Re: [PATCH v12 00/13] Introduce VDUSE - vDPA Device in Userspace

From: Michael S. Tsirkin
Date: Tue Jan 11 2022 - 06:55:05 EST

On Tue, Jan 11, 2022 at 11:31:37AM +0800, Yongji Xie wrote:
> On Mon, Jan 10, 2022 at 11:44 PM Michael S. Tsirkin <mst@xxxxxxxxxx> wrote:
> >
> > On Mon, Jan 10, 2022 at 11:24:40PM +0800, Yongji Xie wrote:
> > > On Mon, Jan 10, 2022 at 11:10 PM Michael S. Tsirkin <mst@xxxxxxxxxx> wrote:
> > > >
> > > > On Mon, Jan 10, 2022 at 09:54:08PM +0800, Yongji Xie wrote:
> > > > > On Mon, Jan 10, 2022 at 8:57 PM Michael S. Tsirkin <mst@xxxxxxxxxx> wrote:
> > > > > >
> > > > > > On Mon, Aug 30, 2021 at 10:17:24PM +0800, Xie Yongji wrote:
> > > > > > > This series introduces a framework that makes it possible to implement
> > > > > > > software-emulated vDPA devices in userspace. And to make the device
> > > > > > > emulation more secure, the emulated vDPA device's control path is handled
> > > > > > > in the kernel and only the data path is implemented in the userspace.
> > > > > > >
> > > > > > > Since the emuldated vDPA device's control path is handled in the kernel,
> > > > > > > a message mechnism is introduced to make userspace be aware of the data
> > > > > > > path related changes. Userspace can use read()/write() to receive/reply
> > > > > > > the control messages.
> > > > > > >
> > > > > > > In the data path, the core is mapping dma buffer into VDUSE daemon's
> > > > > > > address space, which can be implemented in different ways depending on
> > > > > > > the vdpa bus to which the vDPA device is attached.
> > > > > > >
> > > > > > > In virtio-vdpa case, we implements a MMU-based software IOTLB with
> > > > > > > bounce-buffering mechanism to achieve that. And in vhost-vdpa case, the dma
> > > > > > > buffer is reside in a userspace memory region which can be shared to the
> > > > > > > VDUSE userspace processs via transferring the shmfd.
> > > > > > >
> > > > > > > The details and our user case is shown below:
> > > > > > >
> > > > > > > ------------------------ ------------------------- ----------------------------------------------
> > > > > > > | Container | | QEMU(VM) | | VDUSE daemon |
> > > > > > > | --------- | | ------------------- | | ------------------------- ---------------- |
> > > > > > > | |dev/vdx| | | |/dev/vhost-vdpa-x| | | | vDPA device emulation | | block driver | |
> > > > > > > ------------+----------- -----------+------------ -------------+----------------------+---------
> > > > > > > | | | |
> > > > > > > | | | |
> > > > > > > ------------+---------------------------+----------------------------+----------------------+---------
> > > > > > > | | block device | | vhost device | | vduse driver | | TCP/IP | |
> > > > > > > | -------+-------- --------+-------- -------+-------- -----+---- |
> > > > > > > | | | | | |
> > > > > > > | ----------+---------- ----------+----------- -------+------- | |
> > > > > > > | | virtio-blk driver | | vhost-vdpa driver | | vdpa device | | |
> > > > > > > | ----------+---------- ----------+----------- -------+------- | |
> > > > > > > | | virtio bus | | | |
> > > > > > > | --------+----+----------- | | | |
> > > > > > > | | | | | |
> > > > > > > | ----------+---------- | | | |
> > > > > > > | | virtio-blk device | | | | |
> > > > > > > | ----------+---------- | | | |
> > > > > > > | | | | | |
> > > > > > > | -----------+----------- | | | |
> > > > > > > | | virtio-vdpa driver | | | | |
> > > > > > > | -----------+----------- | | | |
> > > > > > > | | | | vdpa bus | |
> > > > > > > | -----------+----------------------+---------------------------+------------ | |
> > > > > > > | ---+--- |
> > > > > > > -----------------------------------------------------------------------------------------| NIC |------
> > > > > > > ---+---
> > > > > > > |
> > > > > > > ---------+---------
> > > > > > > | Remote Storages |
> > > > > > > -------------------
> > > > > > >
> > > > > > > We make use of it to implement a block device connecting to
> > > > > > > our distributed storage, which can be used both in containers and
> > > > > > > VMs. Thus, we can have an unified technology stack in this two cases.
> > > > > > >
> > > > > > > To test it with null-blk:
> > > > > > >
> > > > > > > $ qemu-storage-daemon \
> > > > > > > --chardev socket,id=charmonitor,path=/tmp/qmp.sock,server,nowait \
> > > > > > > --monitor chardev=charmonitor \
> > > > > > > --blockdev driver=host_device,cache.direct=on,aio=native,filename=/dev/nullb0,node-name=disk0 \
> > > > > > > --export type=vduse-blk,id=test,node-name=disk0,writable=on,name=vduse-null,num-queues=16,queue-size=128
> > > > > > >
> > > > > > > The qemu-storage-daemon can be found at https://github.com/bytedance/qemu/tree/vduse
> > > > > >
> > > > > > It's been half a year - any plans to upstream this?
> > > > >
> > > > > Yeah, this is on my to-do list this month.
> > > > >
> > > > > Sorry for taking so long... I've been working on another project
> > > > > enabling userspace RDMA with VDUSE for the past few months. So I
> > > > > didn't have much time for this. Anyway, I will submit the first
> > > > > version as soon as possible.
> > > > >
> > > > > Thanks,
> > > > > Yongji
> > > >
> > > > Oh fun. You mean like virtio-rdma? Or RDMA as a backend for regular
> > > > virtio?
> > > >
> > >
> > > Yes, like virtio-rdma. Then we can develop something like userspace
> > > rxe、siw or custom protocol with VDUSE.
> > >
> > > Thanks,
> > > Yongji
> >
> > Would be interesting to see the spec for that.
>
> Will send it ASAP.
>
> > The issues with RDMA revolved around the fact that current
> > apps tend to either use non-standard propocols for connection
> > establishment or use UD where there's IIRC no standard
> > at all. So QP numbers are hard to virtualize.
> > Similarly many use LIDs directly with the same effect.
> > GUIDs might be virtualizeable but no one went to the effort.
> >
>
> Actually we aimed at emulating a soft RDMA with normal NIC (not use
> RDMA capability) rather than virtualizing a physical RDMA NIC into
> several vRDMA devices. If so, I think we won't have those issues,
> right?

Right, maybe you won't.

> > To say nothing about the interaction with memory overcommit.
> >
>
> I don't get you here. Could you give me more details?
>
> Thanks,
> Yongji

RDMA devices tend to want to pin the memory under DMA.

--
MST