Commit a5d6d3a1 authored by Eric Dumazet's avatar Eric Dumazet Committed by Paul E. McKenney

softirq: A single rcu_bh_qs() call per softirq set is enough

Calling rcu_bh_qs() after every softirq action is not really needed.
What RCU needs is at least one rcu_bh_qs() per softirq round to note a
quiescent state was passed for rcu_bh.

Note for Paul and myself : this could be inlined as a single instruction
and avoid smp_processor_id()
(sone this_cpu_write(rcu_bh_data.passed_quiesce, 1))
Signed-off-by: default avatarEric Dumazet <edumazet@google.com>
Signed-off-by: default avatarPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: default avatarJosh Triplett <josh@joshtriplett.org>
parent fa07a58f
...@@ -232,7 +232,6 @@ asmlinkage void __do_softirq(void) ...@@ -232,7 +232,6 @@ asmlinkage void __do_softirq(void)
bool in_hardirq; bool in_hardirq;
__u32 pending; __u32 pending;
int softirq_bit; int softirq_bit;
int cpu;
/* /*
* Mask out PF_MEMALLOC s current task context is borrowed for the * Mask out PF_MEMALLOC s current task context is borrowed for the
...@@ -247,7 +246,6 @@ asmlinkage void __do_softirq(void) ...@@ -247,7 +246,6 @@ asmlinkage void __do_softirq(void)
__local_bh_disable_ip(_RET_IP_, SOFTIRQ_OFFSET); __local_bh_disable_ip(_RET_IP_, SOFTIRQ_OFFSET);
in_hardirq = lockdep_softirq_start(); in_hardirq = lockdep_softirq_start();
cpu = smp_processor_id();
restart: restart:
/* Reset the pending bitmask before enabling irqs */ /* Reset the pending bitmask before enabling irqs */
set_softirq_pending(0); set_softirq_pending(0);
...@@ -276,11 +274,11 @@ asmlinkage void __do_softirq(void) ...@@ -276,11 +274,11 @@ asmlinkage void __do_softirq(void)
prev_count, preempt_count()); prev_count, preempt_count());
preempt_count_set(prev_count); preempt_count_set(prev_count);
} }
rcu_bh_qs(cpu);
h++; h++;
pending >>= softirq_bit; pending >>= softirq_bit;
} }
rcu_bh_qs(smp_processor_id());
local_irq_disable(); local_irq_disable();
pending = local_softirq_pending(); pending = local_softirq_pending();
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment