Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 23 Dec 1997 19:40:54 -0600
From:      David Kelly <dkelly@hiwaay.net>
To:        leifn@image.dk (Leif Neland)
Cc:        freebsd-questions@FreeBSD.ORG, freebsd-scsi@FreeBSD.ORG
Subject:   Re: Adaptec AHA-2842A troubles 
Message-ID:  <199712240140.TAA28552@nospam.hiwaay.net>
In-Reply-To: Message from leifn@image.dk (Leif Neland)  of "23 Dec 1997 19:32:14 %2B0100." <903_9712231954@swimsuit.swimsuit.roskildebc.dk> 

next in thread | previous in thread | raw e-mail | index | archive | help
Leif Neland writes:

>On the third 486, it worked.
>This used to have an AVA-1505 and IDE, but now has the 2842A, and boots
>from IDE.
>
>So now the first 486 boots (dos) from a 1542A, and has the fbsd-disks on
>the AVA-1505.
>
>But I still can't understand why the scsi-bus works differently on
>different boards.

Its not the SCSI bus that's different. Its the PB MB and/or BIOS. Not 
all VL bus slots are bus master capable. Maybe it wasn't until the 3rd 
PC that you happened on a bus master VL bus slot. And then again there 
is the issue of the VL bus clock rate. Some MB's ran the VL bus at full 
CPU clock, a 486DX50 (not a DX/2) will probably run the VL bus at 50 
MHz. Adaptec's plain 2842 would run at 50 MHz, the 2842A would not.

Can multiple 1542's exist in a system at one time? Once Upon A Time I 
tried a 1542CF and NE2100 at the same time. The NE2100 didn't tolerate 
any other bus masters. Sold all my NE2100's for cheap. Am wondering if 
Leif had multiple bus masters on his failure systems and that caused 
the failures?

I have a cheapo VL/PCI 486MB with AMD 5X86/133-P75. Not sure why I can't
boot my 2842 non-A. It scans and detects the SCSI devices but attempting
to boot ends up with a message that claims the A: disk isn't readable.
Never made it to BootEasy. There wasn't a floppy in the A: drive.
Putting a floppy in the A: drive doesn't work if A: is not the first
device in the BIOS boot order. Moving the SCSI cable back to the 2940 in
the same system and everything works (see dmesg snippet below).

Is there a BIOS issue with the Adaptec 2842? I have one of the early 
ones, one the prior owner said was run at 50 MHz VL bus speed. Am 
pretty sure I'm not trying to run it over 33 MHz. Prior owner said, 
"I never could boot the 2842 either." Ain't PC's just dandy?

If FreeBSD discovers the card during probe, has it passed the the bus 
master test? I can't hook anything to the 2842 until I get new/more 
cables or snag an external HD or tape from another system:

FreeBSD 2.2.5-STABLE #0: Thu Dec 18 15:50:41 CST 1997
    root@Grumpy.tbe.com:/usr3/src/sys/compile/GRUMPY
CPU: AMD Am5x86 Write-Back (486-class CPU)
  Origin = "AuthenticAMD"  Id = 0x4f4  Stepping=4
  Features=0x1<FPU>
real memory  = 33554432 (32768K bytes)
avail memory = 30494720 (29780K bytes)
ahc0: <Adaptec 284X SCSI host adapter> at 0x1c00-0x1cff irq 15 on eisa0 slot 1
ahc0: aic7770 <= Rev C, Single Channel, SCSI Id=7, 4 SCBs
ahc0 waiting for scsi devices to settle
...
ahc1 <Adaptec 2940 SCSI host adapter> rev 0 int a irq 11 on pci0:13
ahc1: aic7870 Single Channel, SCSI Id=7, 16/255 SCBs
ahc1 waiting for scsi devices to settle
ahc1: target 0 Tagged Queuing Device
(ahc1:0:0): "IBM DCAS-34330 S61A" type 0 fixed SCSI 2
sd0(ahc1:0:0): Direct-Access 4134MB (8467200 512 byte sectors)
(ahc1:3:0): "ARCHIVE ANCDA 2750 28077 -003" type 1 removable SCSI 2
st0(ahc1:3:0): Sequential-Access density code 0x15, 512-byte blocks, write-enabl
ed

Other thing to note is the 2842 probes before the 2940, both in BIOS
boot time and FreeBSD kernel probe time. I *think* I've built a kernel
with sd0 wired to ahc1 (haven't tried it yet). Am thinking if I put a HD
on the 2842 the BIOS is going to attempt to boot it (and fail) and my
new kernel will go for naught?

If I can make the 2842 work in this system (my only VL bus slot) then 
the 2940 can go to work in another box.

--
David Kelly N4HHE, dkelly@nospam.hiwaay.net
=====================================================================
The human mind ordinarily operates at only ten percent of its
capacity -- the rest is overhead for the operating system.





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199712240140.TAA28552>