scsi initiator framework and driver expectations!

Adrian Chadd <[email protected]>
Newsgroups gmane.os.netbsd.devel.kernel
Message-ID <CAJ-Vmokm4dw0ucSr2dSjQs0McJdW4CRm_ytpEqYL+NxqE8ANjg@mail.gmail.com>
hi!

I'm debugging fun edge cases with the WD33C93B on my SGI Indy.
(They're all from a ZuluSCSI 1.2 whose firmware panics the first time
netBSD sends a write6 command for a 64k buffer w/ SYNC transfers
enabled, but I digress.)

I have questions about the expectations between the driver and the scsi layer.

* is the expectation that anything submitted to the driver is
completed, error or otherwise?
* what handles transfer timeouts? the WD33C93B driver has some error
paths where the transfer just isn't ever completed (error or
otherwise), and as with the above question, I'd like to understand
what the lifecycle should be.
* in the case of a transfer timeout, should we retry the transfer with
SYNC disabled and see if it succeeds? where would be the best place
for that to happen?

Thanks!



-adrian
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.