blob: 4919b2dfd1b391c4169de1b66f5e78bd9ba2d9cc [file] [log] [blame]
Mikulas Patockafd2ed4d2013-08-16 10:54:23 -04001DM statistics
2=============
3
4Device Mapper supports the collection of I/O statistics on user-defined
5regions of a DM device. If no regions are defined no statistics are
6collected so there isn't any performance impact. Only bio-based DM
7devices are currently supported.
8
9Each user-defined region specifies a starting sector, length and step.
10Individual statistics will be collected for each step-sized area within
11the range specified.
12
13The I/O statistics counters for each step-sized area of a region are
14in the same format as /sys/block/*/stat or /proc/diskstats (see:
15Documentation/iostats.txt). But two extra counters (12 and 13) are
Mikulas Patockadfcfac32015-06-09 17:22:05 -040016provided: total time spent reading and writing. When the histogram
17argument is used, the 14th parameter is reported that represents the
18histogram of latencies. All these counters may be accessed by sending
19the @stats_print message to the appropriate DM device via dmsetup.
Mikulas Patockac96aec32015-06-09 17:21:39 -040020
21The reported times are in milliseconds and the granularity depends on
22the kernel ticks. When the option precise_timestamps is used, the
23reported times are in nanoseconds.
Mikulas Patockafd2ed4d2013-08-16 10:54:23 -040024
25Each region has a corresponding unique identifier, which we call a
26region_id, that is assigned when the region is created. The region_id
27must be supplied when querying statistics about the region, deleting the
28region, etc. Unique region_ids enable multiple userspace programs to
29request and process statistics for the same DM device without stepping
30on each other's data.
31
32The creation of DM statistics will allocate memory via kmalloc or
33fallback to using vmalloc space. At most, 1/4 of the overall system
34memory may be allocated by DM statistics. The admin can see how much
35memory is used by reading
36/sys/module/dm_mod/parameters/stats_current_allocated_bytes
37
38Messages
39========
40
Mikulas Patockac96aec32015-06-09 17:21:39 -040041 @stats_create <range> <step>
42 [<number_of_optional_arguments> <optional_arguments>...]
43 [<program_id> [<aux_data>]]
Mikulas Patockafd2ed4d2013-08-16 10:54:23 -040044
45 Create a new region and return the region_id.
46
47 <range>
48 "-" - whole device
49 "<start_sector>+<length>" - a range of <length> 512-byte sectors
50 starting with <start_sector>.
51
52 <step>
53 "<area_size>" - the range is subdivided into areas each containing
54 <area_size> sectors.
55 "/<number_of_areas>" - the range is subdivided into the specified
56 number of areas.
57
Mikulas Patockac96aec32015-06-09 17:21:39 -040058 <number_of_optional_arguments>
59 The number of optional arguments
60
61 <optional_arguments>
62 The following optional arguments are supported
63 precise_timestamps - use precise timer with nanosecond resolution
64 instead of the "jiffies" variable. When this argument is
65 used, the resulting times are in nanoseconds instead of
66 milliseconds. Precise timestamps are a little bit slower
67 to obtain than jiffies-based timestamps.
Mikulas Patockadfcfac32015-06-09 17:22:05 -040068 histogram:n1,n2,n3,n4,... - collect histogram of latencies. The
69 numbers n1, n2, etc are times that represent the boundaries
70 of the histogram. If precise_timestamps is not used, the
71 times are in milliseconds, otherwise they are in
72 nanoseconds. For each range, the kernel will report the
73 number of requests that completed within this range. For
74 example, if we use "histogram:10,20,30", the kernel will
75 report four numbers a:b:c:d. a is the number of requests
76 that took 0-10 ms to complete, b is the number of requests
77 that took 10-20 ms to complete, c is the number of requests
78 that took 20-30 ms to complete and d is the number of
79 requests that took more than 30 ms to complete.
Mikulas Patockac96aec32015-06-09 17:21:39 -040080
Mikulas Patockafd2ed4d2013-08-16 10:54:23 -040081 <program_id>
82 An optional parameter. A name that uniquely identifies
83 the userspace owner of the range. This groups ranges together
84 so that userspace programs can identify the ranges they
85 created and ignore those created by others.
86 The kernel returns this string back in the output of
87 @stats_list message, but it doesn't use it for anything else.
Mikulas Patockac96aec32015-06-09 17:21:39 -040088 If we omit the number of optional arguments, program id must not
89 be a number, otherwise it would be interpreted as the number of
90 optional arguments.
Mikulas Patockafd2ed4d2013-08-16 10:54:23 -040091
92 <aux_data>
93 An optional parameter. A word that provides auxiliary data
94 that is useful to the client program that created the range.
95 The kernel returns this string back in the output of
96 @stats_list message, but it doesn't use this value for anything.
97
98 @stats_delete <region_id>
99
100 Delete the region with the specified id.
101
102 <region_id>
103 region_id returned from @stats_create
104
105 @stats_clear <region_id>
106
107 Clear all the counters except the in-flight i/o counters.
108
109 <region_id>
110 region_id returned from @stats_create
111
112 @stats_list [<program_id>]
113
114 List all regions registered with @stats_create.
115
116 <program_id>
117 An optional parameter.
118 If this parameter is specified, only matching regions
119 are returned.
120 If it is not specified, all regions are returned.
121
122 Output format:
123 <region_id>: <start_sector>+<length> <step> <program_id> <aux_data>
124
125 @stats_print <region_id> [<starting_line> <number_of_lines>]
126
127 Print counters for each step-sized area of a region.
128
129 <region_id>
130 region_id returned from @stats_create
131
132 <starting_line>
133 The index of the starting line in the output.
134 If omitted, all lines are returned.
135
136 <number_of_lines>
137 The number of lines to include in the output.
138 If omitted, all lines are returned.
139
140 Output format for each step-sized area of a region:
141
142 <start_sector>+<length> counters
143
144 The first 11 counters have the same meaning as
145 /sys/block/*/stat or /proc/diskstats.
146
147 Please refer to Documentation/iostats.txt for details.
148
149 1. the number of reads completed
150 2. the number of reads merged
151 3. the number of sectors read
152 4. the number of milliseconds spent reading
153 5. the number of writes completed
154 6. the number of writes merged
155 7. the number of sectors written
156 8. the number of milliseconds spent writing
157 9. the number of I/Os currently in progress
158 10. the number of milliseconds spent doing I/Os
159 11. the weighted number of milliseconds spent doing I/Os
160
161 Additional counters:
162 12. the total time spent reading in milliseconds
163 13. the total time spent writing in milliseconds
164
165 @stats_print_clear <region_id> [<starting_line> <number_of_lines>]
166
167 Atomically print and then clear all the counters except the
168 in-flight i/o counters. Useful when the client consuming the
169 statistics does not want to lose any statistics (those updated
170 between printing and clearing).
171
172 <region_id>
173 region_id returned from @stats_create
174
175 <starting_line>
176 The index of the starting line in the output.
177 If omitted, all lines are printed and then cleared.
178
179 <number_of_lines>
180 The number of lines to process.
181 If omitted, all lines are printed and then cleared.
182
183 @stats_set_aux <region_id> <aux_data>
184
185 Store auxiliary data aux_data for the specified region.
186
187 <region_id>
188 region_id returned from @stats_create
189
190 <aux_data>
191 The string that identifies data which is useful to the client
192 program that created the range. The kernel returns this
193 string back in the output of @stats_list message, but it
194 doesn't use this value for anything.
195
196Examples
197========
198
199Subdivide the DM device 'vol' into 100 pieces and start collecting
200statistics on them:
201
202 dmsetup message vol 0 @stats_create - /100
203
204Set the auxillary data string to "foo bar baz" (the escape for each
205space must also be escaped, otherwise the shell will consume them):
206
207 dmsetup message vol 0 @stats_set_aux 0 foo\\ bar\\ baz
208
209List the statistics:
210
211 dmsetup message vol 0 @stats_list
212
213Print the statistics:
214
215 dmsetup message vol 0 @stats_print 0
216
217Delete the statistics:
218
219 dmsetup message vol 0 @stats_delete 0