Re: [PATCH] shows Active/Inactive on per-node meminfo

[mita akinobu]

On Saturday 21 August 2004 03:48, Jesse Barnes wrote:
> Just FYI, loops like this are going to be very slow on a large machine.
> Iterating over every node in the system involves a TLB miss on every
> iteration along with an offnode reference and possibly cacheline demotion.

get_zone_counts() is used by max_sane_readahead(), and
max_sane_readahead() is often called in filemap_nopage().

If iterating over every node is going to be very slow, the following change
would have a little bit of improvement on a large machine?


--- linux-2.6.8.1-mm3/mm/readahead.c.orig	2004-08-21 15:18:08.924273720 +0900
+++ linux-2.6.8.1-mm3/mm/readahead.c	2004-08-21 15:22:31.123413392 +0900
@@ -572,6 +572,6 @@ unsigned long max_sane_readahead(unsigne
 	unsigned long inactive;
 	unsigned long free;
 
-	get_zone_counts(&active, &inactive, &free);
+	__get_zone_counts(&active, &inactive, &free, NODE_DATA(numa_node_id()));
 	return min(nr, (inactive + free) / 2);
 }



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/