Hi,
On Thu, 10 Feb 2000 17:44:54 -0800, Bruce Thompson
<bruce@otherother.com> said:
> Sorry, but I must respectfully disagree. The time to service the
> request specifically is _not_ unbounded, it's bounded by the size of
> the disk.
... which may be several hours of IO. That's effectively unbounded as
far as the user is concerned, and still implies that we need extra
fairness policies above that provided by a plain CSCAN IO scheduler.
> In general I would assert that the probability of the user process
> swamping the driver in this manner is extremely low, or else the
> relative priority between the driver and the user process is mixed
> up.
No, unfortunately it is a common pattern --- if a user process starts
writing to a large file sequentially, then the blocks hitting the IO
layers typically end up being queued in ascending order of block number,
which is exactly what is required to provoke maximal unfairness in an
unmodified CSCAN.
--Stephen
-
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