The return value of virtio_blk_handle_vq() is no longer used. Get rid of
it. This is a step towards unifying the dataplane and non-dataplane
virtqueue handler functions.
Prepare virtio_blk_handle_output() to be used by both dataplane and
non-dataplane by making the condition for starting ioeventfd more
specific. This way it won't trigger when dataplane has already been
started.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20211207132336.36627-4-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>
With more recent versions of Meson, the build.ninja file is more selective
as to what is built by default, and not building the modules results in test
failures.
Mark the modules as built-by-default and, to make the dependencies more
precise, also require them to be up-to-date before running tests.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/801
Tested-by: Li Zhang <lizhang@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If roms/seabios/Makefile is not present, the configure script
is not creating the roms/seabios directory anymore (commit
5dce7b8d8c, "configure: remove DIRS", 2021-12-18); thus, creating
roms/seabios/config.mak fails.
The easiest thing to do is to not create the file, since it will not
be used.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This is unnecessary, because the interrupt would be retrieved and queued
anyway by KVM_GET_VCPU_EVENTS and KVM_SET_VCPU_EVENTS respectively,
and it makes the flow more similar to the one for KVM_GET/SET_SREGS2.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This ensures that the file is regenerated properly whenever config-target.h
or config-devices.h files change.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This will be needed as soon as config-poison.h moves from configure to
a meson custom_target (which is built at "ninja" time).
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This makes it easier to identify candidates for moving to Meson.
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Always include the STRIP variable in config-host.mak (it's only used
by the s390-ccw firmware build, and it adds a default if configure
omitted it), and use meson-buildoptions.sh to turn
--enable/--disable-strip into -Dstrip.
The default is now not to strip the binaries like for almost every other
package that has a configure script.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
These files have been removed for more than year in the best
case, or for more than ten years for some really old TCG files.
Remove any traces of it.
Acked-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Similar to other optional features, leave the variables empty and compute
the actual value later. Use the existence of include or source directories
to detect whether an OS or CPU supports respectively bsd-user and linux-user.
For now, BSD user-mode emulation is buildable even on TCI-only
architectures. This probably will change once safe signals are
brought over from linux-user.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
--dynamic-list is present on all supported ELF (not Windows or Darwin)
platforms, since it dates back to 2006; -exported_symbols_list is
likewise present on all supported versions of macOS. Do not bother
doing a functional test in configure.
Remove the file creation from configure as well: for Darwin, move the
the creation of the Darwin-formatted symbols to meson; for ELF, use the
file in the source path directly and switch from -Wl, to -Xlinker to
not break weird paths that include a comma.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The handling for the XFS_IOC_DIOINFO ioctl is currently quite excessive:
This is not a "real" feature like the other features that we provide with
the "--enable-xxx" and "--disable-xxx" switches for the configure script,
since this does not influence lots of code (it's only about one call to
xfsctl() in file-posix.c), so people don't gain much with the ability to
disable this with "--disable-xfsctl".
It's also unfortunate that the ioctl will be disabled on Linux in case
the user did not install the right xfsprogs-devel package before running
configure. Thus let's simplify this by providing the ioctl definition
on our own, so we can completely get rid of the header dependency and
thus the related code in the configure script.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20211215125824.250091-1-thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It is not necessary to have a separate static_library just for common_user
files; using the one that already covers the rest of common_ss is enough
unless you need to reuse some source files between emulators and tests.
Just place common files for all user-mode emulators in common_ss,
similar to what is already done for softmmu_ss in full system emulators.
The only disadvantage is that the include_directories under bsd-user/include/
and linux-user/include/ are now enabled for all targets rather than only
user mode emulators. This however is not different from how include/sysemu/
is available when building user mode emulators.
Tested-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Avoid polluting the compilation of common-user/ with local include files;
making an include file available to common-user/ should be a deliberate
decision in order to keep a clear interface that can be used by both
bsd-user/ and linux-user/.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
pnv_pec_stk_pci_xscom_write() is pnv_pec_stk_pci_xscom_ops write
callback. It writes values into regs in the stack->nest_regs[] array.
The pnv_pec_stk_pci_xscom_read read callback, on the other hand, returns
values of the stack->pci_regs[]. In fact, at this moment, the only use
of stack->pci_regs[] is in pnv_pec_stk_pci_xscom_read(). There's no code
that is written anything in stack->pci_regs[], which is suspicious.
Considering that stack->nest_regs[] is widely used by the nested
MemoryOps pnv_pec_stk_nest_xscom_ops, in both read and write callbacks,
the conclusion is that we're writing the wrong array in
pnv_pec_stk_pci_xscom_write(). This function should write stack->pci_regs[]
instead.
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: Frederic Barrat <fbarrat@linux.ibm.com>
Message-Id: <20220111200132.633896-2-danielhb413@gmail.com>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
This patch introduces pnv-phb4 user creatable devices that are created
in a similar manner as pnv-phb3 devices, allowing the user to interact
with the PHBs directly instead of creating PCI Express Controllers that
will create a certain amount of PHBs per controller index.
We accomplish this by doing the following:
- add a pnv_phb4_get_stack() helper to retrieve which stack an user
created phb4 would occupy;
- when dealing with an user created pnv-phb4 (detected by checking if
phb->stack is NULL at the start of phb4_realize()), retrieve its stack
and initialize its properties as done in stk_realize();
- use 'defaults_enabled()' in stk_realize() to avoid creating and
initializing a 'stack->phb' qdev that might be overwritten by an user
created pnv-phb4 device. This process is wrapped into a new helper
called pnv_pec_stk_default_phb_realize().
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-Id: <20220111131027.599784-5-danielhb413@gmail.com>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Stefan Hajnoczi
Volker RümelinandPaolo Bonzini
Philippe Mathieu-DaudéandPaolo Bonzini
Maxim LevitskyandPaolo Bonzini
Thomas HuthandPaolo Bonzini
Peter Maydell
Daniel Henrique BarbozaandCédric Le Goater