* ClockPro in Linux MM
@ 2012-03-12 15:19 Zheng Da
2012-03-12 16:04 ` Christoph Lameter
0 siblings, 1 reply; 3+ messages in thread
From: Zheng Da @ 2012-03-12 15:19 UTC (permalink / raw)
To: linux-mm
[-- Attachment #1: Type: text/plain, Size: 517 bytes --]
Hello,
I try to understand the Linux memory management. I was told Linux uses
ClockPro to manage page cache
and http://linux-mm.org/PageReplacementDesign also says so for file pages.
But when I read the ClockPro paper,
it doesn't look the same. The Linux implementation doesn't have
non-resident pages. Other than
that, it doesn't have the same test period mentioned in the paper. I wonder
if the Linux implementation
have the same effect as ClockPro. Could anyone confirm Linux is still using
ClockPro?
Thanks,
Da
[-- Attachment #2: Type: text/html, Size: 778 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: ClockPro in Linux MM
2012-03-12 15:19 ClockPro in Linux MM Zheng Da
@ 2012-03-12 16:04 ` Christoph Lameter
2012-03-12 17:32 ` Zheng Da
0 siblings, 1 reply; 3+ messages in thread
From: Christoph Lameter @ 2012-03-12 16:04 UTC (permalink / raw)
To: Zheng Da; +Cc: linux-mm
On Mon, 12 Mar 2012, Zheng Da wrote:
> I try to understand the Linux memory management. I was told Linux uses
> ClockPro to manage page cache
> and http://linux-mm.org/PageReplacementDesign also says so for file pages.
> But when I read the ClockPro paper,
> it doesn't look the same. The Linux implementation doesn't have
> non-resident pages. Other than
> that, it doesn't have the same test period mentioned in the paper. I wonder
> if the Linux implementation
> have the same effect as ClockPro. Could anyone confirm Linux is still using
> ClockPro?
That Linux is using Clockpro is news to me. Linux Memory management uses
some ideas from Clockpro to improve reclaim etc but it does not implement ClockPro.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: ClockPro in Linux MM
2012-03-12 16:04 ` Christoph Lameter
@ 2012-03-12 17:32 ` Zheng Da
0 siblings, 0 replies; 3+ messages in thread
From: Zheng Da @ 2012-03-12 17:32 UTC (permalink / raw)
To: Christoph Lameter; +Cc: linux-mm
[-- Attachment #1: Type: text/plain, Size: 1029 bytes --]
I know the implementation is different. Linux implementation doesn't have
the clock heads.
But I get a feel that the essence is the same except that Linux version
doesn't have non-resident pages.
Da
On Mon, Mar 12, 2012 at 12:04 PM, Christoph Lameter <cl@linux.com> wrote:
> On Mon, 12 Mar 2012, Zheng Da wrote:
>
> > I try to understand the Linux memory management. I was told Linux uses
> > ClockPro to manage page cache
> > and http://linux-mm.org/PageReplacementDesign also says so for file
> pages.
> > But when I read the ClockPro paper,
> > it doesn't look the same. The Linux implementation doesn't have
> > non-resident pages. Other than
> > that, it doesn't have the same test period mentioned in the paper. I
> wonder
> > if the Linux implementation
> > have the same effect as ClockPro. Could anyone confirm Linux is still
> using
> > ClockPro?
>
> That Linux is using Clockpro is news to me. Linux Memory management uses
> some ideas from Clockpro to improve reclaim etc but it does not implement
> ClockPro.
>
>
[-- Attachment #2: Type: text/html, Size: 1501 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2012-03-12 17:32 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-03-12 15:19 ClockPro in Linux MM Zheng Da
2012-03-12 16:04 ` Christoph Lameter
2012-03-12 17:32 ` Zheng Da
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox