On Thu, Apr 13, 2017 at 05:25:56PM +0100, Tvrtko Ursulin wrote:
> From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
>
> Move the BUILD_BUG_ONs for busy-wait duration outside the
> _wait_for_atomic macro as discussed on the mailing list.
>
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
> Suggested-by: Michal Wajdeczko <michal.wajdeczko@xxxxxxxxx>
> Fixes: 1d1a9774e404 ("drm/i915: Extend intel_wait_for_register_fw() with fast timeout")
> Cc: Michal Wajdeczko <michal.wajdeczko@xxxxxxxxx>
> Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> ---
> drivers/gpu/drm/i915/intel_drv.h | 13 ++++++++++---
> 1 file changed, 10 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
> index f78d7c5f3805..4363852f8500 100644
> --- a/drivers/gpu/drm/i915/intel_drv.h
> +++ b/drivers/gpu/drm/i915/intel_drv.h
> @@ -88,7 +88,6 @@
> int cpu, ret, timeout = (US) * 1000; \
> u64 base; \
> _WAIT_FOR_ATOMIC_CHECK(ATOMIC); \
> - BUILD_BUG_ON((US) > 50000); \
> if (!(ATOMIC)) { \
> preempt_disable(); \
> cpu = smp_processor_id(); \
> @@ -130,8 +129,16 @@
> ret__; \
> })
>
> -#define wait_for_atomic(COND, MS) _wait_for_atomic((COND), (MS) * 1000, 1)
> -#define wait_for_atomic_us(COND, US) _wait_for_atomic((COND), (US), 1)
> +#define wait_for_atomic_us(COND, US) \
> +({ \
> + int ret__; \
> + BUILD_BUG_ON(!__builtin_constant_p(US)); \
> + BUILD_BUG_ON((US) > 50000); \
> + ret__ = _wait_for_atomic((COND), (US), 1); \
> + ret__; \
> +})
Do we need ret__ here? Doesn't _wait_for_atomic() act as the expression
providing the value to the caller?
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
