Next patches in this series will no longer perform an
immediate poll and check of the device's used buffers
for each CVQ state load command. Instead, they will
send CVQ state load commands in parallel by polling
multiple pending buffers at once.
To achieve this, this patch refactoring vhost_svq_poll()
to accept a new argument `num`, which allows vhost_svq_poll()
to wait for the device to use multiple elements,
rather than polling for a single element.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <950b3bfcfc5d446168b9d6a249d554a013a691d4.1693287885.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Now that we have add migration blockers if the device does not support
all the needed features, remove the general blocker applied to all net
devices with CVQ.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20230822085330.3978829-6-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Doing that way allows CVQ to be enabled before the dataplane vqs,
restoring the state as MQ or MAC addresses properly in the case of a
migration.
The patch does it by defining a ->load NetClientInfo callback also for
dataplane. Ideally, this should be done by an independent patch, but
the function is already static so it would only add an empty
vhost_vdpa_net_data_load stub.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230822085330.3978829-5-eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Next patches will add the corresponding data load.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20230822085330.3978829-4-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Previous to this patch the only way CVQ would be shadowed is if it does
support to isolate CVQ group or if all vqs were shadowed from the
beginning. The second condition was checked at the beginning, and no
more configuration was done.
After this series we need to check if data queues are shadowed because
they are in the middle of the migration. As checking if they are
shadowed already covers the previous case, let's just mimic it.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20230822085330.3978829-2-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Enable SVQ with VIRTIO_NET_F_CTRL_VLAN feature.
Co-developed-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <38dc63102a42c31c72fd293d0e6e2828fd54c86e.1690106284.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch introduces vhost_vdpa_net_load_single_vlan()
and vhost_vdpa_net_load_vlan() to restore the vlan
filtering state at device's startup.
Co-developed-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <e76a29f77bb3f386e4a643c8af94b77b775d1752.1690106284.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Use a heap allocation instead of a variable length array in
tap_receive_iov().
The codebase has very few VLAs, and if we can get rid of them all we
can make the compiler error on new additions. This is a defensive
measure against security bugs where an on-stack dynamic allocation
isn't correctly size-checked (e.g. CVE-2021-3527).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Use a g_autofree heap allocation instead of a variable length
array in dump_receive_iov().
The codebase has very few VLAs, and if we can get rid of them all we
can make the compiler error on new additions. This is a defensive
measure against security bugs where an on-stack dynamic allocation
isn't correctly size-checked (e.g. CVE-2021-3527).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
AF_XDP is a network socket family that allows communication directly
with the network device driver in the kernel, bypassing most or all
of the kernel networking stack. In the essence, the technology is
pretty similar to netmap. But, unlike netmap, AF_XDP is Linux-native
and works with any network interfaces without driver modifications.
Unlike vhost-based backends (kernel, user, vdpa), AF_XDP doesn't
require access to character devices or unix sockets. Only access to
the network interface itself is necessary.
This patch implements a network backend that communicates with the
kernel by creating an AF_XDP socket. A chunk of userspace memory
is shared between QEMU and the host kernel. 4 ring buffers (Tx, Rx,
Fill and Completion) are placed in that memory along with a pool of
memory buffers for the packet data. Data transmission is done by
allocating one of the buffers, copying packet data into it and
placing the pointer into Tx ring. After transmission, device will
return the buffer via Completion ring. On Rx, device will take
a buffer form a pre-populated Fill ring, write the packet data into
it and place the buffer into Rx ring.
AF_XDP network backend takes on the communication with the host
kernel and the network interface and forwards packets to/from the
peer device in QEMU.
Usage example:
-device virtio-net-pci,netdev=guest1,mac=00:16:35:AF:AA:5C
-netdev af-xdp,ifname=ens6f1np1,id=guest1,mode=native,queues=1
XDP program bridges the socket with a network interface. It can be
attached to the interface in 2 different modes:
1. skb - this mode should work for any interface and doesn't require
driver support. With a caveat of lower performance.
2. native - this does require support from the driver and allows to
bypass skb allocation in the kernel and potentially use
zero-copy while getting packets in/out userspace.
By default, QEMU will try to use native mode and fall back to skb.
Mode can be forced via 'mode' option. To force 'copy' even in native
mode, use 'force-copy=on' option. This might be useful if there is
some issue with the driver.
Option 'queues=N' allows to specify how many device queues should
be open. Note that all the queues that are not open are still
functional and can receive traffic, but it will not be delivered to
QEMU. So, the number of device queues should generally match the
QEMU configuration, unless the device is shared with something
else and the traffic re-direction to appropriate queues is correctly
configured on a device level (e.g. with ethtool -N).
'start-queue=M' option can be used to specify from which queue id
QEMU should start configuring 'N' queues. It might also be necessary
to use this option with certain NICs, e.g. MLX5 NICs. See the docs
for examples.
In a general case QEMU will need CAP_NET_ADMIN and CAP_SYS_ADMIN
or CAP_BPF capabilities in order to load default XSK/XDP programs to
the network interface and configure BPF maps. It is possible, however,
to run with no capabilities. For that to work, an external process
with enough capabilities will need to pre-load default XSK program,
create AF_XDP sockets and pass their file descriptors to QEMU process
on startup via 'sock-fds' option. Network backend will need to be
configured with 'inhibit=on' to avoid loading of the program.
QEMU will need 32 MB of locked memory (RLIMIT_MEMLOCK) per queue
or CAP_IPC_LOCK.
There are few performance challenges with the current network backends.
First is that they do not support IO threads. This means that data
path is handled by the main thread in QEMU and may slow down other
work or may be slowed down by some other work. This also means that
taking advantage of multi-queue is generally not possible today.
Another thing is that data path is going through the device emulation
code, which is not really optimized for performance. The fastest
"frontend" device is virtio-net. But it's not optimized for heavy
traffic either, because it expects such use-cases to be handled via
some implementation of vhost (user, kernel, vdpa). In practice, we
have virtio notifications and rcu lock/unlock on a per-packet basis
and not very efficient accesses to the guest memory. Communication
channels between backend and frontend devices do not allow passing
more than one packet at a time as well.
Some of these challenges can be avoided in the future by adding better
batching into device emulation or by implementing vhost-af-xdp variant.
There are also a few kernel limitations. AF_XDP sockets do not
support any kinds of checksum or segmentation offloading. Buffers
are limited to a page size (4K), i.e. MTU is limited. Multi-buffer
support implementation for AF_XDP is in progress, but not ready yet.
Also, transmission in all non-zero-copy modes is synchronous, i.e.
done in a syscall. That doesn't allow high packet rates on virtual
interfaces.
However, keeping in mind all of these challenges, current implementation
of the AF_XDP backend shows a decent performance while running on top
of a physical NIC with zero-copy support.
Test setup:
2 VMs running on 2 physical hosts connected via ConnectX6-Dx card.
Network backend is configured to open the NIC directly in native mode.
The driver supports zero-copy. NIC is configured to use 1 queue.
Inside a VM - iperf3 for basic TCP performance testing and dpdk-testpmd
for PPS testing.
iperf3 result:
TCP stream : 19.1 Gbps
dpdk-testpmd (single queue, single CPU core, 64 B packets) results:
Tx only : 3.4 Mpps
Rx only : 2.0 Mpps
L2 FWD Loopback : 1.5 Mpps
In skb mode the same setup shows much lower performance, similar to
the setup where pair of physical NICs is replaced with veth pair:
iperf3 result:
TCP stream : 9 Gbps
dpdk-testpmd (single queue, single CPU core, 64 B packets) results:
Tx only : 1.2 Mpps
Rx only : 1.0 Mpps
L2 FWD Loopback : 0.7 Mpps
Results in skb mode or over the veth are close to results of a tap
backend with vhost=on and disabled segmentation offloading bridged
with a NIC.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> (docker/lcitool)
Signed-off-by: Jason Wang <jasowang@redhat.com>
New features are subject to check with vhost-user and vdpa.
Signed-off-by: Yuri Benditovich <yuri.benditovich@daynix.com>
Signed-off-by: Andrew Melnychenko <andrew@daynix.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Tap indicates support for USO features according to
capabilities of current kernel module.
Signed-off-by: Yuri Benditovich <yuri.benditovich@daynix.com>
Signed-off-by: Andrew Melnychecnko <andrew@daynix.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Passing additional parameters (USOv4 and USOv6 offloads) when
setting TAP offloads
Signed-off-by: Yuri Benditovich <yuri.benditovich@daynix.com>
Signed-off-by: Andrew Melnychenko <andrew@daynix.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
qemu 8.1.0 breaks on illumos platforms due to _XOPEN_SOURCE and others no longer being set correctly, leading to breakage such as:
https://us-central.manta.mnx.io/pkgsrc/public/reports/trunk/tools/20230908.1404/qemu-8.1.0/build.log
This is a result of meson conversion which incorrectly matches against 'solaris' instead of 'sunos' for uname.
First time submitting a patch here, hope I did it correctly. Thanks.
Signed-off-by: Jonathan Perkin <jonathan@perkin.org.uk>
Message-ID: <ZPtdxtum9UVPy58J@perkin.org.uk>
Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
* target/i386: fix BQL handling of the legacy FERR interrupts
* target/i386: fix memory operand size for CVTPS2PD
* target/i386: Add support for AMX-COMPLEX in CPUID enumeration
* compile plugins on Darwin
* configure and meson cleanups
* drop mkvenv support for Python 3.7 and Debian10
* add wrap file for libblkio
* tweak KVM stubs
-----BEGIN PGP SIGNATURE-----
iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmT5t6UUHHBib256aW5p
QHJlZGhhdC5jb20ACgkQv/vSX3jHroMmjwf+MpvVuq+nn+3PqGUXgnzJx5ccA5ne
O9Xy8+1GdlQPzBw/tPovxXDSKn3HQtBfxObn2CCE1tu/4uHWpBA1Vksn++NHdUf2
P0yoHxGskJu5iYYTtIcNw5cH2i+AizdiXuEjhfNjqD5Y234cFoHnUApt9e3zBvVO
cwGD7WpPuSb4g38hHkV6nKcx72o7b4ejDToqUVZJ2N+RkddSqB03fSdrOru0hR7x
V+lay0DYdFszNDFm05LJzfDbcrHuSryGA91wtty7Fzj6QhR/HBHQCUZJxMB5PI7F
Zy4Zdpu60zxtSxUqeKgIi7UhNFgMcax2Hf9QEqdc/B4ARoBbboh4q4u8kQ==
=dH7/
-----END PGP SIGNATURE-----
Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging
* only build util/async-teardown.c when system build is requested
* target/i386: fix BQL handling of the legacy FERR interrupts
* target/i386: fix memory operand size for CVTPS2PD
* target/i386: Add support for AMX-COMPLEX in CPUID enumeration
* compile plugins on Darwin
* configure and meson cleanups
* drop mkvenv support for Python 3.7 and Debian10
* add wrap file for libblkio
* tweak KVM stubs
# -----BEGIN PGP SIGNATURE-----
#
# iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmT5t6UUHHBib256aW5p
# QHJlZGhhdC5jb20ACgkQv/vSX3jHroMmjwf+MpvVuq+nn+3PqGUXgnzJx5ccA5ne
# O9Xy8+1GdlQPzBw/tPovxXDSKn3HQtBfxObn2CCE1tu/4uHWpBA1Vksn++NHdUf2
# P0yoHxGskJu5iYYTtIcNw5cH2i+AizdiXuEjhfNjqD5Y234cFoHnUApt9e3zBvVO
# cwGD7WpPuSb4g38hHkV6nKcx72o7b4ejDToqUVZJ2N+RkddSqB03fSdrOru0hR7x
# V+lay0DYdFszNDFm05LJzfDbcrHuSryGA91wtty7Fzj6QhR/HBHQCUZJxMB5PI7F
# Zy4Zdpu60zxtSxUqeKgIi7UhNFgMcax2Hf9QEqdc/B4ARoBbboh4q4u8kQ==
# =dH7/
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 07 Sep 2023 07:44:37 EDT
# gpg: using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg: issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg: aka "Paolo Bonzini <pbonzini@redhat.com>" [full]
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4 E2F7 7E15 100C CD36 69B1
# Subkey fingerprint: F133 3857 4B66 2389 866C 7682 BFFB D25F 78C7 AE83
* tag 'for-upstream' of https://gitlab.com/bonzini/qemu: (51 commits)
docs/system/replay: do not show removed command line option
subprojects: add wrap file for libblkio
sysemu/kvm: Restrict kvm_pc_setup_irq_routing() to x86 targets
sysemu/kvm: Restrict kvm_has_pit_state2() to x86 targets
sysemu/kvm: Restrict kvm_get_apic_state() to x86 targets
sysemu/kvm: Restrict kvm_arch_get_supported_cpuid/msr() to x86 targets
target/i386: Restrict declarations specific to CONFIG_KVM
target/i386: Allow elision of kvm_hv_vpindex_settable()
target/i386: Allow elision of kvm_enable_x2apic()
target/i386: Remove unused KVM stubs
target/i386/cpu-sysemu: Inline kvm_apic_in_kernel()
target/i386/helper: Restrict KVM declarations to system emulation
hw/i386/fw_cfg: Include missing 'cpu.h' header
hw/i386/pc: Include missing 'cpu.h' header
hw/i386/pc: Include missing 'sysemu/tcg.h' header
Revert "mkvenv: work around broken pip installations on Debian 10"
mkvenv: assume presence of importlib.metadata
Python: Drop support for Python 3.7
configure: remove dead code
meson: list leftover CONFIG_* symbols
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
CONFIG_SOLARIS is only used to pick tap implementations. But the
target OS is invariant and does not depend on the configuration, so move
away from config_host and just use unconditional rules in softmmu_ss.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
GLib recommend to use G_SOURCE_REMOVE / G_SOURCE_CONTINUE
for GSourceFunc callbacks. Our FEWatchFunc is a GSourceFunc
returning such value. Use such definitions which are
"more memorable" [*].
[*] https://docs.gtk.org/glib/callback.SourceFunc.html#return-value
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230705133139.54419-5-philmd@linaro.org>
Enable SVQ with VIRTIO_NET_F_CTRL_RX_EXTRA feature.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <15ecc49975f9b8d1316ed4296879564a18abf31e.1688797728.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch refactors vhost_vdpa_net_load_rx() to
restore the packet receive filtering state in relation to
VIRTIO_NET_F_CTRL_RX_EXTRA feature at device's startup.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <abddc477a476f756de6e3d24c0e9f7b21c99a4c1.1688797728.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Enable SVQ with VIRTIO_NET_F_CTRL_RX feature.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <5d6173a6d7c4c514c98362b404c019f52d73b06c.1688743107.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Due to the size limitation of the out buffer sent to the vdpa device,
which is determined by vhost_vdpa_net_cvq_cmd_len(), excessive CVQ
command is truncated in QEMU. As a result, the vdpa device rejects
this flawd CVQ command.
However, the problem is that, the VIRTIO_NET_CTRL_MAC_TABLE_SET
CVQ command has a variable length, which may exceed
vhost_vdpa_net_cvq_cmd_len() if the guest sets more than
`MAC_TABLE_ENTRIES` MAC addresses for the filter table.
This patch solves this problem by following steps:
* Increase the out buffer size to vhost_vdpa_net_cvq_cmd_page_len(),
which represents the size of the buffer that is allocated and mmaped.
This ensures that everything works correctly as long as the guest
sets fewer than `(vhost_vdpa_net_cvq_cmd_page_len() -
sizeof(struct virtio_net_ctrl_hdr)
- 2 * sizeof(struct virtio_net_ctrl_mac)) / ETH_ALEN` MAC addresses.
Considering the highly unlikely scenario for the guest setting
more than that number of MAC addresses for the filter table, this
should work fine for the majority of cases.
* If the CVQ command exceeds vhost_vdpa_net_cvq_cmd_page_len(),
instead of directly sending this CVQ command, QEMU should send
a VIRTIO_NET_CTRL_RX_PROMISC CVQ command to vdpa device. Addtionally,
a fake VIRTIO_NET_CTRL_MAC_TABLE_SET command including
(`MAC_TABLE_ENTRIES` + 1) non-multicast MAC addresses and
(`MAC_TABLE_ENTRIES` + 1) multicast MAC addresses should be provided
to the device model.
By doing so, the vdpa device turns promiscuous mode on, aligning
with the VirtIO standard. The device model marks
`n->mac_table.uni_overflow` and `n->mac_table.multi_overflow`,
which aligns with the state of the vdpa device.
Note that the bug cannot be triggered at the moment, since
VIRTIO_NET_F_CTRL_RX feature is not enabled for SVQ.
Fixes: 7a7f87e94c ("vdpa: Move command buffers map to start of net device")
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <267e15e4eed2d7aeb9887f193da99a13d22a2f1d.1688743107.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We can access the CVQ header through `struct virtio_net_ctrl_hdr`,
instead of accessing it through a `uint8_t` pointer,
which improves the code's readability and maintainability.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <cd522e06a4371e9d6b8a1c1a86f90a92401d56e8.1688743107.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch introduces vhost_vdpa_net_load_rx_mode()
and vhost_vdpa_net_load_rx() to restore the packet
receive filtering state in relation to
VIRTIO_NET_F_CTRL_RX feature at device's startup.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <804cedac93e19ba3b810d52b274ca5ec11469f09.1688743107.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch refactors vhost_vdpa_net_load_mac() to
restore the MAC address filtering state at device's startup.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <4b9550c14bc8c98c8f48e04dbf3d3ac41489d3fd.1688743107.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
According to VirtIO standard, "The driver MUST follow
the VIRTIO_NET_CTRL_MAC_TABLE_SET command by a le32 number,
followed by that number of non-multicast MAC addresses,
followed by another le32 number, followed by that number
of multicast addresses."
Considering that these data is not stored in contiguous memory,
this patch refactors vhost_vdpa_net_load_cmd() to accept
scattered data, eliminating the need for an addtional data copy or
packing the data into s->cvq_cmd_out_buffer outside of
vhost_vdpa_net_load_cmd().
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <3482cc50eebd13db4140b8b5dec9d0cc25b20b1b.1688743107.git.yin31149@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
QEMU uses vhost_handle_guest_kick() to forward guest's available
buffers to the vdpa device in SVQ avail ring.
In vhost_handle_guest_kick(), a `g_autofree` `elem` is used to
iterate through the available VirtQueueElements. This `elem` is
then passed to `svq->ops->avail_handler`, specifically to the
vhost_vdpa_net_handle_ctrl_avail(). If this handler fails to
process the CVQ command, vhost_handle_guest_kick() regains
ownership of the `elem`, and either frees it or requeues it.
Yet the problem is that, vhost_vdpa_net_handle_ctrl_avail()
mistakenly frees the `elem`, even if it fails to forward the
CVQ command to vdpa device. This can result in a use-after-free
for the `elem` in vhost_handle_guest_kick().
This patch solves this problem by refactoring
vhost_vdpa_net_handle_ctrl_avail() to only freeing the `elem` if
it owns it.
Fixes: bd907ae4b0 ("vdpa: manual forward CVQ buffers")
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <e3f2d7db477734afe5c6a5ab3fa8b8317514ea34.1688746840.git.yin31149@gmail.com>
Reviewed-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
According to VirtIO standard, "The class, command and
command-specific-data are set by the driver,
and the device sets the ack byte.
There is little it can do except issue a diagnostic
if ack is not VIRTIO_NET_OK."
Therefore, QEMU should stop sending the queued SVQ commands and
cancel the device startup if the device's ack is not VIRTIO_NET_OK.
Yet the problem is that, vhost_vdpa_net_load_offloads() returns 1 based on
`*s->status != VIRTIO_NET_OK` when the device's ack is VIRTIO_NET_ERR.
As a result, net->nc->info->load() also returns 1, this makes
vhost_net_start_one() incorrectly assume the device state is
successfully loaded by vhost_vdpa_net_load() and return 0, instead of
goto `fail` label to cancel the device startup, as vhost_net_start_one()
only cancels the device startup when net->nc->info->load() returns a
negative value.
This patch fixes this problem by returning -EIO when the device's
ack is not VIRTIO_NET_OK.
Fixes: 0b58d3686a ("vdpa: Add vhost_vdpa_net_load_offloads()")
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <b0396b80e96322b86f1a0b10c098fc1edd947d72.1688438055.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
According to VirtIO standard, "The class, command and
command-specific-data are set by the driver,
and the device sets the ack byte.
There is little it can do except issue a diagnostic
if ack is not VIRTIO_NET_OK."
Therefore, QEMU should stop sending the queued SVQ commands and
cancel the device startup if the device's ack is not VIRTIO_NET_OK.
Yet the problem is that, vhost_vdpa_net_load_mq() returns 1 based on
`*s->status != VIRTIO_NET_OK` when the device's ack is VIRTIO_NET_ERR.
As a result, net->nc->info->load() also returns 1, this makes
vhost_net_start_one() incorrectly assume the device state is
successfully loaded by vhost_vdpa_net_load() and return 0, instead of
goto `fail` label to cancel the device startup, as vhost_net_start_one()
only cancels the device startup when net->nc->info->load() returns a
negative value.
This patch fixes this problem by returning -EIO when the device's
ack is not VIRTIO_NET_OK.
Fixes: f64c7cda69 ("vdpa: Add vhost_vdpa_net_load_mq")
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <ec515ebb0b4f56368751b9e318e245a5d994fa72.1688438055.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
According to VirtIO standard, "The class, command and
command-specific-data are set by the driver,
and the device sets the ack byte.
There is little it can do except issue a diagnostic
if ack is not VIRTIO_NET_OK."
Therefore, QEMU should stop sending the queued SVQ commands and
cancel the device startup if the device's ack is not VIRTIO_NET_OK.
Yet the problem is that, vhost_vdpa_net_load_mac() returns 1 based on
`*s->status != VIRTIO_NET_OK` when the device's ack is VIRTIO_NET_ERR.
As a result, net->nc->info->load() also returns 1, this makes
vhost_net_start_one() incorrectly assume the device state is
successfully loaded by vhost_vdpa_net_load() and return 0, instead of
goto `fail` label to cancel the device startup, as vhost_net_start_one()
only cancels the device startup when net->nc->info->load() returns a
negative value.
This patch fixes this problem by returning -EIO when the device's
ack is not VIRTIO_NET_OK.
Fixes: f73c0c43ac ("vdpa: extract vhost_vdpa_net_load_mac from vhost_vdpa_net_load")
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <a21731518644abbd0c495c5b7960527c5911f80d.1688438055.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch sorts the vdpa_feature_bits array
alphabetically in ascending order to avoid future duplicates.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
This entry was duplicated on referenced commit. Removing it.
Fixes: 402378407d ("vhost-vdpa: multiqueue support")
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Move the file descriptor type checking before doing anything with it.
If it's not usable, don't close it as it could be in use by another
part of QEMU, only fail and report an error.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Use directly net_socket_fd_init_stream() and net_socket_fd_init_dgram()
when the socket type is already known.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
When a peer nic is still attached to the vdpa backend, it is too early to free
up the vhost-net and vdpa structures. If these structures are freed here, then
QEMU crashes when the guest is being shut down. The following call chain
would result in an assertion failure since the pointer returned from
vhost_vdpa_get_vhost_net() would be NULL:
do_vm_stop() -> vm_state_notify() -> virtio_set_status() ->
virtio_net_vhost_status() -> get_vhost_net().
Therefore, we defer freeing up the structures until at guest shutdown
time when qemu_cleanup() calls net_cleanup() which then calls
qemu_del_net_client() which would eventually call vhost_vdpa_cleanup()
again to free up the structures. This time, the loop in net_cleanup()
ensures that vhost_vdpa_cleanup() will be called one last time when
all the peer nics are detached and freed.
All unit tests pass with this change.
CC: imammedo@redhat.com
CC: jusual@redhat.com
CC: mst@redhat.com
Fixes: CVE-2023-3301
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2128929
Signed-off-by: Ani Sinha <anisinha@redhat.com>
Message-Id: <20230619065209.442185-1-anisinha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Bug introducing when refactoring. Otherway, the guest never received
the used buffer.
Fixes: be4278b65f ("vdpa: extract vhost_vdpa_net_cvq_add from vhost_vdpa_net_handle_ctrl_avail")
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230602173451.1917999-1-eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
QEMU does not emulate it so it must be disabled as long as the backend
does not support it.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230602173328.1917385-1-eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Enable SVQ with VIRTIO_NET_F_CTRL_GUEST_OFFLOADS feature.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <778d642ecae6deed8a218b0e6232e4d7bb96b439.1685704856.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Eugenio Pérez <eperezma@redhat.com>
Tested-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch introduces vhost_vdpa_net_load_offloads() to
restore offloads state at device's startup.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Message-Id: <7e2b5cad9c48c917df53d80dec27dbfeb513e1a3.1685704856.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Eugenio Pérez <eperezma@redhat.com>
Tested-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We can use virtio_vdev_has_feature() instead of manually
accessing the features.
Signed-off-by: Hawkins Jiawei <yin31149@gmail.com>
Acked-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <ff838d30206209fd865511b16ffb34cc0d5e8d8f.1685704856.git.yin31149@gmail.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Eugenio Pérez <eperezma@redhat.com>
Tested-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The vdpa devices that use va addresses neeeds these maps shared.
Otherwise, vhost_vdpa checks will refuse to accept the maps.
The mmap call will always return a page aligned address, so removing the
qemu_memalign call. Keeping the ROUND_UP for the size as we still need
to DMA-map them in full.
Not applying fixes tag as it never worked with va devices.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230602143854.1879091-4-eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We need to call it from resource cleanup context, as munmap needs the
size of the mappings.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230602143854.1879091-3-eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
It was a mistake to forbid in all cases, as SVQ is already able to send
all the CVQ messages before start forwarding data vqs. It actually
caused a regression, making impossible to migrate device previously
migratable.
Fixes: 36e4647247 ("vdpa: add vhost_vdpa_net_valid_svq_features")
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230602143854.1879091-2-eperezma@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Evaluating it at start time instead of initialization time may make the
guest capable of dynamically adding or removing migration blockers.
Also, moving to initialization reduces the number of ioctls in the
migration, reducing failure possibilities.
As a drawback we need to check for CVQ isolation twice: one time with no
MQ negotiated and another one acking it, as long as the device supports
it. This is because Vring ASID / group management is based on vq
indexes, but we don't know the index of CVQ before negotiating MQ.
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230526153143.470745-3-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
We need to tell in the caller, as some errors are expected in a normal
workflow. In particular, parent drivers in recent kernels with
VHOST_BACKEND_F_IOTLB_ASID may not support vring groups. In that case,
-ENOTSUP is returned.
This is the case of vp_vdpa in Linux 6.2.
Next patches in this series will use that information to know if it must
abort or not. Also, next patches return properly an errp instead of
printing with error_report.
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Message-Id: <20230526153143.470745-2-eperezma@redhat.com>
Tested-by: Lei Yang <leiyang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We use the user_ss[] array to hold the user emulation sources,
and the softmmu_ss[] array to hold the system emulation ones.
Hold the latter in the 'system_ss[]' array for parity with user
emulation.
Mechanical change doing:
$ sed -i -e s/softmmu_ss/system_ss/g $(git grep -l softmmu_ss)
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230613133347.82210-10-philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Fix the following Cocoa trivial warnings:
C compiler for the host machine: cc (clang 14.0.0 "Apple clang version 14.0.0 (clang-1400.0.29.202)")
Objective-C compiler for the host machine: clang (clang 14.0.0)
[100/334] Compiling Objective-C object libcommon.fa.p/net_vmnet-bridged.m.o
net/vmnet-bridged.m:40:31: warning: a function declaration without a prototype is deprecated in all versions of C [-Wstrict-prototypes]
static char* get_valid_ifnames()
^
void
[742/1436] Compiling Objective-C object libcommon.fa.p/ui_cocoa.m.o
ui/cocoa.m:1937:22: warning: a function declaration without a prototype is deprecated in all versions of C [-Wstrict-prototypes]
static int cocoa_main()
^
void
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-Id: <20230425192820.34063-1-philmd@linaro.org>
