From: John Hubbard <jhubbard@nvidia.com>
To: "Christian König" <christian.koenig@amd.com>,
"Jason Gunthorpe" <jgg@ziepe.ca>
Cc: Gal Pressman <galpress@amazon.com>,
Daniel Vetter <daniel@ffwll.ch>,
Sumit Semwal <sumit.semwal@linaro.org>,
Doug Ledford <dledford@redhat.com>,
"open list:DMA BUFFER SHARING FRAMEWORK"
<linux-media@vger.kernel.org>,
dri-devel <dri-devel@lists.freedesktop.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-rdma <linux-rdma@vger.kernel.org>,
Oded Gabbay <ogabbay@habana.ai>, Tomer Tayar <ttayar@habana.ai>,
Yossi Leybovich <sleybo@amazon.com>,
Alexander Matushevsky <matua@amazon.com>,
Leon Romanovsky <leonro@nvidia.com>,
Jianxin Xiong <jianxin.xiong@intel.com>
Subject: Re: [RFC] Make use of non-dynamic dmabuf in RDMA
Date: Tue, 24 Aug 2021 23:47:51 -0700 [thread overview]
Message-ID: <f049576e-4f36-6407-883d-24fac47c4491@nvidia.com> (raw)
In-Reply-To: <a716ae41-2d8c-c75a-a779-cc85b189fea2@amd.com>
On 8/24/21 11:17 PM, Christian König wrote:
...
>> I think it depends on the user, if the user creates memory which is
>> permanently located on the GPU then it should be pinnable in this way
>> without force migration. But if the memory is inherently migratable
>> then it just cannot be pinned in the GPU at all as we can't
>> indefinately block migration from happening eg if the CPU touches it
>> later or something.
>
> Yes, exactly that's the point. Especially GPUs have a great variety of setups.
>
> For example we have APUs where the local memory is just stolen system memory and all buffers must be
> migrate-able because you might need all of this stolen memory for scanout or page tables. In this
> case P2P only makes sense to avoid the migration overhead in the first place.
>
> Then you got dGPUs where only a fraction of the VRAM is accessible from the PCIe BUS. Here you also
> absolutely don't want to pin any buffers because that can easily crash when we need to migrate
> something into the visible window for CPU access.
>
> The only real option where you could do P2P with buffer pinning are those compute boards where we
> know that everything is always accessible to everybody and we will never need to migrate anything.
> But even then you want some mechanism like cgroups to take care of limiting this. Otherwise any
> runaway process can bring down your whole system.
>
> Key question at least for me as GPU maintainer is if we are going to see modern compute boards
> together with old non-ODP setups. Since those compute boards are usually used with new hardware
> (like PCIe v4 for example) the answer I think is most likely "no".
>
That is a really good point. Times have changed and I guess ODP is on most (all?) of
the new Infiniband products now, and maybe we don't need to worry so much about
providing first-class support for non-ODP setups.
I've got to drag my brain into 2021+! :)
thanks,
--
John Hubbard
NVIDIA
next prev parent reply other threads:[~2021-08-25 6:47 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-18 7:43 [RFC] Make use of non-dynamic dmabuf in RDMA Gal Pressman
2021-08-18 8:00 ` Christian König
2021-08-18 8:37 ` Gal Pressman
2021-08-18 9:34 ` Daniel Vetter
2021-08-19 23:06 ` Jason Gunthorpe
2021-08-20 7:25 ` Daniel Vetter
2021-08-20 12:33 ` Jason Gunthorpe
2021-08-20 12:58 ` Gal Pressman
2021-08-20 14:32 ` Jason Gunthorpe
2021-08-21 9:16 ` Gal Pressman
2021-08-23 10:43 ` Christian König
2021-08-24 9:06 ` Gal Pressman
2021-08-24 9:32 ` Christian König
2021-08-24 17:27 ` John Hubbard
2021-08-24 17:32 ` Jason Gunthorpe
2021-08-24 17:35 ` John Hubbard
2021-08-24 19:15 ` Dave Airlie
2021-08-24 19:30 ` Jason Gunthorpe
2021-08-24 19:43 ` Alex Deucher
2021-08-24 20:00 ` Xiong, Jianxin
2021-08-25 6:17 ` Christian König
2021-08-25 6:47 ` John Hubbard [this message]
2021-08-25 12:18 ` Jason Gunthorpe
2021-08-25 12:27 ` Christian König
2021-08-25 12:38 ` Jason Gunthorpe
2021-08-25 13:51 ` Christian König
2021-08-25 14:47 ` Jason Gunthorpe
2021-08-25 15:14 ` Christian König
2021-08-25 15:49 ` Jason Gunthorpe
2021-08-25 16:02 ` Oded Gabbay
2021-09-01 11:20 ` Gal Pressman
2021-09-01 11:24 ` Christian König
2021-09-02 6:56 ` Gal Pressman
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=f049576e-4f36-6407-883d-24fac47c4491@nvidia.com \
--to=jhubbard@nvidia.com \
--cc=christian.koenig@amd.com \
--cc=daniel@ffwll.ch \
--cc=dledford@redhat.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=galpress@amazon.com \
--cc=jgg@ziepe.ca \
--cc=jianxin.xiong@intel.com \
--cc=leonro@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=matua@amazon.com \
--cc=ogabbay@habana.ai \
--cc=sleybo@amazon.com \
--cc=sumit.semwal@linaro.org \
--cc=ttayar@habana.ai \
/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).