On Wed, Nov 04, 2020 at 02:06:23PM +0100, Thomas Gleixner wrote:
> When an exception/interrupt hits kernel space and the kernel is not
> currently in the idle task then RCU must be watching.
>
> irqentry_enter() validates this via rcu_irq_enter_check_tick(), which in
> turn invokes lockdep when taking a lock. But at that point lockdep does not
> yet know about the fact that interrupts have been disabled by the CPU,
> which triggers a lockdep splat complaining about inconsistent state.
>
> Invoking trace_hardirqs_off() before rcu_irq_enter_check_tick() defeats the
> point of rcu_irq_enter_check_tick() because trace_hardirqs_off() uses RCU.
>
> So use the same sequence as for the idle case and tell lockdep about the
> irq state change first, invoke the RCU check and then do the lockdep and
> tracer update.
>
> Fixes: a5497bab5f72 ("entry: Provide generic interrupt entry/exit code")
> Reported-by: Mark Rutland <mark.rutl...@arm.com>
> Signed-off-by: Thomas Gleixner <t...@linutronix.de>
> Cc: sta...@vger.kernel.org
I just gave this a spin on x86_64 defconfig + PROVE_LOCKING +
DEBUG_LOCKDEP + NO_HZ_FULL + CONTEXT_TRACKING_FORCE, and it gets rid of
the boot-time splat. So FWIW:
Tested-by: Mark Rutland <mark.rutl...@arm.com>
Mark.
> ---
> kernel/entry/common.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> --- a/kernel/entry/common.c
> +++ b/kernel/entry/common.c
> @@ -337,10 +337,10 @@ noinstr irqentry_state_t irqentry_enter(
> * already contains a warning when RCU is not watching, so no point
> * in having another one here.
> */
> + lockdep_hardirqs_off(CALLER_ADDR0);
> instrumentation_begin();
> rcu_irq_enter_check_tick();
> - /* Use the combo lockdep/tracing function */
> - trace_hardirqs_off();
> + trace_hardirqs_off_finish();
> instrumentation_end();
>
> return ret;
