On Sun, Jan 17, 1999 at 09:29:19PM +0000, Karl Pielorz wrote: > > > Lee Cremeans wrote: > > > > The machine is a dual P90 (running SMP) - I get the problems even with a > > > single CPU kernel... It's running Neptune chipset (quite old) - and only > > > has 1 IDE channel. > > > > I would say that this has something to do with the DMA support, but since > > Neptune's IDE controller isn't DMA capable and PCI, that doesn't follow. > > Chances are that there's a not-quite-bad spot on your drive -- does it > > always seem to happen in the same place on the disk? > > Nope, it's more 'load' oriented than 'spot' oriented (i.e. the harder you push > them - the more of the above you get)... Both drives are brand new out the box > (for what it's worth), so I doub't they're 'bad'... DMA is disabled in the > kernel config as well, and the dmesg shows no hints of either multi-block or > DMA transfers... > > Thinking that way I've tried another cable since (brand new out the bag - > again 'for what it's worth') - and that has the same symptoms... So does > running only 1 drive on the bus... :-(
I dunno then; it sounds like the IDE port is broken and starts dropping interrupts. This is a bit over my head -- any interrupt-service-routine gurus want to jump in here? I don't want to condemn the motherboard just yet. -- +--------------------------------------------------------------------+ | Lee Cremeans -- Manassas, VA, USA (WakkyMouse on DALnet and WTnet)| | lcrem...@tidalwave.net| http://st-lcremean.tidalwave.net/~lee | To Unsubscribe: send mail to majord...@freebsd.org with "unsubscribe freebsd-current" in the body of the message