From: David Miller Subject: [ofa-general] Re: [PATCH 4/9] mlx4_en: Verify number of RX rings doesn't exceed MAX_RX_RINGS Date: Thu, 25 Dec 2008 18:18:48 -0800 (PST) Message-ID: <20081225.181848.211823343.davem@davemloft.net> References: <494F6551.3040500@mellanox.co.il> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, rdreier@cisco.com, general@lists.openfabrics.org, jeff@garzik.org To: yevgenyp@mellanox.co.il Return-path: In-Reply-To: <494F6551.3040500@mellanox.co.il> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: general-bounces@lists.openfabrics.org Errors-To: general-bounces@lists.openfabrics.org List-Id: netdev.vger.kernel.org From: Yevgeny Petrilin Date: Mon, 22 Dec 2008 12:00:49 +0200 > Required in cases were dev->caps.num_comp_vectors > MAX_RX_RINGS. > For current values this would happen on machines that have more > then 16 cores. > > Signed-off-by: Yevgeny Petrilin This patch does not apply to the tree. In fact, there is no reference at all to num_comp_vectors in the mlx4_en driver at all. The code there currently reads: if (!mdev->profile.prof[i].rx_ring_num) { mdev->profile.prof[i].rx_ring_num = 1; mlx4_info(mdev, "Defaulting to %d rx rings for port:%d\n", 1, i); So there is some other patch applied to your copy of the driver already that adds all of that num_comp_vectors stuff. It isn't in the upstream sources, and it certainly isn't in the networking development GIT tree(s), that is for sure. Please, please (did I say please?), please be more careful in the future and don't send patches that do not apply properly to the tree. This wastes a lot of my time and discourages my handling your patches efficiently in the future.