Greetings all. Last night a friend asked me to pull some data off an old Mac driver ( Conner 128 meg). So follwoing the thread on scsi support raised by Dee W. I figured i would give this a try. Not having used scsi in a very long time, I dug up an old adaptec 1542c isa card and proceeded with the installation. After approx an hour I had the mac drive being recognized by the card (totally misconfigured) and got the bios loaded signal from the scsi bios reporting sequence. Next linux loaded as usual. However doing an insmod aha1542 gives me the following error root@gunnel:/home/chris > insmod aha1542 Using /lib/modules/2.4.4-4GB/kernel/drivers/scsi/aha1542.o /lib/modules/2.4.4-4GB/kernel/drivers/scsi/aha1542.o: unresolved symbol isapnp_find_dev now dmesg is informative after running modprobe aha1542 Configuring Adaptec (SCSI-ID 7) at IO:330, IRQ 11, DMA priority 0 aha1542.c: Using extended bios translation scsi0 : Adaptec 1542 aha1542.c: Unable to abort command for target 0 aha1542.c: Trying device reset for target 0 aha1542.c: Unable to abort command for target 0 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 0 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 0 lun 0 Vendor: CONNER Model: CP30175E-170Mb Rev: 5.23 Type: Direct-Access ANSI SCSI revision: 01 CCS Detected scsi disk sda at scsi0, channel 0, id 0, lun 0 aha1542.c: Unable to abort command for target 0 aha1542.c: Trying device reset for target 0 aha1542.c: Unable to abort command for target 0 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 0 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 0 lun 1 aha1542.c: Unable to abort command for target 1 aha1542.c: Trying device reset for target 1 aha1542.c: Unable to abort command for target 1 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 1 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 1 lun 0 aha1542.c: Unable to abort command for target 2 aha1542.c: Trying device reset for target 2 aha1542.c: Unable to abort command for target 2 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 2 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 2 lun 0 aha1542.c: Unable to abort command for target 3 aha1542.c: Trying device reset for target 3 aha1542.c: Unable to abort command for target 3 \ and this continues through to target 6 So my question is how do I address the unresolved symbol issue? Many thanks in advance Chris H.
On Sunday 02 September 2001 1:07 pm, Chris Herrnberger wrote:
Greetings all.
Last night a friend asked me to pull some data off an old Mac driver ( Conner 128 meg). So follwoing the thread on scsi support raised by Dee W. I figured i would give this a try. Not having used scsi in a very long time, I dug up an old adaptec 1542c isa card and proceeded with the installation.
After approx an hour I had the mac drive being recognized by the card (totally misconfigured) and got the bios loaded signal from the scsi bios reporting sequence. Next linux loaded as usual.
However doing an insmod aha1542 gives me the following error root@gunnel:/home/chris > insmod aha1542 Using /lib/modules/2.4.4-4GB/kernel/drivers/scsi/aha1542.o /lib/modules/2.4.4-4GB/kernel/drivers/scsi/aha1542.o: unresolved symbol isapnp_find_dev
now dmesg is informative after running modprobe aha1542
Configuring Adaptec (SCSI-ID 7) at IO:330, IRQ 11, DMA priority 0 aha1542.c: Using extended bios translation scsi0 : Adaptec 1542 aha1542.c: Unable to abort command for target 0 aha1542.c: Trying device reset for target 0 aha1542.c: Unable to abort command for target 0 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 0 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 0 lun 0 Vendor: CONNER Model: CP30175E-170Mb Rev: 5.23 Type: Direct-Access ANSI SCSI revision: 01 CCS Detected scsi disk sda at scsi0, channel 0, id 0, lun 0 aha1542.c: Unable to abort command for target 0 aha1542.c: Trying device reset for target 0 aha1542.c: Unable to abort command for target 0 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 0 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 0 lun 1 aha1542.c: Unable to abort command for target 1 aha1542.c: Trying device reset for target 1 aha1542.c: Unable to abort command for target 1 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 1 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 1 lun 0 aha1542.c: Unable to abort command for target 2 aha1542.c: Trying device reset for target 2 aha1542.c: Unable to abort command for target 2 Sent BUS RESET to scsi host 0 aha1542.c: Unable to abort command for target 2 scsi: device set offline - not ready or command retry failed after host reset: host 0 channel 0 id 2 lun 0 aha1542.c: Unable to abort command for target 3 aha1542.c: Trying device reset for target 3 aha1542.c: Unable to abort command for target 3 \ and this continues through to target 6 So my question is how do I address the unresolved symbol issue?
Did you type insmod aha1542=0x330? M
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter: Info: PNP bios disabled in main bios Scsi bios loads and the driver is reported ISAPNP enabled in linux dmesg reports: isapnp: Scanning for PnP cards... isapnp: Card '3Com 3C509B EtherLink III' isapnp: 1 Plug & Play card detected total now insmod aha1542=0x330 reports root@gunnel:/home/chris > insmod aha1542=0x330 Using /lib/modules/2.4.4-4GB/kernel/abi/common/abi-common.o/aha1542=0x330 insmod: /lib/modules/2.4.4-4GB/kernel/abi/common/abi-common.o/aha1542=0x330: Not a directory Any insights would be helpfull and appreciated. Chris
On Monday 03 September 2001 11:50 am, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
Info:
PNP bios disabled in main bios Scsi bios loads and the driver is reported ISAPNP enabled in linux
dmesg reports: isapnp: Scanning for PnP cards... isapnp: Card '3Com 3C509B EtherLink III' isapnp: 1 Plug & Play card detected total
now insmod aha1542=0x330 reports root@gunnel:/home/chris > insmod aha1542=0x330 Using /lib/modules/2.4.4-4GB/kernel/abi/common/abi-common.o/aha1542=0x330 insmod: /lib/modules/2.4.4-4GB/kernel/abi/common/abi-common.o/aha1542=0x330: Not a directory
Any insights would be helpfull and appreciated.
Chris
I believe it should be: insmod aha154x aha1542x=0x0330 The first parm tells insmod what to load and the second parm is passed to the loaded module. You might want to add the irq also: aha1542x=0x0330,<irq> -- +----------------------------------------------------------------------------+ + Bruce S. Marshall bmarsh@bmarsh.com Bellaire, MI 09/03/01 12:10 + +----------------------------------------------------------------------------+ "The fault lies not with our technologies but with our systems." - Roger Levian
On Mon, Sep 03, 2001 at 12:12:14PM -0400, Bruce Marshall wrote:
On Monday 03 September 2001 11:50 am, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
I believe it should be:
insmod aha154x aha1542x=0x0330
The first parm tells insmod what to load and the second parm is passed to the loaded module. You might want to add the irq also: aha1542x=0x0330,<irq>
A couple of points here: 1. It is usually best to use the 'modprobe' command, rather than 'insmod'. 'modprobe' will handle all module dependencies automatically (and some other stuff) while 'insmod' won't. 2. The aha1542 module uses 0x330 as its base address by default. 3. The aha1542 module does not take an IRQ as a parameter. 4. The systax is incorrect in the above example: aha1542=0x330 can be used in that form when it is passed as a kernel parameter, for example at a LILO prompt or as part of an append= line in the lilo.conf file. When using insmod or modprobe, the module should be loaded as follows (using modprobe as an example): modprobe aha1542 io=0x330 Sorry, I missed the start of this thread and so can't be of much more help, Chris (a different one...) -- __ _ -o)/ / (_)__ __ ____ __ Chris Reeves /\\ /__/ / _ \/ // /\ \/ / ICQ# 22219005 _\_v __/_/_//_/\_,_/ /_/\_\
On Monday 03 September 2001 12:33 pm, Chris Reeves wrote:
On Mon, Sep 03, 2001 at 12:12:14PM -0400, Bruce Marshall wrote:
On Monday 03 September 2001 11:50 am, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
I believe it should be:
insmod aha154x aha1542x=0x0330
The first parm tells insmod what to load and the second parm is passed to the loaded module. You might want to add the irq also: aha1542x=0x0330,<irq>
A couple of points here: 1. It is usually best to use the 'modprobe' command, rather than 'insmod'. 'modprobe' will handle all module dependencies automatically (and some other stuff) while 'insmod' won't. 2. The aha1542 module uses 0x330 as its base address by default. 3. The aha1542 module does not take an IRQ as a parameter. 4. The systax is incorrect in the above example:
I beg to differ: 1) Yes, modprobe is better but I wasn't sure what environment or problem he was dealing with. IAE, the information given would have helped him 2) Yes.. (and your point?) 3) You're right on this one. (I have only used 152x's). 4) And your syntax is incorrect.
From the 'Using OpenLinux' book which lists the parms for all modules:
Boot Parameters: Format: aha1542=<IO port>,<bus on>,<bus off>,<DMA speed> IO port 0x0130,0x0134,0x0230,0x0330,0x0334, (probed at 330 and 334) 0x234 bus on has a value 2-15 with 11 as default (microsec) bus off has a value 1-64 with 4 as a default DMA speed 5,6,7,8,10 (MB/sec) Note the use of aha1542=<parms> The adaptec drivers use this format specifically. Same for the aha152x module. NOTE: The above are BOOT parameters. The book notes that "no MODULE parameters are provided for this driver".
aha1542=0x330 can be used in that form when it is passed as a kernel parameter, for example at a LILO prompt or as part of an append= line in the lilo.conf file. When using insmod or modprobe, the module should be loaded as follows (using modprobe as an example): modprobe aha1542 io=0x330
Sorry, I missed the start of this thread and so can't be of much more help, Chris (a different one...)
-- +----------------------------------------------------------------------------+ + Bruce S. Marshall bmarsh@bmarsh.com Bellaire, MI 09/03/01 13:16 + +----------------------------------------------------------------------------+ "You might be a high-tech Red-neck if: you have ever purchased an electronic appliance "as-is"
On Mon, Sep 03, 2001 at 01:27:43PM -0400, Bruce Marshall wrote:
On Monday 03 September 2001 12:33 pm, Chris Reeves wrote:
On Mon, Sep 03, 2001 at 12:12:14PM -0400, Bruce Marshall wrote:
I believe it should be:
insmod aha154x aha1542x=0x0330
The first parm tells insmod what to load and the second parm is passed to the loaded module. You might want to add the irq also: aha1542x=0x0330,<irq>
A couple of points here: 1. It is usually best to use the 'modprobe' command, rather than 'insmod'. 'modprobe' will handle all module dependencies automatically (and some other stuff) while 'insmod' won't. 2. The aha1542 module uses 0x330 as its base address by default. 4. The systax is incorrect in the above example:
I beg to differ:
1) Yes, modprobe is better but I wasn't sure what environment or problem he was dealing with. IAE, the information given would have helped him
Fair enough - I was adding more general information while I was at it. BTW, I didn't mean any of that in a bad way. :-/ Accept my humble apologies ;-)
2) Yes.. (and your point?)
What can I say... I was on a roll ;-)
4) And your syntax is incorrect.
From the 'Using OpenLinux' book which lists the parms for all modules:
Boot Parameters:
Format:
aha1542=<IO port>,<bus on>,<bus off>,<DMA speed>
IO port 0x0130,0x0134,0x0230,0x0330,0x0334, (probed at 330 and 334) 0x234
bus on has a value 2-15 with 11 as default (microsec)
bus off has a value 1-64 with 4 as a default
DMA speed 5,6,7,8,10 (MB/sec)
Note the use of aha1542=<parms>
The adaptec drivers use this format specifically. Same for the aha152x module.
NOTE: The above are BOOT parameters. The book notes that "no MODULE parameters are provided for this driver".
That's right, they're boot parameters, as I said below.
aha1542=0x330 can be used in that form when it is passed as a kernel parameter, for example at a LILO prompt or as part of an append= line in the lilo.conf file. When using insmod or modprobe, the module should be loaded as follows (using modprobe as an example): modprobe aha1542 io=0x330
My bad - I misread the source. :( If I hadn't thought that bit was incorrect, I wouldn't have bothered with the rest of the message (since what you had was, well, correct)... I'll read slower next time. Is it me, or am I seriously overusing smileys here? As you can probably tell, I'm in a happy, jolly, bouncy mood. It's great to be back! Not a great start though. :'( Have fun, Chris ;-) :-) :-p -- __ _ -o)/ / (_)__ __ ____ __ Chris Reeves /\\ /__/ / _ \/ // /\ \/ / ICQ# 22219005 _\_v __/_/_//_/\_,_/ /_/\_\
On Monday 03 September 2001 02:18 pm, Chris Reeves wrote:
Is it me, or am I seriously overusing smileys here? As you can probably tell, I'm in a happy, jolly, bouncy mood. It's great to be back! Not a great start though. :'(
Good mood eh?? You'll get over it... <grin> Just kidding. Hope the rest of the week goes as well.... -- +----------------------------------------------------------------------------+ + Bruce S. Marshall bmarsh@bmarsh.com Bellaire, MI 09/03/01 15:30 + +----------------------------------------------------------------------------+ "Time is nature's way of making sure that everything doesn't happen at once."
On Mon, Sep 03, 2001 at 01:27:43PM -0400, Bruce Marshall wrote:
NOTE: The above are BOOT parameters. The book notes that "no MODULE parameters are provided for this driver".
I forgot to mention in that last mail (and was reminded by your other mail that I just read) that the source disagree with that statement. :-/ Maybe I'm misinterpreting what it says, but here is the relevant comment: * LILO/Module params: aha1542=<PORTBASE>[,<BUSON>,<BUSOFF>[,<DMASPEED>]] I'm assuming this means the parameters will work with insmod/modprobe. No harm in trying ;-) Bye, Chris -- __ _ -o)/ / (_)__ __ ____ __ Chris Reeves /\\ /__/ / _ \/ // /\ \/ / ICQ# 22219005 _\_v __/_/_//_/\_,_/ /_/\_\
On Monday 03 September 2001 02:24 pm, Chris Reeves wrote:
I forgot to mention in that last mail (and was reminded by your other mail that I just read) that the source disagree with that statement. :-/ Maybe I'm misinterpreting what it says, but here is the relevant comment:
* LILO/Module params: aha1542=<PORTBASE>[,<BUSON>,<BUSOFF>[,<DMASPEED>]]
I'm assuming this means the parameters will work with insmod/modprobe. No harm in trying ;-)
Does a module really know the difference between boot time and insmod time? Loading is loading...... and parms are parms I would think. -- +----------------------------------------------------------------------------+ + Bruce S. Marshall bmarsh@bmarsh.com Bellaire, MI 09/03/01 15:33 + +----------------------------------------------------------------------------+ "Mistakes are oft the stepping stones to failure."
On Monday 03 September 2001 3:50 pm, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
Yes, I should have said insmod aha1542 aha1542=0x330. There are some optional parameters but I'd try this first. If this works you can add boot support later. M
On Monday 03 September 2001 03:06 pm, Martin Webster wrote:
On Monday 03 September 2001 3:50 pm, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
Yes, I should have said insmod aha1542 aha1542=0x330. There are some optional parameters but I'd try this first. If this works you can add boot support later.
M
My 'somewhat authoritative' book says there "are no module parameters" for aha1542, only boot parameters.... Strange if it is true but that what it says. -- +----------------------------------------------------------------------------+ + Bruce S. Marshall bmarsh@bmarsh.com Bellaire, MI 09/03/01 14:13 + +----------------------------------------------------------------------------+ "Graffiti has changed deface of the nation."
On Monday 03 September 2001 20.14, Bruce Marshall wrote:
On Monday 03 September 2001 03:06 pm, Martin Webster wrote:
On Monday 03 September 2001 3:50 pm, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
Yes, I should have said insmod aha1542 aha1542=0x330. There are some optional parameters but I'd try this first. If this works you can add boot support later.
M
My 'somewhat authoritative' book says there "are no module parameters" for aha1542, only boot parameters.... Strange if it is true but that what it says.
The source says differently: (from /usr/src/linux/drivers/scsi/aha1542.c) * LILO/Module params: aha1542=<PORTBASE>[,<BUSON>,<BUSOFF>[,<DMASPEED>]] ^^^^^^^^^^^^^ * * Where: <PORTBASE> is any of the valid AHA addresses: * 0x130, 0x134, 0x230, 0x234, 0x330, 0x334 * <BUSON> is the time (in microsecs) that AHA spends on the AT-bus * when transferring data. 1542A power-on default is 11us, * valid values are in range: 2..15 (decimal) * <BUSOFF> is the time that AHA spends OFF THE BUS after while * it is transferring data (not to monopolize the bus). * Power-on default is 4us, valid range: 1..64 microseconds. * <DMASPEED> Default is jumper selected (1542A: on the J1), * but experimenter can alter it with this. * Valid values: 5, 6, 7, 8, 10 (MB/s) * Factory default is 5 MB/s. */ #if defined(MODULE) int isapnp = 0; int aha1542[] = {0x330, 11, 4, -1}; MODULE_PARM(aha1542, "1-4i"); MODULE_PARM(isapnp, "i"); regards Anders
On Monday 03 September 2001 02:24 pm, Anders Johansson wrote:
On Monday 03 September 2001 20.14, Bruce Marshall wrote:
On Monday 03 September 2001 03:06 pm, Martin Webster wrote:
On Monday 03 September 2001 3:50 pm, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
Yes, I should have said insmod aha1542 aha1542=0x330. There are some optional parameters but I'd try this first. If this works you can add boot support later.
M
My 'somewhat authoritative' book says there "are no module parameters" for aha1542, only boot parameters.... Strange if it is true but that what it says.
The source says differently: (from /usr/src/linux/drivers/scsi/aha1542.c)
* LILO/Module params: aha1542=<PORTBASE>[,<BUSON>,<BUSOFF>[,<DMASPEED>]] ^^^^^^^^^^^^^ * * Where: <PORTBASE> is any of the valid AHA addresses: * 0x130, 0x134, 0x230, 0x234, 0x330, 0x334 * <BUSON> is the time (in microsecs) that AHA spends on the AT-bus * when transferring data. 1542A power-on default is 11us, * valid values are in range: 2..15 (decimal) * <BUSOFF> is the time that AHA spends OFF THE BUS after while * it is transferring data (not to monopolize the bus). * Power-on default is 4us, valid range: 1..64 microseconds. * <DMASPEED> Default is jumper selected (1542A: on the J1), * but experimenter can alter it with this. * Valid values: 5, 6, 7, 8, 10 (MB/s) * Factory default is 5 MB/s. */
#if defined(MODULE) int isapnp = 0; int aha1542[] = {0x330, 11, 4, -1}; MODULE_PARM(aha1542, "1-4i"); MODULE_PARM(isapnp, "i");
regards Anders
Well, I would agree with you.... it would be terribly strange if the module could handle parms at boot time and not any other. I was just going by the book. -- +----------------------------------------------------------------------------+ + Bruce S. Marshall bmarsh@bmarsh.com Bellaire, MI 09/03/01 15:31 + +----------------------------------------------------------------------------+ "The best way to become acquainted with a subject is to write a book about it." - Benjamin Disraeli
On Monday 03 September 2001 6:14 pm, Bruce Marshall wrote:
On Monday 03 September 2001 03:06 pm, Martin Webster wrote:
On Monday 03 September 2001 3:50 pm, Chris Herrnberger wrote:
On September 2, 2001 09:27 am, Martin Webster wrote:
Did you type insmod aha1542=0x330?
Thanks for responding this is driving me nuts. Are you sure the syntax is correct. I understand what you are doing but Im not sure that this is the correct manner to enter the parameter:
Yes, I should have said insmod aha1542 aha1542=0x330. There are some optional parameters but I'd try this first. If this works you can add boot support later.
M
My 'somewhat authoritative' book says there "are no module parameters" for aha1542, only boot parameters.... Strange if it is true but that what it says.
Hey, you're right! ;-) The driver is compiled directly in the kernel. So, we need to do one of the following: 1. At the LILO prompt type: aha1542=0x330 2. Add kernel parameters in /etc/lilo.conf or /etc/modules.conf. For the latter, you should be able to enter: alias scsi_hostadapter aha1542=0x330 in /etc/modules.conf or possibly: alias block-major-11 aha1542=0x330 (I've read that this works) M
participants (5)
-
Anders Johansson -
Bruce Marshall -
Chris Herrnberger -
Chris Reeves -
Martin Webster