Subject: Re: 2940 & netbsd-current
To: Justin T. Gibbs <gibbs@freefall.freebsd.org>
From: Jan-Hinrich Fessel <oskar@zappa.unna.ping.de>
List: current-users
Date: 03/29/1996 09:11:48
In message <199603281539.HAA08099@freefall.freebsd.org>you write:
> >Uh, are there more than one Adaptec 2940's out there?  (Other than the 2940W
> >Fast/Wide appelation, that is)
> 
> There are cards based on the aic7870 Rev B, aic7870 Rev C, and the aic7880.
> The aic7880 supports Ultra SCSI (20MHz max sync).
Mar 28 23:39:02 zippo.unna.ping.de /netbsd: sd1(ahc0:1:0): timed out
Mar 28 23:52:16 zippo.unna.ping.de /netbsd: NetBSD 1.1B (ZIPPO) #6: Wed Mar 27 04:40:59 MET 1996
Mar 28 23:52:16 zippo.unna.ping.de /netbsd:     root@zippo:/test/sys/arch/i386/compile/ZIPPO
Mar 28 23:52:16 zippo.unna.ping.de /netbsd: pci0 at mainbus0 bus 0: configuration mode 1
Mar 28 23:52:16 zippo.unna.ping.de /netbsd: vendor 0x8086 product 0x122d (class bridge, subclass host, revision 0x02) at pci0 dev 0 function 0 not configured
Mar 28 23:52:16 zippo.unna.ping.de /netbsd: vendor 0x8086 product 0x122e (class bridge, subclass ISA, revision 0x02) at pci0 dev 7 function 0 not configured
Mar 28 23:52:17 zippo.unna.ping.de /netbsd: ep0 at pci0 dev 10 function 0: aui/bnc/utp address 00:20:af:f9:a4:1d
Mar 28 23:52:17 zippo.unna.ping.de /netbsd: pci_map_int: pin A mapped to line 12
Mar 28 23:52:17 zippo.unna.ping.de /netbsd: ahc0 at pci0 dev 12 function 0ahc0: 294x Single Channel, SCSI Id=7, aic7870, 16 SCBs
Mar 28 23:52:17 zippo.unna.ping.de /netbsd: ahc0: Using Edge Triggered Interrupts

This one times out on my ccd every so often.  Even worse, the machine will 
not come back to normal operation.  Even more worse, it handles write access 
to my HP DAT much worse, i.e. times out and halts the system, than the ncr 
in my old board (which burnt down last week due to fan failure).

> It really depends on load, type of drives, etc.  But since I was the one
> who fixed the firmware bugs, I know just what kind of destruction they
> can cause. 8-)  The aic7xxx cards do seem to be significantly faster than
> the NCR 8XX cards especially with something like a wide Quantum Atlas.
> The new driver also has some significant performance improvements when
> compared to the ancient code NetBSD-current is using.

So, who is going to import it?
I have trouble buying an NCR controller, because al resellers say it didn't 
pass the CE-regulation.

Gruesse
       Oskar

 ==============================================================================
 					Tragbar ist, was nicht herunterfaellt.
						    oskar@zappa.unna.ping.de
 ==============================================================================