Re: [patch] Re: MP3 skippety skip skipageness

Rik van Riel (riel@nl.linux.org)
Thu, 18 Mar 1999 19:43:38 +0100 (CET)


On Thu, 18 Mar 1999, Stephen C. Tweedie wrote:
> On Tue, 16 Mar 1999 00:40:57 +0100 (CET), Rik van Riel
> <riel@nl.linux.org> said:
>
> > With the (very) agressive readahead we're doing
> > now, a request queue of 64 entries really won't
> > be big enough any more. In fact, it'll only be
> > big enough for 2 to 4 'actual' requests.
>
> No, we automatically merge sequential IOs into a single request so the
> readahead will typically only occupy a fairly small fraction of the
> request queue. Only if you encounter really excessive filesystem
> fragmentation will you start to hammer the request queue.

MP3 playing during FS reads work great, yes. But the
problem is that clustered readins also cause larger
amounts of FS writing on smaller machines, and that's
where things start to break down.

Another possible cause is that the mp3 player has to
free memory itself when it is already in need of new
data from disk...

cheers,

Rik -- Open Source: you deserve to be in control of your data.
+-------------------------------------------------------------------+
| Linux Memory Management site: http://humbolt.geo.uu.nl/Linux-MM/ |
| Nederlandse Linux documentatie: http://www.nl.linux.org/ |
+-------------------------------------------------------------------+

-
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/