virtio-blk and virtio-scsi invoke virtio_irqfd_notify() to send Used
Buffer Notifications from an IOThread. This involves an eventfd
write(2) syscall. Calling this repeatedly when completing multiple I/O
requests in a row is wasteful.
Use the defer_call() API to batch together virtio_irqfd_notify() calls
made during thread pool (aio=threads), Linux AIO (aio=native), and
io_uring (aio=io_uring) completion processing.
Behavior is unchanged for emulated devices that do not use
defer_call_begin()/defer_call_end() since defer_call() immediately
invokes the callback when called outside a
defer_call_begin()/defer_call_end() region.
fio rw=randread bs=4k iodepth=64 numjobs=8 IOPS increases by ~9% with a
single IOThread and 8 vCPUs. iodepth=1 decreases by ~1% but this could
be noise. Detailed performance data and configuration specifics are
available here:
https://gitlab.com/stefanha/virt-playbooks/-/tree/blk_io_plug-irqfd
This duplicates the BH that virtio-blk uses for batching. The next
commit will remove it.
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20230913200045.1024233-4-stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
vhost-user-scsi has a VirtioDeviceClass->reset() function that calls
->vhost_reset_device(). The other vhost devices don't notify the vhost
device upon reset.
Stateful vhost devices may need to handle device reset in order to free
resources or prevent stale device state from interfering after reset.
Call ->vhost_device_reset() from virtio_reset() so that that vhost
devices are notified of device reset.
This patch affects behavior as follows:
- vhost-kernel: No change in behavior since ->vhost_reset_device() is
not implemented.
- vhost-user: back-ends that negotiate
VHOST_USER_PROTOCOL_F_RESET_DEVICE now receive a
VHOST_USER_DEVICE_RESET message upon device reset. Otherwise there is
no change in behavior. DPDK, SPDK, libvhost-user, and the
vhost-user-backend crate do not negotiate
VHOST_USER_PROTOCOL_F_RESET_DEVICE automatically.
- vhost-vdpa: an extra SET_STATUS 0 call is made during device reset.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20231004014532.1228637-4-stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Raphael Norwitz <raphael.norwitz@nutanix.com>
Reviewed-by: Hanna Czenczek <hreitz@redhat.com>
The 'next' was converted from a local variable to an output parameter
in commit:
412e0e81b1 ("virtio: handle virtqueue_read_next_desc() errors")
But all the actual uses of the 'i/next' as an output were removed a few
months prior in commit:
aa570d6fb6 ("virtio: combine the read of a descriptor")
Remove the unused argument to simplify the code.
Also, adding a comment to the function to describe what it is actually
doing, as it is not obvious that the 'desc' is both an input and an
output argument.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Message-Id: <20230927140016.2317404-3-i.maximets@ovn.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
It was supposed to be a compiler barrier and it was a compiler barrier
initially called 'wmb' when virtio core support was introduced.
Later all the instances of 'wmb' were switched to smp_wmb to fix memory
ordering issues on non-x86 platforms. However, this one doesn't need
to be an actual barrier, as its only purpose was to ensure that the
value is not read twice.
And since commit aa570d6fb6 ("virtio: combine the read of a descriptor")
there is no need for a barrier at all, since we're no longer reading
guest memory here, but accessing a local structure.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Message-Id: <20230927140016.2317404-2-i.maximets@ovn.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We do not need the most up to date number of heads, we only want to
know if there is at least one.
Use shadow variable as long as it is not equal to the last available
index checked. This avoids expensive qatomic dereference of the
RCU-protected memory region cache as well as the memory access itself.
The change improves performance of the af-xdp network backend by 2-3%.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Message-Id: <20230927135157.2316982-1-i.maximets@ovn.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The virtio_list duplicates information about virtio devices that already
exist in the QOM composition tree. Instead of creating this list of
realized virtio devices, search the QOM composition tree instead.
This patch modifies the QMP command qmp_x_query_virtio to instead
recursively search the QOM composition tree for devices of type
'TYPE_VIRTIO_DEVICE'. The device is also checked to ensure it's
realized.
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230926224107.2951144-2-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Lots of virtio functions that are on a hot path in data transmission
are initializing indirect descriptor cache at the point of stack
allocation. It's a 112 byte structure that is getting zeroed out on
each call adding unnecessary overhead. It's going to be correctly
initialized later via special init function. The only reason to
actually initialize right away is the ability to safely destruct it.
Replacing a designated initializer with a function to only initialize
what is necessary.
Removal of the unnecessary stack initializations improves throughput
of virtio-net devices in terms of 64B packets per second by 6-14 %
depending on the case. Tested with a proposed af-xdp network backend
and a dpdk testpmd application in the guest, but should be beneficial
for other virtio devices as well.
Signed-off-by: Ilya Maximets <i.maximets@ovn.org>
Message-Id: <20230811143423.3258788-1-i.maximets@ovn.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
virtio_load() as a whole should run in coroutine context because it
reads from the migration stream and we don't want this to block.
However, it calls virtio_set_features_nocheck() and devices don't
expect their .set_features callback to run in a coroutine and therefore
call functions that may not be called in coroutine context. To fix this,
drop out of coroutine context for calling virtio_set_features_nocheck().
Without this fix, the following crash was reported:
#0 __pthread_kill_implementation (threadid=<optimized out>, signo=signo@entry=6, no_tid=no_tid@entry=0) at pthread_kill.c:44
#1 0x00007efc738c05d3 in __pthread_kill_internal (signo=6, threadid=<optimized out>) at pthread_kill.c:78
#2 0x00007efc73873d26 in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/raise.c:26
#3 0x00007efc738477f3 in __GI_abort () at abort.c:79
#4 0x00007efc7384771b in __assert_fail_base (fmt=0x7efc739dbcb8 "", assertion=assertion@entry=0x560aebfbf5cf "!qemu_in_coroutine()",
file=file@entry=0x560aebfcd2d4 "../block/graph-lock.c", line=line@entry=275, function=function@entry=0x560aebfcd34d "void bdrv_graph_rdlock_main_loop(void)") at assert.c:92
#5 0x00007efc7386ccc6 in __assert_fail (assertion=0x560aebfbf5cf "!qemu_in_coroutine()", file=0x560aebfcd2d4 "../block/graph-lock.c", line=275,
function=0x560aebfcd34d "void bdrv_graph_rdlock_main_loop(void)") at assert.c:101
#6 0x0000560aebcd8dd6 in bdrv_register_buf ()
#7 0x0000560aeb97ed97 in ram_block_added.llvm ()
#8 0x0000560aebb8303f in ram_block_add.llvm ()
#9 0x0000560aebb834fa in qemu_ram_alloc_internal.llvm ()
#10 0x0000560aebb2ac98 in vfio_region_mmap ()
#11 0x0000560aebb3ea0f in vfio_bars_register ()
#12 0x0000560aebb3c628 in vfio_realize ()
#13 0x0000560aeb90f0c2 in pci_qdev_realize ()
#14 0x0000560aebc40305 in device_set_realized ()
#15 0x0000560aebc48e07 in property_set_bool.llvm ()
#16 0x0000560aebc46582 in object_property_set ()
#17 0x0000560aebc4cd58 in object_property_set_qobject ()
#18 0x0000560aebc46ba7 in object_property_set_bool ()
#19 0x0000560aeb98b3ca in qdev_device_add_from_qdict ()
#20 0x0000560aebb1fbaf in virtio_net_set_features ()
#21 0x0000560aebb46b51 in virtio_set_features_nocheck ()
#22 0x0000560aebb47107 in virtio_load ()
#23 0x0000560aeb9ae7ce in vmstate_load_state ()
#24 0x0000560aeb9d2ee9 in qemu_loadvm_state_main ()
#25 0x0000560aeb9d45e1 in qemu_loadvm_state ()
#26 0x0000560aeb9bc32c in process_incoming_migration_co.llvm ()
#27 0x0000560aebeace56 in coroutine_trampoline.llvm ()
Cc: qemu-stable@nongnu.org
Buglink: https://issues.redhat.com/browse/RHEL-832
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20230905145002.46391-3-kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
virtio_queue_packed_set_last_avail_idx() is used by vhost devices to set
the internal queue indices to what has been reported by the vhost
back-end through GET_VRING_BASE. For packed virtqueues, this
32-bit value is expected to contain both the device's internal avail and
used indices, as well as their respective wrap counters.
To get the used index, we shift the 32-bit value right by 16, and then
apply a mask of 0x7ffff. That seems to be a typo, because it should be
0x7fff; first of all, the virtio specification says that the maximum
queue size for packed virt queues is 2^15, so the indices cannot exceed
2^15 - 1 anyway, making 0x7fff the correct mask. Second, the mask
clearly is wrong from context, too, given that (A) `idx & 0x70000` must
be 0 at this point (`idx` is 32 bit and was shifted to the right by 16
already), (B) `idx & 0x8000` is the used_wrap_counter, so should not be
part of the used index, and (C) `vq->used_idx` is a `uint16_t`, so
cannot fit the 0x70000 part of the mask anyway.
This most likely never produced any guest-visible bugs, though, because
for a vhost device, qemu will probably not evaluate the used index
outside of virtio_queue_packed_get_last_avail_idx(), where we
reconstruct the 32-bit value from avail and used indices and their wrap
counters again. There, it does not matter whether the highest bit of
the used_idx is the used index wrap counter, because we put the wrap
counter exactly in that position anyway.
Signed-off-by: Hanna Czenczek <hreitz@redhat.com>
Message-Id: <20230721134945.26967-1-hreitz@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: German Maglione <gmaglione@redhat.com>
All callers now pass is_external=false to aio_set_fd_handler() and
aio_set_event_notifier(). The aio_disable_external() API that
temporarily disables fd handlers that were registered is_external=true
is therefore dead code.
Remove aio_disable_external(), aio_enable_external(), and the
is_external arguments to aio_set_fd_handler() and
aio_set_event_notifier().
The entire test-fdmon-epoll test is removed because its sole purpose was
testing aio_disable_external().
Parts of this patch were generated using the following coccinelle
(https://coccinelle.lip6.fr/) semantic patch:
@@
expression ctx, fd, is_external, io_read, io_write, io_poll, io_poll_ready, opaque;
@@
- aio_set_fd_handler(ctx, fd, is_external, io_read, io_write, io_poll, io_poll_ready, opaque)
+ aio_set_fd_handler(ctx, fd, io_read, io_write, io_poll, io_poll_ready, opaque)
@@
expression ctx, notifier, is_external, io_read, io_poll, io_poll_ready;
@@
- aio_set_event_notifier(ctx, notifier, is_external, io_read, io_poll, io_poll_ready)
+ aio_set_event_notifier(ctx, notifier, io_read, io_poll, io_poll_ready)
Reviewed-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20230516190238.8401-21-stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Host notifiers can now use is_external=false since virtio-blk and
virtio-scsi no longer rely on is_external=true for drained sections.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20230516190238.8401-20-stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
virtio_queue_aio_detach_host_notifier() does two things:
1. It removes the fd handler from the event loop.
2. It processes the virtqueue one last time.
The first step can be peformed by any thread and without taking the
AioContext lock.
The second step may need the AioContext lock (depending on the device
implementation) and runs in the thread where request processing takes
place. virtio-blk and virtio-scsi therefore call
virtio_queue_aio_detach_host_notifier() from a BH that is scheduled in
AioContext.
The next patch will introduce a .drained_begin() function that needs to
call virtio_queue_aio_detach_host_notifier(). .drained_begin() functions
cannot call aio_poll() to wait synchronously for the BH. It is possible
for a .drained_poll() callback to asynchronously wait for the BH, but
that is more complex than necessary here.
Move the virtqueue processing out to the callers of
virtio_queue_aio_detach_host_notifier() so that the function can be
called from any thread. This is in preparation for the next patch.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20230516190238.8401-17-stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
When a virtqueue size is changed by the guest via
virtio_queue_set_num(), its region cache is not automatically updated.
If the size was increased, this could lead to accessing the cache out
of bounds. For example, in vring_get_used_event():
static inline uint16_t vring_get_used_event(VirtQueue *vq)
{
return vring_avail_ring(vq, vq->vring.num);
}
static inline uint16_t vring_avail_ring(VirtQueue *vq, int i)
{
VRingMemoryRegionCaches *caches = vring_get_region_caches(vq);
hwaddr pa = offsetof(VRingAvail, ring[i]);
if (!caches) {
return 0;
}
return virtio_lduw_phys_cached(vq->vdev, &caches->avail, pa);
}
vq->vring.num will be greater than caches->avail.len, which will
trigger a failed assertion down the call path of
virtio_lduw_phys_cached().
Fix this by calling virtio_init_region_cache() after
virtio_queue_set_num() if we are not already calling
virtio_queue_set_rings(). In the legacy path this is already done by
virtio_queue_update_rings().
Signed-off-by: Carlos López <clopez@suse.de>
Message-Id: <20230317002749.27379-1-clopez@suse.de>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Acked-by: Halil Pasic <pasic@linux.ibm.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In virtqueue_{split,packed}_get_avail_bytes() descriptors are read
in a loop via MemoryRegionCache regions and calls to
vring_{split,packed}_desc_read() - these take a region cache and the
index of the descriptor to be read.
For direct descriptors we use a cache provided by the caller, whose
size matches that of the virtqueue vring. We limit the number of
descriptors we can read by the size of that vring:
max = vq->vring.num;
...
MemoryRegionCache *desc_cache = &caches->desc;
For indirect descriptors, we initialize a new cache and limit the
number of descriptors by the size of the intermediate descriptor:
len = address_space_cache_init(&indirect_desc_cache,
vdev->dma_as,
desc.addr, desc.len, false);
desc_cache = &indirect_desc_cache;
...
max = desc.len / sizeof(VRingDesc);
However, the first initialization of `max` is done outside the loop
where we process guest descriptors, while the second one is done
inside. This means that a sequence of an indirect descriptor followed
by a direct one will leave a stale value in `max`. If the second
descriptor's `next` field is smaller than the stale value, but
greater than the size of the virtqueue ring (and thus the cached
region), a failed assertion will be triggered in
address_space_read_cached() down the call chain.
Fix this by initializing `max` inside the loop in both functions.
Fixes: 9796d0ac8f ("virtio: use address_space_map/unmap to access descriptors")
Signed-off-by: Carlos López <clopez@suse.de>
Message-Id: <20230302100358.3613-1-clopez@suse.de>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
hw/virtio/virtio.h and hw/virtio/vhost.h include each other. The
former doesn't actually need the latter, so drop that inclusion to
break the loop.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20221222120813.727830-2-armbru@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>
Reviewed-by: Edgar E. Iglesias <edgar@zeroasic.com>
virtio.c is big enough, extract more QMP related code to virtio-qmp.c.
To do so, expose qmp_find_virtio_device() and declar virtio_list in
the internal virtio-qmp.h header.
Note we have to leave qmp_x_query_virtio_queue_status() and
qmp_x_query_virtio_queue_element(), because they access VirtQueue
internal fields, and VirtQueue is only declared within virtio.c.
Suggested-by: Jonah Palmer <jonah.palmer@oracle.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221222080005.27616-3-philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
To emphasize this function is QMP related, rename it.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221222080005.27616-2-philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Add the functions to support the configure interrupt in virtio
The function virtio_config_guest_notifier_read will notify the
guest if there is an configure interrupt.
The function virtio_config_set_guest_notifier_fd_handler is
to set the fd hander for the notifier
Signed-off-by: Cindy Lu <lulu@redhat.com>
Message-Id: <20221222070451.936503-7-lulu@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The monitor decoders are the only functions using the CONFIG_xxx
definitions declared in the target specific CONFIG_DEVICES header.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221213111707.34921-7-philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <a class="moz-txt-link-rfc2396E" href="mailto:philmd@linaro.org"><philmd@linaro.org></a>
These config helpers use the target-dependent LD/ST API.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221213111707.34921-6-philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
These arrays are only accessed read-only, move them to .rodata.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221213111707.34921-5-philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jonah Palmer<jonah.palmer@oracle.com>
Signed-off-by: Philippe Mathieu-Daudé <a class="moz-txt-link-rfc2396E" href="mailto:philmd@linaro.org"><philmd@linaro.org></a>
Commit f3034ad71f ("qmp: decode feature & status bits in
virtio-status") did not guard all qmp_virtio_feature_map_t
arrays with the corresponding #ifdef'ry used in
qmp_decode_features(). Fix that and reduce the arrays scope
by declaring them static.
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221213111707.34921-4-philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jonah Palmer<jonah.palmer@oracle.com>
Suggested-by: Richard Henderson <a class="moz-txt-link-rfc2396E" href="mailto:richard.henderson@linaro.org"><richard.henderson@linaro.org></a>
Signed-off-by: Philippe Mathieu-Daudé <a class="moz-txt-link-rfc2396E" href="mailto:philmd@linaro.org"><philmd@linaro.org></a>
virtio.c uses target_words_bigendian() which is declared in
"hw/core/cpu.h". Add the missing header to avoid when refactoring:
hw/virtio/virtio.c:2451:9: error: implicit declaration of function 'target_words_bigendian' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
if (target_words_bigendian()) {
^
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221213111707.34921-2-philmd@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The has_FOO for pointer-valued FOO are redundant, except for arrays.
They are also a nuisance to work with. Recent commit "qapi: Start to
elide redundant has_FOO in generated C" provided the means to elide
them step by step. This is the step for qapi/virtio.json.
Said commit explains the transformation in more detail. The invariant
violations mentioned there do not occur here.
Cc: Laurent Vivier <lvivier@redhat.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221104160712.3005652-29-armbru@redhat.com>
Virtio 1.0 is pretty clear that features have to be
negotiated before enabling VQs. Unfortunately Seabios
ignored this ever since gaining 1.0 support (UEFI is ok).
Comment the error out for now, and add a TODO.
Fixes: 3c37f8b8d1 ("virtio: introduce virtio_queue_enable()")
Cc: "Kangjie Xu" <kangjie.xu@linux.alibaba.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Message-Id: <20221121200339.362452-1-mst@redhat.com>
Introduce the interface queue_enable() in VirtioDeviceClass and the
fucntion virtio_queue_enable() in virtio, it can be called when
VIRTIO_PCI_COMMON_Q_ENABLE is written and related virtqueue can be
started. It only supports the devices of virtio 1 or later. The
not-supported devices can only start the virtqueue when DRIVER_OK.
Signed-off-by: Kangjie Xu <kangjie.xu@linux.alibaba.com>
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20221017092558.111082-4-xuanzhuo@linux.alibaba.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Introduce a new interface function virtio_queue_reset() to implement
reset for vq.
Add a new callback to VirtioDeviceClass for queue reset operation for
each child device.
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20221017092558.111082-3-xuanzhuo@linux.alibaba.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Separate the logic of vq reset. This logic will be called directly
later.
Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20221017092558.111082-2-xuanzhuo@linux.alibaba.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This new command shows the information of a VirtQueue element.
[Note: Up until v10 of this patch series, virtio.json had many (15+)
enums defined (e.g. decoded device features, statuses, etc.). In v10
most of these enums were removed and replaced with string literals.
By doing this we get (1) simpler schema, (2) smaller generated code,
and (3) less maintenance burden for when new things are added (e.g.
devices, device features, etc.).]
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <1660220684-24909-6-git-send-email-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
These new commands show the internal status of a VirtIODevice's
VirtQueue and a vhost device's vhost_virtqueue (if active).
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <1660220684-24909-5-git-send-email-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Display feature names instead of bitmaps for host, guest, and
backend for VirtIODevices.
Display status names instead of bitmaps for VirtIODevices.
Display feature names instead of bitmaps for backend, protocol,
acked, and features (hdev->features) for vhost devices.
Decode features according to device ID. Decode statuses
according to configuration status bitmap (config_status_map).
Decode vhost user protocol features according to vhost user
protocol bitmap (vhost_user_protocol_map).
Transport features are on the first line. Undecoded bits (if
any) are stored in a separate field.
[Jonah: Several changes made to this patch from prev. version (v14):
- Moved all device features mappings to hw/virtio/virtio.c
- Renamed device features mappings (less generic)
- Generalized @FEATURE_ENTRY macro for all device mappings
- Virtio device feature map definitions include descriptions of
feature bits
- Moved @VHOST_USER_F_PROTOCOL_FEATURES feature bit from transport
feature map to vhost-user-supported device feature mappings
(blk, fs, i2c, rng, net, gpu, input, scsi, vsock)
- New feature bit added for virtio-vsock: @VIRTIO_VSOCK_F_SEQPACKET
- New feature bit added for virtio-iommu: @VIRTIO_IOMMU_F_BYPASS_CONFIG
- New feature bit added for virtio-mem: @VIRTIO_MEM_F_UNPLUGGED_INACCESSIBLE
- New virtio transport feature bit added: @VIRTIO_F_IN_ORDER
- Added device feature map definition for virtio-rng
]
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <1660220684-24909-4-git-send-email-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This new command shows the status of a VirtIODevice, including
its corresponding vhost device's status (if active).
Next patch will improve output by decoding feature bits, including
vhost device's feature bits (backend, protocol, acked, and features).
Also will decode status bits of a VirtIODevice.
[Jonah: From patch v12; added a check to @virtio_device_find to ensure
synchronicity between @virtio_list and the devices in the QOM
composition tree.]
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <1660220684-24909-3-git-send-email-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This new command lists all the instances of VirtIODevices with
their canonical QOM path and name.
[Jonah: @virtio_list duplicates information that already exists in
the QOM composition tree. However, extracting necessary information
from this tree seems to be a bit convoluted.
Instead, we still create our own list of realized virtio devices
but use @qmp_qom_get with the device's canonical QOM path to confirm
that the device exists and is realized. If the device exists but
is actually not realized, then we remove it from our list (for
synchronicity to the QOM composition tree).
Also, the QMP command @x-query-virtio is redundant as @qom-list
and @qom-get are sufficient to search '/machine/' for realized
virtio devices. However, @x-query-virtio is much more convenient
in listing realized virtio devices.]
Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <1660220684-24909-2-git-send-email-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This is the first step towards moving all device config size calculation
logic into the virtio core code. In particular, this adds a struct that
contains all the necessary information for common virtio code to be able
to calculate the final config size for a device. This is expected to be
used with the new virtio_get_config_size helper, which calculates the
final length based on the provided host features.
This builds on top of already existing code like VirtIOFeature and
virtio_feature_get_config_size(), but adds additional fields, as well as
sanity checking so that device-specifc code doesn't have to duplicate it.
An example usage would be:
static const VirtIOFeature dev_features[] = {
{.flags = 1ULL << FEATURE_1_BIT,
.end = endof(struct virtio_dev_config, feature_1)},
{.flags = 1ULL << FEATURE_2_BIT,
.end = endof(struct virtio_dev_config, feature_2)},
{}
};
static const VirtIOConfigSizeParams dev_cfg_size_params = {
.min_size = DEV_BASE_CONFIG_SIZE,
.max_size = sizeof(struct virtio_dev_config),
.feature_sizes = dev_features
};
// code inside my_dev_device_realize()
size_t config_size = virtio_get_config_size(&dev_cfg_size_params,
host_features);
virtio_init(vdev, VIRTIO_ID_MYDEV, config_size);
Currently every device is expected to write its own boilerplate from the
example above in device_realize(), however, the next step of this
transition is moving VirtIOConfigSizeParams into VirtioDeviceClass,
so that it can be done automatically by the virtio initialization code.
All of the users of virtio_feature_get_config_size have been converted
to use virtio_get_config_size so it's no longer needed and is removed
with this commit.
Signed-off-by: Daniil Tatianin <d-tatianin@yandex-team.ru>
Message-Id: <20220906073111.353245-2-d-tatianin@yandex-team.ru>
Reviewed-by: Raphael Norwitz <raphael.norwitz@nutanix.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
If the guest driver attempts to use the UNUSED(30) bit it is
potentially buggy as 6.3 Legacy Interface: Reserved Feature Bits
states it "SHOULD NOT be negotiated". For now just log this guest
error.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20220802095010.3330793-9-alex.bennee@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch adds a get_vhost() callback function for VirtIODevices that
returns the device's corresponding vhost_dev structure, if the vhost
device is running. This patch also adds a vhost_started flag for
VirtIODevices.
Previously, a VirtIODevice wouldn't be able to tell if its corresponding
vhost device was active or not.
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <1648819405-25696-3-git-send-email-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This patch drops the name parameter for the virtio_init function.
The pair between the numeric device ID and the string device ID
(name) of a virtio device already exists, but not in a way that
lets us map between them.
This patch lets us do this and removes the need for the name
parameter in the virtio_init function.
Signed-off-by: Jonah Palmer <jonah.palmer@oracle.com>
Message-Id: <1648819405-25696-2-git-send-email-jonah.palmer@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The virtio-scsi event virtqueue is not emptied by its handler function.
This is typical for rx virtqueues where the device uses buffers when
some event occurs (e.g. a packet is received, an error condition
happens, etc).
Polling non-empty virtqueues wastes CPU cycles. We are not waiting for
new buffers to become available, we are waiting for an event to occur,
so it's a misuse of CPU resources to poll for buffers.
Introduce the new virtio_queue_aio_attach_host_notifier_no_poll() API,
which is identical to virtio_queue_aio_attach_host_notifier() except
that it does not poll the virtqueue.
Before this patch the following command-line consumed 100% CPU in the
IOThread polling and calling virtio_scsi_handle_event():
$ qemu-system-x86_64 -M accel=kvm -m 1G -cpu host \
--object iothread,id=iothread0 \
--device virtio-scsi-pci,iothread=iothread0 \
--blockdev file,filename=test.img,aio=native,cache.direct=on,node-name=drive0 \
--device scsi-hd,drive=drive0
After this patch CPU is no longer wasted.
Reported-by: Nir Soffer <nsoffer@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Tested-by: Nir Soffer <nsoffer@redhat.com>
Message-id: 20220427143541.119567-3-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
One less qemu-specific macro. It also helps to make some headers/units
only depend on glib, and thus moved in standalone projects eventually.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
g_new(T, n) is neater than g_malloc(sizeof(T) * n). It's also safer,
for two reasons. One, it catches multiplication overflowing size_t.
Two, it returns T * rather than void *, which lets the compiler catch
more type errors.
This commit only touches allocations with size arguments of the form
sizeof(T).
Patch created mechanically with:
$ spatch --in-place --sp-file scripts/coccinelle/use-g_new-etc.cocci \
--macro-file scripts/cocci-macro-file.h FILES...
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20220315144156.1595462-4-armbru@redhat.com>
Reviewed-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
The migration code will not look at a VMStateDescription's
minimum_version_id_old field unless that VMSD has set the
load_state_old field to something non-NULL. (The purpose of
minimum_version_id_old is to specify what migration version is needed
for the code in the function pointed to by load_state_old to be able
to handle it on incoming migration.)
We have exactly one VMSD which still has a load_state_old,
in the PPC CPU; every other VMSD which sets minimum_version_id_old
is doing so unnecessarily. Delete all the unnecessary ones.
Commit created with:
sed -i '/\.minimum_version_id_old/d' $(git grep -l '\.minimum_version_id_old')
with the one legitimate use then hand-edited back in.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
---
It missed vmstate_ppc_cpu.
Now that virtio-blk and virtio-scsi are ready, get rid of
the handle_aio_output() callback. It's no longer needed.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20211207132336.36627-7-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
The difference between ->handle_output() and ->handle_aio_output() was
that ->handle_aio_output() returned a bool return value indicating
progress. This was needed by the old polling API but now that the bool
return value is gone, the two functions can be unified.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20211207132336.36627-6-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
The virtqueue host notifier API
virtio_queue_aio_set_host_notifier_handler() polls the virtqueue for new
buffers. AioContext previously required a bool progress return value
indicating whether an event was handled or not. This is no longer
necessary because the AioContext polling API has been split into a poll
check function and an event handler function. The event handler is only
run when we know there is work to do, so it doesn't return bool.
The VirtIOHandleAIOOutput function signature is now the same as
VirtIOHandleOutput. Get rid of the bool return value.
Further simplifications will be made for virtio-blk and virtio-scsi in
the next patch.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20211207132336.36627-3-stefanha@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Adaptive polling measures the execution time of the polling check plus
handlers called when a polled event becomes ready. Handlers can take a
significant amount of time, making it look like polling was running for
a long time when in fact the event handler was running for a long time.
For example, on Linux the io_submit(2) syscall invoked when a virtio-blk
device's virtqueue becomes ready can take 10s of microseconds. This
can exceed the default polling interval (32 microseconds) and cause
adaptive polling to stop polling.
By excluding the handler's execution time from the polling check we make
the adaptive polling calculation more accurate. As a result, the event
loop now stays in polling mode where previously it would have fallen
back to file descriptor monitoring.
The following data was collected with virtio-blk num-queues=2
event_idx=off using an IOThread. Before:
168k IOPS, IOThread syscalls:
9837.115 ( 0.020 ms): IO iothread1/620155 io_submit(ctx_id: 140512552468480, nr: 16, iocbpp: 0x7fcb9f937db0) = 16
9837.158 ( 0.002 ms): IO iothread1/620155 write(fd: 103, buf: 0x556a2ef71b88, count: 8) = 8
9837.161 ( 0.001 ms): IO iothread1/620155 write(fd: 104, buf: 0x556a2ef71b88, count: 8) = 8
9837.163 ( 0.001 ms): IO iothread1/620155 ppoll(ufds: 0x7fcb90002800, nfds: 4, tsp: 0x7fcb9f1342d0, sigsetsize: 8) = 3
9837.164 ( 0.001 ms): IO iothread1/620155 read(fd: 107, buf: 0x7fcb9f939cc0, count: 512) = 8
9837.174 ( 0.001 ms): IO iothread1/620155 read(fd: 105, buf: 0x7fcb9f939cc0, count: 512) = 8
9837.176 ( 0.001 ms): IO iothread1/620155 read(fd: 106, buf: 0x7fcb9f939cc0, count: 512) = 8
9837.209 ( 0.035 ms): IO iothread1/620155 io_submit(ctx_id: 140512552468480, nr: 32, iocbpp: 0x7fca7d0cebe0) = 32
174k IOPS (+3.6%), IOThread syscalls:
9809.566 ( 0.036 ms): IO iothread1/623061 io_submit(ctx_id: 140539805028352, nr: 32, iocbpp: 0x7fd0cdd62be0) = 32
9809.625 ( 0.001 ms): IO iothread1/623061 write(fd: 103, buf: 0x5647cfba5f58, count: 8) = 8
9809.627 ( 0.002 ms): IO iothread1/623061 write(fd: 104, buf: 0x5647cfba5f58, count: 8) = 8
9809.663 ( 0.036 ms): IO iothread1/623061 io_submit(ctx_id: 140539805028352, nr: 32, iocbpp: 0x7fd0d0388b50) = 32
Notice that ppoll(2) and eventfd read(2) syscalls are eliminated because
the IOThread stays in polling mode instead of falling back to file
descriptor monitoring.
As usual, polling is not implemented on Windows so this patch ignores
the new io_poll_read() callback in aio-win32.c.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20211207132336.36627-2-stefanha@redhat.com
[Fixed up aio_set_event_notifier() calls in
tests/unit/test-fdmon-epoll.c added after this series was queued.
--Stefan]
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
This reverts commit 081f864f56.
Fixes: 081f864f56 ("virtio: add support for configure interrupt")
Cc: "Cindy Lu" <lulu@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Packed Virtqueues wrap used_idx instead of letting it run freely like
Split Virtqueues do. If the used ring wraps more than once there is no
way to compare vq->signalled_used and vq->used_idx in
virtio_packed_should_notify() since they are modulo vq->vring.num.
This causes the device to stop sending used buffer notifications when
when virtio_packed_should_notify() is called less than once each time
around the used ring.
It is possible to trigger this with virtio-blk's dataplane
notify_guest_bh() irq coalescing optimization. The call to
virtio_notify_irqfd() (and virtio_packed_should_notify()) is deferred to
a BH. If the guest driver is polling it can complete and submit more
requests before the BH executes, causing the used ring to wrap more than
once. The result is that the virtio-blk device ceases to raise
interrupts and I/O hangs.
Cc: Tiwei Bie <tiwei.bie@intel.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20211130134510.267382-1-stefanha@redhat.com>
Fixes: 86044b24e8 ("virtio: basic packed virtqueue support")
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Add the functions to support the configure interrupt in virtio
The function virtio_config_guest_notifier_read will notify the
guest if there is an configure interrupt.
The function virtio_config_set_guest_notifier_fd_handler is
to set the fd hander for the notifier
Signed-off-by: Cindy Lu <lulu@redhat.com>
Message-Id: <20211104164827.21911-7-lulu@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Let devices specify transaction attributes when calling
dma_memory_map().
Patch created mechanically using spatch with this script:
@@
expression E1, E2, E3, E4;
@@
- dma_memory_map(E1, E2, E3, E4)
+ dma_memory_map(E1, E2, E3, E4, MEMTXATTRS_UNSPECIFIED)
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20211223115554.3155328-7-philmd@redhat.com>
