From: Andrew Morton <akpm@osdl.org>
To: Peter Osterlund <petero2@telia.com>
Cc: pavel@ucw.cz, linux-kernel@vger.kernel.org
Subject: Re: Software suspend testing in 2.6.0-test1
Date: Fri, 18 Jul 2003 13:15:27 -0700 [thread overview]
Message-ID: <20030718131527.7cf4ca5e.akpm@osdl.org> (raw)
In-Reply-To: <m2oezrppxo.fsf@telia.com>
Peter Osterlund <petero2@telia.com> wrote:
>
> The patch below works, but doesn't really solve anything, because it
> is just as slow as the original code.
That's OK - it's a step in the right direction.
> This is not surprising because
> it is still balance_pgdat() that does the page freeing, the only
> difference being that it is now called from kswapd which is woken up
> by the alloc_page() call.
Yup. Probaby we should not be throttling kswapd if it is making good
progress, but this needs broad testing and thought. Something like this,
untested:
--- 25/mm/vmscan.c~less-kswapd-throttling Thu Jul 17 15:35:09 2003
+++ 25-akpm/mm/vmscan.c Thu Jul 17 15:35:12 2003
@@ -930,7 +930,8 @@ static int balance_pgdat(pg_data_t *pgda
}
if (all_zones_ok)
break;
- blk_congestion_wait(WRITE, HZ/10);
+ if (to_free)
+ blk_congestion_wait(WRITE, HZ/10);
}
return nr_pages - to_free;
}
> Note that I had to use HZ/5 in free_some_memory() or else the loop
> would terminate too early. The main problem seems to be that
> balance_pgdat() is too slow when freeing memory. The function can fail
> to free memory in the inner loop either because the disk is congested
> or because too few pages are scanned, but in both cases the function
> calls blk_congestion_wait(), and in the second case the disk may be
> idle and then blk_congestion_wait() doesn't return until the timeout
> expires.
hm, perhaps because kswapd is throttled rather than doing work.
Try the above change, see if you still need the extended timeout in
free_some_memory().
Also, play around with using __GFP_WAIT|__GFP_HIGH|__GFP_NORETRY rather
than GFP_ATOMIC.
next prev parent reply other threads:[~2003-07-18 20:08 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-07-17 19:46 Software suspend testing in 2.6.0-test1 Peter Osterlund
2003-07-17 20:00 ` Pavel Machek
2003-07-17 20:09 ` Andrew Morton
2003-07-18 9:59 ` Peter Osterlund
2003-07-18 10:24 ` Andrew Morton
2003-07-18 15:22 ` Pavel Machek
2003-07-18 15:55 ` Peter Osterlund
2003-07-18 16:45 ` Andrew Morton
2003-07-18 17:50 ` Pavel Machek
2003-07-18 18:02 ` Patrick Mochel
2003-07-18 18:04 ` Pavel Machek
2003-07-18 21:05 ` Nigel Cunningham
2003-07-18 19:37 ` Andrew Morton
2003-07-18 19:58 ` Peter Osterlund
2003-07-18 20:15 ` Andrew Morton [this message]
2003-07-18 22:13 ` Pavel Machek
2003-07-20 0:22 ` Peter Osterlund
2003-07-20 1:01 ` Andrew Morton
2003-07-20 7:45 ` Peter Osterlund
2003-07-18 21:05 ` Nigel Cunningham
2003-07-21 10:00 ` Peter Osterlund
2003-07-21 12:58 ` Pavel Machek
2003-07-21 14:36 ` Peter Osterlund
2003-07-21 21:28 ` Pavel Machek
2003-07-21 23:46 ` Peter Osterlund
2003-07-22 11:04 ` Pavel Machek
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20030718131527.7cf4ca5e.akpm@osdl.org \
--to=akpm@osdl.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pavel@ucw.cz \
--cc=petero2@telia.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).