[PATCH] sh: Don't enable interrupts in the page fault path

[PATCH] sh: Don't enable interrupts in the page fault path

[Matt Fleming]

There's already code in do_page_fault() to conditionally enable
interrupts, so we don't need to unconditonally enable them before
calling it. This fixes a lockdep warning where we called
trace_hardirqs_off() but with irqs still enabled.

Signed-off-by: Matt Fleming <matt@console-pimps.org>
---
 arch/sh/kernel/cpu/sh3/entry.S |    2 --
 1 files changed, 0 insertions(+), 2 deletions(-)

diff --git a/arch/sh/kernel/cpu/sh3/entry.S b/arch/sh/kernel/cpu/sh3/entry.S
index 0151933..bb407ef 100644
--- a/arch/sh/kernel/cpu/sh3/entry.S
+++ b/arch/sh/kernel/cpu/sh3/entry.S
@@ -152,8 +152,6 @@ call_do_page_fault:
 	mov.l	1f, r0
 	mov.l	@r0, r6
 
-	sti
-
 	mov.l	3f, r0
 	mov.l	4f, r1
 	mov	r15, r4
-- 
1.6.3.3

Re: [PATCH] sh: Don't enable interrupts in the page fault path

[Paul Mundt]

On Sat, Sep 26, 2009 at 08:44:05PM +0100, Matt Fleming wrote:
> There's already code in do_page_fault() to conditionally enable
> interrupts, so we don't need to unconditonally enable them before
> calling it. This fixes a lockdep warning where we called
> trace_hardirqs_off() but with irqs still enabled.
> 
> Signed-off-by: Matt Fleming <matt@console-pimps.org>

This is a relic from the days where do_page_fault() did not do the
conditional re-enabling. Applied, thanks.