Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-28 Thread Michael S. Tsirkin
On Thu, Feb 21, 2013 at 11:02:31AM -0600, Anthony Liguori wrote: > "Michael S. Tsirkin" writes: > > > On Thu, Feb 21, 2013 at 10:15:58AM -0600, Anthony Liguori wrote: > >> "Michael S. Tsirkin" writes: > >> > >> > On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: > >> >> This is a em

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-27 Thread Paolo Bonzini
Il 27/02/2013 13:08, Markus Armbruster ha scritto: >>> >> The main blocker to this series seems to be hubs, because they interact >>> >> with multiple NetClients and thus could span multiple AioContexts. >>> >> Adding proper locking there is going to be interesting. :) >> > >> > I think folks are p

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-27 Thread Markus Armbruster
Stefan Hajnoczi writes: > On Mon, Feb 25, 2013 at 06:53:04PM +0100, Paolo Bonzini wrote: >> Il 25/02/2013 18:35, mdroth ha scritto: >> > Yup, don't mean to get ahead of things, my main interest is just in how >> > we might deal with the interaction between NetClients and virtio-net >> > dataplane

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-27 Thread Michael S. Tsirkin
On Wed, Feb 27, 2013 at 05:41:40PM +0800, liu ping fan wrote: > On Fri, Feb 22, 2013 at 12:51 AM, Michael S. Tsirkin wrote: > > On Thu, Feb 21, 2013 at 06:45:54PM +0200, Michael S. Tsirkin wrote: > >> On Thu, Feb 21, 2013 at 10:15:58AM -0600, Anthony Liguori wrote: > >> > "Michael S. Tsirkin" wri

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-27 Thread liu ping fan
On Fri, Feb 22, 2013 at 12:51 AM, Michael S. Tsirkin wrote: > On Thu, Feb 21, 2013 at 06:45:54PM +0200, Michael S. Tsirkin wrote: >> On Thu, Feb 21, 2013 at 10:15:58AM -0600, Anthony Liguori wrote: >> > "Michael S. Tsirkin" writes: >> > >> > > On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fa

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-27 Thread liu ping fan
On Tue, Feb 26, 2013 at 1:53 AM, Paolo Bonzini wrote: > Il 25/02/2013 18:35, mdroth ha scritto: >>> > Moving more of the os_host_main_loop_wait to AioContext would be >>> > possible (timers are on the todo list, in fact), but we should only do >>> > it as need arises. >> Were you planning on hangi

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-26 Thread Stefan Hajnoczi
On Mon, Feb 25, 2013 at 06:53:04PM +0100, Paolo Bonzini wrote: > Il 25/02/2013 18:35, mdroth ha scritto: > > Yup, don't mean to get ahead of things, my main interest is just in how > > we might deal with the interaction between NetClients and virtio-net > > dataplane threads without introducing ad-

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-25 Thread Paolo Bonzini
Il 25/02/2013 18:35, mdroth ha scritto: >> > Moving more of the os_host_main_loop_wait to AioContext would be >> > possible (timers are on the todo list, in fact), but we should only do >> > it as need arises. > Were you planning on hanging another GSource off of AioContext to handle > timers? No,

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-25 Thread mdroth
On Fri, Feb 22, 2013 at 09:53:11AM +0100, Paolo Bonzini wrote: > Il 22/02/2013 00:38, mdroth ha scritto: > > On Thu, Feb 21, 2013 at 11:12:01PM +0100, Paolo Bonzini wrote: > >> Il 21/02/2013 22:07, mdroth ha scritto: > > > > 100% agree. In particular hw/dataplane/event-poll.c should be the

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-22 Thread Paolo Bonzini
Il 22/02/2013 12:06, Kevin Wolf ha scritto: > or example think of > the read-only qcow2 snapshot BDS that we've been talking about before. > > So it would have to be more like one AioContext per connected component > in the BDS graph Yes. For now it means one AioContext per top-level BDS, but th

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-22 Thread Paolo Bonzini
Il 22/02/2013 11:43, Stefan Hajnoczi ha scritto: >> > >> > It can and should stay. :) BHs are tied to an AioContext. Running >> > thread-creation BHs from the AioContext's main thread ensures that the >> > worker threads inherit the right attributes for e.g. real-time. > When I say "this can sta

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-22 Thread Kevin Wolf
Am 21.02.2013 um 18:05 hat Paolo Bonzini geschrieben: > Il 21/02/2013 17:33, Stefan Hajnoczi ha scritto: > > * BlockDriverState <-> AioContext > > > >It probably makes sense to bind a BlockDriverState to an AioContext > >in which its file descriptors and aio activity happens. > > ... and

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-22 Thread Stefan Hajnoczi
On Thu, Feb 21, 2013 at 05:38:04PM -0600, mdroth wrote: > On Thu, Feb 21, 2013 at 11:12:01PM +0100, Paolo Bonzini wrote: > > Il 21/02/2013 22:07, mdroth ha scritto: > > >> > > > >> > 100% agree. In particular hw/dataplane/event-poll.c should be the > > >> > first > > >> > to go away, but AioCont

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-22 Thread Stefan Hajnoczi
On Thu, Feb 21, 2013 at 06:05:44PM +0100, Paolo Bonzini wrote: > Il 21/02/2013 17:33, Stefan Hajnoczi ha scritto: > > This especially means things like adding live > > migration support or solving other limitations. The limitations are > > there to motivate core QEMU refactoring, so that core QEMU

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-22 Thread Paolo Bonzini
Il 22/02/2013 00:38, mdroth ha scritto: > On Thu, Feb 21, 2013 at 11:12:01PM +0100, Paolo Bonzini wrote: >> Il 21/02/2013 22:07, mdroth ha scritto: > > 100% agree. In particular hw/dataplane/event-poll.c should be the first > to go away, but AioContext provides the functionality that P

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread mdroth
On Thu, Feb 21, 2013 at 11:12:01PM +0100, Paolo Bonzini wrote: > Il 21/02/2013 22:07, mdroth ha scritto: > >> > > >> > 100% agree. In particular hw/dataplane/event-poll.c should be the first > >> > to go away, but AioContext provides the functionality that Ping Fan > >> > needs. But hw/dataplane

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Paolo Bonzini
Il 21/02/2013 22:07, mdroth ha scritto: >> > >> > 100% agree. In particular hw/dataplane/event-poll.c should be the first >> > to go away, but AioContext provides the functionality that Ping Fan >> > needs. But hw/dataplane/vring.c will probably be here for a longer > Has there been any discussi

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread mdroth
On Thu, Feb 21, 2013 at 06:05:44PM +0100, Paolo Bonzini wrote: > Il 21/02/2013 17:33, Stefan Hajnoczi ha scritto: > > Interesting patch series. I want to share my thoughts on the status of > > dataplane and what I'm currently working on. There will probably be > > some overlap that we can coordin

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Michael S. Tsirkin
On Thu, Feb 21, 2013 at 06:45:54PM +0200, Michael S. Tsirkin wrote: > On Thu, Feb 21, 2013 at 10:15:58AM -0600, Anthony Liguori wrote: > > "Michael S. Tsirkin" writes: > > > > > On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: > > >> This is a emulation to virtio-blk dataplane, which

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Michael S. Tsirkin
On Thu, Feb 21, 2013 at 11:02:31AM -0600, Anthony Liguori wrote: > "Michael S. Tsirkin" writes: > > > On Thu, Feb 21, 2013 at 10:15:58AM -0600, Anthony Liguori wrote: > >> "Michael S. Tsirkin" writes: > >> > >> > On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: > >> >> This is a em

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Paolo Bonzini
Il 21/02/2013 17:33, Stefan Hajnoczi ha scritto: > Interesting patch series. I want to share my thoughts on the status of > dataplane and what I'm currently working on. There will probably be > some overlap that we can coordinate. > > First, I want to eventually delete hw/dataplane/ :). That is

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Anthony Liguori
"Michael S. Tsirkin" writes: > On Thu, Feb 21, 2013 at 10:15:58AM -0600, Anthony Liguori wrote: >> "Michael S. Tsirkin" writes: >> >> > On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: >> >> This is a emulation to virtio-blk dataplane, which push the data >> >> handling out of bigl

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Michael S. Tsirkin
On Thu, Feb 21, 2013 at 10:15:58AM -0600, Anthony Liguori wrote: > "Michael S. Tsirkin" writes: > > > On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: > >> This is a emulation to virtio-blk dataplane, which push the data > >> handling out of biglock. And it is a try to implement this

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Stefan Hajnoczi
On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: > This is a emulation to virtio-blk dataplane, which push the data > handling out of biglock. And it is a try to implement this process > in userspace, while vhost-net in kernel. > > The iperf's result show it improving the perfermance

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Anthony Liguori
"Michael S. Tsirkin" writes: > On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: >> This is a emulation to virtio-blk dataplane, which push the data >> handling out of biglock. And it is a try to implement this process >> in userspace, while vhost-net in kernel. > > What's the motivat

Re: [Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Michael S. Tsirkin
On Thu, Feb 21, 2013 at 08:54:44PM +0800, Liu Ping Fan wrote: > This is a emulation to virtio-blk dataplane, which push the data > handling out of biglock. And it is a try to implement this process > in userspace, while vhost-net in kernel. What's the motivation for doing this? Speed up the slirp

[Qemu-devel] [PATCH 0/9] introduce virtio net dataplane

2013-02-21 Thread Liu Ping Fan
This is a emulation to virtio-blk dataplane, which push the data handling out of biglock. And it is a try to implement this process in userspace, while vhost-net in kernel. The iperf's result show it improving the perfermance of base line, but still slower than vhost-net (maybe the rx path need op