From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sagi Grimberg Subject: Re: [PATCH 1/1] IB/iser: Remove hard coded values for cqe and send_wr Date: Wed, 08 Oct 2014 08:58:10 +0300 Message-ID: <5434D272.8090007@dev.mellanox.co.il> References: <1412728888-13100-1-git-send-email-jkallickal@emulex.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1412728888-13100-1-git-send-email-jkallickal@emulex.com> Sender: linux-scsi-owner@vger.kernel.org To: Jay Kallickal , linux-scsi@vger.kernel.org, michaelc@cs.wisc.edu, linux-rdma@vger.kernel.org Cc: Minh Tran , Jayamohan Kallickal List-Id: linux-rdma@vger.kernel.org On 10/8/2014 3:41 AM, Jay Kallickal wrote: > From: Jayamohan Kallickal > > This patch allows the underlying hardware to choose > values other than hard coded max values for cqe and send_wr > while preventing them from exceeding max supported values. Hi Minh and Jayamohan, So I agree that we would want to take device capabilities into account here, but we need to be able to adjust scsi_cmds_max (can_queue) in case the max wqe supported is lower than scsi_cmds_max * num_posts_per_cmd. So generally I agree with this approach, but we need to take care of stuff later when the session is created. One more thing, this is not rebased on the latest iser patches please send v1 on top of: http://marc.info/?l=linux-rdma&m=141216135013146&w=2 P.S. What device did you test with (that supports less than iSER needs)? Thanks! Sagi. > > Signed-off-by: Minh Tran > Signed-off-by: Jayamohan Kallickal > --- > drivers/infiniband/ulp/iser/iser_verbs.c | 18 ++++++++++++++---- > 1 file changed, 14 insertions(+), 4 deletions(-) >