[lkml]   [1998]   [Nov]   [30]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [2.1.130-3] Page cache DEFINATELY too persistant... feature?
On 30 Nov 1998, Zlatko Calusic wrote:
> Rik van Riel <> writes:

> > that (or abolish the percentages completely) kswapd
> > doesn't have an incentive to switch from a succesful
> > round of swap_out() -- which btw doesn't free any
> > actual memory so kswapd just continues doing that --
> > to shrink_mmap().
> Yep, this is the conclusion of my experiments, too.

> I made the following change in do_try_to_free_page():


> Unfortunately, this really killed swapout performance, so I dropped
> the idea. Even letting swap_out do more passes, before changing state,
> didn't feel good.
> One other idea I had, was to replace (code at the very beginning of
> do_try_to_free_page()):
> if (buffer_over_borrow() || pgcache_over_borrow())
> shrink_mmap(i, gfp_mask);
> with:
> if (buffer_over_borrow() || pgcache_over_borrow())
> state = 0;

I am now trying:
if (buffer_over_borrow() || pgcache_over_borrow() ||
shrink_mmap(i, gfp_mask);

Note that this doesn't stop kswapd from swapping out so
swapout performance shouldn't suffer. It does however
free up memory so kswapd should _terminate_ and keep the
amount of I/O done to a sane level.

Note that I'm running with my experimentas swapin readahead
patch enabled so the system should be stressed even more
than normally :)


Rik -- now completely used to dvorak kbd layout...
| Linux memory management tour guide. |
| Scouting Vries cubscout leader. |

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:45    [W:0.054 / U:0.248 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site