> --- vmscan.c 1998/11/18 07:17:02 1.56
> +++ vmscan.c 1998/11/21 09:12:37
> @@ -570,6 +583,10 @@
> if (!retval)
> break;
> count--;
> + if (!(gfp_mask & __GFP_DMA) &&
> + (nr_free_pages > freepages.high + SWAP_CLUSTER_MAX
> + || atomic_read(&nr_async_pages) > SWAP_CLUSTER_MAX))
I'd really appreciate it if SWAP_CLUSTER_MAX would be replaced
with pager_daemon.swap_cluster. I have observed quite an increase
in performance when I set pager_daemon.swap_cluster to
SWAP_CLUSTER_MAX * nr_swap_partitions... There's a reason for
a sysctl tuneable kswapd :)
> + break;
> } while (count > 0);
> current-> flags &= ~PF_MEMALLOC;
> }
Apart from that little detail (that also should be fixed
in page_io.c) I think that this patch should go in.
Linus, if you don't like the second part of this test, we
can also replace the break with a run_task_queue(&tq_disk)...
cheers,
Rik -- slowly getting used to dvorak kbd layout...
+-------------------------------------------------------------------+
| Linux memory management tour guide. H.H.vanRiel@phys.uu.nl |
| Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ |
+-------------------------------------------------------------------+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/