Commit f62508f68d04adefc4cf9b0177ba02c8818b3eec
Committed by
Ingo Molnar
1 parent
dc7acbb251
Documentation: Add statistics about nested locks
Explain what the trailing "/1" on some lock class names of lock_stat output means. Reviewed-by: Yong Zhang <yong.zhang0@gmail.com> Signed-off-by: Juri Lelli <juri.lelli@gmail.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/4DD4F6C1.5090701@gmail.com Signed-off-by: Ingo Molnar <mingo@elte.hu>
Showing 1 changed file with 34 additions and 2 deletions Side-by-side Diff
Documentation/lockstat.txt
... | ... | @@ -12,8 +12,9 @@ |
12 | 12 | - HOW |
13 | 13 | |
14 | 14 | Lockdep already has hooks in the lock functions and maps lock instances to |
15 | -lock classes. We build on that. The graph below shows the relation between | |
16 | -the lock functions and the various hooks therein. | |
15 | +lock classes. We build on that (see Documentation/lockdep-design.txt). | |
16 | +The graph below shows the relation between the lock functions and the various | |
17 | +hooks therein. | |
17 | 18 | |
18 | 19 | __acquire |
19 | 20 | | |
... | ... | @@ -127,6 +128,37 @@ |
127 | 128 | points are the points we're contending with. |
128 | 129 | |
129 | 130 | The integer part of the time values is in us. |
131 | + | |
132 | +Dealing with nested locks, subclasses may appear: | |
133 | + | |
134 | +32............................................................................................................................................................................................... | |
135 | +33 | |
136 | +34 &rq->lock: 13128 13128 0.43 190.53 103881.26 97454 3453404 0.00 401.11 13224683.11 | |
137 | +35 --------- | |
138 | +36 &rq->lock 645 [<ffffffff8103bfc4>] task_rq_lock+0x43/0x75 | |
139 | +37 &rq->lock 297 [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a | |
140 | +38 &rq->lock 360 [<ffffffff8103c4c5>] select_task_rq_fair+0x1f0/0x74a | |
141 | +39 &rq->lock 428 [<ffffffff81045f98>] scheduler_tick+0x46/0x1fb | |
142 | +40 --------- | |
143 | +41 &rq->lock 77 [<ffffffff8103bfc4>] task_rq_lock+0x43/0x75 | |
144 | +42 &rq->lock 174 [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a | |
145 | +43 &rq->lock 4715 [<ffffffff8103ed4b>] double_rq_lock+0x42/0x54 | |
146 | +44 &rq->lock 893 [<ffffffff81340524>] schedule+0x157/0x7b8 | |
147 | +45 | |
148 | +46............................................................................................................................................................................................... | |
149 | +47 | |
150 | +48 &rq->lock/1: 11526 11488 0.33 388.73 136294.31 21461 38404 0.00 37.93 109388.53 | |
151 | +49 ----------- | |
152 | +50 &rq->lock/1 11526 [<ffffffff8103ed58>] double_rq_lock+0x4f/0x54 | |
153 | +51 ----------- | |
154 | +52 &rq->lock/1 5645 [<ffffffff8103ed4b>] double_rq_lock+0x42/0x54 | |
155 | +53 &rq->lock/1 1224 [<ffffffff81340524>] schedule+0x157/0x7b8 | |
156 | +54 &rq->lock/1 4336 [<ffffffff8103ed58>] double_rq_lock+0x4f/0x54 | |
157 | +55 &rq->lock/1 181 [<ffffffff8104ba65>] try_to_wake_up+0x127/0x25a | |
158 | + | |
159 | +Line 48 shows statistics for the second subclass (/1) of &rq->lock class | |
160 | +(subclass starts from 0), since in this case, as line 50 suggests, | |
161 | +double_rq_lock actually acquires a nested lock of two spinlocks. | |
130 | 162 | |
131 | 163 | View the top contending locks: |
132 | 164 |