rcu: Eliminate __rcu_pending() false positives
authorPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Sat, 14 Nov 2009 03:51:39 +0000 (19:51 -0800)
committerIngo Molnar <mingo@elte.hu>
Sat, 14 Nov 2009 09:31:42 +0000 (10:31 +0100)
commit2f51f9884f6a36b0fe9636d5a1937e5cbd25723b
tree0505e111a9714b2bd022322f76e6c4de26861107
parent560d4bc0df9a5e63b980432282d8c2bd3559ec74
rcu: Eliminate __rcu_pending() false positives

Now that there are both ->gpnum and ->completed fields in the
rcu_node structure, __rcu_pending() should check rdp->gpnum and
rdp->completed against rnp->gpnum and rdp->completed, respectively,
instead of the prior comparison against the rcu_state fields
rsp->gpnum and rsp->completed.

Given the old comparison, __rcu_pending() could return 1, resulting
in a needless raise_softirq(RCU_SOFTIRQ).  This useless work would
happen if RCU responded to a scheduling-clock interrupt after the
rcu_state fields had been updated, but before the rcu_node fields
had been updated.

Changing the comparison from the rcu_state fields to the rcu_node
fields prevents this useless work from happening.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: mathieu.desnoyers@polymtl.ca
Cc: josh@joshtriplett.org
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
Cc: Valdis.Kletnieks@vt.edu
Cc: dhowells@redhat.com
LKML-Reference: <12581706991966-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/rcutree.c