Re: [PATCH v3 01/11] PCI/P2PDMA: Support peer-to-peer memory

From: Logan Gunthorpe <logang@deltatee.com>
To: Jason Gunthorpe <jgg@mellanox.com>,
	Jonathan Cameron <Jonathan.Cameron@huawei.com>
Cc: "Sinan Kaya" <okaya@codeaurora.org>,
	linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
	linux-nvme@lists.infradead.org, linux-rdma@vger.kernel.org,
	linux-nvdimm@lists.01.org, linux-block@vger.kernel.org,
	"Stephen Bates" <sbates@raithlin.com>,
	"Christoph Hellwig" <hch@lst.de>, "Jens Axboe" <axboe@kernel.dk>,
	"Keith Busch" <keith.busch@intel.com>,
	"Sagi Grimberg" <sagi@grimberg.me>,
	"Bjorn Helgaas" <bhelgaas@google.com>,
	"Max Gurtovoy" <maxg@mellanox.com>,
	"Dan Williams" <dan.j.williams@intel.com>,
	"Jérôme Glisse" <jglisse@redhat.com>,
	"Benjamin Herrenschmidt" <benh@kernel.crashing.org>,
	"Alex Williamson" <alex.williamson@redhat.com>,
	"Eric Wehage" <Eric.Wehage@huawei.com>
Subject: Re: [PATCH v3 01/11] PCI/P2PDMA: Support peer-to-peer memory
Date: Mon, 26 Mar 2018 11:30:38 -0600	[thread overview]
Message-ID: <6ead7f19-c0e5-5584-2518-1e1889a007e4@deltatee.com> (raw)
In-Reply-To: <20180326164129.GF15530@mellanox.com>

On 26/03/18 10:41 AM, Jason Gunthorpe wrote:
> On Mon, Mar 26, 2018 at 12:11:38PM +0100, Jonathan Cameron wrote:
>> On Tue, 13 Mar 2018 10:43:55 -0600
>> Logan Gunthorpe <logang@deltatee.com> wrote:
>>
>>> On 12/03/18 09:28 PM, Sinan Kaya wrote:
>>>> On 3/12/2018 3:35 PM, Logan Gunthorpe wrote:
>>>> Regarding the switch business, It is amazing how much trouble you went into
>>>> limit this functionality into very specific hardware.
>>>>
>>>> I thought that we reached to an agreement that code would not impose
>>>> any limits on what user wants.
>>>>
>>>> What happened to all the emails we exchanged?  
>>>
>>> It turns out that root ports that support P2P are far less common than 
>>> anyone thought. So it will likely have to be a white list.
>>
>> This came as a bit of a surprise to our PCIe architect.
> 
> I don't think it is a hardware problem.

The latest and greatest Power9 CPUs still explicitly do not support
this. And, if I recall correctly, the ARM64 device we played with did
not either -- but I suspect that will differ depending on vendor.

The latest Intel devices of course do support it, but go back 3-4 years
or so and the performance is pretty much unusable (at least for the
purposes of what this patchset implements). Even older CPUs did not
support it.

We haven't done any testing on AMD devices but I assume they are similar
to Intel.

In any case, even if every CPU manufactured today supported it well,
there are still older devices out there without support that we need to
ensure are handled properly.

Logan