[dpdk-dev] [PATCH v2 1/1] vhost: fix leak of fds and mmaps

[Rich Lane]

On Sun, Jan 17, 2016 at 11:58 PM, Yuanhan Liu <yuanhan.liu at linux.intel.com>
wrote:

> On Sun, Jan 17, 2016 at 11:57:18AM -0800, Rich Lane wrote:
> > The common vhost code only supported a single mmap per device. vhost-user
> > worked around this by saving the address/length/fd of each mmap after
> the end
> > of the rte_virtio_memory struct. This only works if the vhost-user code
> frees
> > dev->mem, since the common code is unaware of the extra info. The
> > VHOST_USER_RESET_OWNER message is one situation where the common code
> frees
> > dev->mem and leaks the fds and mappings. This happens every time I shut
> down a
> > VM.
> >
> > The new code calls back into the implementation (vhost-user or
> vhost-cuse) to
> > clean up these resources.
> >
> > The vhost-cuse changes are only compile tested.
> >
> > Signed-off-by: Rich Lane <rlane at bigswitch.com>
> > ---
> > v1->v2:
> > - Call into vhost-user/vhost-cuse to free mmaps.
> >
> >  lib/librte_vhost/vhost-net.h                  |  6 ++++++
> >  lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 12 ++++++++++++
> >  lib/librte_vhost/vhost_user/vhost-net-user.c  |  1 -
> >  lib/librte_vhost/vhost_user/virtio-net-user.c | 25
> ++++++++++---------------
> >  lib/librte_vhost/vhost_user/virtio-net-user.h |  1 -
> >  lib/librte_vhost/virtio-net.c                 |  8 +-------
> >  6 files changed, 29 insertions(+), 24 deletions(-)
> >
> > diff --git a/lib/librte_vhost/vhost-net.h b/lib/librte_vhost/vhost-net.h
> > index c69b60b..e8d7477 100644
> > --- a/lib/librte_vhost/vhost-net.h
> > +++ b/lib/librte_vhost/vhost-net.h
> > @@ -115,4 +115,10 @@ struct vhost_net_device_ops {
> >
> >
> >  struct vhost_net_device_ops const *get_virtio_net_callbacks(void);
> > +
> > +/*
> > + * Implementation-specific cleanup. Defined by vhost-cuse and
> vhost-user.
> > + */
> > +void vhost_impl_cleanup(struct virtio_net *dev);
>
> TBH, I am not quite like "_impl_"; maybe "_backend_" is better?
>

If you have a strong preference I will change it. Let me know.


> OTOH, what I thought of has slight difference than yours: not
> necessary to export a function, but instead, call the vhost
> backend specific unmap function inside the backend itself. Say,
> call vhost_user_unmap() on RESET_OWNER and connection close.
> What do you think of that?


The munmap must be done after the notify_ops->destroy_device callback. That
means
the backend can't call it before reset_owner() or destroy_device(). The
munmap could
be done afterwards, but that requires saving dev->mem in the caller in the
case of
destroy_device. The cleanest solution is for the vhost common code to ask
the
backend to clean up at the correct time.