ARM: unwind: set frame.pc correctly for current-thread unwinding

When e.g. a WARN_ON() is encountered, we attempt to unwind the current
thread. To do this, we set frame.pc to unwind_backtrace, which means it
points at the beginning of the function. However, the rest of the state
is initialised from within the function, which means the function
prologue has already been run.

This can be confusing, and with a recent patch from Ard, can result in
the unwinder misbehaving if we want to be strict about the PC value.

If we correctly initialise the state so it is self-consistent (in other
words, set frame.pc to the location we are initialising it) then we
eliminate this confusion, and avoid possible future issues.

Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>

1 files changed, 6 insertions, 1 deletions

diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c
index 04ccff9d9793..3cd8892ed52b 100644
--- a/arch/arm/kernel/unwind.c
+++ b/arch/arm/kernel/unwind.c
@@ -501,7 +501,12 @@ void unwind_backtrace(struct pt_regs *regs, struct task_struct *tsk,
 		frame.fp = (unsigned long)__builtin_frame_address(0);
 		frame.sp = current_stack_pointer;
 		frame.lr = (unsigned long)__builtin_return_address(0);
-		frame.pc = (unsigned long)unwind_backtrace;
+		/* We are saving the stack and execution state at this
+		 * point, so we should ensure that frame.pc is within
+		 * this block of code.
+		 */
+here:
+		frame.pc = (unsigned long)&&here;
 	} else {
 		/* task blocked in __switch_to */
 		frame.fp = thread_saved_fp(tsk);