On Mon, Nov 10, 2014 at 09:47:27PM +0800, Herbert Xu wrote: > On Mon, Nov 03, 2014 at 11:56:24PM +0800, Amos Kong wrote: > > > > @@ -98,6 +99,8 @@ static inline void cleanup_rng(struct kref *kref) > > > > if (rng->cleanup) > > rng->cleanup(rng); > > You need a compiler barrier here to prevent reordering. Michael Büsch pointed out that we should actually have a memory barrier here. I thought we didn't need it because I was only worried about the code in my original complaint. However, expecting driver writers to use correct synchronisation primitives is surely asking too much. So let's add an smp_wmb() here to ensure all side-effects of cleanup is visible. Thanks, -- Email: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt _______________________________________________ Virtualization mailing list Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linuxfoundation.org/mailman/listinfo/virtualization
