On Wed, Dec 15, 2021 at 12:18:44PM +0100, Uladzislau Rezki (Sony) wrote:
> Instead of invoking a synchronize_rcu() to free a pointer
> after a grace period we can directly make use of new API
> that does the same but in more efficient way.
>
> TO: linux-rdma@xxxxxxxxxxxxxxx
> TO: Jason Gunthorpe <jgg@xxxxxxxxxx>
> TO: Mike Marciniszyn <mike.marciniszyn@xxxxxxxxxxxxxxxxxxxx>
> Acked-by: Mike Marciniszyn <mike.marciniszyn@xxxxxxxxxxxxxxxxxxxx>
> Signed-off-by: Uladzislau Rezki (Sony) <urezki@xxxxxxxxx>
> ---
> drivers/infiniband/hw/hfi1/sdma.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/infiniband/hw/hfi1/sdma.c b/drivers/infiniband/hw/hfi1/sdma.c
> index f07d328689d3..7264a35e8f4c 100644
> --- a/drivers/infiniband/hw/hfi1/sdma.c
> +++ b/drivers/infiniband/hw/hfi1/sdma.c
> @@ -1292,8 +1292,7 @@ void sdma_clean(struct hfi1_devdata *dd, size_t num_engines)
> sdma_map_free(rcu_access_pointer(dd->sdma_map));
> RCU_INIT_POINTER(dd->sdma_map, NULL);
> spin_unlock_irq(&dd->sde_map_lock);
> - synchronize_rcu();
> - kfree(dd->per_sdma);
> + kvfree_rcu(dd->per_sdma);
> dd->per_sdma = NULL;
>
> if (dd->sdma_rht) {
> --
> 2.30.2
>
+ linux-rdma@xxxxxxxxxxxxxxx
+ Jason Gunthorpe <jgg@xxxxxxxxxx>
+ Mike Marciniszyn <mike.marciniszyn@xxxxxxxxxxxxxxxxxxxx>
--
Vlad Rezki
