Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH] nvme-multipath: implement active-active round-robin path
 selector
To:     Christoph Hellwig <hch@lst.de>, Baegjae Sung <baegjae@gmail.com>
Cc:     keith.busch@intel.com, axboe@fb.com,
        linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org
References: <20180327043851.6640-1-baegjae@gmail.com>
 <20180328080646.GB20373@lst.de>
From:   Sagi Grimberg <sagi@grimberg.me>
Message-ID: <a8574e1c-d749-3abd-bc7d-2d6010087993@grimberg.me>
Date:   Wed, 4 Apr 2018 15:36:45 +0300
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.6.0
MIME-Version: 1.0
In-Reply-To: <20180328080646.GB20373@lst.de>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> For PCIe devices the right policy is not a round robin but to use
> the pcie device closer to the node.  I did a prototype for that
> long ago and the concept can work.  Can you look into that and
> also make that policy used automatically for PCIe devices?

I think that active/active makes sense for fabrics (link throughput
aggregation) but also for dual-ported pci devices (given that
this is a real use-case).

I agree that the default can be a home-node path selection.