Chris Mason wrote:
>
> On Thursday, December 28, 2000 11:29:01 AM -0800 Linus Torvalds
> <torvalds@transmeta.com> wrote:
> [ skipping io on the first walk in page_launder ]
> >
> > There are some arguments for starting the writeout early, but there are
> > tons of arguments against it too (the main one being "avoid doing IO if
> > you can do so"), so your patch is probably fine. In the end, the
> > performance characteristics are what matters. Does the patch make for
> > smoother behaviour and better performance?
>
> My dbench speeds have always varied from run to run, but the average speed
> went up about 9% with the anon space mapping patch and the page_launder
> change. I could not find much difference in a pure test13-pre4, probably
> because dbench doesn't generate much swap on my machine. I'll do more
> tests when I get back on Monday night.
>
> Daniel, sounds like dbench varies less on your machine, what did the patch
> do for you?
>
> BTW, the last anon space mapping patch I sent also works on test13-pre5.
> The block_truncate_page fix does help my patch, since I have bdflush
> locking pages ( thanks Marcelo )
Yes it does, but the little additional patch you posted no longer
applies. Your patch is suffering badly under pre5 here, reducing
throughput from around 10 MB/sec to 5-6 MB/sec, and highly variable. If
you're not seeing this you should probably try to get a few others to
run it.
-- Daniel - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Dec 31 2000 - 21:00:12 EST