rcu: Remove redundant offline check
authorPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tue, 10 Mar 2015 00:00:56 +0000 (17:00 -0700)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Wed, 27 May 2015 19:58:38 +0000 (12:58 -0700)
Because offline CPUs are propagated up the rcu_node tree's ->qsmaskinit
bits just before each grace period starts, the ->qsmaskinit bit cannot
be clear when the corresponding ->qsmask bit is set.  Furthermore, this
condition used to correspond to a CPU that was on its way offline, and
making RCU's notion of an offline CPU more precise has eliminated this
situation.  This commit therefore removes the now-redundant offline
check from force_qs_rnp().

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
kernel/rcu/tree.c

index e338a12c3a1bac3d0ccf6acb2636975d8ec9786c..a1df68fce545d9f2d82cdfb1bbf1a2c72b27ccb0 100644 (file)
@@ -2749,8 +2749,6 @@ static void force_qs_rnp(struct rcu_state *rsp,
                bit = 1;
                for (; cpu <= rnp->grphi; cpu++, bit <<= 1) {
                        if ((rnp->qsmask & bit) != 0) {
-                               if ((rnp->qsmaskinit & bit) == 0)
-                                       *isidle = false; /* Pending hotplug. */
                                if (f(per_cpu_ptr(rsp->rda, cpu), isidle, maxj))
                                        mask |= bit;
                        }