aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/iostats.txt
diff options
context:
space:
mode:
Diffstat (limited to 'Documentation/iostats.txt')
-rw-r--r--Documentation/iostats.txt17
1 files changed, 8 insertions, 9 deletions
diff --git a/Documentation/iostats.txt b/Documentation/iostats.txt
index f6dece5b7014..c76c21d87e85 100644
--- a/Documentation/iostats.txt
+++ b/Documentation/iostats.txt
@@ -1,8 +1,6 @@
1I/O statistics fields 1I/O statistics fields
2--------------- 2---------------
3 3
4Last modified Sep 30, 2003
5
6Since 2.4.20 (and some versions before, with patches), and 2.5.45, 4Since 2.4.20 (and some versions before, with patches), and 2.5.45,
7more extensive disk statistics have been introduced to help measure disk 5more extensive disk statistics have been introduced to help measure disk
8activity. Tools such as sar and iostat typically interpret these and do 6activity. Tools such as sar and iostat typically interpret these and do
@@ -46,11 +44,12 @@ the above example, the first field of statistics would be 446216.
46By contrast, in 2.6 if you look at /sys/block/hda/stat, you'll 44By contrast, in 2.6 if you look at /sys/block/hda/stat, you'll
47find just the eleven fields, beginning with 446216. If you look at 45find just the eleven fields, beginning with 446216. If you look at
48/proc/diskstats, the eleven fields will be preceded by the major and 46/proc/diskstats, the eleven fields will be preceded by the major and
49minor device numbers, and device name. Each of these formats provide 47minor device numbers, and device name. Each of these formats provides
50eleven fields of statistics, each meaning exactly the same things. 48eleven fields of statistics, each meaning exactly the same things.
51All fields except field 9 are cumulative since boot. Field 9 should 49All fields except field 9 are cumulative since boot. Field 9 should
52go to zero as I/Os complete; all others only increase. Yes, these are 50go to zero as I/Os complete; all others only increase (unless they
5332 bit unsigned numbers, and on a very busy or long-lived system they 51overflow and wrap). Yes, these are (32-bit or 64-bit) unsigned long
52(native word size) numbers, and on a very busy or long-lived system they
54may wrap. Applications should be prepared to deal with that; unless 53may wrap. Applications should be prepared to deal with that; unless
55your observations are measured in large numbers of minutes or hours, 54your observations are measured in large numbers of minutes or hours,
56they should not wrap twice before you notice them. 55they should not wrap twice before you notice them.
@@ -96,11 +95,11 @@ introduced when changes collide, so (for instance) adding up all the
96read I/Os issued per partition should equal those made to the disks ... 95read I/Os issued per partition should equal those made to the disks ...
97but due to the lack of locking it may only be very close. 96but due to the lack of locking it may only be very close.
98 97
99In 2.6, there are counters for each cpu, which made the lack of locking 98In 2.6, there are counters for each CPU, which make the lack of locking
100almost a non-issue. When the statistics are read, the per-cpu counters 99almost a non-issue. When the statistics are read, the per-CPU counters
101are summed (possibly overflowing the unsigned 32-bit variable they are 100are summed (possibly overflowing the unsigned long variable they are
102summed to) and the result given to the user. There is no convenient 101summed to) and the result given to the user. There is no convenient
103user interface for accessing the per-cpu counters themselves. 102user interface for accessing the per-CPU counters themselves.
104 103
105Disks vs Partitions 104Disks vs Partitions
106------------------- 105-------------------