SCSI problem nearly solved

Ok, after letting the problem get some rest I figured out why I see those strange IRQ 0. There is a bug in the driver, which leads to enabling IRQ 0 instead of IRQ 1, when a scsi_abort() occurs. Easy fix, but only part of the problem. The lockup happened because of a bug in the generic wd33c93 driver part. If the command, which should be aborted, was at the head of the input queue, the driver messed up the input queue, which leads to a endless loop on the next abort. Solved !

Now the only remaining problem is, that the the wd93 reset code, isn't able to reset the bus, because the chip is busy (looks like it doesn't get ready again in that situation). I guess I have to use the HPC3 channel reset, but my first try to do it wasn't quite successful. I guess one of my next tries will succeed:-)

Thomas.

   This device has completely bogus header. Compaq scores again :-|
It's a host bridge, but it should be called ghost bridge instead ;^)
                                        [Martin `MJ' Mares on linux-kernel]