[patch, -rc5-mm1] lock validator: irqflags-trace entry.S fix

[Ingo Molnar <mingo@elte.hu>]

* Michal Piotrowski <michal.k.k.piotrowski@gmail.com> wrote:

> >here's the fix for the irqs-off iret irqflags-tracing problem. Does this
> >fix the bug(s) on your box?
> 
> Yes. Thanks!

great! Here's the cleaned up fix for Andrew:

----------------------------
Subject: lock validator: irqflags-trace entry.S fix
From: Ingo Molnar <mingo@elte.hu>

this fixes the irqflags-tracing bug reported (and relentlessly
debugged) by Michal Piotrowski: if we took a fault while interrupts
were disabled (for example of a vmalloc area) then irqflags-tracing
mistakenly assumed that the iret would re-enable interrupts.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
---
 arch/i386/kernel/entry.S |   13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

Index: linux/arch/i386/kernel/entry.S
===================================================================
--- linux.orig/arch/i386/kernel/entry.S
+++ linux/arch/i386/kernel/entry.S
@@ -84,6 +84,15 @@ VM_MASK		= 0x00020000
 #define resume_kernel		restore_nocheck
 #endif
 
+.macro TRACE_IRQS_IRET
+#ifdef CONFIG_TRACE_IRQFLAGS
+	testl $IF_MASK,EFLAGS(%esp)     # interrupts off?
+	jz 1f
+	TRACE_IRQS_ON
+1:
+#endif
+.endm
+
 #ifdef CONFIG_VM86
 #define resume_userspace_sig	check_userspace
 #else
@@ -364,7 +373,7 @@ restore_all:
 	CFI_REMEMBER_STATE
 	je ldt_ss			# returning to user-space with LDT SS
 restore_nocheck:
-	TRACE_IRQS_ON
+	TRACE_IRQS_IRET
 restore_nocheck_notrace:
 	RESTORE_REGS
 	addl $4, %esp
@@ -404,7 +413,7 @@ ldt_ss:
 	 * and a switch16 pointer on top of the current frame. */
 	call setup_x86_bogus_stack
 	CFI_ADJUST_CFA_OFFSET -8	# frame has moved
-	TRACE_IRQS_ON
+	TRACE_IRQS_IRET
 	RESTORE_REGS
 	lss 20+4(%esp), %esp	# switch to 16bit stack
 1:	iret