Revision 494de90098784b8e2797598cefdd34188884ec2e authored by Mel Gorman on 03 July 2008, 04:27:51 UTC, committed by Linus Torvalds on 03 July 2008, 16:22:59 UTC
The non-NUMA case of build_zonelist_cache() would initialize the zlcache_ptr for both node_zonelists[] to NULL. Which is problematic, since non-NUMA only has a single node_zonelists[] entry, and trying to zero the non-existent second one just overwrote the nr_zones field instead. As kswapd uses this value to determine what reclaim work is necessary, the result is that kswapd never reclaims. This causes processes to stall frequently in low-memory situations as they always direct reclaim. This patch initialises zlcache_ptr correctly. Signed-off-by: Mel Gorman <mel@csn.ul.ie> Tested-by: Dan Williams <dan.j.williams@intel.com> [ Simplified patch a bit ] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent c461a97
| File | Mode | Size |
|---|---|---|
| netfilter | ||
| Kconfig | -rw-r--r-- | 1.6 KB |
| Makefile | -rw-r--r-- | 270 bytes |
| README | -rw-r--r-- | 299 bytes |
| TODO | -rw-r--r-- | 1.2 KB |
| af_decnet.c | -rw-r--r-- | 53.4 KB |
| dn_dev.c | -rw-r--r-- | 33.0 KB |
| dn_fib.c | -rw-r--r-- | 18.0 KB |
| dn_neigh.c | -rw-r--r-- | 15.5 KB |
| dn_nsp_in.c | -rw-r--r-- | 21.4 KB |
| dn_nsp_out.c | -rw-r--r-- | 17.6 KB |
| dn_route.c | -rw-r--r-- | 43.7 KB |
| dn_rules.c | -rw-r--r-- | 5.6 KB |
| dn_table.c | -rw-r--r-- | 20.0 KB |
| dn_timer.c | -rw-r--r-- | 3.1 KB |
| sysctl_net_decnet.c | -rw-r--r-- | 9.8 KB |
Computing file changes ...
