[technical points elided]
Hmm; this was a long time ago and it's certainly possible that
I misunderstood the mechanism used (or the developer was simply
yanking my chain with a bogus description). Or maybe I heard it
right, but he exploited something specific to the Interactive
One horrid possibility that I should mention, if only for the
inevitably amusing misquotes, is the possibility of running NFS
over SCSI. Obviously this could be done with full-blown support
of IP-over-SCSI, but that might be overkill. Maybe the real
mechanism used was a "SCSI transport" at the VFS layer. With a
virtual disk, the VFS calls are mapped to generic SCSI commands.
The host reads them, does the same VFS call, then returns the
results. This should reduce caching issues to roughly the same
scale as multiple writing processes on a single system.
Unfortunately, if that's the case the nature of the "disk" won't
be totally transparent to the embedded system.
-- Bear Giles firstname.lastname@example.org
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to email@example.com Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:18 EST