RE: IDEA: multiple dirty lists in buffer.c

Kai Harrekilde-Petersen (KHP@olicom.dk)
Wed, 31 Mar 1999 08:46:10 +0200


> From: Gerard Roudier [mailto:groudier@club-internet.fr]
>
> On Tue, 30 Mar 1999, Rik van Riel wrote:
>
> > On Mon, 29 Mar 1999, Gerard Roudier wrote:

> > > Basically, for SCSI disks, using for example 8 dirty lru lists and
> > > the following hash: [SNIP]
> >
> > But why would we ever want to use an LRU list for write-outs?
>
> Because we are unable to ask applications about buffers that will be
> rewritten soon and those that will not. ;-)
>
> The LRU is just some minimal and simple aging heuristic that
> can be used
> when we have no other info about the relevance of caching or
> not. It is
> usually better that just random and fairly simple.

</lurk>

I am in no way an expert on the kernel, but the discussion about
ordering just made this pop into my head: Have you considered
multi-disk partitions (ie: raid systems)? - I assume that the current
single-dirty-list design does some "default" ordering between the
disks, which is necessary for multi-disk partitions, and the
dirty-list-per-disk design might remove it without realizing it.

<lurk>

Kai

-- 
Kai Harrekilde-Petersen <khp@olicom.dk> #include <std/disclaimer.h> 

- 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/