Description of Problem:
An updated kernel 2.4.9-31 inherited an afliction reported elsewhere but
not against 7.1.  Namely while booting UP1100 with IDE drives the following
happens:

hda: 20010816 sectors (10246 MB) w/512KiB Cache, CHS=19852/16/63
hdb: 17803440 sectors (9115 MB) w/1966KiB Cache, CHS=17662/16/63, UDMA(33)
hdc: 30015216 sectors (15368 MB) w/2048KiB Cache, CHS=29777/16/63
Partition check:
 hda: hda1 hda2 hda3 hda4 hda5
 hdb:hdb: timeout waiting for DMA
ide_dmaproc: chipset supported ide_dma_timeout func only: 14
hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
hdb: drive not ready for command
 hdb1 hdb2 hdb3 hdb4 <hdb: timeout waiting for DMA
ide_dmaproc: chipset supported ide_dma_timeout func only: 14
hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
hdb: drive not ready for command
 hdb5hdb: timeout waiting for DMA
ide_dmaproc: chipset supported ide_dma_timeout func only: 14
hdb: status error: status=0x58 { DriveReady SeekComplete DataRequest }
hdb: drive not ready for command
 hdb6 >
 hdc: hdc1 hdc2 hdc3 hdc4 hdc5 hdc6

I can boot the machine because it happens to have a root file system on
/dev/hda and neither /dev/hdb nor /dev/hdc is really mounted (a pure
accident) in the test configuration but otherwise I would be in a big
trouble.  After boot everything is verrry slooow....

Of course other kernels, configured for this Alpha, do not have
even a whiff of the problem.

The real layout of IDE disks on the machine in question happens to be:

 hda: hda1 hda2 hda3 hda4 hda5
 hdb: hdb1 hdb2 hdb3 hdb4 < hdb5 hdb6 >
 hdc: hdc1 hdc2 hdc3 hdc4 hdc5 hdc6

(Yes, hdb indeed has a FAT type partition table while hda and hdc a "BSD
label").

Version-Release number of selected component (if applicable):
kernel-2.4.9-31

How Reproducible:
On every boot.