Re: elevator-starvation-4 (2.2.14 && 2.3.42)

From: Andrea Arcangeli (andrea@suse.de)
Date: Fri Feb 11 2000 - 10:51:48 EST


On Thu, 10 Feb 2000, Bruce Thompson wrote:

>[..] In a case where
>requests are being generated faster than the driver can deal with
>them then no matter how we order the request queue, we've got a
>problem.

Just think when one user grabs TV or when an user writes on an NFS server
with a gigabit network. Or think a CPU intensive program that generate a
flood of output on disk (a common strace -o /tmp/o can do exactly that
with a fast enough CPU and a slow enough HD).

I use `cp /dev/zero .` to do that testing just to emulate the effect of
the possible real world scenarios.

`cp /dev/zero .` usually shows the hang because usually CPU is faster than
HD.

Andrea

-
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/



This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:20 EST