MIME-Version: 1.0
In-Reply-To: <20150416164418.GC22946@obsidianresearch.com>
References: <20150413192701.GA19112@obsidianresearch.com> <20150413194602.GA21467@phlsvsds.ph.intel.com>
 <20150413200138.GC19112@obsidianresearch.com> <20150414141806.GA7354@phlsvsds.ph.intel.com>
 <20150414172514.GB7682@obsidianresearch.com> <20150414174311.GA28111@phlsvsds.ph.intel.com>
 <20150414175904.GB13896@obsidianresearch.com> <1828884A29C6694DAF28B7E6B8A82373A8FC127C@ORSMSX109.amr.corp.intel.com>
 <20150414182137.GD7682@obsidianresearch.com> <552E1A1A.4090704@profitbricks.com>
 <20150416164418.GC22946@obsidianresearch.com>
From: Roland Dreier <roland@kernel.org>
Date: Thu, 16 Apr 2015 10:02:46 -0700
Message-ID: <CAG4TOxPMaRAR_B69J1_SHaHh8YMW2Unx786t2FbrA8g=05BAgA@mail.gmail.com>
Subject: Re: [PATCH v3 07/28] IB/Verbs: Reform IB-ulp ipoib
To: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>
Cc: Michael Wang <yun.wang@profitbricks.com>,
        "Hefty, Sean" <sean.hefty@intel.com>,
        "Weiny, Ira" <ira.weiny@intel.com>,
        Hal Rosenstock <hal.rosenstock@gmail.com>,
        "linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Tom Tucker <tom@opengridcomputing.com>,
        Steve Wise <swise@opengridcomputing.com>,
        Hoang-Nam Nguyen <hnguyen@de.ibm.com>,
        Christoph Raisch <raisch@de.ibm.com>,
        infinipath <infinipath@intel.com>, Eli Cohen <eli@mellanox.com>,
        "Latif, Faisal" <faisal.latif@intel.com>,
        Jack Morgenstein <jackm@dev.mellanox.co.il>,
        Or Gerlitz <ogerlitz@mellanox.com>, Haggai Eran <haggaie@mellanox.com>,
        Tom Talpey <tom@talpey.com>, Doug Ledford <dledford@redhat.com>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 912
Lines: 20

On Thu, Apr 16, 2015 at 9:44 AM, Jason Gunthorpe
<jgunthorpe@obsidianresearch.com> wrote:
>> We can give client->add() callback a return value and make
>> ib_register_device() return -ENOMEM when it failed, just wondering
>> why we don't do this at first, any special reason?

> No idea, but having ib_register_device fail and unwind if a client
> fails to attach makes sense to me.

It seems a bit unfriendly to fail an entire device if one ULP has a
problem.  Let's say you have a system whose main network connection is
IPoIB.  Would you want that connection to come up even if, say, the
NFS/RDMA server fails to find the memory registration type it likes?

 - R.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/