commit 2e6d46d7 (vhost: add
vhost_get_features and vhost_ack_features) removes the step that
initializes the acked_features to backend_features.

As this field is now uninitialized, vhost initialization will sometimes
fail.

To fix, initialize acked_features on each ack.
Tested-by: NAndrey Korolyov <andrey@xdel.ru>
Cc: Nikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: NJason Wang <jasowang@redhat.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

commit a9f98bb5 "vhost: multiqueue
support" changed the order of stopping the device. Previously
vhost_dev_stop would disable backend and only afterwards, unset guest
notifiers. We now unset guest notifiers while vhost is still
active. This can lose interrupts causing guest networking to fail. In
particular, this has been observed during migration.

To fix this, several other changes are needed:
- remove the hdev->started assertion in vhost.c since we may want to
start the guest notifiers before vhost starts and stop the guest
notifiers after vhost is stopped.
- introduce the vhost_net_set_vq_index() and call it before setting
guest notifiers. This is to guarantee vhost_net has the correct
virtqueue index when setting guest notifiers.

MST: fix up error handling.

Cc: qemu-stable@nongnu.org
Cc: Jason Wang <jasowang@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Tested-by: NAndrey Korolyov <andrey@xdel.ru>
Reported-by: N"Zhangjie (HZ)" <zhangjie14@huawei.com>
Tested-by: NWilliam Dauchy <william@gandi.net>
Signed-off-by: NJason Wang <jasowang@redhat.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>

Checking vhost device internal state in vhost_net looks like
a layering violation since vhost_net does not
set this flag: it is set and tested by vhost.c.
There seems to be no reason to check this:
caller in virtio net uses its own flag,
vhost_started, to ensure vhost is started/stopped
as appropriate.
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Reviewed-by: NAmos Kong <akong@redhat.com>

As of today, vhost assumes guest and host have the same endianness.
This is definitely not compatible with modern PPC64 and ARM that
can change endianness at runtime. Let's disable vhost-net and print
an error message when we detect such a case:

qemu-system-ppc64: vhost-net does not support cross-endian
qemu-system-ppc64: unable to start vhost net: 38: falling back on userspace virtio

This way users can continue to run VMs without changing their setup and
have a chance to know that performance will be impacted.
Suggested-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NGreg Kurz <gkurz@linux.vnet.ibm.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

The supplied chardev id will be inspected for supported options. Only
a socket backend, with a set path (i.e. a Unix socket) and optionally
the server parameter set, will be allowed. Other options (nowait, telnet)
will make the chardev unusable and the netdev will not be initialised.

Additional checks for validity:
  - requires `-numa node,memdev=..`
  - requires `-device virtio-net-*`

The `vhostforce` option is used to force vhost-net when we deal with
non-MSIX guests.
Signed-off-by: NAntonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Acked-by: NLuiz Capitulino <lcapitulino@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>

Handle the feature bits negotiation when using vhost-user. Allow
the underlying implementation to have a finer control over all the
bits except the VIRTIO_NET_F_MAC.
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Use vhost_set_backend_type to initialise a proper vhost_ops structure.
In vhost_net_init and vhost_net_start_one call conditionally TAP related
initialisation depending on the vhost backend type.
Signed-off-by: NAntonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Decouple vhost from the Linux kernel by introducing vhost_ops. The
intention is to provide different backends - a 'kernel' backend based on
the ioctl interface, and an 'user' backend based on a UNIX domain socket
and shared memory interface.
Signed-off-by: NAntonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

vhost_dev_init will replace devfd and devpath with a single opaque argument.
This is initialised with a file descriptor. When TAP is used (through
vhost_net), open /dev/vhost-net and pass the fd as an opaque parameter in
VhostNetOptions. The same applies to vhost-scsi - open /dev/vhost-scsi and
pass the fd.
Signed-off-by: NAntonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

This decouples virtio-net from the TAP netdev backend and allows support
for other backends to be implemented.
Signed-off-by: NAntonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

The poll callback needs to be called when bringing up or down
the vhost_net instance. As it is not mandatory for an NetClient
to implement it, invoke it only when it is set.
Signed-off-by: NAntonios Motakis <a.motakis@virtualopensystems.com>
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Generalize the features get/ack to be used for both vhost-net and vhost-scsi.
In vhost-net add vhost_net_get_feature_bits to select the feature bit set
depending on the NetClient kind.
Signed-off-by: NNikolay Nikolaev <n.nikolaev@virtualopensystems.com>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

There is no need to access backend->info->has_vnet_hdr() and friends
anymore.  Use the qemu_has_vnet_hdr() API instead.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

With this patch, virtio-net and vmxnet3 frontends make
use of the qemu_peer_* API for backend offloadings manipulations,
instead of calling TAP-specific functions directly.
We also remove the existing checks which prevent those frontends
from using offloadings with backends different from TAP (e.g. netmap).
Signed-off-by: NVincenzo Maffione <v.maffione@gmail.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

cgcc complains that -ENOSYS is not a good value for 'bool'.

A dummy virtio will never have pending queue entries, so let us return
false.
Signed-off-by: NStefan Weil <sw@weilnetz.de>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

This remove virtio-bindings, and use class instead.
Signed-off-by: NKONRAD Frederic <fred.konrad@greensocs.com>
Message-id: 1366791683-5350-6-git-send-email-fred.konrad@greensocs.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Many of these should be cleaned up with proper qdev-/QOM-ification.
Right now there are many catch-all headers in include/hw/ARCH depending
on cpu.h, and this makes it necessary to compile these files per-target.
However, fixing this does not belong in these patches.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Done with this script:

cd hw
for i in `find . -name '*.h' | sed 's/^..//'`; do
  echo '\,^#.*include.*["<]'$i'[">], s,'$i',hw/&,'
done | sed -i -f - `find . -type f`

This is so that paths remain valid as files are moved.

Instead, files in hw/dataplane are referenced with the relative path.
We know they are not going to move to include/, and they are the only
include files that are in subdirectories _and_ move.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Commit 6daf194d and be62a2eb got rid of a bunch, but they keep coming
back.  Tracked down with this Coccinelle semantic patch:

    @r@
	expression err, eno, cls, fmt;
	position p;
    @@
    (
	error_report(fmt, ...)@p
    |
	error_set(err, cls, fmt, ...)@p
    |
	error_set_errno(err, eno, cls, fmt, ...)@p
    |
	error_setg(err, fmt, ...)@p
    |
	error_setg_errno(err, eno, fmt, ...)@p
    )
    @script:python@
	fmt << r.fmt;
	p << r.p;
    @@
    if "\\n" in str(fmt):
	print "%s:%s:%s:%s" % (p[0].file, p[0].line, p[0].column, fmt)
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Message-id: 1360354939-10994-4-git-send-email-armbru@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch lets vhost support multiqueue. The idea is simple, just launching
multiple threads of vhost and let each of vhost thread processing a subset of
the virtqueues of the device. After this change each emulated device can have
multiple vhost threads as its backend.

To do this, a virtqueue index were introduced to record to first virtqueue that
will be handled by this vhost_net device. Based on this and nvqs, vhost could
calculate its relative index to setup vhost_net device.

Since we may have many vhost/net devices for a virtio-net device. The setting of
guest notifiers were moved out of the starting/stopping of a specific vhost
thread. The vhost_net_{start|stop}() were renamed to
vhost_net_{start|stop}_one(), and a new vhost_net_{start|stop}() were introduced
to configure the guest notifiers and start/stop all vhost/vhost_net devices.
Signed-off-by: NJason Wang <jasowang@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Support backend guest notifier masking in vhost-net:
create eventfd at device init, when masked,
make vhost use that as eventfd instead of
sending an interrupt.
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Move public headers to include/net, and leave private headers in net/.
Put the virtio headers in include/net/tap.h, removing the multiple copies
that existed.  Leave include/net/tap.h as the interface for NICs, and
net/tap_int.h as the interface for OS-specific parts of the tap backend.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Modern linux supports arbitrary header size,
which makes it possible to pass mrg buf header
to tap directly without iovec mangling.
Use this capability when it is there.

This removes the need to deal with it in
vhost-net as we do now.
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

The path to /dev/vhost-net is currently hardcoded in vhost_dev_init().
This needs to be changed so that /dev/vhost-scsi can be used.  Pass in
the device path instead of hardcoding it.
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: NNicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Now that VLANClientState has been renamed to NetClientState all 'vc'
local variables should be 'nc'.  Much of the code already used 'nc' but
there are places where 'vc' needs to be renamed.
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: NZhi Yong Wu <wuzhy@linux.vnet.ibm.com>
Reviewed-by: NLaszlo Ersek <lersek@redhat.com>

The vlan feature is no longer part of net core.  Rename VLANClientState
to NetClientState because net clients are not explicitly associated with
a vlan at all, instead they have a peer net client to which they are
connected.

This patch is a mechanical search-and-replace except for a few
whitespace fixups where changing VLANClientState to NetClientState
misaligned whitespace.
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: NZhi Yong Wu <wuzhy@linux.vnet.ibm.com>
Reviewed-by: NLaszlo Ersek <lersek@redhat.com>

NET_CLIENT_TYPE_ -> NET_CLIENT_OPTIONS_KIND_
Signed-off-by: NLaszlo Ersek <lersek@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>

All files under GPLv2 will get GPLv2+ changes starting tomorrow.
event_notifier.c and exec-obsolete.h were only ever touched by Red Hat
employees and can be relicensed now.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

When the vhost notifier is disabled, the userspace handler runs
immediately: virtio_pci_set_host_notifier_internal might
call virtio_queue_notify_vq.
Since the VQ state and the tap backend state aren't
recovered yet, this causes
"Guest moved used index from XXX to YYY" assertions.

The solution is to split out host notifier handling
from vhost VQ setup and disable notifiers as our last step
when we stop vhost-net. For symmetry enable them first thing
on start.
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

qemu_malloc/qemu_free no longer exist after this commit.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

When qemu gets compiled without support of vhost-net, any attempt
to use it fails with a very clear error message:

 qemu-system-x86_64: -netdev ...,vhost=on: vhost-net requested but could not be initialized

there's absolutely no reason given _why_ it coult not be
initialized, and even strace'ing the process in question
does not reveal any errors.  So print a message telling
what's going on.
Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

Add support for event_idx feature, and utilize it to
reduce the number of interrupts and exits for the guest.
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

When MSI is off, each interrupt needs to be bounced through the io
thread when it's set/cleared, so vhost-net causes more context switches and
higher CPU utilization than userspace virtio which handles networking in
the same thread.

We'll need to fix this by adding level irq support in kvm irqfd,
for now disable vhost-net in these configurations.

Added a vhostforce flag to force vhost-net back on.
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

file.index is unsigned, hence 'while (--file.index >= 0)'
will loop > forever. Change to while (file.index-- > 0).
Reported-by: NJes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

use the new tap APIs to set header length
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>

This header is not present on my system and causes a build
failure, but is also not used in these files, so remove it.
Signed-off-by: NMike McCormack <mikem@ring3k.org>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Fix build failure introduced by 0bfcd599

The format statement expects unsigned long on x86_64, but receives
unsigned long long, so gcc exits with an error.
Signed-off-by: NJes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>