On Thu Feb 13, 2020 at 2:40 PM, Marcelo Diop-Gonzalez wrote:
> If kref_get_unless_zero() fails, we should keep looking for the
> next service, since the callers of this function expect that a NULL
> return value means there are no more.
>
> Signed-off-by: Marcelo Diop-Gonzalez <marcgonzalez@xxxxxxxxxx>
Reviewed-by: Nicolas Saenz Julienne <nsaenzjulienne@xxxxxxx>
Tested-by: Nicolas Saenz Julienne <nsaenzjulienne@xxxxxxx>
Note that, as Dan says, picking up the Fixes tag would be nice.
> ---
> .../vc04_services/interface/vchiq_arm/vchiq_core.c | 14 +++++++++-----
> 1 file changed, 9 insertions(+), 5 deletions(-)
>
> diff --git
> a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> index d7d7f4d9d57f..edcd97373809 100644
> --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> @@ -252,11 +252,15 @@ next_service_by_instance(struct vchiq_state
> *state,
> struct vchiq_service *service;
>
> rcu_read_lock();
> - service = __next_service_by_instance(state, instance, pidx);
> - if (service && kref_get_unless_zero(&service->ref_count))
> - service = rcu_pointer_handoff(service);
> - else
> - service = NULL;
> + while (1) {
> + service = __next_service_by_instance(state, instance, pidx);
> + if (!service)
> + break;
> + if (kref_get_unless_zero(&service->ref_count)) {
> + service = rcu_pointer_handoff(service);
> + break;
> + }
> + }
> rcu_read_unlock();
> return service;
> }
> --
> 2.25.0.225.g125e21ebc7-goog
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
