* Swapping out old pages
@ 1999-05-18 10:18 Karl Vogel
1999-05-19 18:46 ` Rik van Riel
0 siblings, 1 reply; 2+ messages in thread
From: Karl Vogel @ 1999-05-18 10:18 UTC (permalink / raw)
To: linux-mm
I recently became interested in all this VM stuff, so I'm not what you would
call an expert (more like a newbie), but when browsing through the 2.2 mm
sources, the following questions popped into my mind:
- why does the swap out algorithm select the task with the largest RSS to
free pages. If I'm not mistaken, the age of a page isn't considered?! Why is
that? Am I overlooking something? (doesn't this mean that if I start a new
large process, it's pages immediately get swapped out even though there are
other processes that haven't done anything in the past couple of hours)
- wouldn't it be beneficial if there is a parameter that allows you to
specify that after a certain age, a page is swapped out to make room for the
buffer cache. (even if the system has plenty of ram left - the idea is that:
if your system has alot of RAM, old/unused pages (e.g. init code from
daemons etc) are never swapped out and take away ram that can be used for
better things).
--
To unsubscribe, send a message with 'unsubscribe linux-mm my@address'
in the body to majordomo@kvack.org. For more info on Linux MM,
see: http://humbolt.geo.uu.nl/Linux-MM/
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Swapping out old pages
1999-05-18 10:18 Swapping out old pages Karl Vogel
@ 1999-05-19 18:46 ` Rik van Riel
0 siblings, 0 replies; 2+ messages in thread
From: Rik van Riel @ 1999-05-19 18:46 UTC (permalink / raw)
To: Karl Vogel; +Cc: linux-mm
On Tue, 18 May 1999, Karl Vogel wrote:
> - why does the swap out algorithm select the task with the largest
> RSS to free pages. If I'm not mistaken, the age of a page isn't
> considered?! Why is that? Am I overlooking something?
You're overlooking the fact that p->swap_cnt is only
recalculated when all processes have been scanned.
A new sweep starts at the process with the largest
swap_cnt -- but all processes are scanned eventually.
> - wouldn't it be beneficial if there is a parameter that allows
> you to specify that after a certain age, a page is swapped out to
> make room for the buffer cache.
The system already does that, and has been doing so
rather agressively since 2.1.89...
cheers,
Rik -- Open Source: you deserve to be in control of your data.
+-------------------------------------------------------------------+
| Le Reseau netwerksystemen BV: http://www.reseau.nl/ |
| Linux Memory Management site: http://www.linux.eu.org/Linux-MM/ |
| Nederlandse Linux documentatie: http://www.nl.linux.org/ |
+-------------------------------------------------------------------+
--
To unsubscribe, send a message with 'unsubscribe linux-mm my@address'
in the body to majordomo@kvack.org. For more info on Linux MM,
see: http://humbolt.geo.uu.nl/Linux-MM/
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~1999-05-19 19:04 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1999-05-18 10:18 Swapping out old pages Karl Vogel
1999-05-19 18:46 ` Rik van Riel
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox