Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 29 Apr 1998 12:55:23 -0700
From:      Julian Elischer <julian@whistle.com>
To:        schofiel@xs4all.nl
Cc:        freebsd-hardware@FreeBSD.ORG
Subject:   Re: [Target Mode] Was: Ooops - sorry
Message-ID:  <354785AB.63DECDAD@whistle.com>
References:  <3547389F.121E@xs4all.nl>

next in thread | previous in thread | raw e-mail | index | archive | help
Rob Schofield wrote:
> 
> Julian, I suspect I may well have grievously offended you. If this is
> the case, I apologise, as it was not the intention to do so (late at
> night, half-busy with work). I hope you can forgive me - certainly
> cheesed up everything else in that letter!

no worries..

> 
> As regards my comments on SCSI:

[says that the chipsets themselves can support both target  and
initiator modes]

well, yes of course.
and those cards that are based on a raw chip can do it.
the drivers hopwever are not likely to be heavily tested in target mode
if they support it at all. 

> Ah, now I can correct you here; I have successfully worked on two
> PC-based projects where a 1510 and a 1740 were used in target mode, so
> the *hardware at least* does. Whether, as you rightly say, the 
> PC- based BIOS provided with these cards actually implements this 
> mode or not (and
> for any other SCSI device class other than Disk, at that) is another
> question; we didn't use it - simply disabled the BIOS, got a register
> map and the right data sheet for the chips on the cards, and wrote our
> own drivers, which were then successfully tested under the operating
> system we were using (OS-9000).
t it, so we'd have to add that support..
> 

the 1510 was a card that was ony suitable for low load applications 
(or dedicated use) because the processor had to do an aweful lot of
work for it's data.  Of course as it was chip-level you could do
anything with the driver.

The 1540 (or 1520) series was more suitable as they added a 
coprocessor to do all the hard work. The firmware that this 
processor ran was capable of doing some target mode work but 
they broke that feature regularly and didn't always notice.
 ;-(

As you could not bypass the coprocessor, if they didn't do target mode
correctly, neither did you.


The 1740 was supposed to support target mode too, but we 
eventually gave up on them ever getting it to work correctly.
(talk to peter dufault on this).

It was like the 1540 in that it had a coprocessor, but they 
never got the target mode reliable during the time that we 
tried to use it. They eventually told us that they were going to
stop saying that they supported target mode on that and that we 
should stop asking them to fix it.

Rumour was that there was one release of the firmware that
worked, but they quickly corrected that oversight and we could never
get one. Once again since you couldn't bypass the coprocessor
you were dependent on it doing it correctly.

Peter had changes to the SCSI system to make allowances for target
mode operation of adapters, but as we couldn't get any adapters
that could actually do it at that time (1993?) we never checked them
in.
> 
[...]

> >we are NOT using it as they won't give source..
> 
> No defintiely, they won't *give* you the ASPI sources for their
> commercially available, main income source. Would you? ;^) However, if
> you asked them for the developer's packages and ASPI specs, well,
> they'll let you have that. Maybe I was just lucky when I asked, I got
> all my stuff for free in '95....
> 
> >and 'MOST' is not enough.
> 
> ... but it could form the solid basis of an good, original piece 
> of work performed by you, based on commercially-accepted 
> standards? What do you want them to do, write it all for you?

With the new CAM based system that Justin is working on, there 
may be an oportunity to make better use of externally written
modules.

[...]

> 
> Tell me, why use the firmware? I thought this was U**X? Why on earth do
> you need to use the card firmware?
> 

Only on intelligent cards where the firmware is executed by the
coprocessor. Since the advent of efficient chip-level solutions
such as the adaptec and NCR/symbios/adaptec (??) chips
it depends more on the drivers and we are freed of the firmware
(thank god) but we still need to have people working on them who
want and need target mode because of course even if the 
state-machines support target mode, we never use/test it.

I'm basically out of the SCSI business these days.
We don't use it here and I don't have the time anyhow.
Justin is now the person doing all the work and he has made
an engineering decision that in the long run freeBSD is better
served by rewriting the SCSI system to be based on CAM rather than
trying to patch the old one more.
I guess that there will eventually be someone who wants to patch 
target mode into the CAM based version and I'm sure that Justin will 
have made allowances for that. (CAM supports it, so his 
implimentation should at least have the hooks in the right places 
for it)

julian

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hardware" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?354785AB.63DECDAD>