,4,5,7,9,10,12,15 polling interval = 1000 ms
cs: IO port probe 0x1000-0x17ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x330-0x337 0x388-0x38f 0x398-0x39f 0x480-0x48f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0xa0000000-0xa0ffffff: clean.
aha152x: processing commandline: ok
aha152x: BIOS test: passed, detected 1 controller(s)
aha152x0: vital data: PORTBASE=0x100, IRQ=3, SCSI ID=7, reconnect=enabled, parity=enabled, synchronous=disabled, delay=100, extended translation=disabled
aha152x: trying software interrupt, ok.
scsi0 : Adaptec 152x SCSI driver; $Revision: 1.7 $
scsi : 1 host.
scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 0, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 0, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 1, scsi0, channel 0, id 1, lun 0 Test Unit Ready 00 00 00 00 00
SCSI device not inactive - rq_status=65535, target=1, pid=1, state=4099, owner=258.
Device busy??? <---- I took out the card
scsi : aborting command due to timeout : pid 1, scsi0, channel 0, id 1, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 2, scsi0, channel 0, id 2, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 2, scsi0, channel 0, id 2, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 3, scsi0, channel 0, id 3, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 3, scsi0, channel 0, id 3, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 4, scsi0, channel 0, id 4, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 4, scsi0, channel 0, id 4, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 5, scsi0, channel 0, id 5, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 5, scsi0, channel 0, id 5, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 6, scsi0, channel 0, id 6, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 6, scsi0, channel 0, id 6, lun 0 Test Unit Ready 00 00 00 00 00
aha152x_cs: no SCSI devices found
...
... second try
...
aha152x: processing commandline: ok
aha152x: BIOS test: passed, detected 1 controller(s)
aha152x0: vital data: PORTBASE=0x100, IRQ=5, SCSI ID=7, reconnect=enabled, parity=enabled, synchronous=disabled, delay=100, extended translation=disabled
aha152x: trying software interrupt, lost.
aha152x: IRQ 5 possibly wrong.  Please verify.
aha152x_cs: no SCSI devices found
scsi : 0 hosts.
...
... third try
...
aha152x: processing commandline: ok
aha152x: BIOS test: passed, detected 1 controller(s)
aha152x0: vital data: PORTBASE=0x100, IRQ=3, SCSI ID=7, reconnect=enabled, parity=enabled, synchronous=disabled, delay=100, extended translation=disabled
aha152x: trying software interrupt, ok.
scsi0 : Adaptec 152x SCSI driver; $Revision: 1.7 $
scsi : 1 host.
scsi : aborting command due to timeout : pid 7, scsi0, channel 0, id 0, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 7, scsi0, channel 0, id 0, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 8, scsi0, channel 0, id 1, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 8, scsi0, channel 0, id 1, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 9, scsi0, channel 0, id 2, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 9, scsi0, channel 0, id 2, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 10, scsi0, channel 0, id 3, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 10, scsi0, channel 0, id 3, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 11, scsi0, channel 0, id 4, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 11, scsi0, channel 0, id 4, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 12, scsi0, channel 0, id 5, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 12, scsi0, channel 0, id 5, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 13, scsi0, channel 0, id 6, lun 0 Test Unit Ready 00 00 00 00 00
scsi : aborting command due to timeout : pid 13, scsi0, channel 0, id 6, lun 0 Test Unit Ready 00 00 00 00 00
aha152x_cs: no SCSI devices found
scsi : 0 hosts.

None Maybe an interrupt conflict?

Re: Question bus toaster problem with aha152x (Ted Herman)
Date: 1999, Dec 14
From: David Hinds <dhinds@pcmcia.sourceforge.org>

On Tue, Dec 14, 1999 at 10:23:41AM -0800, Ted Herman wrote:

> Linux PCMCIA Card Services 3.0.9

The PCMCIA-HOWTO will tell you that "aborting command due to timeout"
normally indicates an interrupt conflict, in this case, on irq 3.
This is also a pretty old PCMCIA distribution.  So I'd first try
excluding interrupts in /etc/pcmcia/config.opts, and if that does not
help, then do an upgrade.

-- Dave

Warning Verify the kind of Bus Toaster card

Re: Question bus toaster problem with aha152x (Ted Herman)
Date: 2000, Jan 06
From: Francisco Escarpa PacoEscarpa

Hello:

Are you sure that the card chipset is Aha-152X ?. I had the same problem when I first test my Bus Toaster card. At last it resulted that bus toaster cards newer than 1997 uses a symbios chipset, that are only partialy supported ( I still can't use my card with Linux ).

If the part number ( over the barcode ) is 404-15-002A, it uses symbios chipset.

Good luck, and happy new year.

Paco

Question Problem with configure HP 820 Rewriter

Date: 2000, Jan 07
From: Antonio Trujillo Carmona trujo

Recetly I buy a rewriter HP M820 in order to use with my TI notebok, it is a scsi rewriter of CD with a cardbus to scsi card.

Ican use this writer in a desktop box (of course with Linux) but I cant use it in the notebok. I ask to HP and they say me that what I was ask is reserver information ;-)

I insert the pcmcia card and and obtain:
 unsoporter card
 "KME    ",  "SCSI-card", "1"
so I look at /etc/pcmcia/config and I can see some "KME " (only with 1 space)
and all use qlogic controler.
then I add one item with "KME    "  ...
restart psmcia and load in memory qlogicisp module with insmod
after that Iinsert the card and obtain that

Jan  2 22:18:29 localhost cardmgr[630]: initializing socket 1
Jan  2 22:18:29 localhost cardmgr[630]: socket 1: HP
Jan  2 22:18:29 localhost cardmgr[630]: module /lib/modules/2.2.12-4/pcmcia/qlogicisp.o not available
Jan  2 22:18:29 localhost kernel: qlogic_cs: GetFirstTuple: No more items
Jan  2 22:18:29 localhost kernel: scsi : 0 hosts.
Jan  2 22:18:29 localhost kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000
Jan  2 22:18:29 localhost kernel: current->tss.cr3 = 01d23000, %cr3 = 01d23000
Jan  2 22:18:29 localhost kernel: *pde = 00000000
Jan  2 22:18:29 localhost kernel: Oops: 0000
Jan  2 22:18:29 localhost kernel: CPU:    0
Jan  2 22:18:29 localhost kernel: EIP:    0010:[rw_intr+1009/1812]
Jan  2 22:18:29 localhost kernel: EFLAGS: 00010246
Jan  2 22:18:29 localhost kernel: eax: 00000000   ebx: 00000000   ecx: 8304d8a0   edx: 00000000
Jan  2 22:18:29 localhost kernel: esi: 00000000   edi: 00000000   ebp: 81d15900   esp: 81d158cc
Jan  2 22:18:29 localhost kernel: ds: 0018   es: 0018   ss: 0018
Jan  2 22:18:29 localhost kernel: Process cardmgr (pid: 630, process nr: 16, stackpage=81d15000)
Jan  2 22:18:29 localhost kernel: Stack: 81d15b08 81d15900 0000001d 8302ef20 8303668c 830363f2 801b88ba 8304d8a0
Jan  2 22:18:29 localhost kernel:        8304c60e 00000001 8304d8a0 0000000f 820826c0 81d15b28 8304c5da 8139b200
Jan  2 22:18:29 localhost kernel:        8304c6c0 8139b200 82082718 00d15ac8 81d15960 8302f7f4 8212c800 00000000
Jan  2 22:18:29 localhost kernel: Call Trace: [<8302ef20>] [<8303668c>] [<830363f2>] [rw_intr+1646/1812] [<8304d8a0>] [<830
4c60e>] [<8304d8a0>] Jan 2 22:18:29 localhost kernel: [<8304c5da>] [<8304c6c0>] [<8302f7f4>] [<8302fc2d>] [<8302fc55>] [<830312cd>] [ste xt_lock+1990/5528] [inflate_codes+994/1128] Jan 2 22:18:29 localhost kernel: [vesa_powerdown_screen+5/24] [<8304c74b>] [<8304c6c0>] [<8302d81f>] [sys_rt_sigtim edwait+263/892] [stext_lock+4631/5528] [remap_area_pages+310/588] [<83036a7e>] Jan 2 22:18:29 localhost kernel: [sys_sched_getparam+62/276] [<8302ec3a>] [<8304c160>] [<8304d920>] [<8304c6c0>] [< 8302edf9>] [<830473bd>] [<83047bda>] Jan 2 22:18:29 localhost kernel: [stext_lock+1990/5528] [inflate_codes+994/1128] [do_blank_screen+12/328] [inflate_ dynamic+588/1476] [inflate_dynamic+605/1476] [clear_buffer_attributes+68/124] [inflate_block+223/240] [__kstrtab_mtrr_del+1 2/44] Jan 2 22:18:29 localhost kernel: [proc_dointvec_minmax+408/712] [proc_dointvec_jiffies+24/36] [stext_lock+4631/5528 ] [<83048639>] [kmem_cache_grow+373/876] [do_acct_process+494/664] [sys_capset+73/548] [do_pipe+153/352]
Jan  2 22:18:29 localhost kernel:        [show_regs+44/168]
Jan  2 22:18:29 localhost kernel: Code: 66 8b 16 81 fa 3d 01 00 00 7f 24 52 a1 20 40 26 80 50 e8 88

Can some one help me?

thank

None You can't use the qlogic_cs driver with a CardBus device

Re: Question Problem with configure HP 820 Rewriter (Antonio Trujillo Carmona)
Date: 2000, Jan 07
From: David Hinds <dhinds@pcmcia.sourceforge.org>

What PCMCIA driver release are you using?  (it will be reported higher
up in your system log).

If this is really a CardBus-to-SCSI adapter, the qlogic driver will
never work.  I would then guess that you're using a pre-3.1.0 driver
release, since those older drivers could cause kernel traps if you
tried to load a 16-bit driver for a CardBus device.

Only one 32-bit CardBus-to-SCSI adapter is currently supported under
Linux: the Adaptec APA-1480 card.

-- Dave

More More info...

Re: None You can't use the qlogic_cs driver with a CardBus device (David Hinds)
Date: 2000, Feb 02
From: Dylan Thurston dthurston

I had a similar problem with the hp 820 rewritable. The card has a little switch on it between CardBus and 16-bit; I believe (but I'm not positive) I had it set to 16 bit. I was using version 3.1.8 of the PCMCIA modules. I can post the log if desired.

On further inspection, it looks like the SCSI modules were not loaded, which is odd, since I have the kernel module loader enabled. I'll try again. But it still shouldn't crash like that.

Question Cannot install SCSI card supplied with HP m820e

Re: Question Problem with configure HP 820 Rewriter (Antonio Trujillo Carmona)
Date: 2000, Jan 31
From: Stefan Sitter <Stefan.Sitter@lrz.uni-muenchen.de>

I'm also trying to get the HP m820e up (with the SCSI card
supplied by HP), but I fail totally.
I'm still using kernel 2.0.36, pcmcia version 2.9.6.
Was anybody more successful? Is my problem related to
my old (but otherwise stable) installation? Should I upgrade?

/var/log/messages says:
....      cardmgr{70]: unsupported card in socket 0
....      cardmgr[70]: version info: "KME    ", "SCSI-CARD-001", "1"
....      cardmgr[70]: manfid: 0x0032, 0x2604


Thank you for your attention.
-Stefan

Question Addition

Re: Question Cannot install SCSI card supplied with HP m820e (Stefan Sitter)
Date: 2000, Jan 31
From: Stefan Sitter <Stefan.Sitter@lrz.uni-muenchen.de>

I forgot to mention what the PCMCIA card looks like:

- A switch "CardBus/16 bit"

- Text: "Hewlett Packard, CD-RW PC Card".

None Not a supported card

Re: Question Cannot install SCSI card supplied with HP m820e (Stefan Sitter)
Date: 2000, Jan 31
From: David Hinds <dhinds@pcmcia.sourceforge.org>

This card is not supported.  I have no idea what chipset it uses.

-- Dave

Idea It is "supported"

Re: Question Cannot install SCSI card supplied with HP m820e (Stefan Sitter)
Date: 2000, Feb 05
From: Fortunato Navarro fortu

The way to make _the PCMCIA card_ works and why:

It seems HP sells this SCSI CD toaster with a SCSI pcmcia chipset
manufactured by KME (Matshusita).

If installed the drivers in Windoze you can read in
windows\inf\other\KMEUJDB310A.INF:

[KME]
;%*NJScsi.DeviceDesc%=NJScsi.Device, PCMCIA\WORKBIT-ULTRANINJA-16-8F2C
%*NJScsi.DeviceDesc%=NJScsi.Device, PCMCIA\KME____-SCSI-CARD-001-3406

and in windows\inf\other\KMEUJDB310B.INF:

[KME]
;%*WB_NINJA_SCSI.DeviceDesc%=WB_NINJA_SCSI.Device,PCI\VEN_1145&DEV_8007
%*WB_NINJA_SCSI.DeviceDesc%=WB_NINJA_SCSI.Device,PCI\VEN_1145&DEV_F007

It seems to be the drivers for the 16-bit and the CardBus configs (choosable
through the small switch on the card)

The "WB" refers to WorkBit. WorkBit make a SCSI pcmcia controller named
NinjaSCSI3 with the chipset of KME which has a driver in the contrib section,
of pcmcia-modules: "nin_cs".

I've compiled that module and installed with the version of pcmcia-modules
the author says to complain with. To make it works with the pcmcia shipped
with M820e you also have to put the following lines in your
/etc/pcmcia/config:

device "nin_cs"
  class "scsi" module "nin_cs"

  card "Workbit NinjaSCSI-3"
  version "KME ","SCSI-CARD-001"
  bind "nin_cs"

The module is a port from FreeBSD and Linux source code given by Workbit
(http://www.workbit.co.jp/workbit/products/driver/sc3-unix.html) and it's
latest version can be downloaded from:

http://www.netlab.is.tsukuba.ac.jp/~yokota/izumi/ninja/
     
About the switch: as far as I've tested it doesn't work in CardBus mode,
and I think it's because pcmcia-modules doesnt provide support for it. My
laptop neither support CardBus so I can tell more about it.

Well, now the bad news....

I've managed to use the pcmcia controller with a Fujitsu 4GB SCSI2 hard-disk
in 16 bit mode under linux 2.2, and it works great (a little slow due to the
lack of CardBus support). But when I plug the M820e strange things happen.

Usually I can't mount a CD and I have the following logs:

Jan 16 23:25:55 cid kernel: scsi : aborting command due to timeout : pid 199, scsi0, channel 0, id 4, lun 0 Read (10) 00 00 04 79 75 00 00 01 00
Jan 16 23:25:59 cid kernel: scsi0 channel 0 : resetting for second half of retries.
Jan 16 23:25:59 cid kernel: SCSI bus is being reset for host 0 channel 0.
Jan 16 23:25:59 cid kernel: scsi0: device driver called scsi_done() for a syncronous reset.

in a no-end loop. Sometimes it can mount the CD but when it attemps to read
something big it does the same crash. The worse of all is that it hangs the
machine completely.

I have hope someday this works, once I could mount even a CD in udf format!,
but crashed when reading as before.

   (Off-Topic: UDF it's a format wich allow use of CD as a "big floppy",
   you can append and remove (if its a RW media) without making images of
   disk. There is support for it in 2.2.x kernels and its compatable with
   other Windoze software.)

I thought it was a interrupt conflict, but with different irq and other io
port it happens the same. I have a Toshiba laptop (welcome Mr.Murphy ;) so
any experience from other more-normal-laptop user could help.

Somebody in this forum (Antonio Trujillo?, if you're in Spain, please
drop me a mail to fortu@die.upm.es) said he could work with the M820e in
a Desktop PC with a 'normal' SCSI adapter. I tried with no success.

PLEASE, give some feedback of your experience with the configuration above.
The CD Writer seems to be very good hardware, also the pcmcia card. May be
we (HP clients :) could ask together HP for some specs or support
(a month ago they said to support Linux, so ...)

thanks.

More Untitled

Re: Idea It is "supported" (Fortunato Navarro)
Date: 2000, Feb 07
From: Stefan Sitter <Stefan.Sitter@lrz.uni-muenchen.de>

Thank you Fortunato:

This sounds like good news. I'll get the ninja driver package and try.

I can confirm Antonio's report that the HP m820e works with a "normal" SCSI adapter. So far, I tried this from an old box (486 CPU, 50MHz, therefore only double speed because of buffer underrun with quadruple speed; Adaptec 1542C SCSI adapter; Kernel version 2.0.36). The HP writer was the sole SCSI device connected. I was able to use CDR and CDRW media, both with cdrecord and cdrdao. I have not yet tried 700MB CD-R media.

I had asked the German HP support about the HP m820e, and they said that HP themselves do not support Linux and they cannot guarantee that third-party solutions have the full functionality because they have not been provided by HP ;-). They also said that the PCMCIA adapter was NOT an SCSI adaptor but something "separate". Also, media with more than normal capacity were not supported:-(

They would offer some solutions at:
     http://www2.hp.com/cposupport/
(there they say that 80min media ARE supported:-)

However, I do not find any data about the PCMCIA adaptor there.

-Stefan

None More about M820

Re: Idea It is "supported" (Fortunato Navarro)
Date: 2000, Feb 11
From: trujo trujo

 After read what Fortunato wrote I try to recompile pcmcia with ninja 
and in that frist momen all work well, but after the first powerd off I can't
mount it never more, I alwai obtain the some anwers
Feb  9 01:50:04 localhost cardmgr[611]: initializing socket 0 
Feb  9 01:50:04 localhost cardmgr[611]: socket 0: Workbit NinjaSCSI-3 
Feb  9 01:50:04 localhost cardmgr[611]: executing: 'insmod /lib/modules/2.2.11-2mdk/pcmcia/nin_cs.o' 
Feb  9 01:50:04 localhost kernel: scsi0 : NinjaSCSI-3 Driver version 0.2.1, I/Obase 0x240 IRQ  9 ID 7  
Feb  9 01:50:04 localhost kernel: scsi : 1 host.  
Feb  9 01:50:04 localhost kernel:   Vendor: hp        Model: CD-Writer+ M820   Rev: 1.23  
Feb  9 01:50:04 localhost kernel:   Type:   CD-ROM                             ANSI SCSI revision: 02  
Feb  9 01:50:04 localhost kernel: Detected scsi CD-ROM sr0 at scsi0, channel 0, id 4, lun 0  
Feb  9 01:50:04 localhost kernel: Detected scsi CD-ROM sr0 at scsi0, channel 0, id 4, lun 0  
Feb  9 01:50:04 localhost kernel: sr0: scsi3-mmc drive: 20x/20x writer cd/rw xa/form2 cdda pop-up  
Feb  9 01:50:04 localhost kernel: sr0: scsi3-mmc drive: 20x/20x writer cd/rw xa/form2 cdda pop-up  
Feb  9 01:50:04 localhost cardmgr[611]: executing: './scsi start scd0' 

When I try to mount  I get

Feb  9 01:50:24 localhost kernel: attempt to access beyond end of device  
Feb  9 01:50:24 localhost kernel: 0b:00: rw=0, want=33, limit=2  
Feb  9 01:50:24 localhost kernel: dev 0b:00 blksize=1024 blocknr=32 sector=64 size=1024 count=1  
Feb  9 01:50:24 localhost kernel: isofs_read_super: bread failed, dev=0b:00, iso_blknum=16, block=32  

I think is a bug in ninja drivers, can some one can helpme in the way of debug
it?

Question What PCMCIA SCSI card works with M820E?

Re: Question Cannot install SCSI card supplied with HP m820e (Stefan Sitter)
Date: 2000, Jun 13
From: christopher snow snowchr

I need to get my M820E working quickly with Linux and have resigned myself to the fact that I will probably have to buy a another PCMCIA scsi card to do so. I am thinking of the Adaptec 1460 SlimSCSI. Has anyone had any experiences of this (or any other) card with the M820E? Will it work?

Thanks, Chris Snow

None SCSI card that works with HP M820e

Re: Question What PCMCIA SCSI card works with M820E? (christopher snow)
Date: 2000, Aug 01
From: christopher snow snowchr

The HP M820e works using the adaptec 1460D SlimSCSI card under Mandrake 7.1. No configuration was necessary, the devices were detected during the installation of Mandrake. Note: so far I have only tested CDCOPY and multi-session modes using X-CDRoast and Kisocd.

Regards, Chris

Question M820e with the 2.4.0-test15 kernel; driver almost works?

Re: None SCSI card that works with HP M820e (christopher snow)
Date: 2000, Aug 05
From: Johan Larsson jola

Hi.

I have problems with M820e and the latest NinjaSCSI3 alpha-000615. The kernel reports in /var/log/messages:

Aug 5 16:05:44 voyager kernel: cs: IO port probe 0x1000-0x17ff: clean.
Aug 5 16:05:44 voyager kernel: cs: IO port probe 0x0100-0x04ff: excluding
0x300-0x307 0x378-0x37f 0x4d0-0x4d7
Aug 5 16:05:44 voyager kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Aug 5 16:05:44 voyager kernel: cs: memory probe
0xa0000000-0xa0ffffff: clean.
Aug 5 16:05:44 voyager kernel: ttyS01 at port 0x02f8 (irq = 3) is a
16550A
Aug 5 16:05:44 voyager kernel: scsi0 : NinjaSCSI-3/32Bi Driver version
1.0, I/O 0x0100-0x0110 IRQ 9 ID 7, Chip ID 1/Rev 1
Aug 5 16:05:44 voyager kernel: scsi : 1 host.
Aug 5 16:05:44 voyager keytable: Loading keymap:
Aug 5 16:05:44 voyager kernel: ninja3_cs: no SCSI devices found
Aug 5 16:05:44 voyager kernel: ninja3: index 0x01: Vcc 5.0, irq 9, io
0x0100-0x010f
Aug 5 16:05:44 voyager kernel: scsi : 0 hosts.
Aug 5 16:05:44 voyager kernel: Trying to free nonexistent resource
<0100-010f>

Can anyone help me fix this?

Perhaps there is problem with my configuration of the 2.4.0-test5 kernel? (Sometimes the pcmcia compilation
reports errors in cs_internal).

PS. The NinjaSCSI3 driver appeared to work ok. with the original kernel for the Red Hat 6.2 distribution (detected
by cdrecord, mounted drives etc).

Johan

None No SCSI devices found = kernel configuration problem

Re: Question M820e with the 2.4.0-test15 kernel; driver almost works? (Johan Larsson)
Date: 2000, Aug 07
From: David Hinds <dhinds@pcmcia.sourceforge.org>

> Aug 5 16:05:44 voyager kernel: ninja3_cs: no SCSI devices found

Reconfigure your kernel, and this time turn on CONFIG_BLK_DEV_SR
(maybe also CONFIG_BLK_DEV_SD, I think).  The SCSI host adapter driver
will not see any devices that your kernel doesn't have drivers for.

-- Dave

Warning Untitled

Re: Question Problem with configure HP 820 Rewriter (Antonio Trujillo Carmona)
Date: 2000, Aug 15
From: Yannis Skipper <skipper@performance.gr>

I have HP Omnibook 5700CT (PCMCIA card controler TI 1131) and HP M830e.

And I have problems, too. Right now I build new kernel 2.4.0-test6 with last pcmcia lib and last NinjaSCSI3 drivers for new kernels.

My Best Regards

Skipper

Question Help w/ programming EigerLabs SCSI (fuji600) card?

Date: 2000, Jan 19
From: Josko Plazonic jplazo

Hi,

I was successful w/ reverse engineering eigerlabs scsi fuji600 based scsi card - i.e. I have a working(!) windows scsi miniport driver source. Unfortunately I don't own a laptop any more (and I am about to sell the card as well) nor do I know nearly enough about scsi drivers in linux to finish the job (and most importantly - I don't have that much free time anymore...).

Anyone outthere who wants to try? Any suggestions David?

Thanks,

JP

Question Critical bug ? (pcmcia-cs SCSI apa1460A)

Date: 2000, Jan 22
From: Maya Tamiya maya

Hi,

I'm having trouble trying to use SCSI hard disk drive with a PCMCIA
card, and it seems to be a PCMCIA problem to me.

- On using pcmcia-cs SCSI (APA-1460A), the filesystem corrupts.

- I did all the tests on both linux 2.0.36 and 2.0.38, and the results 
  were the same.
  The tests were done on pcmcia-cs-3.0.[5-9], 3.0.{13,14}, 3.1.{7,8},
  and they succeeded (no error occured) only in the case of 3.0.5. 

- It seems that the error tends to occure when the load average gets
  high. 

- Symptoms

        * "e2fsck -f -v /dev/sda1" sometimes fails.

        * Even if the first e2fsck try is successful, it ends up
          failing if I repeat some more right after that.

        * If I terminate it with C-c instead of inputting <y> after
          failing and try again, then sometimes it does not fail.
          However, it ends up failing if I repeat some more. 

        * I didn't do any "mount" during the test, of course.

        * In the case of 3.0.5, I got no fail out of more than 100
          e2fsck's on exactly the same file system (without any
          fixing).

        * I've given it a total of more than 150 tests with 3 HDDs,
          and the only case it didn't fail was the case of 3.0.5. 

        * I believe there is no problem with quality and setting of my
          hardware, because it never failed at all in the case of
          3.0.5, in which I replaced only pcmcia-cs without touching
          anything but the keyboad.  Replacing it with other versions
          again and giving some more tries would get it a failure.

- Error messages      

     [dmesg]
     aha152x: more data than expected (2 bytes)
     aha152x: data ( 0  0 )
     (*** every time the same)

     [fsck]
     Inode 209163 has illegal block(s).  Clear<y>?
     or
     Deleted inode 245289 has zero dtime.  Fix<y>?
     etc.
     (the numbers (209163 and 245289) change every time.)

- PCMCIA log

Linux PCMCIA Card Services 3.0.5
  kernel build: 2.0.38 #1 Tue Jan 18 08:16:13 JST 2000
  options:  [pci] [cardbus] [apm]
Intel PCIC probe: 
  Cirrus PD6832 CardBus at mem 0x68000000, 2 sockets
    host opts [0]: [ring] [0/7/4] [3/17/4] [no pci irq] [lat 168/176] [bus 32/34
]
    host opts [1]: [ring] [0/7/4] [3/17/4] [no pci irq] [lat 168/176] [bus 35/37
]
    ISA irqs (default) = 3,4,5,7,11,12 status change on irq 11
cs: IO port probe 0x0100-0x03ff: excluding 0x220-0x22f 0x2f8-0x2ff 0x398-0x39f 0
x3f8-0x3ff
cs: IO port probe 0x0a20-0x0a27: clean.
Swansea University Computer Society IPX 0.34 for NET3.035
IPX Portions Copyright (c) 1995 Caldera, Inc.
Appletalk 0.17 for Linux NET3.035
--- 8< --- 8< --- 8< --- (snip) --- 8< --- 8< --- 8< ---
cs: warning: no high memory space available!
cs: memory probe 0x0d0000-0x0dffff: excluding 0xdc000-0xdffff
cs: memory probe 0x0e0000-0x0effff: excluding 0xe0000-0xeffff
cs: memory probe 0x0c0000-0x0cffff: excluding 0xc0000-0xcbfff
eth0: 3Com 3c589, port 0x300, irq 3, Auto port, hw_addr 00:60:97:93:1A:AB
Swansea University Computer Society IPX 0.34 for NET3.035
IPX Portions Copyright (c) 1995 Caldera, Inc.
Appletalk 0.17 for Linux NET3.035
aha152x: processing commandline: ok
aha152x: BIOS test: passed, detected 1 controller(s)
eth0: autodetected 10baseT
aha152x0: vital data: PORTBASE=0x340, IRQ=5, SCSI ID=7, reconnect=enabled, parit
y=enabled, synchronous=disabled, delay=100, extended translation=disabled
aha152x: trying software interrupt, ok.
scsi0 : Adaptec 152x SCSI driver; $Revision: 1.18 $
scsi : 1 host.
  Vendor: MELCO     Model: DSC-UE8.3G        Rev: 1.05
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sda at scsi0, channel 0, id 0, lun 0
  Vendor: MELCO     Model: DSC-UE8.3G        Rev: 1.05
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sdb at scsi0, channel 0, id 1, lun 0
  Vendor: MELCO     Model: DSC-UE8.3G        Rev: 1.04
  Type:   Direct-Access                      ANSI SCSI revision: 02
Detected scsi disk sdc at scsi0, channel 0, id 2, lun 0
SCSI device sda: hdwr sector= 512 bytes. Sectors= 16405632 [8010 MB] [8.0 GB]
 sda: sda1
SCSI device sdb: hdwr sector= 512 bytes. Sectors= 16405632 [8010 MB] [8.0 GB]
 sdb: sdb1
SCSI device sdc: hdwr sector= 512 bytes. Sectors= 16405632 [8010 MB] [8.0 GB]
 sdc: sdc1 sdc2

        * The line "cs: warning: no high memory space available!"
          disappears after I eject and insert the card.

        * 3.0.5 does not recognize insertion of the card, so I do
          "cardctl insert" after I insert the card.



Thank you very much!

--
Thank you, 
maya

None I think it has to do with the CardBus bridge setup

Re: Question Critical bug ? (pcmcia-cs SCSI apa1460A) (Maya Tamiya)
Date: 2000, Jan 24
From: David Hinds <dhinds@pcmcia.sourceforge.org>

>   The tests were done on pcmcia-cs-3.0.[5-9], 3.0.{13,14}, 3.1.{7,8},
>   and they succeeded (no error occured) only in the case of 3.0.5. 

I think this is the clue.  Use a fairly recent package (3.0.14 or
3.1.8) and put:

   PCIC_OPTS="cb_write_post=0"

in /etc/sysconfig/pcmcia and see if that helps.

-- Dave

Question Yes, thank you and with 3COM's CardBus NIC 3CCFE575CT?

Re: None I think it has to do with the CardBus bridge setup (David Hinds)
Date: 2000, Jan 27
From: Maya Tamiya maya

> >  The tests were done on pcmcia-cs-3.0.[5-9], 3.0.{13,14},
> >  3.1.{7,8}, and they succeeded (no error occured) only in the 
> >  case of 3.0.5.
>
> I think this is the clue.  Use a fairly recent package (3.0.14 or
> 3.1.8) and put:
> 
>    PCIC_OPTS="cb_write_post=0"
> 
> in /etc/sysconfig/pcmcia and see if that helps.

Yes, that helped! Thank you very much!

I've run a test script for more than 12 hours and there seems to be no problem.

But I have one thing I'm concerned with. I'm not using a CardBus Card now, but I'm thinking about using 3COM's CardBus NIC 3CCFE575CT along with apa1460 (at the same time). Would that option affect performance or something?

Thank you very much.

 --
 Thanks,
 maya

None I'm afraid so

Re: Question Yes, thank you and with 3COM's CardBus NIC 3CCFE575CT? (Maya Tamiya)
Date: 2000, Jan 27
From: David Hinds <dhinds@pcmcia.sourceforge.org>

Yes, this should impact performance; I'm not sure how large the impact
will be.

-- Dave

Question Backup Freezing Problems with APA-1480A & Sony DAT Tape Drive

Date: 2000, Jan 31
From: <jung_p@vptt.ch>

Hello,

I am experiencing systematic scsi transfer freezings after 4-8 min (400-800MB) of backing up with the configuration below. This happens as well when transfering data via tar as with the bought Arkeia backup software. Again, it happens as well with the APA-1480 as with the APA-1460 card.

With the APA-1480, it is at least possible to shut down the application. With the APA-1460, the whole Laptop freezes.

Dave Hinds suggested for the APA-1460 to disable disconnection / reconnection support (module "aha152x_cs" opts "reconnect=0"), which works great for the APA-1460A somebody gave me for testing. Now, since I have the same problem with my APA-1480A, I tried this setting on module "aha1480_cb" ..., but then, no SCSI device shows up anymore ! Without the latter setting on aha1480_cb, I can see and address all the connected SCSI devices but have the tape freezing problem. Any hint here ? Is the disconnection / reconnection managed differently ?

Btw. when I start the laptop with the APA-1480 adapter plugged in without any connected scsi device, the laptop freezes upon detecting the APA-1480 itself. However, booting with the adapter and at least one SCSI device connected, there is no problem. Could this be linked to the above problem ?

Generally, when connecting a SCSI disk instead of the tape drive however, there seems to be no problem, even when transferring large amounts of data. The cable is OK, the termination is the same. The tape drive works perfectly in this configuration under Windows NT4.0 (on the same laptop).

Would you have a hint at which settings/logs to look at ? Is it a problem to specifically connect a SCSI DAT tape drive to the APA-1480A ?

Thank you in advance, Best regards, Pierre.

My configuration: Hardware: IBM Thinkpad 600 / 128 MB RAM, Adaptec APA-1480B, Cardbus PCMCIA-SCSI adapter or Adaptec APA-1460A Slim SCSI adapter, Sony SDT-S9000 DAT DDS3 tape drive with active termination

Software: SuSE Linux 6.3, Kernel (new, compiled) 2.2.14, PCMCIA 3.19 (18.01.2000), tar Arkeia v4.2 backup SW

Idea load st-module by changing file /etc/pcmcia/config

Re: Question Backup Freezing Problems with APA-1480A & Sony DAT Tape Drive
Date: 2000, Feb 15
From: Peter Sondhauss <peter@roentgen.physik.uni-jena.de>

Dear Pierre,

I had very similar problems using DDS3 DAT drives with a Adaptec SlimSCSI 1460D adaptor under SuSE-Linux 6.2. I could read tapes, but when I tried to write, the system freezed earlier or later

In my case the reason was, that the system seemed to lack tape-support, although the kernel was compiled with built-in tape-support. It helped to load the st-module before the device driver module (in my case aha152x_cs) by modifying the file /etc/pcmcia/config:

 device "aha152x_cs"
   class "scsi" module "scsi/st", "aha152x_cs"

(see also the PCMCIA-Howto section 4.5 "PCMCIA SCSI adapters").

Probably this helps you, too.

-- Peter

Question SlimSCSI 1480A with pcmcia-3.1.6

Date: 2000, Feb 06
From: Alex Ma alexma7

Hi,
I have a Adaptec SlimSCSI 1480a and can not get it to work with Thinkpad i1460 properly.  

My configraion is,
	RedHat 6.1, kernel 2.2.12-20
	pcmcia 3.1.6
below is the messages from boot up,

pcmcia: Starting PCMCIA services:
pcmcia:  modules
kernel: Linux PCMCIA Card Services 3.1.6 
kernel:   kernel build: 2.2.12-20 #1 Mon Sep 27 10:40:35 EDT 1999 
kernel:   options:  [pci] [cardbus] [apm] 
kernel: Intel PCIC probe:  
kernel:   O2Micro OZ6812 PCI-to-CardBus at bus 0 slot 19, mem 0x68000000, 1 socket 
kernel:     host opts [0]: [a 24] [b 8d] [c 00] [d 83] [mhpg 18] [fifo 00] [e 2a] [no pci irq] [lat 32/176] [bus 32/34] 
kernel:     ISA irqs (default) = 3,4,7,11,12 polling interval = 1000 ms 
pcmcia:  cardmgr.
rc: Starting pcmcia succeeded
cardmgr[376]: starting, version is 3.1.6
cardmgr[376]: watching 1 sockets
kernel: cs: IO port probe 0x1000-0x17ff: clean. 
kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x3b8-0x3df 0x408-0x40f 0x480-0x48f 0x4d0-0x4d7 
kernel: cs: IO port probe 0x0a00-0x0aff: clean. 
kernel: cs: cb_alloc(bus 32): vendor 0x9004, device 0x6075
cardmgr[376]: initializing socket 0
cardmgr[376]: socket 0: Adaptec APA-1480 SCSI Host Adapter
inet: inetd startup succeeded
cardmgr[376]: executing: 'insmod /lib/modules/2.2.12-20/pcmcia/cb_enabler.o'
cardmgr[376]: executing: 'insmod /lib/modules/2.2.12-20/pcmcia/apa1480_cb.o reset=1 ultra=1'
kernel: cs: cb_config(bus 32) 
kernel:   fn 0 bar 1: io 0x200-0x2ff 
kernel:   fn 0 bar 2: mem 0x60030000-0x60030fff 
kernel:   fn 0 rom: mem 0x60020000-0x6002ffff 
kernel: apa1480_attach(bus 32, function 0) 
kernel: (scsi0) <Adaptec PCMCIA SCSI controller> found at PCI 0/0 
kernel: (scsi0) Narrow Channel, SCSI ID=7, 3/255 SCBs 
kernel: (scsi0) Downloading sequencer code... 413 instructions downloaded 
kernel: scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.20/3.2.4 
kernel:        <Adaptec PCMCIA SCSI controller> 
kernel: scsi : 1 host. 

That's it.  Non of the devices (Jaz, Zip and a hard drive) was detected after that.  Can anyone please help?
Thank you in advance.  -Alex

None My first suggestion is to upgrade to 3.1.10

Re: Question SlimSCSI 1480A with pcmcia-3.1.6 (Alex Ma)
Date: 2000, Feb 07
From: David Hinds <dhinds@pcmcia.sourceforge.org>

I suspect that this is a PCI interrupt configuration problem that is
fixed in the current (3.1.10) drivers.

-- Dave

Ok It is working, you're cool.

Re: None My first suggestion is to upgrade to 3.1.10 (David Hinds)
Date: 2000, Feb 08
From: Alex Ma alexma7

Thank you Dave,

I upgraded to 3.1.10 and that solve the problem.  Now I can access jaz, zip and a external HD.  I am thinking about getting a external scsi cd-rw.  Do you have any suggestion?
Thanks again for your work on pcmcia.

-Alex

News Success with Panasonic KXL-RW10A (aka HP 820)

Date: 2000, Mar 02
From: Bill Randle billr

Thanks to all the hints in this newsgroup regarding the HP 820 and the Workbit Ninja SCSI card, I was able to get my new Panasonic KXL-RW10A up and working this afternoon. This uses the same KME scsi card used by the HP 820. [In fact, I read somewhere that the HP is simply a rebadged Panasonic, but I don't know for sure.]

In any event, I used kernel 2.2.14, pcmcia-cs-3.1.11 and the Ninja driver package from http://www.netlab.is.tsukuba.ac.jp/~yokota/izumi/ninja.

I used the supplied config info from the Ninja package, changing the version line to:

	version "KME    ","SCSI-CARD-001","1"

Note that the four spaces after KME are very important, otherwise it won't get recognized. [The version line in Fortunato Navaro's posting about the HP 820 had only a single space.]

It does work only in 16-bit mode, so you must make sure the little slide switch on the PCMCIA card is set to "16bit".

Thanks to all here, and especially Dave, for a great package.

	-Bill Randle
	Tektronix, Inc.
	billr@exgate.tek.com

Question Which PC card controller chipset do you have?

Re: News Success with Panasonic KXL-RW10A (aka HP 820) (Bill Randle)
Date: 2000, Aug 04
From: Karl Oberjohn kwoberjohn

Hello, Bill,

Another PCMCIA-SCSI newsgroup user (Alberto Gonzalez Palomo) and I also have the Panasonic KXL-RW10A SCSI cd-rw drive, but we are having limited success with it under Linux (see our posts to this newsgroup). Both of us have laptops with a TI 1220 pc-card controller chipset. I am curious, which controller does your notebook have?

Thanks for the info,

Karl Oberjohn

Question ADA 1480 and other 32 bit card?

Date: 2000, Mar 21
From: Jim Dai dai

I have a toshiba Portege 7200 and 2GB Jaz drive. I tried SlimSCSI 1480 about year ago, giving it up after many tries. I have been using SlimSCSi 1460B since then, and are disturbed by slow speed the card gives.

I would like to move to a 32 bit card. ADA 1480 is the first one comes to my mind. I have seen successfull stories with this card. Is the lastest pcmcia support for this card reliable? (I have redhat 6.1 + pmccia 3.1.13.)

Of course, I do not limit myself to Ada 1480. Are there any other cards that give me fast access to jaz drive that have been working reliably?

Thanks

Please email response to dai@isye.gatech.edu

None The APA1480 is the only supported CardBus SCSI card

Re: Question ADA 1480 and other 32 bit card? (Jim Dai)
Date: 2000, Mar 21
From: David Hinds <dhinds@pcmcia.sourceforge.org>

The APA1480 card is the only CardBus SCSI card that is supported under
Linux at this time.

It works reliably for most people.  I have not received any problem
reports for it in recent memory.

However, I should also warn you that Toshiba laptops have generally
been fairly troublesome, and if you had problems before (depending on
what those problems were), you may still have problems now.

-- Dave

Sad SlimSCSI 1480A failed on Toshiba Portge 7020

Re: None The APA1480 is the only supported CardBus SCSI card (David Hinds)
Date: 2000, Mar 25
From: Jim Dai dai

Slim-SCSI failed on Toshiba Portege under pmcica-3.1.13 kernel 2.2.14, at least under default configuration. The only thing that I have not tried is edit /etc/sysconfig/pmcia and add PCIC_OPT="pci_int=0". I also see report of sucess under pcmcia 3.1.3 (same configuration, Portege 7020 + SlimSCSI 1480A).

Linux PCMCIA Card Services 3.1.13
  kernel build: 2.2.14 #1 Sun Mar 19 14:24:36 EST 2000
  options:  [pci] [cardbus] [apm]
PCI routing table version 1.0 at 0xf0180
  00:0b.0 -> irq 11
  00:0b.1 -> irq 11
Intel PCIC probe: 
  Toshiba ToPIC97 PCI-to-CardBus at slot 00:0b, mem 0x68000000
    host opts [0]: [slot 0xf0] [ccr 0x10] [cdr 0x86] [rcr 0x02] [pci irq 11] [lat 168/176] [bus 20/20]
    host opts [1]: [slot 0xf0] [ccr 0x20] [cdr 0x86] [rcr 0x02] [pci irq 11] [lat 168/176] [bus 21/21]
    ISA irqs (<6>    PCI irq 5 seems to be wedged!
    PCI irq 7 seems to be wedged!
    PCI irq 10 seems to be wedged!
    PCI irq 12 seems to be wedged!
    PCI irq 15 seems to be wedged!
default) = 4,5,7,9,10,12,15 PCI status changes
cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff
cs: IO port probe 0x0800-0x08ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x170-0x177 0x370-0x37f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0x0d0000-0x0dffff: clean.
eth0: 3Com 3c589, io 0x300, irq 5, hw_addr 00:20:AF:0D:FA:86

Ok SlimSCSI 1480 works on Toshiba Portege 7020

Re: Question ADA 1480 and other 32 bit card? (Jim Dai)
Date: 2000, Mar 26
From: Jim Dai dai

Redhat 6.1, kernel 2.2.14 and pcmcia-3.1.13

Adaptec SlimSCSI 1480 (32 bit card) works now after turning PCIC_OPTS="pci_int=0" in /etc/sysconfig/pmcia

Cheers

Question Panasonic KXL-783A: Let's try this again...

Date: 2000, Mar 28
From: Mike Bilow mbilow

I am thinking of taking yet another stab at a driver for the Panasonic KXL-783A CD-ROM drive. I looked into this about a year ago and gave up any hope of getting reasonable documentation, but several months later a package arrived in the mail from QLogic. So, in theory, I have some documentation on hand.

Things seem to have progressed a great deal since I last looked at this. I am running on Debian Potato, which uses a 2.2.14 kernel and 3.1.8 of PCMCIA card services.

The driver comes up and reports:

    scsi0: Qlogicfas Driver version 0.46, chip 50 at 250, IRQ 5, TPdma:1
    scsi: 1 host

Then, after a few seconds, it reports:

    scsi: 0 hosts

Apparently, the card is recognized by something, as "cardctl ident" reports:

    product info: "KME ", "KXLC003", "00"
    manfid: 0x0032, 0x0504
    function: 8 (SCSI)

Some strangeness... if I "exclude port=0x250-0x25f" (which is often used for power management stuff), the card comes up and reports "chip 50 at A50, IRQ 9" instead. But the card is not supposed to end up on 0xA50, as "dump_cis" reports in part:

    * * *
    cftable_entry 0x25
      io 0x240-0x25f [lines=10] [16bit] [range]
    cftable_entry 0x27
      io 0x280-0x29f [lines=10] [16bit] [range]
    cftable_entry 0x26
      io 0x260-0x27f [lines=10] [16bit] [range]
    cftable_entry 0x24
      io 0x220-0x23f [lines=10] [16bit] [range]

All Qlogic ISA cards are on 0x230, according to the driver source code comments. This does not apply to PCMCIA.

Is anyone else working on this? Does anyone care about the KXL-783A anymore? Why is the KXLC003 controller listed in the config file if it is unsupported, or is there something else wrong?

The only device on the SCSI chain is the KXL-783A CD drive, and it works under Windows 9x.

-- Mike

None It should actually work

Re: Question Panasonic KXL-783A: Let's try this again... (Mike Bilow)
Date: 2000, Mar 28
From: David Hinds <dhinds@pcmcia.sourceforge.org>

The 0xa50 IO address is not a problem.  This:

>     cftable_entry 0x25
>       io 0x240-0x25f [lines=10] [16bit] [range]

means that the card actually only decodes the lower 10 address lines,
so if X is ok, then X+0x400, X+0x800, etc are all ok.

Did you have the SCSI cd-rom driver loaded when you tested the card?
(if not, the CD-ROM will be ignored, consistent with your messages)

-- Dave

None Not quite what I meant...

Re: None It should actually work (David Hinds)
Date: 2000, Mar 28
From: Mike Bilow mbilow

No, no... if I am deliberately trying to exclude 0x250-0x25f because of conflicts with, say, power management, then the last thing I want is for the SCSI card to land on 0xA50 instead of 0x250. I understand about the 10-bit address issue, but suppose the power management stuff built into the notebook decodes only 10 bits? Then accesses to the SCSI card at 0xA50 will be misread by the power management hardware as accesses intended for it. Even if the power management hardware understands 16 bits, then the SCSI card would misread accesses to the power management hardware as intended for it, too. Only if both components "sharing" the lower 10 bits of the address are capable of decoding all 16 bits will such sharing work.

What you seem to imply is that, in order to protect 10-bit hardware, it is necessary to exclude all equivalent ranges? If that is the case, then perhaps a directive should be added to the "exclude" parameter to do this automatically.

What I should probably do is explicitly tell the driver what port range to use, although I am not sure exactly how to do that without looking into the driver source again. This machine happens to be an IBM ThinkPad 701CS (remember them?) so it also has some funny doings on the parallel port addresses (0x278-0x27f and 0x378x-0x37f) because one of the parallel ports is used for the printer port and the other is used for the disconnectable 3.5-inch floppy drive. In fact, I think there is some sound hardware at 0x220-0x22f also, so probably the only range at which the SCSI controller might be supported is 0x280-0x29f.

I thought of the issue about kernel support for SCSI CD-ROM devices, but I am running on the Debian installation kernel from the 2.2.9 Potato test release. (Note that Debian release numbers have no connection with kernel version numbers, despite looking similar by coincidence.) Until I can get the CD-ROM drive visible to the system, installing enough of Debian to build a kernel on that machine is going to be hard. I am fairly confident that SCSI CD-ROM support is built into the installation kernel, as this would be critically important and a major bug if not. The installation root disk does provide /dev/scd* nodes, too. I will see if I can double check this, however, as I am not absolutely certain.

If the KXLC003 controller really is supposed to work now, then I will put some effort into resolving this properly.

-- Mike

None The card decodes 10 bits, the bridge decodes 16 bits

Re: None Not quite what I meant... (Mike Bilow)
Date: 2000, Mar 28
From: David Hinds <dhinds@pcmcia.sourceforge.org>

The card itself decodes 10 bits of address space.  The PCMCIA bridge
decodes all 16 bits, so the card will never see accesses to aliases
like 0x250-0x25f if the bridge was configured for 0xa50-0xa5f.

If your power management device is particularly goofy and only decodes
10 bits of address, then yes, you would need to also exclude all
aliases of 0x250-0x25f.

There is no interface for directly forcing the IO port range for a
card.  I decided to not provide such an option because it should never
be needed and I expected it to be generally misused.

I'm not sure if it is a good idea to assume that your kernel has SCSI
CD-ROM support built in.  At least, check your /lib/modules/*/scsi
directory to see if there is an sr_mod module.  If so, try loading
that before inserting your card.

-- Dave

Question adaptec1480 scsi & RH6.2

Date: 2000, Apr 03
From: matthias caraco caraco

after upgrading to RH6.2 my adaptec 1480 scsicard does not work anymore it used to work fine under RH6.1

here is a part of /var/log/messages:

Apr  3 11:41:02 localhost pcmcia: Starting PCMCIA services:
Apr  3 11:41:02 localhost pcmcia:  modules
Apr  3 11:41:02 localhost kernel: Linux PCMCIA Card Services 3.1.8 
Apr  3 11:41:02 localhost kernel:   kernel build: 2.2.14-5.0 #1 Tue Mar 7 21:07:39 EST 2000 
Apr  3 11:41:02 localhost kernel:   options:  [pci] [cardbus] [apm] 
Apr  3 11:41:02 localhost kernel: Intel PCIC probe:  
Apr  3 11:41:02 localhost kernel:   TI 1220 PCI-to-CardBus at bus 0 slot 4, mem 0x68000000, 2 sockets 
Apr  3 11:41:02 localhost kernel:     host opts [0]: [ring] [pwr save] [serial pci & irq] [no pci irq] [lat 64/176] [bus 32/34] 
Apr  3 11:41:02 localhost kernel:     host opts [1]: [ring] [pwr save] [serial pci & irq] [no pci irq] [lat 64/176] [bus 35/37] 
Apr  3 11:41:02 localhost kernel:     ISA irqs (scanned) = 3,4,9 polling interval = 1000 ms 
Apr  3 11:41:02 localhost pcmcia:  cardmgr.
Apr  3 11:41:02 localhost rc: Starting pcmcia succeeded
Apr  3 11:41:02 localhost cardmgr[273]: starting, version is 3.1.8
Apr  3 11:41:02 localhost kernel: cs: cb_alloc(bus 35): vendor 0x9004, device 0x6075 
Apr  3 11:41:03 localhost cardmgr[273]: watching 2 sockets
Apr  3 11:41:03 localhost kernel: cs: IO port probe 0x1000-0x17ff: clean. 
Apr  3 11:41:03 localhost kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x320-0x327 0x378-0x37f 0x4d0-0x4d7 
Apr  3 11:41:03 localhost kernel: cs: IO port probe 0x0a00-0x0aff: clean. 
Apr  3 11:41:03 localhost cardmgr[273]: initializing socket 1
Apr  3 11:41:03 localhost cardmgr[273]: socket 1: Adaptec APA-1480 SCSI Host Adapter
Apr  3 11:41:03 localhost cardmgr[273]: executing: 'insmod /lib/modules/2.2.14-5.0/pcmcia/cb_enabler.o'
Apr  3 11:41:03 localhost cardmgr[273]: executing: 'insmod /lib/modules/2.2.14-5.0/pcmcia/apa1480_cb.o'
Apr  3 11:41:03 localhost kernel: cs: cb_config(bus 35) 
Apr  3 11:41:03 localhost kernel:   fn 0 bar 1: io 0xa00-0xaff 
Apr  3 11:41:03 localhost kernel:   fn 0 bar 2: mem 0x60030000-0x60030fff 
Apr  3 11:41:03 localhost kernel:   fn 0 rom: mem 0x60020000-0x6002ffff 
Apr  3 11:41:03 localhost kernel: apa1480_attach(bus 35, function 0) 
Apr  3 11:41:03 localhost kernel: aic7xxx: <Adaptec PCMCIA SCSI controller> at PCI 35/0/0 
Apr  3 11:41:03 localhost kernel: aic7xxx: I/O ports already in use, ignoring. 
Apr  3 11:41:03 localhost kernel: apa1480_cs: no SCSI devices found 
Apr  3 11:41:03 localhost kernel: scsi : 0 hosts. 
Apr  3 11:41:03 localhost cardmgr[273]: get dev info on socket 1 failed: No such device

any suggestions?

None An aic7xxx driver update broke the APA1480

Re: Question adaptec1480 scsi & RH6.2 (matthias caraco)
Date: 2000, Apr 03
From: David Hinds <dhinds@pcmcia.sourceforge.org>

This will be fixed in the next 3.1.14 beta; I'll put it on
sourceforge.org in /pcmcia/NEW later this afternoon.

-- Dave

Note SlimSCSI 1480 w/ Red Hat 6.2: I/O port conflict

Re: None An aic7xxx driver update broke the APA1480 (David Hinds)
Date: 2000, Jun 06
From: Eric Smith brouhaha

If this is the same problem I just had on Saturday night, it's caused by the Cardbus enabler and the aic7xxx drivers both trying to allocate the ports. RH 6.2 shipped with pcmcia-cs-3.1.8 IIRC. I didn't have an internet connection at the time I was messing with this, but I'd previously downloaded pcmcia-cs-3.1.11, so I built that. It didn't help. I ended up commenting out the check_region() call in the aic7xxx driver, and everything was happy.

I'll have to try downloading the latest version of pcmcia-cs.

None It is fixed in current drivers

Re: Note SlimSCSI 1480 w/ Red Hat 6.2: I/O port conflict (Eric Smith)
Date: 2000, Jun 06
From: David Hinds <dhinds@pcmcia.sourceforge.org>

It was fixed in 3.1.14.

-- Dave

Sad small problem with adaptec 1480 and pcmcia 3.1.14

Date: 2000, Apr 16
From: Jerome Limozin jlimozin

My config : mandrake 7.0, kernel 2.2.14 (plain, not the mandrake's one), pcmcia modules 3.1.14

HP omnibook 4100 + adaptec slimSCSI 1480 cardbus

I have a problem with newest pcmcia modules (3.1.14), I don't have it with 3.1.13. When I boot my laptop with the scsi card inserted, but SCSI cable unplugged, or with scsi cable plugged, but with all external scsi devices off, the boot process freezes when pcmcia tries to start. I have to power cycle to restart. If external scsi devices are on when I boot, it works.

with 3.1.13, I could boot even if no external devices are plugged and powered. I prefer this behaviour.

Is it a bug or a feature :-) ? Thanks!! Jerome

None It's a bug in the kernel aic7xxx driver

Re: Sad small problem with adaptec 1480 and pcmcia 3.1.14 (Jerome Limozin)
Date: 2000, Apr 17
From: David Hinds <dhinds@pcmcia.sourceforge.org>

For now, you just have to live with it.  It is a bug in the aic7xxx
driver source, and I have described the problem and provided a fix to
the driver maintainer.

-- Dave

Question Re : It's a bug in the kernel aic7xxx driver

Re: None It's a bug in the kernel aic7xxx driver (David Hinds)
Date: 2000, Apr 18
From: Jerome Limozin jlimozin

Hi David,

thanks for the answer. I'll keep 1.1.13 for the moment....

Do you know if this bug will be fixed in the coming kernel 2.2.15, or if I shall wait longer?

Many thanks! Jerome

None I think 2.2.15 is unlikely

Re: Question Re : It's a bug in the kernel aic7xxx driver (Jerome Limozin)
Date: 2000, Apr 20
From: David Hinds <dhinds@pcmcia.sourceforge.org>

But you can prod dledford@redhat.com about it if you like ;)

-- Dave

Question Panasonic CD Rom KXL-D720

Date: 2000, May 04
From: Steve Thomas <stevet@sb.net>

I've successfully installed RedHat 6.1 on my Toshiba Satellite 110CS and have most everything working. However, I can't get it to recognize my CD ROM which is attached to a scsi pcmcia card. The CD is on the list of devices that should work (tho not directly supported.) It is a Panasonic KXL-D720.

When I insert the card, the following message comes up:

parport0: detected irq 7; use procfs to enable interrupt-driven operation.

Is "procfs" a utility that I can use to make the fix? I can't find it anywhere in my linux set-up. Also, when I run the following commands for the card services, I get the results noted:

#cardctl config
Socket 0:
  Vcc = 5.0, Vpp1 = 0.0, Vpp2 = 0.0
  Interface type is memory and I/O
  IRQ 3 is exclusive, level mode, enabled
  Function 0:
    Config register base = 0x0200
      Option = 0x68, status = 0000, pin = 0000, copy = 0000
    I/O window 1: 0x0100 to 0x011f, 8 bit
Socket 1:
  not configured

#cardctl ident
Socket 0:
  product info: "KME", "KXLC001", "00"
  manfid: 0x0032, 0x0104
  function: 3 (parallel)
Socket 1
  no product info available

What do I need to do to get this thing working? I'm a REAL novice here and, despite reading the PCMCIA How-To, I still haven't a clue as to what to do.

Thanks for the help.

Steve

None You're misunderstanding the "not supported" bit

Re: Question Panasonic CD Rom KXL-D720 (Steve Thomas)
Date: 2000, May 04
From: David Hinds <dhinds@pcmcia.sourceforge.org>

Sorry.  When I say the KXL-D720 is "not supported", I mean just that:
it does not work with the existing Linux drivers, period.

-- Dave

More Not Supported

Re: Question Panasonic CD Rom KXL-D720 (Steve Thomas)
Date: 2000, May 04
From: Steve Thomas <stevet@sb.net>

Well, I just re-read the list and I see that it is not on it now. But, I swear to you that it once was. I was initially working on this a couple of months ago and had to drop it for a while.

This drive was listed under the QLogic_cs driver the last I checked.

Steve

None Not the KXL-D720

Re: More Not Supported (Steve Thomas)
Date: 2000, May 04
From: David Hinds <dhinds@pcmcia.sourceforge.org>

The KXL-D720 has been on the "not supported" list for years.  The
KXL-D740 uses the qlogic_cs driver; that's completely different.

-- Dave

Ok OK

Re: None Not the KXL-D720 (David Hinds)
Date: 2000, May 04
From: Steve Thomas <stevet@sb.net>

Hey,

Sorry for bothering you. I should have been more careful. Linux does still intimidate me a Great Deal! Thanks.

Steve

Question Adaptec 1480 + LinuxPPC: possible?

Date: 2000, May 11
From: Albrecht Dress Albrecht.Dress

Hello!

I am new to this list, so please excuse me if this is the wrong place for this question...

I want to use an Adaptec 1480 CardBus card on an Apple PowerBook/500 (which does not have internal SCSI any more) under LinuxPPC 2.2.15pre20. The "generic" aic7xxx driver is reported to work with LinuxPPC (e.g. with the '2940), so I tried the pcmcia package. The installation went fine, I fixed the config files, the card is detected, but at some point there is a message "disabled by BIOS, ignoring" (see the log, below), and no scsi works. The card is o.k., as it works with MacOS. Does anyone know what went wrong?

Thanks for any help, Albrecht.

---from /var/log/messages

 kernel: Linux PCMCIA Card Services 3.1.14 
 kernel:   kernel build: 2.2.15pre20 #3 Thu May 11 11:18:51 CEST 2000 
 kernel:   options:  [pci] [cardbus] 
 kernel: Intel PCIC probe:  
 kernel:   TI 1211 rev 00 PCI-to-CardBus at slot 00:1a, mem 0x80080000 
 kernel:     host opts [0]: [serial pci & irq] [pci irq 58] [lat 16/32] [bus 1/1] 
 kernel: pci_scan... 
 kernel:   testing PCI irq 58 
 kernel:     PCI card interrupts, polling interval = 1000 ms 
 kernel: cs: cb_alloc(bus 1): vendor 0x9004, device 0x6075 
 cardmgr[607]: starting, version is 3.1.14
 cardmgr[607]: watching 1 sockets
 cardmgr[607]: initializing socket 0
 cardmgr[607]: socket 0: Adaptec APA-1480 SCSI Host Adapter
 cardmgr[607]: executing: 'insmod /lib/modules/2.2.15pre20/pcmcia/cb_enabler.o'
 cardmgr[607]: executing: 'insmod /lib/modules/2.2.15pre20/pcmcia/apa1480_cb.o'
 kernel: cs: cb_config(bus 1) 
 kernel:   fn 0 bar 1: io 0x0-0xff 
 kernel:   fn 0 bar 2: mem 0x90030000-0x90030fff 
 kernel:   fn 0 rom: mem 0x90020000-0x9002ffff 
 kernel: apa1480_attach(bus 1, function 0) 
 kernel: aic7xxx: <Adaptec PCMCIA SCSI controller> at PCI 1/0/0 
 kernel: aic7xxx: Initial PCI_COMMAND value was 0x7 
 kernel: aic7xxx: Initial DEVCONFIG value was 0x1000 
 kernel: aic7xxx: <Adaptec PCMCIA SCSI controller> at PCI 1/0/0 
 kernel: aic7xxx: Controller disabled by BIOS, ignoring. 
 kernel: apa1480_cs: no SCSI devices found 
 kernel: scsi : 0 hosts. 
 cardmgr[607]: get dev info on socket 0 failed: No such device

None IO window allocation glitch

Re: Question Adaptec 1480 + LinuxPPC: possible? (Albrecht Dress)
Date: 2000, May 11
From: David Hinds <dhinds@pcmcia.sourceforge.org>

In /etc/pcmcia/config.opts, change your IO port window to not start at
0x0000.  The resource allocator is trying to put the card's IO window
at base 0x0000, which is bad for PCI devices, because this means the
IO window is disabled.  I'll see if I can fix the allocator to avoid
doing this, but the config file fix is a good workaround for now.

-- Dave

Sad This did not work...

Re: None IO window allocation glitch (David Hinds)
Date: 2000, May 17
From: Albrecht Dress Albrecht.Dress

I tried a number of io ranges starting at != 0x0 (e.g. 0x1000), but the result was the same, does not work... *sigh*

Thanks anyway, Albrecht

Question Adaptec SlimSCSI & Mustek Scanner

Date: 2000, May 24
From: Nils Pickert <nils@_nospam_mipi.de>

Hi, I am using a Network PII-233 Notebook (which is made by Gericom and sold as Network 1100MT by Mediamarkt in Germany) with an Adaptec SlimSCSI-Controller. The Controller works fine when using a ZIP-Drive, but does not recognize a Mustek 6200 scanner. (Only: 1 host found, Zipdrive with ID 6...)

I have SCSI support compiled in, my Kernel is 2.2.15, SCSI Disk and Generic-Device support are compiled in as well. The SCSI-IDs are set correctly (Zip: 6, Scanner 3 or 7), the bus is terminated. Windows detects the Scanner as "Unknown Device", but gives the nearly correct ID "Mustek yyy6200üüü" in the Hardware-Info thingy.

I already tried different combinations for the SCSI-Bus-topology: first ZIP, then Scanner and vice versa, Scanner only, Zip only. None of them seems to work, the computer never finds the scanner, the ZIP works perfectly.

What am I doing wrong, or is it something serious?

Thanks in advance,

nils pickert

Question Ricoh MP8040SE

Date: 2000, Jun 02
From: Raoul Golan raoul

Has anyone had any luck with the Ricoh MP8040SE? It's a slim notebook SCSI cd burner, and comes with its own SCSI PCMCIA adapter.

What sort of adapter is it? Will it be supported?

Idea Try the nin_cs driver.

Re: Question Ricoh MP8040SE (Raoul Golan)
Date: 2000, Aug 04
From: Alberto Gonzalez Palomo matmota

  This drive seems to be the same as the Panasonic
KXL-RW10A and Hewlett Packard m820e. Read the "Panasonic
KXL-RW10A CD-RW almost working" thread for more info.

  In short, it doesn't work yet. You can use it
for audio CDs without problems, but data transfer has
trouble. Seems that past a certain amount of data,
the driver gets blocked.

  If you have any different results, please tell
us. Karl has the Panasonic, I the HP, and you the
Ricoh. Perhaps we could figure out what's happening
by joining efforts.

  Also, what pccard chipset your notebook has?
  Karl and I have the TI 1220.

Alberto González Palomo
matmota@linuxfreak.com

Ok that's ok, bought an HP instead :)

Re: Idea Try the nin_cs driver. (Alberto Gonzalez Palomo)
Date: 2000, Aug 10
From: Raoul Golan raoul

Hola Alberto!

Thanks for your reply... since I didn't recieve any replies for a long time, I decided against the Ricoh and went for the HP (only slightly more expensive) because I heard that some people had had success with it. I had my suspicions that the Ricoh and the HP were identical, but I thought I'd better stick to HP.

So far, I've been able to burn some cds with it without any trouble... occasionally I'll get an I/O error, especially if I'm creating an image from it by doing a simple dd, sometimes even a hang (if I increase the dd blocksize to something large). On the whole, not too bad! Audio CDs do not work for me though - opening it with something like a cd player gives me a bad ioctl error.

My chipset is also TI 220. Kernel is 2.2.15, with pcmcia-cs-3.1.15, on Mandrake 7.1.

So far so good...

Thanks for the response, if I find anything else out I'll be sure to let you know.

Raoul

More and I almost forgot...

Re: Ok that's ok, bought an HP instead :) (Raoul Golan)
Date: 2000, Aug 10
From: Raoul Golan raoul

And I forgot to add: it's with the 1.0.2 ninja drivers

Ok Ricoh MP8040SE works for writing

Re: Question Ricoh MP8040SE (Raoul Golan)
Date: 2000, Nov 05
From: Paulo R. Cavalcanti roma

The Ricoh MP8040SE uses the same pcmcia scsi card of the HP CDW m820e and Panassonic KXL-RW10A, indeed. All of them work with the Ninja driver (http://www.netlab.is.tsukuba.ac.jp/~yokota/izumi/ninja/). Unfortunately, all of the sites about the scsi card are written in Japanese. I tried the latest version contained in NinjaSCSI3-1.0.3.tar.gz and had no success at all. Then I used the previous version, NinjaSCSI3-1.0.2.tar.gz, and got promissing results in 16 bits mode.

I have a Dell Inspiron 7500 laptop running Windows 98 and RedHat 6.2 (Linux 2.2.14.5). My current pcmcia package is pcmcia-cs-3.1.19. The reason is because I bought a D-Link DFE-650 recently, and I had to upgrade the package for making the new chipset of D-Link work with Linux.

I compiled everything together with no problem. I strongly recommend installing forms first, so you will have cardinfo working, which is very helpful to check your pcmcia card status (put them up and down, etc...).

I mounted the CD-RW in read-only mode, using scd0. However, if I use a file manager or even netscape for reading a large file, the computer hangs.

Surprisingly enough, it seems to be no problem writing CD's using either xcdroast or kisocd. In my opinion, kisocd is a better choice (does anyone know why xcdroast allows the dangerous partition option for creating images?).

I recorded a CD-RW in 2x speed. Erase it, recorded againg and everything was fine. I did not try any audio yet.

Actually, despite of the fact that I do not like Windows, Adaptec Easy-CD creator is much better. I have been using mkisofs to create CD images on my Windows partition, and all I have to do is reboot the computer and burn the CD in Windows.

I have also installed the UDF (udf-0.9.2.1.tar.gz) patch for supporting CD's formatted this way. It works well (although a little bit slow yet) with my DVD Rom. I would like to test it with my Ricoh CD-RW also, but had no time yet.

I would like to thank David Hinds for making all these damn cards working. I really believe it is not easy. Maybe soon we will have very good quality drivers for all devices in Linux too, and I will not fell like a Kamikaze burning CDs in Linux.

If anyone find out anything new about this Ninja driver, please let me know.

Paulo R. Cavalcanti roma@ldeo.columbia.edu

Question 1480 + U2W tape

Date: 2000, Jun 20
From: Valentijn Sessink valentyn

Vaio Laptop, kernel 2.2.16, PCMCIA 3.1.16, Adaptec 1480 cardbus card and a HP SureStore DAT40.

Whatever I try, I can't have the laptop communicate with the tape thing.

- tapedrive on, insert card: laptop freezes.
- tapedrive off, insert card. Card mentions that there is no device connected. Then turn on the tapedrive: freeze. Turn tape off: laptop works again.

I tried various look alikes (insert card with "cardctl insert", etc etc, but to no avail.

The thing could be, that we need a 50 -> 68 pen adapter to get the HP40 to connect to the SCSI card. Another tape drive (HP DDS-1 tape) works like it should.

What can I do about this?

Best regards,

Valentijn

None Maybe a termination issue?

Re: Question 1480 + U2W tape (Valentijn Sessink)
Date: 2000, Jun 20
From: David Hinds <dhinds@pcmcia.sourceforge.org>

I am not an expert on SCSI issues.  However I checked some web sources
and wonder if this might be a "high byte termination" problem.  There
are 68 to 50 pin converters that do not terminate the high byte, and
there are 68 to 50 pin adapters that do provide termination.  You need
to use an adapter that has termination.

-- Dave

Question Panasonic KXL-RW10A CD-RW almost working

Date: 2000, Jul 25
From: Karl Oberjohn kwoberjohn

Hello.

I just wanted to share my experience of trying to set up my Panasonic KXL-RW10A CD-RW drive on my notebook, a Mitsubishi AMiTY CN2 with a TI 1220 PC-card controller. I am running kernel 2.4.0-test4 and consider myself to be somewhat of a Linux newbie.

The drive has a SCSI interface and uses a KME SCSI card with a little switch for selecting between 16-bit and 32-bit (Cardbus) operation. This is apparently the same card used with the HP M820e. I believe the Ricoh MP8040SE uses the same card as well.

I have had some promising results with the card in 16-bit mode and using the contributed NinjaSCSI driver (http://www.netlab.is.tsukuba.ac.jp/~yokota/izumi/ninja/), but I haven't got it all working just yet.

The output of cardctl ident is:

  Socket 1:
    product info: "KME ", "SCSI-CARD-001", "1"
    manfid: 0x0032, 0x2604

I compiled the latest version of the NinjaSCSI driver and copied the nin_cs.o file to my /lib/modules/2.4.0-test4/pcmcia directory. Then I added the following to my /etc/pcmcia/config file:

  device "nin_cs"
    class "scsi" module "nin_cs"

  card "KME SCSI Card (16-bit)"
    version "KME ", "SCSI-CARD-001", "1"
    bind "nin_cs"

Upon restarting my notebook, I get the following messages, which seem to indicate that the card is identified correctly:

scsi0 : NinjaSCSI-3/32Bi Driver version 1.0, I/O 0x0100-0x0110 IRQ 9 ID 7, Chip ID 1/Rev 1
scsi : 1 host.
  Vendor: MATSHITA Model: CDRRW01 Rev: 1.35
  Type: CD-ROM ANSI SCSI revision: 02
Detected scsi CD-ROM sr0 at scsi0, channel 0, id 4, lun 0
sr0: scsi3-mmc drive: 20x/20x writer cd/rw xa/form2 cdda pop-up
Uniform CD-ROM driver Revision: 3.11
ninja3: index 0x01: Vcc 5.0, irq 9, io 0x0100-0x010f
VFS: Disk change detected on device sr(11,0)
ISO 9660 Extensions: Microsoft Joliet Level 3
ISO 9660 Extensions: RRIP_1991A
spurious 8259A interrupt: IRQ7.

At this point, I am able to mount the SCSI CD-RW drive. I can also do simple file operations, such as viewing directories and displaying small files. But if I change to a directory with a large number of files (like the RPMs directory of a RedHat CD) or open a file that's more that a few kilobytes long, my notebook hangs for a second and then reboots. Does anyone know why this might be happening? It is frustrating, because it seems I'm very close to getting this to work...

I greatly appreciate any comments or suggestions.

Karl Oberjohn

Feedback Same sympthoms as HP 820e.

Re: Question Panasonic KXL-RW10A CD-RW almost working (Karl Oberjohn)
Date: 2000, Aug 04
From: Alberto Gonzalez Palomo matmota

Hi, Karl.

I have been experiencing exactly the same sympthoms with
the HP CDW m820e since I got the 1.0 driver version from
Hiroshi. Previously, the only thing that worked was audio
CDs.

I have tested with kernel releases 2.2.5, 2.2.12, 2.2.13,
2.2.14, and 2.2.15, and with various pcmcia-cs releases
in the 3.1.8 - 3.1.14 range, with essentially constant
results. Also compiling with gcc 2.72 or 2.95 didn't seem
to make a difference.

I sent some commented log files to Hiroshi. When the
problem seems to begin, the logs show a repetition of
the message "NinjaSCSI3 Interrupt" (or similar), which
seems to indicate a problem with interrupt handling.
I tried to change as many options as I could in the
nin_cs.c file, but without any result.

When I installed it under Windows NT, I had to choose
the "non IRQ" version of the driver, as it wouldn't
work with the default one. I remember reading somewhere,
probably in the driver's installation help, that this
is necessary for some TI pccard chipsets.
My Asus F7400 has the same TI 1220 controller, which
makes me suspect that this could be a IRQ handling
problem in that chipset.

However, in the nin_cs.c file there is a comment stating
that the driver doesn't use interrupts, so I don't know
what more to try.

There is an experimental version for the 2.4 kernel.
¿Have you tried it?

If you get any result, even if it doesn't really work,
please post it here. I would like to know of it.

Alberto González Palomo
matmota@linuxfreak.com

Feedback No problems here

Re: Feedback Same sympthoms as HP 820e. (Alberto Gonzalez Palomo)
Date: 2000, Aug 05
From: Philipp Huenerberg phuenerberg

Hi!

I'm using the Panasonic KXL-RW10A without any problems. I never experienced any, and I tried listing the RPMS directory of a RedHat CD, it worked without any trouble. Even accessing the kernel-source RPM on the cd worked. Kernel is 2.2.1[4-7pre12], pcmcia_cs version 3.1.1[4-9], nin_cs version 1.0.2, controller is a TI1225.

If you have any recipe how to make it definitely fail, let me know and I'll give it a try. Also, if you like to get my configuration, I'm happy to send you any information you like.

I've just one problem: the cardbus driver doesn't work yet, it always times out cause the driver only finds the SCSI controller, but not the drive :|

Cu, Philipp

Ok It's working now--should've stuck with kernel 2.2 :)

Re: Question Panasonic KXL-RW10A CD-RW almost working (Karl Oberjohn)
Date: 2000, Aug 15
From: Karl Oberjohn kwoberjohn

Well, that'll teach me to rely on alpha code for my day-to-day computing... I switched from kernel 2.4-test* and the alpha NinjaSCSI driver back to kernel 2.2.16 and the more-stable NinjaSCSI3-1.0.2 driver. Success! I just finished an X-CD-Roast session.

The Panasonic KXL-RW10A now works (in 16-bit mode) with my Mitsubishi AMiTY CN2 (TI 1220 controller, Slakware 7.1) and my wife's NEC Versa 6050MH (TI 1130 controller, Mandrake 6). I was even able to squeeze the nin_cs.o module onto the Slackware pcmcia disk image so that I could install Slackware from CD.

Thanks to everybody for your helpful responses.

Cheers,

Karl Oberjohn

Question APA-1480 w HD

Date: 2000, Aug 02
From: anewman

Hi

I'm trying to mount a SUN UFS SCSI drive through the adaptec 1480 adapter. PCMCIA services work fine, but I'm having trouble with the SCSI part. I'm using the 2.2.16 kernel and PCMCIA v3.1.16 Here's the log:

Aug  2 17:25:03 assiniboine kernel: Linux PCMCIA Card Services 3.1.16
Aug  2 17:25:03 assiniboine kernel:   kernel build: 2.2.16 #4 Wed Aug 2 16:50:12 PDT 2000
Aug  2 17:25:03 assiniboine kernel:   options:  [pci] [cardbus] [apm] [pnp]
Aug  2 17:25:03 assiniboine kernel: PCI routing table version 1.0 at 0xfe840
Aug  2 17:25:03 assiniboine kernel: PnP: PNP BIOS installation structure at 0xc00ff020
Aug  2 17:25:03 assiniboine kernel: PnP: PNP BIOS version 1.0, entry at ec000:21d5, dseg at ec000
Aug  2 17:25:03 assiniboine kernel: Intel PCIC probe: 
Aug  2 17:25:03 assiniboine kernel:   TI 1225 rev 01 PCI-to-CardBus at slot 00:0a, mem 0x68000000
Aug  2 17:25:03 assiniboine kernel:     host opts [0]: [ring] [serial pci & irq] [pci irq 10] [lat 168/176] [bus 32/34]
Aug  2 17:25:03 assiniboine kernel:     host opts [1]: [ring] [serial pci & irq] [pci irq 10] [lat 168/176] [bus 35/37]
Aug  2 17:25:03 assiniboine kernel:     ISA irqs (default) = 9,11 PCI status changes
Aug  2 17:25:03 assiniboine cardmgr[346]: starting, version is 3.1.16
Aug  2 17:25:03 assiniboine cardmgr[346]: watching 2 sockets
Aug  2 17:25:03 assiniboine kernel: cs: IO port probe 0x1000-0x17ff: clean.
Aug  2 17:25:03 assiniboine kernel: cs: IO port probe 0x0100-0x04ff: clean.
Aug  2 17:25:03 assiniboine kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Aug  2 17:25:03 assiniboine kernel: cs: cb_alloc(bus 35): vendor 0x9004, device 0x6075
Aug  2 17:25:03 assiniboine cardmgr[346]: initializing socket 1
Aug  2 17:25:03 assiniboine cardmgr[346]: socket 1: Adaptec APA-1480 SCSI Host Adapter
Aug  2 17:25:03 assiniboine cardmgr[346]: executing: 'modprobe cb_enabler'
Aug  2 17:25:03 assiniboine cardmgr[346]: executing: 'modprobe apa1480_cb'
Aug  2 17:25:04 assiniboine kernel: cs: cb_config(bus 35)
Aug  2 17:25:04 assiniboine kernel:   fn 0 bar 1: io 0xa00-0xaff
Aug  2 17:25:04 assiniboine kernel:   fn 0 bar 2: mem 0x60030000-0x60030fff
Aug  2 17:25:04 assiniboine kernel:   fn 0 rom: mem 0x60020000-0x6002ffff
Aug  2 17:25:04 assiniboine kernel:   irq 10
Aug  2 17:25:04 assiniboine kernel: apa1480_attach(device 23:00.0)
Aug  2 17:25:04 assiniboine kernel: aic7xxx: <Adaptec PCMCIA SCSI controller> at PCI 35/0/0
Aug  2 17:25:04 assiniboine kernel: aic7xxx: MMAPed I/O failed, reverting to Programmed I/O.
Aug  2 17:25:04 assiniboine kernel: (scsi0) <Adaptec PCMCIA SCSI controller> found at PCI 35/0/0
Aug  2 17:25:04 assiniboine kernel: (scsi0) Narrow Channel, SCSI ID=7, 3/255 SCBs
Aug  2 17:25:04 assiniboine kernel: (scsi0) Downloading sequencer code... 423 instructions downloaded
Aug  2 17:25:05 assiniboine modprobe: can't locate module char-major-6
Aug  2 17:25:09 assiniboine kernel: apa1480_cs: no SCSI devices found
Aug  2 17:25:09 assiniboine cardmgr[346]: get dev info on socket 1 failed: No such device
Aug  2 17:25:09 assiniboine cardmgr[346]: initializing socket 0
Aug  2 17:25:09 assiniboine kernel: cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa0ffffff
Aug  2 17:25:09 assiniboine kernel: cs: memory probe 0x60000000-0x60ffffff: clean.
Aug  2 17:25:09 assiniboine cardmgr[346]: socket 0: Linksys EtherFast LANmodem 56K
Aug  2 17:25:09 assiniboine cardmgr[346]: executing: 'modprobe 8390'
Aug  2 17:25:09 assiniboine cardmgr[346]: executing: 'modprobe pcnet_cs'
Aug  2 17:25:09 assiniboine cardmgr[346]: executing: 'modprobe serial_cs'
Aug  2 17:25:09 assiniboine kernel: eth0: NE2000 (DL10019 rev 05): io 0x300, irq 9, hw_addr 00:E0:98:08:F5:12
Aug  2 17:25:09 assiniboine kernel: tty02 at 0x0af8 (irq = 9) is a 16450
Aug  2 17:25:09 assiniboine cardmgr[346]: executing: './network start eth0'
Aug  2 17:25:09 assiniboine cardmgr[346]: + cat: /var/lib/pcmcia/scheme: No such file or directory
Aug  2 17:25:10 assiniboine cardmgr[346]: executing: './serial start ttyS2'
Aug  2 17:25:10 assiniboine cardmgr[346]: + cat: /var/lib/pcmcia/scheme: No such file or directory
Aug  2 17:25:11 assiniboine kernel: eth0: found link beat

any suggestions would be greatly appreciated....

None Probably no SCSI disk driver

Re: Question APA-1480 w HD
Date: 2000, Aug 02
From: David Hinds <dhinds@pcmcia.sourceforge.org>

> Aug  2 17:25:09 assiniboine kernel: apa1480_cs: no SCSI devices found

Did you build your kernel with the SCSI disk driver enabled?

Or is it maybe compiled as a module?

If not, then SCSI disk devices will be invisible and the apa1480
driver will report "no devices found".  The PCMCIA-HOWTO has
instructions for modifying /etc/pcmcia/config to load the sd_mod
module automatically.

-- Dave

More alas, the driver is there

Re: None Probably no SCSI disk driver (David Hinds)
Date: 2000, Aug 03
From: anewman

hi
  I did compile the kernel with the driver in place - both the
 aic7xx and the aha152x_cs (dox say the aic7xx but I wanted to
 have all bases covered).  Also in the kernel are all the
 supports for the Sun filesystems.

Here's what I put in scsi.opts (based on my best guess - I do
 know that the SCSI ID is 3):

case "$ADDRESS" in
*,sd,*,0,3,0)
    PARTS="1"
    DO_FSTAB="y" ; DO_FSCK="y" ; DO_MOUNT="y"
    FSTYPE="ufs"
    OPTS=""
    MOUNTPT="/mnt/mri6"
    ;;
esac

None No, not that driver...

Re: More alas, the driver is there
Date: 2000, Aug 03
From: David Hinds <dhinds@pcmcia.sourceforge.org>

No... you don't need/want either the aha152x or aic7xxx drivers in
your kernel.  The driver you are missing is the SCSI disk driver
(CONFIG_BLK_DEV_SD), not the driver for the host adapter.

Look at /proc/devices and see if there is an entry for "sd".  That
indicates if the SCSI disk driver is present.

-- Dave

Idea getting closer...?

Re: None No, not that driver... (David Hinds)
Date: 2000, Aug 05
From: anewman

Hi David

I think I may have some more useful diagnostic info after
spending a few hours farting around with it.  It seems that
it wants the IRQ of the SCSI card to be IRQ 10, however, on my
machine (according to Windows) this is assigned to the USB
port. It seems the appropriate IRQs for the PCMCIA slots are 9
& 11.  I put "exclude IRQ 10" in the config.opts file, but
when I do "cardctl config" it still says:
-----------------------------------------
Socket 0:
  Vcc 5.0V  Vpp1 0.0V  Vpp2 0.0V
  interface type is "memory and I/O"
  irq 9 [shared] [level]
  Speaker output is enabled
  function 0:
    config base 0xffe0
      option 0x43 status 0x08 pin 0x00
    io 0x0300-0x031f [auto]
    io 0x0af8-0x0aff [8bit]
Socket 1:
  Vcc 3.3V  Vpp1 3.3V  Vpp2 3.3V
  interface type is "cardbus"
  irq 10 [exclusive] [level]
  function 0:
-----------------------------------------
I tried modifying /etc/pcmcia/config to specify what IRQs to try:
device "apa1480_cb"
  class "scsi" module "cb_enabler", "apa1480_cb" opts "irq_list=9,11"
as suggested in the HOWTO, but when booting it tells me this
is not a valid option for the apa1480_cb module (nor
cb_enabler if I put it before the comma).

The only other diagnostic clue I can provide is that the card is interfering with the system bell, apparently at a very low level - I do get sound through the sound card (KDE sounds, CD, etc), but no beeps e.g., when cardmgr is doing its stuff, or even at boot time before LILO even kicks in.

Something DOES seem to be happening to the drive - it gives a couple of brief token "whirrs" when the cardmgr is doing its thing. Still nothing in /proc/devices though.

Additionally, here's the current relevant dmesg dump (without the IRQ specifications in the config file):
-----------------------------------------
Linux PCMCIA Card Services 3.1.16
  kernel build: 2.2.16 #6 Sat Aug 5 13:17:21 PDT 2000
  options:  [pci] [cardbus] [apm] [pnp]
PCI routing table version 1.0 at 0xfe840
  00:0a.1 -> irq 10
PnP: PNP BIOS installation structure at 0xc00ff020
PnP: PNP BIOS version 1.0, entry at ec000:21d5, dseg at ec000
Intel PCIC probe: 
  TI 1225 rev 01 PCI-to-CardBus at slot 00:0a, mem 0x68000000
    host opts [0]: [ring] [serial pci & irq] [pci irq 10] [lat 168/176] [bus 32/34]
    host opts [1]: [ring] [serial pci & irq] [pci irq 10] [lat 168/176] [bus 35/37]
    ISA irqs (default) = 9,11 PCI status changes
cs: IO port probe 0x1000-0x17ff: clean.
cs: IO port probe 0x0100-0x04ff: clean.
cs: IO port probe 0x0a00-0x0aff: clean.
cs: cb_alloc(bus 35): vendor 0x9004, device 0x6075
ROM image dump:
  image 0: 0x000000-0x0001ff, signature PCIR
  image 1: 0x000200-0x0003ff, signature PCIR
cs: cb_config(bus 35)
  fn 0 bar 1: io 0xa00-0xaff
  fn 0 bar 2: mem 0x60030000-0x60030fff
  fn 0 rom: mem 0x60020000-0x6002ffff
  irq 10
cs: cb_enable(bus 35)
  bridge io map 0 (flags 0x21): 0xa00-0xaff
  bridge mem map 0 (flags 0x1): 0x60020000-0x60030fff
apa1480_attach(device 23:00.0)
(scsi0) <Adaptec PCMCIA SCSI controller> found at PCI 35/0/0
(scsi0) Narrow Channel, SCSI ID=7, 3/255 SCBs
(scsi0) Downloading sequencer code... 423 instructions downloaded
scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.28/3.2.4
       <Adaptec PCMCIA SCSI controller>
scsi : 1 host.
apa1480_cs: no SCSI devices found
scsi : 0 hosts.
cs: cb_disable(bus 35)
cs: cb_release(bus 35)
cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa0ffffff
cs: memory probe 0x60000000-0x60ffffff: clean.
eth0: NE2000 (DL10019 rev 05): io 0x300, irq 9, hw_addr 00:E0:99:88:F5:12
tty02 at 0x0af8 (irq = 9) is a 16450
eth0: found link beat
-----------------------------------------

thanks again for any insights! -aaron

None No, the interrupt is not the problem

Re: Idea getting closer...?
Date: 2000, Aug 07
From: David Hinds <dhinds@pcmcia.sourceforge.org>

No.  Did you try my previous suggestion?

This card must use irq 10.  It is shared with USB; that is not a
problem.  Your 16-bit cards may use irq's 9 and 11, however, CardBus
cards must use a PCI interrupt, and irq 10 is it for you.

> apa1480_cs: no SCSI devices found

This message means you have a kernel configuration problem: your
kernel has no support for the device(s) attached to the card.  The
apa1480_cb driver is working fine.

-- Dave

Sad SCSI disk drivers are there

Re: None No, the interrupt is not the problem (David Hinds)
Date: 2000, Aug 11
From: anewman

Hi again

Yeah, i did make sure the SCSI disk drivers were there.  First
compiled the SCSI support into the kernel, then tried it as a
module exactly as you suggestedin the how-to.  Still no dice.
Here's the dmseg output resulting form booting with the SCSI
disk driver as a module: 
----------------------------------------------------------
Linux PCMCIA Card Services 3.1.16
  kernel build: 2.2.16 #9 Mon Aug 7 20:27:13 PDT 2000
  options:  [pci] [cardbus] [apm] [pnp]
PCI routing table version 1.0 at 0xfe840
  00:0a.1 -> irq 10
PnP: PNP BIOS installation structure at 0xc00ff020
PnP: PNP BIOS version 1.0, entry at ec000:21d5, dseg at ec000
Intel PCIC probe: 
  TI 1225 rev 01 PCI-to-CardBus at slot 00:0a, mem 0x68000000
    host opts [0]: [ring] [serial pci & irq] [pci irq 10] [lat 168/176] [bus 32/34]
    host opts [1]: [ring] [serial pci & irq] [pci irq 10] [lat 168/176] [bus 35/37]
    ISA irqs (default) = 9,11 PCI status changes
cs: IO port probe 0x1000-0x17ff: clean.
cs: IO port probe 0x0100-0x04ff: clean.
cs: IO port probe 0x0a00-0x0aff: clean.
cs: cb_alloc(bus 32): vendor 0x9004, device 0x6075
ROM image dump:
  image 0: 0x000000-0x0001ff, signature PCIR
  image 1: 0x000200-0x0003ff, signature PCIR
cs: cb_config(bus 32)
  fn 0 bar 1: io 0xa00-0xaff
  fn 0 bar 2: mem 0x60030000-0x60030fff
  fn 0 rom: mem 0x60020000-0x6002ffff
  irq 10
cs: cb_enable(bus 32)
  bridge io map 0 (flags 0x21): 0xa00-0xaff
  bridge mem map 0 (flags 0x1): 0x60020000-0x60030fff
apa1480_attach(device 20:00.0)
aic7xxx: <Adaptec PCMCIA SCSI controller> at PCI 32/0/0
aic7xxx: MMAPed I/O failed, reverting to Programmed I/O.
(scsi0) <Adaptec PCMCIA SCSI controller> found at PCI 32/0/0
(scsi0) Narrow Channel, SCSI ID=7, 3/255 SCBs
(scsi0) Downloading sequencer code... 423 instructions downloaded
scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.28/3.2.4
       <Adaptec PCMCIA SCSI controller>
scsi : 1 host.
apa1480_cs: no SCSI devices found
scsi : 0 hosts.
cs: cb_disable(bus 32)
cs: cb_release(bus 32)
-------------------------------------------------------------

Any other thoughts?

Thanks again for all your help.

a

Question APA-1480 problems: maybe reset?

Date: 2000, Aug 15
From: Albrecht Dress Albrecht.Dress

I use an Adaptec APA1480 CardBus card on an Apple PowerBook 2000, running LinuxPPC with a 2.2.15pre13 kernel and pcmcia-cs.25-jul-00. Up to now, I tested a scsi disk at 20 MB/sec, an old Exabyte 8505 tape drive at 5 MB/sec, and a new Exabyte Mammoth tape drive at 10 MB/sec, and I experienced a quite odd behaviour.

If I switch the device on, wait for a long time until the "internal" reset sequence is finished and insert the card after that, then no scsi device will be found. If I insert the card during this sequence (e.g. as long as the disk activity led is on), then the disk and the Exa 8505 are recognised, but the Mamooth gives a number of odd messages in /var/log/messages (see below). I suspect that something with the scsi reset sequence upon card insertion failed. Any ideas?

Thanks for your help, Albrecht.

--- From /var/log/messages ---
 kernel: cs: cb_alloc(bus 1): vendor 0x9004, device 0x6075 
 cardmgr[1922]: initializing socket 0
 cardmgr[1922]: socket 0: Adaptec APA-1480 SCSI Host Adapter
 cardmgr[1922]: executing: 'modprobe cb_enabler'
 cardmgr[1922]: executing: 'modprobe apa1480_cb'
 kernel: cs: cb_config(bus 1) 
 kernel:   fn 0 bar 1: io 0x1000-0x10ff 
 kernel:   fn 0 bar 2: mem 0x90030000-0x90030fff 
 kernel:   fn 0 rom: mem 0x90020000-0x9002ffff 
 kernel:   irq 58 
 kernel: apa1480_attach(device 01:00.0) 
 kernel: (scsi0) <Adaptec PCMCIA SCSI controller> found at PCI 1/0/0 
 kernel: (scsi0) Narrow Channel, SCSI ID=7, 3/255 SCBs 
 kernel: (scsi0) Downloading sequencer code... 422 instructions downloaded 
 kernel: scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.31/3.2.4 
 kernel:        <Adaptec PCMCIA SCSI controller> 
 kernel: scsi : 1 host. 
 kernel: scsi : aborting command due to timeout : pid 23, scsi0, channel 0, id 2, lun 0 Test Unit Ready 00 00 00 00 00  
 kernel: SCSI host 0 abort (pid 23) timed out - resetting 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: SCSI host 0 channel 0 reset (pid 23) timed out - trying harder 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: scsi : aborting command due to timeout : pid 24, scsi0, channel 0, id 3, lun 0 Test Unit Ready 00 00 00 00 00  
 kernel: SCSI host 0 abort (pid 24) timed out - resetting 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: SCSI host 0 channel 0 reset (pid 24) timed out - trying harder 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: scsi : aborting command due to timeout : pid 25, scsi0, channel 0, id 4, lun 0 Test Unit Ready 00 00 00 00 00  
 kernel: SCSI host 0 abort (pid 25) timed out - resetting 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: SCSI host 0 channel 0 reset (pid 25) timed out - trying harder 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: scsi : aborting command due to timeout : pid 26, scsi0, channel 0, id 5, lun 0 Test Unit Ready 00 00 00 00 00  
 kernel: SCSI host 0 abort (pid 26) timed out - resetting 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: SCSI host 0 channel 0 reset (pid 26) timed out - trying harder 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: scsi : aborting command due to timeout : pid 27, scsi0, channel 0, id 6, lun 0 Test Unit Ready 00 00 00 00 00  
 kernel: SCSI host 0 abort (pid 27) timed out - resetting 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: SCSI host 0 channel 0 reset (pid 27) timed out - trying harder 
 kernel: SCSI bus is being reset for host 0 channel 0. 
 kernel: apa1480_cs: no SCSI devices found 
 kernel: scsi : 0 hosts. 
 cardmgr[1922]: get dev info on socket 0 failed: No such device
 cardmgr[1922]: shutting down socket 0
 cardmgr[1922]: executing: 'modprobe -r apa1480_cb'
 cardmgr[1922]: + ds: Device or resource busy
 cardmgr[1922]: executing: 'modprobe -r cb_enabler'
 kernel: cs: cb_free(bus 1) 

None Termination issue?

Re: Question APA-1480 problems: maybe reset? (Albrecht Dress)
Date: 2000, Aug 15
From: David Hinds <dhinds@pcmcia.sourceforge.org>

My guess would be that this is a termination problem.  Have you tried
hooking up all combinations, or just individual devices one at a time?
From what you say, and your error messages, it sounds like the Mammoth
device was not properly terminated.

-- Dave

Feedback Termination is fine...

Re: None Termination issue? (David Hinds)
Date: 2000, Aug 21
From: Albrecht Dress Albrecht.Dress

All devices are in separate external housings, each with an external active terminator. I checked them all separately, both with the '1480A, and with a desktop PC with an Adaptec AHA-294X. All devices worked fine with the latter one.

I am not familiar with the internals of the CardBus bridge and its drivers, but maybe it helps to drive the CardBus Reset pin (CRST)?

Thanks, Albrecht.

Sad No chance to increase reset delay...

Re: Feedback Termination is fine... (Albrecht Dress)
Date: 2000, Aug 21
From: Albrecht Dress Albrecht.Dress

As an addition to the last posting, I tried to increase the reset delay time for the controller (there is even a kernel configuration option CONFIG_AIC7XXX_RESET_DELAY for this), with no result at all.

If you grep through the whole kernel source tree, you will see the reason: it is _defined_, it is _printed_ in the /proc entry, but never used elsewhere. Strange...

Does anyone know who maintains the aic7xxx driver?

Albrecht.

Ok Bad cable

Re: Question APA-1480 problems: maybe reset? (Albrecht Dress)
Date: 2000, Sep 07
From: David Hinds <dhinds@pcmcia.sourceforge.org>

It turned out that the cable to the CardBus card was broken.

-- Dave

PCMCIA SCSI Host Adapter Issues


Add Message to: "PCMCIA SCSI Host Adapter Issues"

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