Re: 2.6.23-git8: Lock dependency engine debugging failure

On Mon, 2007-10-15 at 20:28 -0700, David Hubbard wrote:
I am not subscribed to LKML, so please CC me in replies. I am
reporting a regression when CONFIG_DEBUG_LOCKDEP is enabled in
2.6.23-git8. The error occurs immediately before loading init.
Complete dmesg and kernel config are attached.

[ 28.528074] VFS: Mounted root (ext3 filesystem) readonly.
[ 28.528090] Freeing unused kernel memory: 212k freed
[ 28.537431] Write protecting the kernel read-only data: 1036k
[ 28.622874] WARNING: at kernel/lockdep.c:2658 check_flags()

Could you try if the below patch works for you?

--
Subject: lockdep: fix fault vs irq tracing

Ensure we fixup the IRQ state before we hit any locking code.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxx>
---
arch/x86/mm/fault_32.c | 10 ++++++++++
arch/x86/mm/fault_64.c | 10 ++++++++++
2 files changed, 20 insertions(+)

Index: linux-2.6/arch/x86/mm/fault_32.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/fault_32.c
+++ linux-2.6/arch/x86/mm/fault_32.c
@@ -308,6 +308,16 @@ fastcall void __kprobes do_page_fault(st
int write, si_code;
int fault;

+#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
+ /*
+ * We can fault from pretty much anywhere, fix up the IRQ state.
+ */
+ if (raw_irqs_disabled())
+ trace_hardirqs_off();
+ else
+ trace_hardirqs_on();
+#endif
+
/* get the address */
address = read_cr2();

Index: linux-2.6/arch/x86/mm/fault_64.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/fault_64.c
+++ linux-2.6/arch/x86/mm/fault_64.c
@@ -311,6 +311,16 @@ asmlinkage void __kprobes do_page_fault(
unsigned long flags;
siginfo_t info;

+#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
+ /*
+ * We can fault from pretty much anywhere, fix up the IRQ state.
+ */
+ if (raw_irqs_disabled())
+ trace_hardirqs_off();
+ else
+ trace_hardirqs_on();
+#endif
+
tsk = current;
mm = tsk->mm;
prefetchw(&mm->mmap_sem);


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages