Displaying 15 results from an estimated 15 matches for "vhost_vsock_start".
2020 May 01
0
[PATCH v2] vhost: vsock: kick send_pkt worker once device is started
...=========== ============
> connect() --------> vhost_transport_send_pkt_work() initializing
> | vq->private_data==NULL
> | will not be queued
> V
> schedule_timeout(2s)
> vhost_vsock_start() <--------- device ready
> set vq->private_data
>
> wait for 2s and failed
> connect() again vq->private_data!=NULL recv connecting pkt
>
> Details:
> 1. Host userspace sends a connect pkt, at that time, guest vsock is un...
2020 Apr 30
2
[PATCH] vhost: vsock: don't send pkt when vq is not started
...gt; connect() --------> vhost_transport_send_pkt_work() initializing
> > > | vq->private_data==NULL
> > > | will not be queued
> > > V
> > > schedule_timeout(2s)
> > > vhost_vsock_start() <--------- device ready
> > > set vq->private_data
> > >
> > > wait for 2s and failed
> > >
> > > connect() again vq->private_data!=NULL recv connecting pkt
> > >
> > > 1. host us...
2020 Apr 30
2
[PATCH] vhost: vsock: don't send pkt when vq is not started
...gt; connect() --------> vhost_transport_send_pkt_work() initializing
> > > | vq->private_data==NULL
> > > | will not be queued
> > > V
> > > schedule_timeout(2s)
> > > vhost_vsock_start() <--------- device ready
> > > set vq->private_data
> > >
> > > wait for 2s and failed
> > >
> > > connect() again vq->private_data!=NULL recv connecting pkt
> > >
> > > 1. host us...
2020 Apr 30
0
[PATCH] vhost: vsock: don't send pkt when vq is not started
...=========== ============
> connect() --------> vhost_transport_send_pkt_work() initializing
> | vq->private_data==NULL
> | will not be queued
> V
> schedule_timeout(2s)
> vhost_vsock_start() <--------- device ready
> set vq->private_data
>
> wait for 2s and failed
>
> connect() again vq->private_data!=NULL recv connecting pkt
>
> 1. host userspace sends a connect pkt, at that time, guest vsock is under
>...
2020 Apr 30
0
[PATCH] vhost: vsock: don't send pkt when vq is not started
...-> vhost_transport_send_pkt_work() initializing
> > > > | vq->private_data==NULL
> > > > | will not be queued
> > > > V
> > > > schedule_timeout(2s)
> > > > vhost_vsock_start() <--------- device ready
> > > > set vq->private_data
> > > >
> > > > wait for 2s and failed
> > > >
> > > > connect() again vq->private_data!=NULL recv connecting pkt
> > >...
2023 Jun 06
2
[PATCH 1/1] vhost: Fix crash during early vhost_transport_send_pkt calls
...can we
> be sure that for example `work_list` has been initialized?
>
> Maybe I'm overthinking since we didn't have this problem before or the
> race is really short that it never happened.
Yeah, I dropped the READ/WRITE_ONCE use because I didn't think we needed
it for the vhost_vsock_start case, and for the case you mentioned above, I
wondered the same thing as you but was not sure so I added the same
behavior as before. When I read memory-barriers.txt, it sounds like we've
been getting lucky.
I'll add back the READ/WRITE_ONCE for vtsk access since that's what we are
ke...
2023 Jun 06
1
[PATCH 1/1] vhost: Fix crash during early vhost_transport_send_pkt calls
...xample `work_list` has been initialized?
> >
> > Maybe I'm overthinking since we didn't have this problem before or the
> > race is really short that it never happened.
>
> Yeah, I dropped the READ/WRITE_ONCE use because I didn't think we needed
> it for the vhost_vsock_start case, and for the case you mentioned above, I
> wondered the same thing as you but was not sure so I added the same
> behavior as before. When I read memory-barriers.txt, it sounds like we've
> been getting lucky.
Yea READ/WRITE_ONCE is one of these things. Once you start adding
them...
2023 Jun 06
1
[PATCH 1/1] vhost: Fix crash during early vhost_transport_send_pkt calls
On Mon, Jun 05, 2023 at 01:57:30PM -0500, Mike Christie wrote:
>If userspace does VHOST_VSOCK_SET_GUEST_CID before VHOST_SET_OWNER we
>can race where:
>1. thread0 calls vhost_transport_send_pkt -> vhost_work_queue
>2. thread1 does VHOST_SET_OWNER which calls vhost_worker_create.
>3. vhost_worker_create will set the dev->worker pointer before setting
>the worker->vtsk
2018 Apr 09
2
kernel BUG at drivers/vhost/vhost.c:LINE! (2)
...> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> vhost_update_used_flags+0x3af/0x4a0 drivers/vhost/vhost.c:1723
> vhost_vq_init_access+0x117/0x590 drivers/vhost/vhost.c:1763
> vhost_vsock_start drivers/vhost/vsock.c:446 [inline]
> vhost_vsock_dev_ioctl+0x751/0x920 drivers/vhost/vsock.c:678
> vfs_ioctl fs/ioctl.c:46 [inline]
> file_ioctl fs/ioctl.c:500 [inline]
> do_vfs_ioctl+0x1cf/0x1650 fs/ioctl.c:684
> ksys_ioctl+0xa9/0xd0 fs/ioctl.c:701
> SYSC_ioctl fs/ioctl.c:7...
2018 Apr 09
2
kernel BUG at drivers/vhost/vhost.c:LINE! (2)
...> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> vhost_update_used_flags+0x3af/0x4a0 drivers/vhost/vhost.c:1723
> vhost_vq_init_access+0x117/0x590 drivers/vhost/vhost.c:1763
> vhost_vsock_start drivers/vhost/vsock.c:446 [inline]
> vhost_vsock_dev_ioctl+0x751/0x920 drivers/vhost/vsock.c:678
> vfs_ioctl fs/ioctl.c:46 [inline]
> file_ioctl fs/ioctl.c:500 [inline]
> do_vfs_ioctl+0x1cf/0x1650 fs/ioctl.c:684
> ksys_ioctl+0xa9/0xd0 fs/ioctl.c:701
> SYSC_ioctl fs/ioctl.c:7...
2016 Apr 01
7
[RFC v5 0/5] Add virtio transport for AF_VSOCK
This series is based on Michael Tsirkin's vhost branch (v4.5-rc6).
I'm about to process Claudio Imbrenda's locking fixes for virtio-vsock but
first I want to share the latest version of the code. Several people are
playing with vsock now so sharing the latest code should avoid duplicate work.
v5:
* Transport reset event for live migration support
* Reorder virtqueues, drop unused
2016 Apr 01
7
[RFC v5 0/5] Add virtio transport for AF_VSOCK
This series is based on Michael Tsirkin's vhost branch (v4.5-rc6).
I'm about to process Claudio Imbrenda's locking fixes for virtio-vsock but
first I want to share the latest version of the code. Several people are
playing with vsock now so sharing the latest code should avoid duplicate work.
v5:
* Transport reset event for live migration support
* Reorder virtqueues, drop unused
2016 Jul 28
6
[RFC v6 0/6] Add virtio transport for AF_VSOCK
This series is based on v4.7.
This RFC is the implementation for the new VIRTIO Socket device. It is
developed in parallel with the VIRTIO device specification and proves the
design. Once the specification has been accepted I will send a non-RFC version
of this patch series.
v6:
* Add VHOST_VSOCK_SET_RUNNING ioctl to start/stop vhost cleanly
* Add graceful shutdown to avoid port reuse while
2016 Jul 28
6
[RFC v6 0/6] Add virtio transport for AF_VSOCK
This series is based on v4.7.
This RFC is the implementation for the new VIRTIO Socket device. It is
developed in parallel with the VIRTIO device specification and proves the
design. Once the specification has been accepted I will send a non-RFC version
of this patch series.
v6:
* Add VHOST_VSOCK_SET_RUNNING ioctl to start/stop vhost cleanly
* Add graceful shutdown to avoid port reuse while
2023 Mar 28
12
[PATCH v6 00/11] vhost: multiple worker support
The following patches were built over linux-next which contains various
vhost patches in mst's tree and the vhost_task patchset in Christian
Brauner's tree:
git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git
kernel.user_worker branch:
https://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git/log/?h=kernel.user_worker
The latter patchset handles the review comment