> Ted> Are you using an old IDE controller that does programmed I/O
> Ted> (i.e., no DMA?). If so, look at the man page for hdparm's -u
> Ted> option. That should make your problem go away. (The
> Ted> explanation for the -u option will also explain why this isn't
> Ted> the default.)
>
jim> Ted, I get the same buffer overrun on two modern systems: a brand-new
jim> Dell I3700 laptop, and a SCSI-only P2B-S motherboard. Furthermore, the
jim> problem will occur when there is relatively little disk activity. On
jim> the other hand, the error is much more frequent when there's significant
jim>CPU load.
The system I am running 2.3.28 is an AMD K6-3/400 with a relatively
high performance SCSI subsystem, actually has a 64 PCI SCSI controller,
admittedly plugged into a 32bit slot of an x86 system.
This would eliminate the IDE problem Ted spoke of in a related email.
However you are also having the problem with a SCSI only MB. hmmmm.
I guess you are running into the evil "other" devices that Ted
mentioned.
I've noticed the CPU load related underruns a little. I'll load up
my 2.3.28 systems CPU and see what happens, since it otherwise
works fine.
I think it would be nice if Linux had a serial subsystem that supports
115200 with a standard 16550A. And apparently is does. Hopefully
the "other" devices will work toward this end too.
-- Brian Litzinger <brian@litzinger.com>
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/