lkml.org 
[lkml]   [2008]   [Apr]   [28]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
DateMon, 28 Apr 2008 14:18:36 -0400
FromRik van Riel <>
Subject[PATCH -mm 01/15] FYI: vmstats are "off-by-one"
PATCH - add vmstat_test for NR_WRITEBACK_TEMP

Against:  2.6.25-rc8-mm2 [also missing in '.25-mm1]

/proc/vmstat output after "nr_vmscan_write" are off-by-one
because of missing test for NR_WRITEBACK_TEMP added to 
mmzone.h.  Apparently missing from:

	mm-add-nr_writeback_temp-counter.patch

in -mm tree.

Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>

---
Andrew,
you may already have this patch.  I am posting it in this
series to help other testers of the code.

Rik


Rik, Kosaki-san:

Here's a patch that I'm going to post to Andrew and the mailing lists.
I noticed that the noreclaim events were off by one.  Missing
vmstat_text in 25-rc8-mm* and 25-mm1.

I have this in my tree as a 25-*mm* "hotfix"--i.e., before any of our
patches.  It causes a couple of 1 one offsets but no patch conflicts.

Lee


 mm/vmstat.c |    1 +
 1 file changed, 1 insertion(+)
Index: linux-2.6.25-rc8-mm2/mm/vmstat.c
===================================================================
--- linux-2.6.25-rc8-mm2.orig/mm/vmstat.c	2008-04-15 12:07:18.000000000 -0400
+++ linux-2.6.25-rc8-mm2/mm/vmstat.c	2008-04-15 12:11:48.000000000 -0400
@@ -699,6 +699,7 @@ static const char * const vmstat_text[] 
 	"nr_unstable",
 	"nr_bounce",
 	"nr_vmscan_write",
+	"nr_writeback_temp",
 
 #ifdef CONFIG_NUMA
 	"numa_hit",
-- 
All Rights Reversed



\
 
 \ /
  Last update: 2008-04-28 20:27    [from the cache]
©2003-2008