On Thu, May 19, 2011 at 6:53 PM, Juri Lelli wrote: > Explain what the trailing "/1" on some lock class names of lock_stat output > means. > > Cc: Yong Zhang > Signed-off-by: Juri Lelli > Looks good. Reviewed-by: Yong Zhang > --- >  Documentation/lockstat.txt |   36 ++++++++++++++++++++++++++++++++++-- >  1 files changed, 34 insertions(+), 2 deletions(-) > > diff --git a/Documentation/lockstat.txt b/Documentation/lockstat.txt > index 65f4c79..75eeb65 100644 > --- a/Documentation/lockstat.txt > +++ b/Documentation/lockstat.txt > @@ -12,8 +12,9 @@ Because things like lock contention can severely impact performance. >  - HOW > >  Lockdep already has hooks in the lock functions and maps lock instances to > -lock classes. We build on that. The graph below shows the relation between > -the lock functions and the various hooks therein. > +lock classes. We build on that (see Documentation/lockdep-design.txt). > +The graph below shows the relation between the lock functions and the various > +hooks therein. > >         __acquire >             | > @@ -128,6 +129,37 @@ points are the points we're contending with. > >  The integer part of the time values is in us. > > +Dealing with nested locks, subclasses may appear: > + > +32............................................................................................................................................................................................... > +33 > +34                               &rq->lock:         13128          13128           0.43         190.53      103881.26          97454        3453404           0.00         401.11    13224683.11 > +35                               --------- > +36                               &rq->lock            645          [] task_rq_lock+0x43/0x75 > +37                               &rq->lock            297          [] try_to_wake_up+0x127/0x25a > +38                               &rq->lock            360          [] select_task_rq_fair+0x1f0/0x74a > +39                               &rq->lock            428          [] scheduler_tick+0x46/0x1fb > +40                               --------- > +41                               &rq->lock             77          [] task_rq_lock+0x43/0x75 > +42                               &rq->lock            174          [] try_to_wake_up+0x127/0x25a > +43                               &rq->lock           4715          [] double_rq_lock+0x42/0x54 > +44                               &rq->lock            893          [] schedule+0x157/0x7b8 > +45 > +46............................................................................................................................................................................................... > +47 > +48                             &rq->lock/1:         11526          11488           0.33         388.73      136294.31          21461          38404           0.00          37.93      109388.53 > +49                             ----------- > +50                             &rq->lock/1          11526          [] double_rq_lock+0x4f/0x54 > +51                             ----------- > +52                             &rq->lock/1           5645          [] double_rq_lock+0x42/0x54 > +53                             &rq->lock/1           1224          [] schedule+0x157/0x7b8 > +54                             &rq->lock/1           4336          [] double_rq_lock+0x4f/0x54 > +55                             &rq->lock/1            181          [] try_to_wake_up+0x127/0x25a > + > +Line 48 shows statistics for the second subclass (/1) of &rq->lock class > +(subclass starts from 0), since in this case, as line 50 suggests, > +double_rq_lock actually acquires a nested lock of two spinlocks. > + >  View the top contending locks: > >  # grep : /proc/lock_stat | head > -- > 1.7.4.1 > -- Only stand for myself ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ¥Šwÿº{.nÇ+‰·¥Š{±þG«�éÿŠ{ayºʇڙë,j­¢f£¢·hš�ï�êÿ‘êçz_è®(­階Šݢj"�ú¶m§ÿÿ¾«þG«�éÿ¢¸?™¨è­Ú&£ø§~�á¶iO•æ¬z·švØ^¶m§ÿÿà ÿ¶ìÿ¢¸?–I¥