All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgg@ziepe.ca>
To: Zhiping Zhang <zhipingz@meta.com>
Cc: fengchengwen <fengchengwen@huawei.com>,
	Leon Romanovsky <leon@kernel.org>,
	Michael Guralnik <michaelgur@nvidia.com>,
	Sumit Semwal <sumit.semwal@linaro.org>,
	Christian Konig <christian.koenig@amd.com>,
	Alex Williamson <alex@shazbot.org>,
	Bjorn Helgaas <bhelgaas@google.com>,
	kvm@vger.kernel.org, linux-rdma@vger.kernel.org,
	linux-pci@vger.kernel.org, dri-devel@lists.freedesktop.org
Subject: Re: [PATCH v13 5/5] RDMA/mlx5: get tph for p2p access when registering dma-buf mr
Date: Mon, 28 Sep 2026 14:29:39 -0300	[thread overview]
Message-ID: <20260928172939.GK163130@ziepe.ca> (raw)
In-Reply-To: <CAH3zFs2rhg6-b-pVDps1V8LsM4Dn48t4J4LFVyKOZtT4axdY=Q@mail.gmail.com>

On Thu, Sep 24, 2026 at 11:17:57PM -0700, Zhiping Zhang wrote:
> On Thu, Sep 24, 2026 at 4:31 PM Jason Gunthorpe <jgg@ziepe.ca> wrote:
> >
> > >
> > On Wed, Sep 23, 2026 at 11:24:56PM -0700, Zhiping Zhang wrote:
> > > I don't think this belongs in the importer. Every in-tree dma-buf
> > > caller of pci_p2pdma_distance() is the exporter, in its .attach,
> > > clearing attach->peer2peer -- amdgpu_dma_buf.c, xe_dma_buf.c,
> > > habanalabs memory.c. There is no importer-side caller.
> >
> > Right, and they shouldn't be doing that, but it still has to be
> > checked that the st is going directly to the peer device not the host
> > bridge. Not using the distance, but by the PCI_P2PDMA_MAP_BUS_ADDR
> > indication.
> >
> > I fear you will need some of Leon's series to make that happen.
> >
> > So probably the proposed change to dmabuf ops is far too simple.
> >
> > Jason
> 
> Hi Jason,
> 
> Thanks for the comments.
> 
> Agreed -- the tag should not be handed out unless the routing is
> PCI_P2PDMA_MAP_BUS_ADDR. That can be enforced conservatively on 7.3
> with two changes. I can fold both into patch 4.
> ```
> In drivers/pci/p2pdma.c, make pci_p2pdma_map_type() callable from
> tristate VFIO_PCI_CORE:
> 
>     EXPORT_SYMBOL_GPL(pci_p2pdma_map_type);

No, that's been rejected several times already.
 
> and in drivers/vfio/pci/vfio_pci_dmabuf.c,
> vfio_pci_dma_buf_get_pci_tph() gains:
> 
>     struct dma_buf_attachment *attach;
> 
>     if (list_empty(&dmabuf->attachments))
>         return -EOPNOTSUPP;
> 
>     list_for_each_entry(attach, &dmabuf->attachments, node)
>         if (pci_p2pdma_map_type(priv->provider, attach->dev) !=
>             PCI_P2PDMA_MAP_BUS_ADDR)
>                 return -EOPNOTSUPP;
> ```

Also no, this has to be done via PCI_P2PDMA_MAP_BUS_ADDR which also
enforces putting the determination in the right place in the code
flow..

> Two properties are worth stating explicitly:

Ah! AI!

Jason

  reply	other threads:[~2026-09-28 17:29 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-31 21:15 [PATCH v13 0/5] vfio/dma-buf: add TPH support for peer-to-peer access Zhiping Zhang
2026-07-31 21:15 ` [PATCH v13 1/5] PCI/TPH: treat reserved 0b10 completer encoding as unsupported Zhiping Zhang
2026-07-31 21:38   ` sashiko-bot
2026-09-15 22:19   ` Bjorn Helgaas
2026-09-15 23:14     ` Zhiping Zhang
2026-07-31 21:15 ` [PATCH v13 2/5] PCI/TPH: Add requester/completer type helpers Zhiping Zhang
2026-07-31 21:40   ` sashiko-bot
2026-08-03  1:21   ` fengchengwen
2026-09-15 22:23   ` Bjorn Helgaas
2026-09-15 23:27     ` Zhiping Zhang
2026-07-31 21:15 ` [PATCH v13 3/5] dma-buf: add optional get_pci_tph() callback Zhiping Zhang
2026-07-31 21:37   ` sashiko-bot
2026-07-31 21:15 ` [PATCH v13 4/5] vfio/pci: implement get_pci_tph and DMA_BUF_TPH feature Zhiping Zhang
2026-07-31 21:38   ` sashiko-bot
2026-07-31 21:15 ` [PATCH v13 5/5] RDMA/mlx5: get tph for p2p access when registering dma-buf mr Zhiping Zhang
2026-07-31 21:48   ` sashiko-bot
2026-07-31 22:59     ` Zhiping Zhang
2026-09-23 19:11   ` Zhiping Zhang
2026-09-24  1:13   ` fengchengwen
2026-09-24  6:24     ` Zhiping Zhang
2026-09-24 23:30       ` Jason Gunthorpe
2026-09-25  6:17         ` Zhiping Zhang
2026-09-28 17:29           ` Jason Gunthorpe [this message]
2026-09-28 21:47             ` Zhiping Zhang
2026-09-30 12:13               ` Leon Romanovsky
2026-10-05 21:33                 ` Zhiping Zhang
2026-08-14 18:08 ` [PATCH v13 0/5] vfio/dma-buf: add TPH support for peer-to-peer access Zhiping Zhang
2026-09-02 20:53   ` Zhiping Zhang
2026-09-10 21:46     ` [PATCH v13 0/5] vfio/dma-buf: add TPH support for peer-to-peer access - ping Zhiping Zhang
2026-09-11  7:58       ` Christian König
2026-09-14 16:44         ` Zhiping Zhang
2026-09-15 20:49           ` Alex Williamson
2026-09-16  0:17             ` Zhiping Zhang
2026-09-15 11:37         ` Leon Romanovsky
2026-09-15 12:59           ` Christian König
2026-09-15 13:37             ` Leon Romanovsky

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=20260928172939.GK163130@ziepe.ca \
    --to=jgg@ziepe.ca \
    --cc=alex@shazbot.org \
    --cc=bhelgaas@google.com \
    --cc=christian.koenig@amd.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=fengchengwen@huawei.com \
    --cc=kvm@vger.kernel.org \
    --cc=leon@kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=michaelgur@nvidia.com \
    --cc=sumit.semwal@linaro.org \
    --cc=zhipingz@meta.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.