]> git.openfabrics.org - ~shefty/rdma-dev.git/commitdiff
[S390] cpu hotplug/idle: move cpu_die call to enabled context
authorHeiko Carstens <heiko.carstens@de.ibm.com>
Mon, 25 Oct 2010 14:10:12 +0000 (16:10 +0200)
committerMartin Schwidefsky <sky@mschwide.boeblingen.de.ibm.com>
Mon, 25 Oct 2010 14:10:15 +0000 (16:10 +0200)
There is no difference if cpu_die is called from enabled or disabled
context. Except that the fast_gup code might be called via
cpu_die -> idle_task_exit -> __mm_drop -> crst_table_free. Which in
turn grabs and releases a spinlock using the _bh ops, which is not
allowed in irq disabled context, since spin_unlock_bh will
unconditionally enable interrupts again.
To get rid of the warning emitted by the softirq code just move the
code to enabled context.

In this case this doesn't fix a bug, we just get rid of a warning.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
arch/s390/kernel/process.c

index d3a2d1c6438ee08c1216353c8fefec7f2ee0dc31..ec2e03b22ead24c57210d09df9c514c9dcbe5816 100644 (file)
@@ -76,17 +76,17 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 static void default_idle(void)
 {
        /* CPU is going idle. */
-       local_irq_disable();
-       if (need_resched()) {
-               local_irq_enable();
-               return;
-       }
 #ifdef CONFIG_HOTPLUG_CPU
        if (cpu_is_offline(smp_processor_id())) {
                preempt_enable_no_resched();
                cpu_die();
        }
 #endif
+       local_irq_disable();
+       if (need_resched()) {
+               local_irq_enable();
+               return;
+       }
        local_mcck_disable();
        if (test_thread_flag(TIF_MCCK_PENDING)) {
                local_mcck_enable();