On Tue, Jul 16, 2019 at 02:53:09PM +0200, Andrea Parri wrote:
>
> P1(atomic_t *reorder_objects, spinlock_t *pd_lock, spinlock_t *reorder_lock)
> {
> int r1;
>
> spin_lock(reorder_lock);
> atomic_inc(reorder_objects);
> spin_unlock(reorder_lock);
> //smp_mb();
> r1 = spin_trylock(pd_lock);
> }
Yes we need a matching mb on the other side. However, we can
get away with using smp_mb__after_atomic thanks to the atomic_inc
above.
Daniel, can you please respin the patch with the matching smp_mb?
Thanks,
--
Email: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
