From: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
To: Jason Wang <jasowang@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>,
"Michael S. Tsirkin" <mst@redhat.com>,
Guenter Roeck <linux@roeck-us.net>,
virtualization@lists.linux-foundation.org
Subject: Re: [PATCH vhost v6 08/11] virtio_ring: introduce virtqueue_dma_dev()
Date: Tue, 11 Apr 2023 14:11:17 +0800 [thread overview]
Message-ID: <1681193477.0633929-1-xuanzhuo@linux.alibaba.com> (raw)
In-Reply-To: <CACGkMEtD0BLBXv7S-TLY-Y483_Lx4UskLyPmEC2wyGa8VE_R6w@mail.gmail.com>
On Tue, 11 Apr 2023 11:56:47 +0800, Jason Wang <jasowang@redhat.com> wrote:
> On Tue, Apr 11, 2023 at 11:31 AM Christoph Hellwig <hch@infradead.org> wrote:
> >
> > On Tue, Apr 11, 2023 at 10:09:40AM +0800, Jason Wang wrote:
> > > We want to support AF_XDP for virtio-net. It means AF_XDP needs to
> > > know the dma device to perform DMA mapping. So we introduce a helper
> > > to expose the dma dev of the virtio device.
> >
> > The whole virtio architecture is based around the core code doing
> > the DMA mapping. I can't see how you can just export a helper to
> > expose the dma device. You'd have to complete rework the layering
> > of the virtio code if you want to do it in the upper level drivers.
> > And why would you want to do this? The low-level code is the only
> > piece that can actually know if you need to do a dma mapping. All
> > the kernel subsystems that don't do it inside the low-level drivers
> > or helpers closely associtated are a giant and hard to fix map
> > (see usb for the prime exhibit).
> >
> > So the first question is: why do you want this for XF_ADP,
>
> Xuan, is it possible to set up the DMA mapping inside the virtio
> driver itself? I vaguely remember at least the RX buffer mapping is
> done by the driver. If this is true, we can avoid exposing DMA details
> to the upper layer.
NO, all dma maping is done inside xdp socket. That is done
when setup.
When adding to RX Ring, xdp socket will call DMA SYNC.
>
> > and
> > the next question will be how to do that without making a complete
> > mess.
> >
> > > This works fine as long as VIRTIO_F_ACCESS_PLATFORM is negotiated. But
> > > when it is not, the virtio driver needs to use a physical address so
> > > we want to expose the virtio device without dma_ops in the hope that
> > > it will go for direct mapping where the physical address is used. But
> > > it may not work on some specific setups (arches that assume an IOMMU
> > > or have arch dma ops).
> >
> > The DMA device for virtio_pci is the underlying PCI device, always.
> > !VIRTIO_F_ACCESS_PLATFORM means there is no dma device at all. Because
> > of all these things you can't just expose a pointer to the dma_device
> > as that is just a completely wrong way of thinking about the problem.
>
> Ok, so if there's no DMA at all we should avoid using the DMA API
> completely. This means we should check dma_dev against NULL in
> virtio_has_dma_quirk().
>
> Thanks
>
> >
>
> _______________________________________________
> Virtualization mailing list
> Virtualization@lists.linux-foundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/virtualization
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization
next prev parent reply other threads:[~2023-04-11 6:14 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-27 4:05 [PATCH vhost v6 00/11] virtio core prepares for AF_XDP Xuan Zhuo
2023-03-27 4:05 ` [PATCH vhost v6 01/11] virtio_ring: split: separate dma codes Xuan Zhuo
2023-03-28 6:26 ` Jason Wang
2023-03-27 4:05 ` [PATCH vhost v6 02/11] virtio_ring: packed: " Xuan Zhuo
2023-03-27 4:05 ` [PATCH vhost v6 03/11] virtio_ring: packed-indirect: " Xuan Zhuo
2023-03-27 4:05 ` [PATCH vhost v6 04/11] virtio_ring: split: support premapped Xuan Zhuo
2023-03-28 6:27 ` Jason Wang
2023-03-27 4:05 ` [PATCH vhost v6 05/11] virtio_ring: packed: " Xuan Zhuo
2023-03-28 6:27 ` Jason Wang
2023-03-27 4:05 ` [PATCH vhost v6 06/11] virtio_ring: packed-indirect: " Xuan Zhuo
2023-03-28 6:29 ` Jason Wang
2023-03-27 4:05 ` [PATCH vhost v6 07/11] virtio_ring: update document for virtqueue_add_* Xuan Zhuo
2023-03-28 6:30 ` Jason Wang
2023-03-27 4:05 ` [PATCH vhost v6 08/11] virtio_ring: introduce virtqueue_dma_dev() Xuan Zhuo
2023-04-06 17:20 ` Guenter Roeck
2023-04-07 3:17 ` Xuan Zhuo
2023-04-07 12:34 ` Guenter Roeck
2023-04-07 11:02 ` Michael S. Tsirkin
2023-04-10 3:29 ` Xuan Zhuo
2023-04-10 5:14 ` Jason Wang
2023-04-10 6:03 ` Xuan Zhuo
2023-04-10 6:40 ` Michael S. Tsirkin
2023-04-10 6:42 ` Xuan Zhuo
2023-04-10 15:27 ` Christoph Hellwig
2023-04-11 1:56 ` Xuan Zhuo
2023-04-11 2:09 ` Jason Wang
2023-04-11 3:31 ` Christoph Hellwig
2023-04-11 3:56 ` Jason Wang
2023-04-11 4:09 ` Christoph Hellwig
2023-04-11 4:54 ` Jason Wang
2023-04-11 5:14 ` Christoph Hellwig
2023-04-11 5:19 ` Jason Wang
2023-04-11 8:55 ` Gerd Hoffmann
2023-04-11 6:11 ` Xuan Zhuo [this message]
2023-04-11 6:20 ` Christoph Hellwig
2023-04-11 6:28 ` Xuan Zhuo
2023-04-11 6:46 ` Christoph Hellwig
2023-04-11 6:51 ` Xuan Zhuo
2023-04-11 7:04 ` Xuan Zhuo
2023-04-12 2:03 ` Xuan Zhuo
2023-04-12 11:54 ` Christoph Hellwig
2023-04-11 3:26 ` Christoph Hellwig
2023-04-11 6:23 ` Xuan Zhuo
2023-04-11 6:30 ` Christoph Hellwig
2023-04-11 6:33 ` Xuan Zhuo
2023-04-11 6:45 ` Christoph Hellwig
2023-04-11 7:23 ` Xuan Zhuo
2023-04-11 12:16 ` Christoph Hellwig
2023-04-18 6:18 ` Xuan Zhuo
2023-04-19 5:10 ` Christoph Hellwig
2023-04-19 6:16 ` Xuan Zhuo
2023-04-11 7:12 ` Xuan Zhuo
2023-04-11 8:59 ` Jason Wang
2023-04-11 12:17 ` Christoph Hellwig
2023-03-27 4:05 ` [PATCH vhost v6 09/11] virtio_ring: correct the expression of the description of virtqueue_resize() Xuan Zhuo
2023-03-27 4:05 ` [PATCH vhost v6 10/11] virtio_ring: separate the logic of reset/enable from virtqueue_resize Xuan Zhuo
2023-03-27 4:05 ` [PATCH vhost v6 11/11] virtio_ring: introduce virtqueue_reset() Xuan Zhuo
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=1681193477.0633929-1-xuanzhuo@linux.alibaba.com \
--to=xuanzhuo@linux.alibaba.com \
--cc=hch@infradead.org \
--cc=jasowang@redhat.com \
--cc=linux@roeck-us.net \
--cc=mst@redhat.com \
--cc=virtualization@lists.linux-foundation.org \
/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).