About a year ago I noticed that when backing up to DLT my main
fileserver's clock seemed to slip severely.  I assumed it was a
problem in the adw or isp drivers, hunted for it a bit, and
really found pretty much nothing.  I stopped using the fileserver
as an NTP server and didn't think much of it.

Now I'm sitting at home with a cheezy 5MB/sec SCSI DLT 2500 on
an ancient bha host adapter, trying to restore a tape of some
music files.  I'm noticing not only time slip but a steady stream
of "ohci0: one scheduling overruns".

I don't see anywhere obvious in the st or mid-layer code that
would block interrupts for a Long Time, but now I'm convinced
there must be some such problem.  Does anyone else see an
obvious suspect here?

-- 
 Thor Lancelot Simon	                                      tls@rek.tjls.com
   But as he knew no bad language, he had called him all the names of common
 objects that he could think of, and had screamed: "You lamp!  You towel!  You
 plate!" and so on.              --Sigmund Freud