[lkml]   [1999]   [Mar]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [patch] Re: MP3 skippety skip skipageness
On Sat, 13 Mar 1999, Pavel Machek wrote:

> > am i correct that you carelessly volunteer as a guinea pig? :) I do not
> > have a sound card installed now, but could you try the attached patch,
> > does it change anything? Thanks,
> Do you believe that your patch helps when I make machine go 20meg onto
> swap? I _really_ don't think so, so I'm not testing it.

Being 20 megs into swap (and actively using it) might
be a bit extreme, yes...

Now about the patch:

> [this is not my patch. Ingo wrote it.]
> --- linux/drivers/block/ll_rw_blk.c.orig Fri Mar 12 10:31:53 1999
> +++ linux/drivers/block/ll_rw_blk.c Fri Mar 12 10:32:05 1999
> @@ -536,7 +536,7 @@
> /* if no request available: if rw_ahead, forget it; otherwise try again blocking.. */
> if (!req) {
> - if (rw_ahead)
> + if (0 && rw_ahead)
> goto end_io;
> req = __get_request_wait(max_req, bh->b_rdev);
> }

I really don't believen that this patch is correct,
or even that it's very much better than the original
(clearly braindead) code.

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.

The real fix would be to either increase the number
of requests to something more reasonable (say 128
or 256) and/or to be more intelligent about putting
more disk sectors into one request...

Rik -- Open Source: you deserve to be in control of your data.
| Linux Memory Management site: |
| Nederlandse Linux documentatie: |

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:50    [W:0.066 / U:4.104 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site