[dpdk-dev] [PATCH] lib/rcu: fix possible spurious thread unregister

[David Marchand]

On Mon, Sep 9, 2019 at 3:52 PM Ruifeng Wang <ruifeng.wang at arm.com> wrote:
>
> Thread unregister returns success while unregister not been performed.
> This is due to incorrect thread registration status check.
> Fix this issue by correcting bitmap check.
>
> Fixes: 64994b56cfd7 ("rcu: add RCU library supporting QSBR mechanism")
> Cc: stable at dpdk.org
>
> Signed-off-by: Ruifeng Wang <ruifeng.wang at arm.com>
> Reviewed-by: Gavin Hu <gavin.hu at arm.com>
> ---
>  lib/librte_rcu/rte_rcu_qsbr.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/lib/librte_rcu/rte_rcu_qsbr.c b/lib/librte_rcu/rte_rcu_qsbr.c
> index ce7f93dd3..8c37c88cd 100644
> --- a/lib/librte_rcu/rte_rcu_qsbr.c
> +++ b/lib/librte_rcu/rte_rcu_qsbr.c
> @@ -158,7 +158,7 @@ rte_rcu_qsbr_thread_unregister(struct rte_rcu_qsbr *v, unsigned int thread_id)
>         /* Check if the thread is already unregistered */
>         old_bmap = __atomic_load_n(__RTE_QSBR_THRID_ARRAY_ELM(v, i),
>                                         __ATOMIC_RELAXED);
> -       if (old_bmap & ~(1UL << id))
> +       if (!(old_bmap & (1UL << id)))
>                 return 0;
>
>         do {
> @@ -175,7 +175,7 @@ rte_rcu_qsbr_thread_unregister(struct rte_rcu_qsbr *v, unsigned int thread_id)
>                 if (success)
>                         __atomic_fetch_sub(&v->num_threads,
>                                                 1, __ATOMIC_RELAXED);
> -               else if (old_bmap & ~(1UL << id))
> +               else if (!(old_bmap & (1UL << id)))
>                         /* Someone else unregistered this thread.
>                          * Counter should not be incremented.
>                          */

Reviewed-by: David Marchand <david.marchand at redhat.com>


Honnappa,

While looking at the rcu doxygen, I noted it does not describe what
return values to expect from register/unregister.
We also have typos on s/rte_rcu_thread_offline/rte_rcu_qsbr_thread_offline/g.

Could you send a patch for this?


Thanks.

--
David Marchand