All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: zhenwei pi <pizhenwei@bytedance.com>
Cc: xuanzhuo@linux.alibaba.com, linux-kernel@vger.kernel.org,
	virtualization@lists.linux-foundation.org,
	Christoph Hellwig <hch@infradead.org>,
	stefanha@redhat.com
Subject: Re: Re: [PATCH v2 1/2] virtio: abstract virtqueue related methods
Date: Wed, 17 May 2023 06:39:43 -0400	[thread overview]
Message-ID: <20230517062401-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <cc9f3aea-b8bc-f312-41bb-3fd68083aa99@bytedance.com>

On Wed, May 17, 2023 at 04:35:55PM +0800, zhenwei pi wrote:
> 
> 
> On 5/17/23 15:46, Christoph Hellwig wrote:
> > On Wed, May 17, 2023 at 03:43:03PM +0800, zhenwei pi wrote:
> > > I have a plan to introduce 'Virtio Over Fabrics'(TCP&RDMA) as Virtio
> > > transport, as mentioned in cover letter of this series:
> > > 3 weeks ago, I posted a proposal 'Virtio Over Fabrics':
> > > https://lists.oasis-open.org/archives/virtio-comment/202304/msg00442.html
> > 
> > Just don't do it.  Please define your own protocols over RDMA or TCP
> > for exactly the operations you need (for many they will already exist)
> > instead of piggyg backing on virtio and making everyone else pay the
> > price.
> > 
> 
> Hi
> 
> 1, `virtqueue_add_inbuf` in current version:
> static inline int virtqueue_add_inbuf(struct virtqueue *vq,
>                                       struct scatterlist *sg,
>                                       unsigned int num,
>                                       void *data,
>                                       gfp_t gfp)
> {
>         if (likely(!vq->abstract))
>                 return vring_virtqueue_add_sgs(vq, &sg, num, 0, 1, data,
> NULL, gfp);
> 
>         return vq->add_sgs(vq, &sg, num, 0, 1, data, NULL, gfp);
> }
> 
> And disassemble 'virtinput_queue_evtbuf':
> static void virtinput_queue_evtbuf(struct virtio_input *vi,
>                                    struct virtio_input_event *evtbuf)
> {
>         struct scatterlist sg[1];
> 
>         sg_init_one(sg, evtbuf, sizeof(*evtbuf));
>         virtqueue_add_inbuf(vi->evt, sg, 1, evtbuf, GFP_ATOMIC);
> }
> 
> I notice that two instructions are newly added for vring like:
>  24d:   80 78 35 00             cmpb   $0x0,0x35(%rax)
>  251:   75 3f                   jne    292
> 
> Is it an expensive price...

Can we somehow only override the kick method?
Then take the ring and send it over ...


> 2, Storage/FS specific remote protocol is quite popular, otherwise I'm not
> familiar with other device protocols. For example, I need a remote crypto
> device to accelerate HTTPS ... With Virtio Over Fabrics, I have a chance to
> attach a virtio-crypto device to do this work.
> 
> -- 
> zhenwei pi

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

WARNING: multiple messages have this Message-ID (diff)
From: "Michael S. Tsirkin" <mst@redhat.com>
To: zhenwei pi <pizhenwei@bytedance.com>
Cc: Christoph Hellwig <hch@infradead.org>,
	stefanha@redhat.com, jasowang@redhat.com,
	xuanzhuo@linux.alibaba.com,
	virtualization@lists.linux-foundation.org,
	linux-kernel@vger.kernel.org
Subject: Re: Re: [PATCH v2 1/2] virtio: abstract virtqueue related methods
Date: Wed, 17 May 2023 06:39:43 -0400	[thread overview]
Message-ID: <20230517062401-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <cc9f3aea-b8bc-f312-41bb-3fd68083aa99@bytedance.com>

On Wed, May 17, 2023 at 04:35:55PM +0800, zhenwei pi wrote:
> 
> 
> On 5/17/23 15:46, Christoph Hellwig wrote:
> > On Wed, May 17, 2023 at 03:43:03PM +0800, zhenwei pi wrote:
> > > I have a plan to introduce 'Virtio Over Fabrics'(TCP&RDMA) as Virtio
> > > transport, as mentioned in cover letter of this series:
> > > 3 weeks ago, I posted a proposal 'Virtio Over Fabrics':
> > > https://lists.oasis-open.org/archives/virtio-comment/202304/msg00442.html
> > 
> > Just don't do it.  Please define your own protocols over RDMA or TCP
> > for exactly the operations you need (for many they will already exist)
> > instead of piggyg backing on virtio and making everyone else pay the
> > price.
> > 
> 
> Hi
> 
> 1, `virtqueue_add_inbuf` in current version:
> static inline int virtqueue_add_inbuf(struct virtqueue *vq,
>                                       struct scatterlist *sg,
>                                       unsigned int num,
>                                       void *data,
>                                       gfp_t gfp)
> {
>         if (likely(!vq->abstract))
>                 return vring_virtqueue_add_sgs(vq, &sg, num, 0, 1, data,
> NULL, gfp);
> 
>         return vq->add_sgs(vq, &sg, num, 0, 1, data, NULL, gfp);
> }
> 
> And disassemble 'virtinput_queue_evtbuf':
> static void virtinput_queue_evtbuf(struct virtio_input *vi,
>                                    struct virtio_input_event *evtbuf)
> {
>         struct scatterlist sg[1];
> 
>         sg_init_one(sg, evtbuf, sizeof(*evtbuf));
>         virtqueue_add_inbuf(vi->evt, sg, 1, evtbuf, GFP_ATOMIC);
> }
> 
> I notice that two instructions are newly added for vring like:
>  24d:   80 78 35 00             cmpb   $0x0,0x35(%rax)
>  251:   75 3f                   jne    292
> 
> Is it an expensive price...

Can we somehow only override the kick method?
Then take the ring and send it over ...


> 2, Storage/FS specific remote protocol is quite popular, otherwise I'm not
> familiar with other device protocols. For example, I need a remote crypto
> device to accelerate HTTPS ... With Virtio Over Fabrics, I have a chance to
> attach a virtio-crypto device to do this work.
> 
> -- 
> zhenwei pi


  reply	other threads:[~2023-05-17 10:39 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-17  2:54 [PATCH v2 0/2] virtio: abstract virtqueue related methods zhenwei pi via Virtualization
2023-05-17  2:54 ` zhenwei pi
2023-05-17  2:54 ` [PATCH v2 1/2] " zhenwei pi via Virtualization
2023-05-17  2:54   ` zhenwei pi
2023-05-17  6:57   ` kernel test robot
2023-05-17  6:57     ` kernel test robot
2023-05-17  7:10   ` kernel test robot
2023-05-17  7:10     ` kernel test robot
2023-05-17  7:39   ` Christoph Hellwig
2023-05-17  7:39     ` Christoph Hellwig
2023-05-17  7:43     ` zhenwei pi via Virtualization
2023-05-17  7:43       ` zhenwei pi
2023-05-17  7:46       ` Christoph Hellwig
2023-05-17  7:46         ` Christoph Hellwig
2023-05-17  8:35         ` zhenwei pi via Virtualization
2023-05-17  8:35           ` zhenwei pi
2023-05-17 10:39           ` Michael S. Tsirkin [this message]
2023-05-17 10:39             ` Michael S. Tsirkin
2023-05-18  0:47             ` zhenwei pi via Virtualization
2023-05-18  0:47               ` zhenwei pi
2023-05-18 10:09               ` Michael S. Tsirkin
2023-05-18 10:09                 ` Michael S. Tsirkin
2023-05-18 11:01                 ` zhenwei pi via Virtualization
2023-05-18 11:01                   ` zhenwei pi
2023-05-17 17:59   ` kernel test robot
2023-05-17 17:59     ` kernel test robot
2023-05-17  2:54 ` [PATCH v2 2/2] tools/virtio: implement virtqueue in test zhenwei pi via Virtualization
2023-05-17  2:54   ` zhenwei pi
2023-05-17  3:46 ` [PATCH v2 0/2] virtio: abstract virtqueue related methods Michael S. Tsirkin
2023-05-17  3:46   ` Michael S. Tsirkin
2023-05-17  3:51   ` zhenwei pi via Virtualization
2023-05-17  3:51     ` zhenwei pi
2023-05-17  3:57     ` Michael S. Tsirkin
2023-05-17  3:57       ` Michael S. Tsirkin
2023-05-17  4:58       ` zhenwei pi via Virtualization
2023-05-17  4:58         ` zhenwei pi
2023-05-17  6:10         ` Michael S. Tsirkin
2023-05-17  6:10           ` Michael S. Tsirkin
2023-05-17  6:21           ` zhenwei pi via Virtualization
2023-05-17  6:21             ` zhenwei pi
2023-05-17 10:13             ` Michael S. Tsirkin
2023-05-17 10:13               ` Michael S. Tsirkin

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=20230517062401-mutt-send-email-mst@kernel.org \
    --to=mst@redhat.com \
    --cc=hch@infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pizhenwei@bytedance.com \
    --cc=stefanha@redhat.com \
    --cc=virtualization@lists.linux-foundation.org \
    --cc=xuanzhuo@linux.alibaba.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.