linux 2.0.29 and adaptec 152x on port 0x140

linux 2.0.29 and adaptec 152x on port 0x140

Post by Pascal Quentin Porcupi » Sun, 23 Mar 1997 04:00:00



i recently acquired an adaptec 152x 16-bit ISA SCSI-2 card and a zip
drive.  the zip drive and adaptec card work fine under win'95, but under
linux i can't even get the OS to find the card.  the card itself is on
port 0x140, irq 11 with the BIOS disabled (i need my ISA drives, thank you
very much :) and i have tried the 152x driver both compiled into the
kernel and as a module.

when i have it compiled into the kernel, on bootup it says "no hosts
detected" and then goes on with its merry little business, unless i give
the kernel parameter (via loadlin) aha152x=0x140,11 (the proper parameter
format according to aha152x.h) in which case the kernel finds the card
then goes into an infinite "timeout exceeded" loop.

when i have the driver compiled as module, i just get "Initialization of
aha152x failed" whether i give it parameters or not.  in my
linux/drivers/scsi Makefile i continually experimented with AHA152X = ,
finally giving up with:

AHA152X = -DDEBUG_AHA152X -DSKIP_BIOSTEST -DSETUP0="{ 0x140, 11, 7, 1, \
1, 0, 10, 100, 0 }"

as per instruction of aha152x.h.  is there something wrong that i'm doing
in trying to get linux to see this card?  specifically, *must* the card
actually be on 0x340 (even though aha152x.h claims that 0x140 is just
fine)?  or, must the BIOS be enabled, preventing me from using my hard
disks?  i'd put my card on 0x340 except that i've got various sound
hardware which requires those addresses (namely a PAS16, which has an
MPU-401, and a GUS ACE on 0x240, though i can move the ACE to 0x230 or
0x250, if necessary without having any conflicts created).

or am i just skipping something so obvious that i'm going to have to hurt
myself when someone tells me the answer? :)  (and yes, i am recompiling
everything necessary whenever i change an option.... that's why i went to
module, btw, so i wouldn't have to reboot after every single configuration
change :)

please cc: any replies via email, as i'm not sure if i'll be continually
following this newsgroup.  thanks in advance.

--
Pascal Q. Porcupine

 
 
 

linux 2.0.29 and adaptec 152x on port 0x140

Post by Pascal Quentin Porcupi » Sun, 23 Mar 1997 04:00:00


sorry for following up to my own article, but i forgot to say something:
the card itself is an Iomega Jaz Jet.  i got that as opposed to the Zip
Zoom because i wanted the slightly better performance (SCSI-2 with a cache
as opposed to a cacheless SCSI-1) as well as the potential for better
future expansion (since it's got both internal and external connections).
if this adds any complications to the response (though it shouldn't, as
it's just a generic AHA-152x, according to Win'95 anyway :) then, well,
you know. :)

and btw, i only have win'95 at all because there are a few programs i need
which have no linux equivalents (and they're DOS apps, btw).  i'm in linux
99.9% of the time and win'95 0.02%.  the remaining 0.08% my computer's
turned off. :)

just thought i'd clear that up. :)



>i recently acquired an adaptec 152x 16-bit ISA SCSI-2 card and a zip
>drive.  the zip drive and adaptec card work fine under win'95, but under
>linux i can't even get the OS to find the card.  the card itself is on
>port 0x140, irq 11 with the BIOS disabled (i need my ISA drives, thank you
>very much :) and i have tried the 152x driver both compiled into the
>kernel and as a module.

>when i have it compiled into the kernel, on bootup it says "no hosts
>detected" and then goes on with its merry little business, unless i give
>the kernel parameter (via loadlin) aha152x=0x140,11 (the proper parameter
>format according to aha152x.h) in which case the kernel finds the card
>then goes into an infinite "timeout exceeded" loop.

>when i have the driver compiled as module, i just get "Initialization of
>aha152x failed" whether i give it parameters or not.  in my
>linux/drivers/scsi Makefile i continually experimented with AHA152X = ,
>finally giving up with:

>AHA152X = -DDEBUG_AHA152X -DSKIP_BIOSTEST -DSETUP0="{ 0x140, 11, 7, 1, \
>1, 0, 10, 100, 0 }"

>as per instruction of aha152x.h.  is there something wrong that i'm doing
>in trying to get linux to see this card?  specifically, *must* the card
>actually be on 0x340 (even though aha152x.h claims that 0x140 is just
>fine)?  or, must the BIOS be enabled, preventing me from using my hard
>disks?  i'd put my card on 0x340 except that i've got various sound
>hardware which requires those addresses (namely a PAS16, which has an
>MPU-401, and a GUS ACE on 0x240, though i can move the ACE to 0x230 or
>0x250, if necessary without having any conflicts created).

>or am i just skipping something so obvious that i'm going to have to hurt
>myself when someone tells me the answer? :)  (and yes, i am recompiling
>everything necessary whenever i change an option.... that's why i went to
>module, btw, so i wouldn't have to reboot after every single configuration
>change :)

>please cc: any replies via email, as i'm not sure if i'll be continually
>following this newsgroup.  thanks in advance.

>--
>Pascal Q. Porcupine


--
Pascal Q. Porcupine


 
 
 

linux 2.0.29 and adaptec 152x on port 0x140

Post by Matt Kirsc » Tue, 25 Mar 1997 04:00:00



> sorry for following up to my own article, but i forgot to say something:
> the card itself is an Iomega Jaz Jet.  i got that as opposed to the Zip
> Zoom because i wanted the slightly better performance (SCSI-2 with a cache
> as opposed to a cacheless SCSI-1) as well as the potential for better
> future expansion (since it's got both internal and external connections).
> if this adds any complications to the response (though it shouldn't, as
> it's just a generic AHA-152x, according to Win'95 anyway :) then, well,
> you know. :)

Well then, that is your problem.
The Jaz jet is not an Adaptec 152x compatible card, at least in the
way that Linux looks at it. The Jaz Jet is a PCI card, and the
Adaptec 152x series are all ISA cards. Windows may not understand
the difference, but Linux does.

You don't need the BIOS activated as far as I know. Unfortunately,
I can't remember for the life of me what Linux driver works with
the Jaz Jet, if any. Maybe it is the Future Domain driver?

Geez, it's on the tip of my tongue. I've read about it in the
newsgroup.


http://www.cnct.com/~bwillmot/jaztool/JAZ-drive-mini-howto

I think the Jaz Jet is an Advansys card, or the Adaptec 2940.

Quote:> just thought i'd clear that up. :)

You just helped your cause immensely. :)

--
Mathew E. Kirsch, CLSE (Certifiable Linux Systems Engineer)
*Opinions expressed herein do not reflect those of my employer.

"If you don't have time to read the FAQ, I don't have time to read it to
you."

 
 
 

1. Adaptec 1542CP and Linux 2.0.29

Has anyone had any luck getting a 1542CP working with 2.0.29?  Mine falls
apart starting with:

aha1542.c: interrupt received, but no mail

tries to reset the SCSI bus, etc, etc, and eventually just hangs

It doesn't appear to be termination/cabling (I used the same exact setup
for almost a year under BSDI).  The BIOS extension stuff in the adaptec
bios config util is all off (even though the SCSI HOWTO claims that this
is no longer a needed precaution).  Is there new firmware for the 1542CP
(mine claims to be v1.02).

Any suggestions would be appreciated.

Thanks,

Brian

--

 http://hagbard.ncsa.uiuc.edu/swetland/ |      -- Larry Wall, Perl 5 Configure
 NCSA SDD Vis Group -- Project Hyperion |

2. Reading audio input from SoundBlaster help needed...

3. symbols from kernel 2.0.29 don't match 2.0.29?!?!?

4. pppd locks up Linux

5. 2.0.29 modules cannot be loaded by 2.0.29 kernel??????

6. MS patch crashes NT4

7. Does anybody use adaptec 29/39160?

8. static slip connection with Red Hat 5

9. ERROR - bad inode number 0 - kernel 2.0.29 using Adaptec 2940

10. Adaptec 152x Controller with Linux

11. SCSI-2 adaptec 152x and Linux

12. Linux nearly starting from Adaptec 152x

13. Testing *BSD on a Linux System and Adaptec 152x - Jaz impossibility