Hi, On Wed, May 22, 2002 at 10:50:08AM +1000, Neil Brown wrote: > I also get raid5 to report the device sector number (bh->b_rsector) > and to print out the first 32 bytes of the before and after blocks. > 25 minutes later I got two messages from raid5 about "multiple 1 > requests" (the '1' means 'WRITE', '0' would be 'READ'). I've just put together a loop.c extension that allows me to layer arbitrary debug code on top of a block device. The extension tracks all the outstanding IOs to that device, and searches for duplicates. I'm running it now on a single-disk UP data=journal filesystem, and no problems so far after a couple of million buffer IOs on the device. I'll get that going on a highmem smp box tomorrow to see if that triggers anything, both on a single disk and layered on top of a raid5 array. Cheers, Stephen
