From: Haggai Eran <haggaie@mellanox.com>
To: Dan Williams <dan.j.williams@intel.com>,
Stephen Bates <sbates@raithlin.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-nvdimm@lists.01.org" <linux-nvdimm@ml01.01.org>,
<linux-rdma@vger.kernel.org>, <linux-block@vger.kernel.org>,
Linux MM <linux-mm@kvack.org>,
Ross Zwisler <ross.zwisler@linux.intel.com>,
"Matthew Wilcox" <willy@linux.intel.com>,
<jgunthorpe@obsidianresearch.com>,
"Christoph Hellwig" <hch@infradead.org>,
Jens Axboe <axboe@fb.com>, Jonathan Corbet <corbet@lwn.net>,
<jim.macdonald@everspin.com>, <sbates@raithin.com>,
"Logan Gunthorpe" <logang@deltatee.com>,
David Woodhouse <dwmw2@infradead.org>,
"Raj, Ashok" <ashok.raj@intel.com>
Subject: Re: [PATCH 0/3] iopmem : A block device for PCIe memory
Date: Wed, 26 Oct 2016 11:24:51 +0300 [thread overview]
Message-ID: <a5418089-2615-8c04-aca8-50ceb43978f1@mellanox.com> (raw)
In-Reply-To: <CAPcyv4gJ_c-6s2BUjsu6okR1EF53R+KNuXnOc5jv0fuwJaa3cQ@mail.gmail.com>
On 10/19/2016 6:51 AM, Dan Williams wrote:
> On Tue, Oct 18, 2016 at 2:42 PM, Stephen Bates <sbates@raithlin.com> wrote:
>> 1. Address Translation. Suggestions have been made that in certain
>> architectures and topologies the dma_addr_t passed to the DMA master
>> in a peer-2-peer transfer will not correctly route to the IO memory
>> intended. However in our testing to date we have not seen this to be
>> an issue, even in systems with IOMMUs and PCIe switches. It is our
>> understanding that an IOMMU only maps system memory and would not
>> interfere with device memory regions.
I'm not sure that's the case. I think it works because with ZONE_DEVICE,
the iommu driver will simply treat a dma_map_page call as any other PFN,
and create a mapping as it does for any memory page.
>> (It certainly has no opportunity
>> to do so if the transfer gets routed through a switch).
It can still go through the IOMMU if you enable ACS upstream forwarding.
> There may still be platforms where peer-to-peer cycles are routed up
> through the root bridge and then back down to target device, but we
> can address that when / if it happens.
I agree.
> I wonder if we could (ab)use a
> software-defined 'pasid' as the requester id for a peer-to-peer
> mapping that needs address translation.
Why would you need that? Isn't it enough to map the peer-to-peer
addresses correctly in the iommu driver?
Haggai
next prev parent reply other threads:[~2016-10-26 8:40 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-10-18 21:42 [PATCH 0/3] iopmem : A block device for PCIe memory Stephen Bates
2016-10-18 21:42 ` [PATCH 1/3] memremap.c : Add support for ZONE_DEVICE IO memory with struct pages Stephen Bates
2016-10-19 17:50 ` Dan Williams
2016-10-19 18:40 ` Stephen Bates
2016-10-19 20:01 ` Dan Williams
2016-10-25 11:54 ` Stephen Bates
2016-10-18 21:42 ` [PATCH 2/3] iopmem : Add a block device driver for PCIe attached IO memory Stephen Bates
2016-10-28 6:45 ` Christoph Hellwig
2016-10-28 19:22 ` Logan Gunthorpe
2016-10-18 21:42 ` [PATCH 3/3] iopmem : Add documentation for iopmem driver Stephen Bates
2016-10-28 6:46 ` Christoph Hellwig
2016-10-19 3:51 ` [PATCH 0/3] iopmem : A block device for PCIe memory Dan Williams
2016-10-19 18:48 ` Stephen Bates
2016-10-19 19:58 ` Dan Williams
2016-10-19 22:54 ` Stephen Bates
2016-10-20 23:22 ` Dave Chinner
2016-10-21 9:57 ` Christoph Hellwig
2016-10-21 11:12 ` Dave Chinner
2016-10-25 11:50 ` Stephen Bates
2016-10-25 21:19 ` Dave Chinner
2016-11-06 14:05 ` Stephen Bates
2016-10-27 10:22 ` Sagi Grimberg
2016-10-27 12:32 ` Christoph Hellwig
2016-10-26 8:24 ` Haggai Eran [this message]
2016-10-26 13:39 ` Dan Williams
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=a5418089-2615-8c04-aca8-50ceb43978f1@mellanox.com \
--to=haggaie@mellanox.com \
--cc=ashok.raj@intel.com \
--cc=axboe@fb.com \
--cc=corbet@lwn.net \
--cc=dan.j.williams@intel.com \
--cc=dwmw2@infradead.org \
--cc=hch@infradead.org \
--cc=jgunthorpe@obsidianresearch.com \
--cc=jim.macdonald@everspin.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nvdimm@ml01.01.org \
--cc=linux-rdma@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=ross.zwisler@linux.intel.com \
--cc=sbates@raithin.com \
--cc=sbates@raithlin.com \
--cc=willy@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).