On Fri, 5 May 2000, Shane Shrybman wrote:
>I still see small but consistent swapouts during I/O load. Is this
>correct behaviour? [..]
We can reduce it. Probably something like this could help:
--- 2.3.99-pre7-pre4-VM/mm/vmscan.c Fri May 5 17:06:02 2000
+++ /tmp/vmscan.c Fri May 5 18:58:58 2000
@@ -429,7 +429,7 @@
/* Always trim SLAB caches when memory gets low. */
kmem_cache_reap(gfp_mask);
- priority = 6;
+ priority = 4;
do {
while (shrink_mmap(priority, zone)) {
if (!--count)
>[..] And are the VM tuning params in /proc functional
>again?
There's a lru_cache sysctl that have three params. Only the first param is
active and it endforces a minium of pages that will be kept in the cache
even if freeable. buffermem and pagecache sysclt are gone. I didn't
changed other tuning interfaces.
Andrea
-
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/
This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:18 EST