> > However, multi-lun ATAPI devices do exist. An ATAPI PD-CD device will
> > be detected as a SCSI disk drive on LUN 0 and as an ATAPI cdrom on LUN
> > 1.
>
> Ok, then very few things can be done about it, and probably none of it
> is really ide-scsi specific... Unless there is a hint on these devices
> that the LUN handling is not complete.
According to the specs, new ATAPI devices should report whether they
support multiple luns using a previously reserved field in the ATAPI
IDENTIFY DEVICE command (bits 0-2 of word 126 contain the last lun
identifier).
I preferred not checking the above bits and allowing the multiple-lun probe
in any case, since in the worst case we can just identify a single-lun
device several times and we won't be using the extra devices, but in the
alternative option we might mis-identify a real multi-lun device as a
single-lun device, and not have access to the additional luns.
> With these small sample I can
> see (maybe 2-3 hosts) I can't say it's really a common case.
> Maybe the SCSI subsystem could be prepared to detect identical devices
> by serial numbers? Although that doesn't sound too healthy... I guess
> very few people has broken and non-broken devices (LUN-wise) in the same
> system.
>
> --
> Janos - Don't worry, my address is real. I'm just bored of spam.
I see a blacklist in scsi.c with the "BLIST_SINGLELUN" and "BLIST_FORCELUN"
options, so it looks like there are also real single-lun SCSI devices
which report themself as multi-lun devices, and vice-versa.
Gadi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html