Re: [PATCH] VM: Implements the swap-out page-clustering technique

[Hamid R. Jahanjou]

Rik van Riel wrote:
> On Sun, 07 Sep 2008 18:28:30 -0600
> Zan Lynx <zlynx@xxxxxxx> wrote:
>   
>> Rik van Riel wrote:
>> [cut]
>>     
>>> On the other hand, you do want to avoid evicting data 
>>> that the process is still using, just because you are 
>>> swapping out a not recently referenced page on a
>>> nearby virtual address.
>>>       
>> How about writing the nearby pages to swap anyway and mark the still 
>> in-use pages as SwapCache.
>>     
>
> That is what will happen pretty much automatically.
>   

I agree. As you said, we have two conflicting requirements here: on the
one hand one likes to swap as many pages as to satisfy the backing
storage "proper block IO size," on the other hand, one should not make a
process lose too many pages.  The code needs tuning really, I'm going to
rewrite it making it more flexible according to the suggestion given on
the mailing list.

> Another thing the swap out page clustering code could
> do is move pages that were recently referenced back
> to the active list, so the VM will not have to scan
> those pages again.
>
>   

I think this is a very good idea, actually the code does this already.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/