From: jvs@netcom.com (Jonathan Stockley) Subject: Re: 99pl9 Seagate SCSI tape bug Date: Wed, 5 May 1993 17:36:28 GMT
In article <zafiris.736538021@navy16> zafiris@rtsg.mot.com (John Zafiris) writes:
>
>I tried to compile the 99pl9 linux kernal with the Seagate SCSI tape
>(no disk or CDROM) code but I got an error at the link stage. The
>linker complains that _seagate_st0x_biosparams ((sp?) I'm typing this
>from memory) is undefined. I did some tinkering to get this to work
>since such a function is indeed present but did not have much luck. I
>tried to use the seagate SCSI code from 99pl4 (which is what I am
>upgrading from) but pl9 has apparently changed quite a bit and the old
>code would need alot of work to integrate into the pl9 kernal.
>
I had exactly the same problem a week or so ago. The solution is to tell
make config to configure scsi disks. This is because (from memory) the
seagate_st0x_biosparams only gets defined if the disk support us configured.
But it always gets declared in the hosts.[ch] code.
Another problem you may have (I did) it that if you don't have a cartridge in
the drive when booting linux it doesn't see it. This seems to be because the
scan_scsis() function does a TEST_UNIT_READY before issuing the INQUIRY command.
If there is no tape in the drive the TEST_UNIT_READY returns an error.
I've looked at other systems scsi drivers and they just use the INQUIRY command.
Jo.
-- Jo Stockley jvs@netcom.com jo@88open.org