[PATCH] powerpc: sched fixups

- Re-add a hunk lost during merge: ppc64 is missing the hunk that disables
  preempt on the secondary CPUs before they call cpu_idle().

- ppc's cpu_idle() had the need_resched() test wrong.

Cc: Nick Piggin <nickpiggin@yahoo.com.au>
CC: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 5c330c3..36d67a8 100644 (file)
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -510,6 +510,7 @@ int __devinit start_secondary(void *unused)
 
        smp_store_cpu_info(cpu);
        set_dec(tb_ticks_per_jiffy);
+       preempt_disable();
        cpu_callin_map[cpu] = 1;
 
        smp_ops->setup_cpu(cpu);

diff --git a/arch/ppc/kernel/idle.c b/arch/ppc/kernel/idle.c
index 3c4e4cb..821a75e 100644 (file)
--- a/arch/ppc/kernel/idle.c
+++ b/arch/ppc/kernel/idle.c
@@ -63,7 +63,7 @@ void cpu_idle(void)
        int cpu = smp_processor_id();
 
        for (;;) {
-               while (need_resched()) {
+               while (!need_resched()) {
                        if (ppc_md.idle != NULL)
                                ppc_md.idle();
                        else