From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Jim Schutt" Subject: Re: [PATCH] mlx4: allow for 4K mtu configuration of IB ports Date: Tue, 31 May 2011 10:11:58 -0600 Message-ID: <4DE5134E.6070401@sandia.gov> References: <055701cc1b29$c8fcc6c0$5af65440$@systemfabricworks.com> <4DDDF0B1.6090305@mellanox.com> <17495_1306419903_p4QEMqSb018281_4DDE6291.1090809@sandia.gov> <4DDE63F9.8060502@sandia.gov> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Or Gerlitz Cc: Roland Dreier , Or Gerlitz , Bob Pearson , linux-rdma , Vladimir Sokolovsky , Alex Netes List-Id: linux-rdma@vger.kernel.org Hi Or, Or Gerlitz wrote: > Roland Dreier wrote: > Jim Schutt wrote: >>> no good reason to insist that the VL usage is the same for both >>> interswitch links, and switch-CA links. Do I need to change this? > >> I don't think changing this is a high priority, since it's a pretty small >> slice of the world, and QoS on the edge links probably is important >> to an even smaller slice, but IMHO it would be better to give QoS to >> HCAs that only support 4 VLs by using a different SL2VL table for links to CAs. > > Jim, > > AFAIK, the way opensm applies an SL-to-VL mapping specification (e.g > dictated by the admin or maybe your routing engine) on a specific link > is by modulation on the number of active VLs for that link - e.g say > the ID mapping was required and there are two VLs for that link, so > we'll have SL-to-VL of 0->0 1->1 2->0 3->1 and so on. So in that > respect, I wasn't sure what's the change here for you. Hmmm, can you tell me where such remapping happens? What I know about so far is the code in sl2vl_update_table(), which AFAICS truncates VL values in the SL2VL maps provided by the routing engine to fit into the number of VLs supported by a port. Am I missing something else? torus-2QoS will currently only use VL values 0 and 4 in the SL2VL maps it generates for CA ports; i.e. any SL maps either to VL 0 or VL 4, depending on QoS level. So sl2vl_update_table() would truncate all those VL 4 values to 0 in the case of ports that support fewer than 8 data VLs. What I'm suggesting is that I need to make torus-2QoS generate SL2VL maps that only reference VLs 0 or 1 for CA ports, in order to allow it to support 2 QoS levels for CA ports that don't support 8 data VLs. -- Jim > > Or. > > -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html