The attached patch "fixes" the bug. How bad sectors cause just sector
errors, no more system hangs. The patch is not perfect, but definitively
an improvement:

<<<<<<<
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x01 { AddrMarkNotFound }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
hde: DMA disabled
PDC202XX: Primary channel reset.
ide2: reset: master: error (0x00?)
hde: task_in_intr: status=0x59 { DriveReady SeekComplete DataRequest
Error }
hde: task_in_intr: error=0x40 { UncorrectableError }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
end_request: I/O error, dev 21:00 (hde), sector 43141736
hde: task_in_intr: status=0x59 { DriveReady SeekComplete DataRequest
Error }
hde: task_in_intr: error=0x40 { UncorrectableError }, LBAsect=43141736,
high=2, low=9587304, sector=43141736
<<<

I think "master: error (0x00?)" indicate another bug.
Andre, what do you think?

--
	Manfred