kernel swap accounting...

[Mulyadi Santosa]

hi :)

On Tue, Jan 31, 2012 at 22:38, Vladimir Murzin <murzin.v at gmail.com> wrote:
> Hi!
>
> AFAIK, shared pages usually get in swap cache first and stay there till page reference count reaches limit. As soon as it is reached page is swapped out on disk.

So, what you're trying to say is that: the difference makes sense due
to some pages that end up in swap cache?

AFAIK too, swap cache is pages that once swapout, then brought back to
RAM but still sits in swap too. This is done to save I/O once there's
successive swap out that follows.


-- 
regards,

Mulyadi Santosa
Freelance Linux trainer and consultant

blog: the-hydra.blogspot.com
training: mulyaditraining.blogspot.com