Date:   Fri, 3 Jan 2020 16:36:59 -0400
From:   Jason Gunthorpe <jgg@ziepe.ca>
To:     Leon Romanovsky <leon@kernel.org>
Cc:     Doug Ledford <dledford@redhat.com>,
        Danit Goldberg <danitg@mellanox.com>,
        RDMA mailing list <linux-rdma@vger.kernel.org>,
        Leon Romanovsky <leonro@mellanox.com>
Subject: Re: [PATCH rdma-next] RDMA/cm: Use RCU synchronization mechanism to
 protect cm_id_private xa_load()
Message-ID: <20200103203659.GA29125@ziepe.ca>
References: <20191219134750.413429-1-leon@kernel.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20191219134750.413429-1-leon@kernel.org>
User-Agent: Mutt/1.9.4 (2018-02-28)
Sender: linux-rdma-owner@vger.kernel.org
Precedence: bulk

On Thu, Dec 19, 2019 at 03:47:50PM +0200, Leon Romanovsky wrote:
> From: Danit Goldberg <danitg@mellanox.com>
> 
> The RCU mechanism is optimized for read-mostly scenarios and therefore
> more suitable to protect the cm_id_private to decrease "cm.lock"
> congestion.
> 
> This patch replaces the existing spinlock locking mechanism
> and kfree with RCU mechanism in places where spinlock(cm.lock)
> protected cm_id_priv xa_load.
> 
> In addition, deletes cm_get_id function and use cm_acquire_id directly
> with the correct locking. The patch also removes an open coded version
> of cm_get_id, and replaces it with a call to cm_acquire_id.
> 
> Signed-off-by: Danit Goldberg <danitg@mellanox.com>
> Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
> ---
>  drivers/infiniband/core/cm.c | 43 +++++++++++-------------------------
>  1 file changed, 13 insertions(+), 30 deletions(-)

Applied to for-next, thanks

Jason