[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

JFFS2 before and after patch jitter performance



Since a picture is worth a 1K words, I wrote a little program to mine data from my log file and
I plotted the jitter time greater than a threshold (in this case I choose 500ms) vs disk "du".

Both, before any patches and after your erase.c and nodemgmt.c patch were applied, the same tests were run and data
collected.

I've uploaded the plot at the following URL.

http://www.embeddedlinuxworks.com/jitter.jpg

Pretty interesting results: The typical performance is significantly better, but the worst case is significantly
worse.

This tool should make for easy visual per/post comparisons, so bring on the patches :)

Vipin


David Woodhouse wrote:

> dwmw2@xxxxxxx.org said:
> >  I didn't expect the jitter to go away.
> > You're still garbage-collecting a whole 256KiB block in the timeslice
> > of the unlucky write, rather than doing a little bit each time.
> > That's a separate problem.
>
> We really ought to sort out the reserved space properly, some time soon.
> In the meantime, this may help.
>
> Index: nodemgmt.c
> ===================================================================
> RCS file: /home/cvs/mtd/fs/jffs2/nodemgmt.c,v
> retrieving revision 1.39
> diff -u -r1.39 nodemgmt.c
> --- nodemgmt.c  2001/04/11 22:20:26     1.39
> +++ nodemgmt.c  2001/06/23 23:17:49
> @@ -42,6 +42,35 @@
>  #include <linux/interrupt.h>
>  #include "nodelist.h"
>
> +/* Bias the free block calculations to reduce jitter. If we've already


To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com