FWIW, I'm using an Orinoco PCI adapter with an Orinoco wireless card
   on a i386/1.5.1 box with no problems... 

   cbb0 at pci0 dev 15 function 0: Texas Instruments product 0xac50 (rev. 0x01)
   cbb0: interrupting at irq 9
   cbb0: cacheline 0x8 lattimer 0x20
   cbb0: bhlc 0x22008 lscp 0x20000000
   cardslot0 at cbb0 slot 0 flags 0
   cardbus0 at cardslot0: bus 0 device 0 cacheline 0x8, lattimer 0x20
   pcmcia0 at cardslot0
   pcmcia0: CIS version PC Card Standard 5.0
   pcmcia0: CIS info: Lucent Technologies, WaveLAN/IEEE, Version 01.01, 
   pcmcia0: Manufacturer code 0x156, product 0x2
   pcmcia0: function 0: network adapter, ccr addr 3e0 mask 1
   pcmcia0: function 0, config table entry 1: I/O card; irq mask ffff; iomask 6, io
   space 0-3f; io16 irqpulse irqlevel
   wi0 at pcmcia0 function 0
   wi0: address XX:XX:XX:XX:XX:XX

This got me realizing that my cbb0 probes differently than the one
above.  I tried fussing with a bunch of BIOS options to no avail.
However, moving the board from slot 4 to slot 2 (which has a different
group of interrupts) seems to have corrected the timeout problem.  It
rearranged how cbb0 was probed (see below for a diff of the dmesg
output in the two cases).

Any explanation of why board position (and consequent assignment of
resources--i.e., irq, pci dev #, etc.--to the boards) should make such
a big difference is welcome.  Also, it is not entirely clear to me
where all the pcmcia0 messages above come from.  I have never seen
them on either my 1.5 or 1.5.2 kernels, even with all sorts of *DEBUG
and *VERBOSE kernel options.

Thanks for all the help and suggestions offered by everyone.

Cheers,
Brook

===========================================================================
--- dmesg-1.5.2.GENERIC_WI_slot4	Tue Oct 23 19:04:28 2001
+++ dmesg-1.5.2.GENERIC_WI_slot2	Tue Oct 23 23:09:09 2001
@@ -38,15 +38,15 @@
 pchb1 at pci0 dev 4 function 4
 pchb1: VIA Technologies VT82C686A SMBus Controller (rev. 0x40)
 auvia0 at pci0 dev 4 function 5: VIA VT82C686A AC'97 Audio (rev 0x50)
-auvia0: interrupting at irq 5
+auvia0: interrupting at irq 9
 auvia0: ICE17 codec; headphone, 18 bit DAC, 18 bit ADC, Unknown 3D
 audio0 at auvia0: full duplex, mmap, independent
+cbb0 at pci0 dev 9 function 0: Texas Instruments product 0xac50 (rev. 0x01)
 ahc1 at pci0 dev 10 function 0
 OptionMode = 3
-ahc1: interrupting at irq 5
+ahc1: interrupting at irq 9
 ahc1: aic7892 Wide Channel A, SCSI Id=7, 16/255 SCBs
 scsibus0 at ahc1 channel 0: 16 targets, 8 luns per target
-cbb0 at pci0 dev 11 function 0: Texas Instruments product 0xac50 (rev. 0x01)
 pciide1 at pci0 dev 17 function 0: Promise Ultra100/ATA Bus Master IDE Accelerator (rev. 0x02)
 pciide1: bus-master DMA support present
 pciide1: primary channel configured to native-PCI mode
@@ -73,7 +73,7 @@
 fdc0 at isa0 port 0x3f0-0x3f7 irq 6 drq 2
 fd0 at fdc0 drive 0: 1.44MB, 80 cyl, 2 head, 18 sec
 isapnp0: no ISA Plug 'n Play devices found
-cbb0: interrupting at irq 10
+cbb0: interrupting at irq 9
 cbb0: cacheline 0x8 lattimer 0x20
 cbb0: bhlc 0x22008 lscp 0x20000000
 cardslot0 at cbb0 slot 0 flags 0
@@ -96,4 +96,3 @@
 wsdisplay0: screen 3 added (80x25, vt100 emulation)
 wsdisplay0: screen 4 added (80x25, vt100 emulation)
 wsmux1: connecting to wsdisplay0
-wi0: device timeout