MIME-Version: 1.0
In-Reply-To: <c143c2fd-8bff-595d-8a9e-debfcd56e531@deltatee.com>
References: <20170418210339.GA24257@obsidianresearch.com> <1492564806.25766.124.camel@kernel.crashing.org>
 <20170419155557.GA8497@obsidianresearch.com> <4899b011-bdfb-18d8-ef00-33a1516216a6@deltatee.com>
 <20170419171451.GA10020@obsidianresearch.com> <b0234176-e86b-6e7c-686d-b4fef719aad7@deltatee.com>
 <20170419183247.GA13716@obsidianresearch.com> <21e8099a-d19d-7df0-682d-627d8b81dfde@deltatee.com>
 <20170419193154.GA14340@obsidianresearch.com> <ea03e81f-e3d5-9f5c-db60-0ceaf4366612@deltatee.com>
 <20170419204808.GA15716@obsidianresearch.com> <c143c2fd-8bff-595d-8a9e-debfcd56e531@deltatee.com>
From: Dan Williams <dan.j.williams@gmail.com>
Date: Wed, 19 Apr 2017 17:07:58 -0700
Message-ID: <CAA9_cmdbi0=wS2TFhezzRhdWvO-W75LAG4aHCEwDMHV1gqP2BA@mail.gmail.com>
Subject: Re: [RFC 0/8] Copy Offload with Peer-to-Peer PCI Memory
To: Logan Gunthorpe <logang@deltatee.com>
Cc: Jason Gunthorpe <jgunthorpe@obsidianresearch.com>,
        Jens Axboe <axboe@kernel.dk>,
        "James E.J. Bottomley" <jejb@linux.vnet.ibm.com>,
        "Martin K. Petersen" <martin.petersen@oracle.com>,
        linux-rdma@vger.kernel.org,
        Benjamin Herrenschmidt <benh@kernel.crashing.org>,
        Steve Wise <swise@opengridcomputing.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        linux-nvme@lists.infradead.org, Keith Busch <keith.busch@intel.com>,
        Jerome Glisse <jglisse@redhat.com>, Bjorn Helgaas <helgaas@kernel.org>,
        linux-pci@vger.kernel.org, linux-nvdimm <linux-nvdimm@ml01.01.org>,
        Max Gurtovoy <maxg@mellanox.com>,
        linux-scsi <linux-scsi@vger.kernel.org>,
        Christoph Hellwig <hch@lst.de>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1084
Lines: 25

On Wed, Apr 19, 2017 at 3:55 PM, Logan Gunthorpe <logang@deltatee.com> wrote:
>
>
> On 19/04/17 02:48 PM, Jason Gunthorpe wrote:
>> On Wed, Apr 19, 2017 at 01:41:49PM -0600, Logan Gunthorpe wrote:
>>
>>>> But.. it could point to a GPU and the GPU struct device could have a
>>>> proxy dma_ops like Dan pointed out.
>>>
>>> Seems a bit awkward to me that in order for the intended use case, you
>>> have to proxy the dma_ops. I'd probably still suggest throwing a couple
>>> ops for things like this in the dev_pagemap.
>>
>> Another option is adding a new 'struct completer_dma_ops *' to struct
>> device for this use case.
>>
>> Seems like a waste to expand dev_pagemap when we only need a unique
>> value per struct device?
>
> I feel like expanding dev_pagemap has a much lower impact than expanding
> struct device... dev_pagemap is only one instance per zone device region
> so expanding it shouldn't be a huge issue. Expanding struct device means
> every device struct in the system gets bigger.

Especially since we expect a very small subset of devices will ever support p2p.