Well, this problem seems strange, but what about trying the following:

1.  Try writing down the dmesg output of your machine with both cards
installed.  Specifically, write down the dev and function numbers put out
for each card.

2.  Decide which card you want to be the first card detected, i.e. which
one you want to be ahb0 and which one should be ahb1.

3.  Wire those down in your config file by replacing the "?" marks in the
ahb line with the appropriate numbers, and creating an ahb0 and ahb1 line.
Then, assign scsibus0 and scsibus1 to each card as well.

4.  Wire your disk targets down at the various buses so that they're always
probed in the same order and land on the same number.

5.  Rebuild your kernel.
	I don't know if this will fix it, but it sounds like either you're
choosing a disk that isn't really the one you want as your root disk,
causing you to not have any devices, i.e. /dev/console, to type on when the
kernel is ready to ramp up to single user mode.  Or, there  is some
conflict between the cards that might be ironed out if their isn't as much
wild carding going on.  I know that I have a machine with a 2940uw and a
3940uw running just fine, but that I had to swap the card positions
physically in the machine to get the behavior I wanted and it took me a
while to get scsibus0 assigned to the bus I wanted.  So, you can either
swap the positions around, and/or you can try the suggestions above.  If
the don't work, they might at least produce moree error messages which
might help us or you point you in the right direction.

Good luck.
-Brian
On Nov 18,  5:38pm, Scott Burns wrote:
} Subject: More info on the Dual 2940UW problem
} HELP !
} 
} I am getting a little desperate now.  If I don't get this thing working in 
} the next week or so management is going to say put NT on it as they don't 
} want  $ 12K sitting around doing nothing ;-( I can always start with a 
} single controller but I would really like to use the second one with two 
} drives per controller.
} 
} With the dual 2940UW cards still running into lockups right after the 
} kernel prints out the biomask, ttymask etc.
} 
} I have confirmed the following:
} 0) This is a new Compaq Prosignia 200 - PII 300, 160MB, PCI, ISA etc. BIOS 
} dated 08/17/98 Rev P05.
} 
} 1) No IRQ conflicts, as a matter of fact I have disabled all non-essential 
} devices on the mother board via the BIOS screens (LPT*, COM*, Floppy, both 
} IDE controllers, onboard NIC, etc. etc.). Adaptec's seem to be seating them 
} selves at: ahc0, irq 11, io 6000, and ahc1, irq 5, io 6400.
} 
} 2) Thought it might be some other device probe in the fat kernel corrupting 
} that controller, as it was on IRQ 5 which the stock kernel has hooked to 
} COM3 etc, as it came on my NetBSD CD from NetBSD.COM. So I have stripped 
} the kernel down to only load the 2940 driver, SCSI devices etc. All floppy, 
} com port, lpt port, NICs etc. have been removed.
} 
} 3) I have tried each 2940UW, cable, drive etc. individually and they work 
} fine alone.
} 
} 4) The 2940UW's are V1.34.3 firmware, brand new last week. Drives are 
} Seagate (ST39173W) Barr 9.1 ultra wides. firmware 5958 as per the kernel 
} printout.
} 
} As soon as I stick in the second 2940 the kernel loads and finds everything 
} properly but right after the bio, tty, net mask stuff near the end of the 
} boot, where it prints out the root file system parition, type etc. it hangs 
} hard.
} 
} I tried the PCIVERBOSE option and no real extra info came out that I could 
} see.
} 
} Here is my config file for the lean kernel I built.
} 
} include "arch/i386/conf/std.i386"
} 
} maxusers	32		# estimated number of users
} options 	I586_CPU
} options 	I686_CPU	
} options 	MATH_EMULATE	# floating point emulation
} options 	DUMMY_NOPS
} 
} options 	INSECURE	# disable kernel security levels
} options 	RTC_OFFSET=0	# hardware clock is this many mins. west of GMT
} 
} file-system 	FFS		# UFS
} file-system 	NFS		# Network File System client
} file-system 	CD9660		# ISO 9660 + Rock Ridge file system
} file-system 	MSDOSFS		# MS-DOS file system
} file-system 	KERNFS		# /kern
} 
} options 	FIFO		# FIFOs; RECOMMENDED
} options 	INET		# IP + ICMP + TCP + UDP
} 
} config		netbsd	root on sd0a type ffs
} 
} mainbus0 at root
} pci*	at mainbus? bus ?
} pci*	at pchb? bus ?
} pchb*	at pci? dev ? function ?	# PCI-Host bridges
} pcib*	at pci? dev ? function ?	# PCI-ISA bridges
} isa*	at pcib?
} npx0	at isa? port 0xf0 irq 13	# x86 math coprocessor
} pc0	at isa? port 0x60 irq 1		# pccons generic PC console driver
} ahc*	at pci? dev ? function ?	# Adaptec [23]94x, aic78x0 SCSI
} scsibus* at ahc?
} sd*	at scsibus? target ? lun ?	# SCSI disk drives
} st*	at scsibus? target ? lun ?	# SCSI tape drives
} cd*	at scsibus? target ? lun ?	# SCSI CD-ROM drives
} ch*	at scsibus? target ? lun ?	# SCSI autochangers
} include	"arch/i386/conf/GENERIC.local"
} 
} pseudo-device	loop			# network loopback
} pseudo-device	pty		64	# pseudo-terminals
} 
} Here is the /var/log/messages output from the stripped down kernel when I 
} only have one controller in and pretty much every thing turned off in CMOS. 
} This also had sync nego. disabled on the sd0 drive so I didn't get the 
} 20MHz 16-bit mode, normally I do. When it hangs with the two controllers 
} /var/log/messages is not updated ;-(:
} 
} Nov 18 03:22:56  syslogd: restart
} Nov 18 03:22:56  /netbsd: Copyright (c) 
} 1996, 1997, 1998
} Nov 18 03:22:56  /netbsd:     The NetBSD Foundation, Inc. 
}  All rights reserved.
} Nov 18 03:22:56  /netbsd: Copyright (c) 1982, 1986, 
} 1989, 1991, 1993
} Nov 18 03:22:56  /netbsd:     The Regents of the 
} University of California.  All rights reserved.
} Nov 18 03:22:56  /netbsd: 
} 
} Nov 18 03:22:56  /netbsd: NetBSD 1.3.2 (ncti102_ncti_com) #0: Wed Nov 18 
} 03:21:03 PST 1998
} Nov 18 03:22:56  /netbsd:     root@:/usr/src/sys/arch/  
} i386/compile/ncti102_ncti_com
} Nov 18 03:22:56  /netbsd: cpu0: family 6 
} model 5 step 2
} Nov 18 03:22:56  /netbsd: cpu0: Intel Pentium Pro 
} (686-class)
} Nov 18 03:22:56  /netbsd: real mem  = 167374848
} Nov 18 03:22:56 
}  /netbsd: avail mem = 163381248
} Nov 18 03:22:56  /netbsd: using 2068 
} buffers containing 8470528 bytes of memory
} Nov 18 03:22:56  /netbsd: 
} mainbus0 (root)
} Nov 18 03:22:56  /netbsd: pci0 at mainbus0 bus 0: 
} configuration mode 1
} Nov 18 03:22:56  /netbsd: pchb0 at pci0 dev 0 function 
} 0
} Nov 18 03:22:56  /netbsd: pchb0: vendor 0x8086 product 0x1237 (rev. 
} 0x02)
} Nov 18 03:22:56  /netbsd: vendor 0x1013 product 0x00b8 (VGA display, 
} revision 0x45) at pci0 dev 10 function 0 not configured
} Nov 18 03:22:56 
}  /netbsd: ahc0 at pci0 dev 11 function 0
} Nov 18 03:22:56  /netbsd: ahc0: 
} interrupting at irq 11
} Nov 18 03:22:56  /netbsd: ahc0: aic7880 Wide 
} Channel, SCSI Id=7, 16 SCBs
} Nov 18 03:22:56  /netbsd: scsibus0 at ahc0 
} channel 0: 16 targets
} Nov 18 03:22:57  /netbsd: ahc0: target 0 using 16Bit 
} transfers
} Nov 18 03:22:57  /netbsd: sd0 at scsibus0 targ 0 lun 0: <SEAGATE, 
} ST39173W, 5958> SCSI2 0/direct fixed
} Nov 18 03:22:57  /netbsd: sd0: 8683MB, 
} 7501 cyl, 10 head, 237 sec, 512 bytes/sect x 17783240 sectors
} Nov 18 
} 03:22:57  /netbsd: vendor 0x0e11 product 0xae35 (miscellaneous network, 
} revision 0x10) at pci0 dev 16 function 0 not configured
} Nov 18 03:22:57 
}  /netbsd: pcib0 at pci0 dev 20 function 0
} Nov 18 03:22:57  /netbsd: pcib0: 
} vendor 0x8086 product 0x7000 (rev. 0x01)
} Nov 18 03:22:57  /netbsd: vendor 
} 0x8086 product 0x7010 (IDE mass storage, interface 0x80) at pci0 dev 20 
} function 1 not configured
} Nov 18 03:22:57  /netbsd: isa0 at pcib0
} Nov 18 
} 03:22:57  /netbsd: npx0 at isa0 port 0xf0-0xff: using exception 16
} Nov 18 
} 03:22:57  /netbsd: pc0 at isa0 port 0x60-0x6f irq 1: color
} Nov 18 03:22:57 
}  /netbsd: pc0: console
} Nov 18 03:22:57  /netbsd: biomask 800 netmask 800 
} ttymask 802
} Nov 18 03:22:57  /netbsd: boot device: sd0
} Nov 18 03:22:57    
} /netbsd: root on sd0a dumps on sd0b
} Nov 18 03:22:56  savecore: can't find 
} device 146/48
} Nov 18 03:25:49  syslogd: exiting on signal 15
} 
>-- End of excerpt from Scott Burns