Re: [PATCH] 2.4.1 find_page_nolock fixes

[Marcelo Tosatti]

On Tue, 30 Jan 2001, Rik van Riel wrote:

> Hi Linus,
> 
> the patch below contains 3 small changes to mm/filemap.c:
> 
> 1. replace the aging in __find_page_nolock() with setting
>    PageReferenced(), otherwise a large number of small
>    reads from (or writes to) a page can drive up the page
>    age unfairly
> 
> 2. remove the wakeup of kswapd from __find_page_nolock(),
>    the wakeup condition is complex and leaving out the
>    wakeup has no influence on any workload I tested in
>    the last few weeks
> 
> 3. add a __find_page_simple(), which is like __find_page_nolock()
>    but only needs 2 arguments and doesn't touch the page ... this
>    can be used by IO clustering and other things that really don't
>    want to influence page aging, removing the 3rd argument also
>    keeps things simple

The swapin readahead still makes the page referenced bit set, and it
should not as we discussed previously.