The bug is that the EventNotifiers do have a NULL io_flush callback.
Because _none_ of the callbacks on the dataplane AioContext have such a
callback, aio_poll will simply do nothing. Fixed by adding the callbacks:
the ioeventfd will always be polled (this can change in the future to
pause/resume the processing during live snapshots or similar operations);
the ioqueue will be polled if there are outstanding requests.
I must admit I have screwed up my testing somehow, because commit
2c20e71 does not work even if cherry-picked on top of 1.4.0, and this
patch fixes it there as well.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
* bonzini/hw-dirs:
sh: move files referencing CPU to hw/sh4/
ppc: move more files to hw/ppc
ppc: move files referencing CPU to hw/ppc/
m68k: move files referencing CPU to hw/m68k/
i386: move files referencing CPU to hw/i386/
arm: move files referencing CPU to hw/arm/
hw: move boards and other isolated files to hw/ARCH
ppc: express FDT dependency of pSeries and e500 boards via default-configs/
build: always link device_tree.o into emulators if libfdt available
hw: include hw header files with full paths
ppc: do not use ../ in include files
vt82c686: vt82c686 is not a PCI host bridge
virtio-9p: remove PCI dependencies from hw/9pfs/
virtio-9p: use CONFIG_VIRTFS, not CONFIG_LINUX
hw: move device-hotplug.o to toplevel, compile it once
hw: move qdev-monitor.o to toplevel directory
hw: move fifo.[ch] to libqemuutil
hw: move char backends to backends/
Conflicts:
backends/baum.c
backends/msmouse.c
hw/a15mpcore.c
hw/arm/Makefile.objs
hw/arm/pic_cpu.c
hw/dataplane/event-poll.c
hw/dataplane/virtio-blk.c
include/char/baum.h
include/char/msmouse.h
qemu-char.c
vl.c
Resolve conflicts caused by header movements.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
During the review of the dataplane code, the EventPoll API morphed itself
(not concidentially) into something very very similar to an AioContext.
Thus, it is trivial to convert virtio-blk-dataplane to use AioContext,
and a first baby step towards letting dataplane talk directly to the
QEMU block layer.
The only interesting note is the value-copy of EventNotifiers. At least
in my opinion this is part of the EventNotifier API and is even portable
to Windows. Of course, in this case you should not close the notifier's
underlying file descriptors or handle with event_notifier_cleanup.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@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: Paolo Bonzini <pbonzini@redhat.com>
qdev-monitor.c is the only "core qdev" file that is not used in
user-mode emulation, and it does not define anything that is used
by hardware models. Remove it from the hw/ directory and
remove hw/qdev-monitor.h from hw/qdev.h too; this requires
some files to have some new explicitly includes.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When dataplane is stopping, the s->vdev->binding->set_host_notifier(...,
false) call can invoke the virtqueue handler if an ioeventfd
notification is pending. This causes hw/virtio-blk.c to invoke
virtio_blk_data_plane_start() before virtio_blk_data_plane_stop()
returns!
The result is that we try to restart dataplane while trying to stop it
and the following assertion is raised:
msix_set_mask_notifier: Assertion `!dev->msix_mask_notifier' failed.
Although the code was intended to prevent this scenario, the s->started
boolean isn't enough. Add s->stopping so that we can postpone clearing
s->started until we've completely stopped dataplane.
This way, virtqueue handler calls during virtio_blk_data_plane_stop()
are ignored. When dataplane is legitimately started again later we
already self-kick ourselves to resume processing.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
virtio_pci_set_guest_notifiers() now takes an additional argument to
specify the number of virtqueues to assign a guest notifier for. This
causes a build breakage for CONFIG_VIRTIO_BLK_DATA_PLANE builds:
/home/mdroth/w/qemu2.git/hw/dataplane/virtio-blk.c: In function
‘virtio_blk_data_plane_start’:
/home/mdroth/w/qemu2.git/hw/dataplane/virtio-blk.c:451:47: error: too
few arguments to function ‘s->vdev->binding->set_guest_notifiers’
/home/mdroth/w/qemu2.git/hw/dataplane/virtio-blk.c: In function
‘virtio_blk_data_plane_stop’:
/home/mdroth/w/qemu2.git/hw/dataplane/virtio-blk.c:511:5: error: too few
arguments to function ‘s->vdev->binding->set_guest_notifiers’
make[1]: *** [hw/dataplane/virtio-blk.o] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [subdir-x86_64-softmmu] Error 2
Fix this by passing 1 as the number of virtqueues to assign notifiers
for.
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
O_DIRECT on Linux has alignment requirements on I/O buffers and
misaligned requests result in -EINVAL. The Linux virtio_blk guest
driver usually submits aligned requests so I forgot to handle misaligned
requests.
It turns out that virtio-win guest drivers submit misaligned requests.
Handle them using a bounce buffer that meets alignment requirements.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Extract code for read/write command processing into do_rdwr_cmd(). This
brings together pieces that are spread across process_request().
The real motivation is to set the stage for handling misaligned
requests, which the next patch tackles.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
virtio-blk-data-plane is a subset implementation of virtio-blk. It only
handles read, write, and flush requests. It does this using a dedicated
thread that executes an epoll(2)-based event loop and processes I/O
using Linux AIO.
This approach performs very well but can be used for raw image files
only. The number of IOPS achieved has been reported to be several times
higher than the existing virtio-blk implementation.
Eventually it should be possible to unify virtio-blk-data-plane with the
main body of QEMU code once the block layer and hardware emulation is
able to run outside the global mutex.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
