Since we can't run on anything else, assume for the moment that this is
a FreeBSD target. In the future, we'll need to handle this properly via
some include file in bsd-user/*bsd/x86_64/mumble.h. There's a number of
other diffs that would be needed to make things work on OtherBSD, so it
doesn't make sense to preseve this one detail today.
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Kyle Evans <kevans@FreeBSD.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Since we can't run on anything else, assume for the moment that this is
a FreeBSD target. In the future, we'll need to handle this properly
via some include file in bsd-user/*bsd/arm/mumble.h. There's a number
of other diffs that would be needed to make things work on OtherBSD,
so it doesn't make sense to preseve this one detail today.
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Kyle Evans <kevans@FreeBSD.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Since we don't build on OpenBSD, only do FreeBSD system calls here. In
the future, we'll need to move this to some place like
bsd-user/freebsd/arm/mumble.h, but until then just leave this
inline. This reflects changes to the upstream.
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Kyle Evans <kevans@FreeBSD.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
This doesn't build on openbsd at the moment, and this could
should arguably be in bsd-user/*bsd/i386 somewhere. Until
we refactor to support OpenBSD/NetBSD again, drop it here.
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Kyle Evans <kevans@FreeBSD.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
This doesn't build on openbsd at the moment, and this could
should arguably be in bsd-user/*bsd/x86_64 somewhere. Until
we refactor to support OpenBSD/NetBSD again, drop it here.
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Kyle Evans <kevans@FreeBSD.org>
Although initial versions of NetBSD did use int $80, it was replaced by
syscall before any releases. OpenBSD and FreeBSD always did syscall.
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Kyle Evans <kevans@FreeBSD.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
We've not realistically been able to actually run any bsd program on any
other bsd program. They are too diverged to do this easily. The current
code is setup to do it, but implementing it is hard. Stop pretending
that we can do this.
Signed-off-by: Warner Losh <imp@bsdimp.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
The initrd passed via the command line is loaded into memory. It's
location and size is then added to the device tree so the kernel knows
where to find it.
Signed-off-by: Stafford Horne <shorne@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Using the device tree means that qemu can now directly tell
the kernel what hardware is configured rather than use having
to maintain and update a separate device tree file.
This patch adds automatic device tree generation support for the
OpenRISC simulator. A device tree is built up based on the state of the
configure openrisc simulator.
This is then dumped to memory and the load address is passed to the
kernel in register r3.
Signed-off-by: Stafford Horne <shorne@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
We previously loaded into in1, but in1 is not filled during
disassembly and hence always zero. This leads to an assertion failure:
qemu-system-s390x: /home/nrb/qemu/include/tcg/tcg.h:654: temp_idx:
Assertion `n >= 0 && n < tcg_ctx->nb_temps' failed.`
Instead, use in2_la2_m64a to load from storage into in2 and pass that to
the helper, which matches what we already do for SCKC.
This fixes the SCK test I sent here under TCG:
<https://www.spinics.net/lists/kvm/msg265169.html>
Fixes: 9dc67537 ("s390x/tcg: implement SET CLOCK ")
Suggested-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Nico Boehr <nrb@linux.ibm.com>
Message-Id: <20220126084201.774457-1-nrb@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Now that we no longer have a limit of 2 CPUs due to fixing the
IRQ routing issues we can increase the max. Here we increase
the limit to 4, we could go higher, but currently OMPIC has a
limit of 4, so we align with that.
Signed-off-by: Stafford Horne <shorne@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Currently the OpenRISC SMP configuration only supports 2 cores due to
the UART IRQ routing being limited to 2 cores. As was done in commit
1eeffbeb11 ("hw/openrisc/openrisc_sim: Use IRQ splitter when connecting
IRQ to multiple CPUs") we can use a splitter to wire more than 2 CPUs.
This patch moves serial initialization out to it's own function and
uses a splitter to connect multiple CPU irq lines to the UART.
Signed-off-by: Stafford Horne <shorne@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Move magic numbers to variables and enums. These will be reused for
upcoming fdt initialization.
Signed-off-by: Stafford Horne <shorne@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
This will allow us to attach machine state attributes like
the device tree fdt.
Signed-off-by: Stafford Horne <shorne@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Eduardo Habkost has left Red Hat and has other daily responsibilities to
attend to. In order to stop spamming him on every series, remove him as
"Reviewer" for the python/ library dir and add Beraldo Leal instead.
For the "python scripts" stanza (which is separate due to level of
support), replace Eduardo as maintainer with myself.
(Thanks for all of your hard work, Eduardo!)
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Beraldo Leal <bleal@redhat.com>
Acked-by: Eduardo Habkost <eduardo@habkost.net>
Message-id: 20220208000525.2601011-1-jsnow@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
Setuptools v60 and later include a bundled version of distutils, a
deprecated standard library scheduled for removal in future versions of
Python. Setuptools v60 is only possible to install for Python 3.7 and later.
Python has a distutils.sysconfig.get_python_lib() function that returns
'/usr/lib/pythonX.Y' on posix systems. RPM-based systems actually use
'/usr/lib64/pythonX.Y' instead, so Fedora patches stdlib distutils for
Python 3.7 and Python 3.8 to return the correct value.
Python 3.9 and later introduce a sys.platlibdir property, which returns
the correct value on RPM-based systems.
The change to a distutils package not provided by Fedora on Python 3.7
and 3.8 causes a regression in distutils.sysconfig.get_python_lib() that
ultimately causes false positives to be emitted by pylint, because it
can no longer find the system source libraries.
Many Python tools are fairly aggressive about updating setuptools
packages, and so even though this package is a fair bit newer than
Python 3.7/3.8, it's not entirely unreasonable for a given user to have
such a modern package with a fairly old Python interpreter.
Updates to Python 3.7 and Python 3.8 are being produced for Fedora which
will fix the problem on up-to-date systems. Until then, we can force the
loading of platform-provided distutils when running the pylint
test. This is the least-invasive yet most comprehensive fix.
References:
https://github.com/pypa/setuptools/pull/2896https://github.com/PyCQA/pylint/issues/5704https://github.com/pypa/distutils/issues/110
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20220204221804.2047468-2-jsnow@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
When invoking setup.py directly, the default behavior for 'install' is
to run the bdist_egg installation hook, which is ... actually deprecated
by setuptools. It doesn't seem to work quite right anymore.
By contrast, 'pip install' will invoke the bdist_wheel hook
instead. This leads to differences in behavior for the two approaches. I
advocate using pip in the documentation in this directory, but the
'setup.py' which has been used for quite a long time in the Python world
may deceptively appear to work at first glance.
Add an error message that will save a bit of time and frustration
that points the user towards using the supported installation
invocation.
Reported-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Beraldo Leal <bleal@redhat.com>
Message-id: 20220207213039.2278569-1-jsnow@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
When running QMP commands with very large response payloads, it is often
not easy to spot the info you want. If we can save the response to a
file then tools like 'grep' or 'jq' can be used to extract information.
For convenience of processing, we merge the QMP command and response
dictionaries together:
{
"arguments": {},
"execute": "query-kvm",
"return": {
"enabled": false,
"present": true
}
}
Example usage
$ ./scripts/qmp/qmp-shell-wrap -l q.log -p -- ./build/qemu-system-x86_64 -display none
Welcome to the QMP low-level shell!
Connected
(QEMU) query-kvm
{
"return": {
"enabled": false,
"present": true
}
}
(QEMU) query-mice
{
"return": [
{
"absolute": false,
"current": true,
"index": 2,
"name": "QEMU PS/2 Mouse"
}
]
}
$ jq --slurp '. | to_entries[] | select(.value.execute == "query-kvm") |
.value.return.enabled' < q.log
false
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-id: 20220128161157.36261-3-berrange@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
With the current 'qmp-shell' tool developers must first spawn QEMU with
a suitable -qmp arg and then spawn qmp-shell in a separate terminal
pointing to the right socket.
With 'qmp-shell-wrap' developers can ignore QMP sockets entirely and
just pass the QEMU command and arguments they want. The program will
listen on a UNIX socket and tell QEMU to connect QMP to that.
For example, this:
# qmp-shell-wrap -- qemu-system-x86_64 -display none
Is roughly equivalent of running:
# qemu-system-x86_64 -display none -qmp qmp-shell-1234 &
# qmp-shell qmp-shell-1234
Except that 'qmp-shell-wrap' switches the socket peers around so that
it is the UNIX socket server and QEMU is the socket client. This makes
QEMU reliably go away when qmp-shell-wrap exits, closing the server
socket.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-id: 20220128161157.36261-2-berrange@redhat.com
[Edited for rebase. --js]
Signed-off-by: John Snow <jsnow@redhat.com>
This misc series of changes:
- Improves documentation of SSH fingerprint checking
- Fixes SHA256 fingerprints with non-blockdev usage
- Blocks the clone3, setns, unshare & execveat syscalls
with seccomp
- Blocks process spawning via clone syscall, but allows
threads, with seccomp
- Takes over seccomp maintainer role
- Expands firmware descriptor spec to allow flash
without NVRAM
# gpg: Signature made Thu 17 Feb 2022 11:57:13 GMT
# gpg: using RSA key DAF3A6FDB26B62912D0E8E3FBE86EBB415104FDF
# gpg: Good signature from "Daniel P. Berrange <dan@berrange.com>" [full]
# gpg: aka "Daniel P. Berrange <berrange@redhat.com>" [full]
# Primary key fingerprint: DAF3 A6FD B26B 6291 2D0E 8E3F BE86 EBB4 1510 4FDF
* remotes/berrange-gitlab/tags/misc-next-pull-request:
docs: expand firmware descriptor to allow flash without NVRAM
MAINTAINERS: take over seccomp from Eduardo Otubo
seccomp: block setns, unshare and execveat syscalls
seccomp: block use of clone3 syscall
seccomp: fix blocking of process spawning
seccomp: add unit test for seccomp filtering
seccomp: allow action to be customized per syscall
block: print the server key type and fingerprint on failure
block: support sha256 fingerprint with pre-blockdev options
block: better document SSH host key fingerprint checking
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
* Improve virtio-net failover test
* Some small fixes for the qtests
* Misc header cleanups by Philippe
# gpg: Signature made Mon 21 Feb 2022 11:40:37 GMT
# gpg: using RSA key 27B88847EEE0250118F3EAB92ED9D774FE702DB5
# gpg: issuer "thuth@redhat.com"
# gpg: Good signature from "Thomas Huth <th.huth@gmx.de>" [full]
# gpg: aka "Thomas Huth <thuth@redhat.com>" [full]
# gpg: aka "Thomas Huth <huth@tuxfamily.org>" [full]
# gpg: aka "Thomas Huth <th.huth@posteo.de>" [unknown]
# Primary key fingerprint: 27B8 8847 EEE0 2501 18F3 EAB9 2ED9 D774 FE70 2DB5
* remotes/thuth-gitlab/tags/pull-request-2022-02-21: (25 commits)
hw/tricore: Remove unused and incorrect header
hw/m68k/mcf: Add missing 'exec/hwaddr.h' header
exec/exec-all: Move 'qemu/log.h' include in units requiring it
softmmu/runstate: Clean headers
linux-user: Add missing "qemu/timer.h" include
target: Add missing "qemu/timer.h" include
core/ptimers: Remove unnecessary 'sysemu/cpus.h' include
exec/ramblock: Add missing includes
qtest: Add missing 'hw/qdev-core.h' include
hw/acpi/memory_hotplug: Remove unused 'hw/acpi/pc-hotplug.h' header
hw/remote: Add missing include
hw/tpm: Clean includes
scripts: Remove the old switch-timer-api script
tests/qtest: failover: migration abort test with failover off
tests/qtest: failover: test migration if the guest doesn't support failover
tests/qtest: failover: check migration with failover off
tests/qtest: failover: check missing guest feature
tests/qtest: failover: check the feature is correctly provided
tests/qtest: failover: use a macro for check_one_card()
tests/qtest: failover: clean up pathname of tests
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
For the users, it looks a little bit weird that capstone and slirp are
not disabled automatically if they run the configure script with the
"--without-default-features" option, so let's do that now.
Note: fdt is *not* changed accordingly since this affects the targets
that we can build, so disabling fdt automatically here might have
unexpected side-effects for the users.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220221090647.150184-1-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
arm, cocoa and misc:
* MAINTAINERS file updates
* Mark remaining global TypeInfo instances as const
* checkpatch: Ensure that TypeInfos are const
* arm hvf: Handle unknown ID registers as RES0
* Make KVM -cpu max exactly like -cpu host
* Fix '-cpu max' for HVF
* Support PAuth extension for hvf
* Kconfig: Add I2C_DEVICES device group
* Kconfig: Add 'imply I2C_DEVICES' on boards with available i2c bus
* hw/arm/armv7m: Handle disconnected clock inputs
* osdep.h: pull out various things into new header files
* hw/timer: fix a9gtimer vmstate
* hw/arm: add initial mori-bmc board
* ui/cocoa: Remove allowedFileTypes restriction in SavePanel
* ui/cocoa: Do not alert even without block devices
* ui/cocoa: Fix the leak of qemu_console_get_label
# gpg: Signature made Mon 21 Feb 2022 13:30:45 GMT
# gpg: using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg: issuer "peter.maydell@linaro.org"
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>" [ultimate]
# gpg: aka "Peter Maydell <pmaydell@gmail.com>" [ultimate]
# gpg: aka "Peter Maydell <pmaydell@chiark.greenend.org.uk>" [ultimate]
# Primary key fingerprint: E1A5 C593 CD41 9DE2 8E83 15CF 3C25 25ED 1436 0CDE
* remotes/pmaydell/tags/pull-target-arm-20220221-1: (25 commits)
ui/cocoa: Fix the leak of qemu_console_get_label
ui/cocoa: Do not alert even without block devices
ui/cocoa: Remove allowedFileTypes restriction in SavePanel
hw/arm: add initial mori-bmc board
hw/timer: fix a9gtimer vmstate
MAINTAINERS: Add Akihiko Odaki to macOS-relateds
include: Move hardware version declarations to new qemu/hw-version.h
include: Move qemu_[id]cache_* declarations to new qemu/cacheinfo.h
include: Move QEMU_MAP_* constants to mmap-alloc.h
include: Move qemu_mprotect_*() to new qemu/mprotect.h
include: Move qemu_madvise() and related #defines to new qemu/madvise.h
hw/arm/armv7m: Handle disconnected clock inputs
Kconfig: Add 'imply I2C_DEVICES' on boards with available i2c bus
Kconfig: Add I2C_DEVICES device group
target/arm: Support PAuth extension for hvf
target/arm: Fix '-cpu max' for HVF
target/arm: Unindent unnecessary else-clause
target/arm: Make KVM -cpu max exactly like -cpu host
target/arm: Use aarch64_cpu_register() for 'host' CPU type
target/arm: Move '-cpu host' code to cpu64.c
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
setAllowedFileTypes is deprecated in macOS 12.
Per Akihiko Odaki [*]:
An image file, which is being chosen by the panel, can be a
raw file and have a variety of file extensions and many are not
covered by the provided list (e.g. "udf"). Other platforms like
GTK can provide an option to open a file with an extension not
listed, but Cocoa can't. It forces the user to rename the file
to give an extension in the list. Moreover, Cocoa does not tell
which extensions are in the list so the user needs to read the
source code, which is pretty bad.
Since this code is harming the usability rather than improving it,
simply remove the [NSSavePanel allowedFileTypes:] call, fixing:
[2789/6622] Compiling Objective-C object libcommon.fa.p/ui_cocoa.m.o
ui/cocoa.m:1411:16: error: 'setAllowedFileTypes:' is deprecated: first deprecated in macOS 12.0 - Use -allowedContentTypes instead [-Werror,-Wdeprecated-declarations]
[openPanel setAllowedFileTypes: supportedImageFileTypes];
^
/Library/Developer/CommandLineTools/SDKs/MacOSX.sdk/System/Library/Frameworks/AppKit.framework/Headers/NSSavePanel.h:215:49: note: property 'allowedFileTypes' is declared deprecated here
@property (nullable, copy) NSArray<NSString *> *allowedFileTypes API_DEPRECATED("Use -allowedContentTypes instead", macos(10.3,12.0));
^
/Library/Developer/CommandLineTools/SDKs/MacOSX.sdk/System/Library/Frameworks/AppKit.framework/Headers/NSSavePanel.h:215:49: note: 'setAllowedFileTypes:' has been explicitly marked deprecated here
FAILED: libcommon.fa.p/ui_cocoa.m.o
[*] https://lore.kernel.org/qemu-devel/4dde2e66-63cb-4390-9538-c032310db3e3@gmail.com/
Suggested-by: Akihiko Odaki <akihiko.odaki@gmail.com>
Reviewed-by: Roman Bolshakov <r.bolshakov@yadro.com>
Tested-by: Roman Bolshakov <r.bolshakov@yadro.com>
Reviewed-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20220215080307.69550-11-f4bug@amsat.org
Reviewed by: Cameron Esfahani <dirty@apple.com>
Reviewed-by: Akihiko Odaki <akihiko.odaki@gmail.com>
Tested-by: Akihiko Odaki <akihiko.odaki@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
A9 gtimer includes global control field and number of per-cpu fields.
But only per-cpu ones are migrated. This patch adds a subsection for
global control field migration.
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
Message-id: 164422345976.2186660.1104517592452494510.stgit@pasha-ThinkPad-X280
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
In the armv7m object, handle clock inputs that aren't connected.
This is always an error for 'cpuclk'. For 'refclk' it is OK for this
to be disconnected, but we need to handle it by not trying to connect
a sourceless-clock to the systick device.
This fixes a bug where on the mps2-an521 and similar boards (which
do not have a refclk) the systick device incorrectly reset with
SYST_CSR.CLKSOURCE 0 ("use refclk") rather than 1 ("use CPU clock").
Cc: qemu-stable@nongnu.org
Reported-by: Richard Petri <git@rpls.de>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220208171643.3486277-1-peter.maydell@linaro.org
Currently there is no way for a board model's Kconfig stanza to
say "I have an i2c bus which the user can plug an i2c device into,
build all the free-standing i2c devices". The Kconfig mechanism
for this is the "device group". Add an I2C_DEVICES group along
the same lines as the existing PCI_DEVICES. Simple free-standing
i2c devices which a user might plausibly want to be able to
plug in on the QEMU commandline should have
default y if I2C_DEVICES
and board models which have an i2c bus that is user-accessible
should use
imply I2C_DEVICES
to cause those pluggable devices to be built.
In this commit we mark only a fairly conservative set of i2c devices
as belonging to the I2C_DEVICES group: the simple sensors and RTCs
(not including PMBus devices or devices which need GPIO lines to be
connected).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Hao Wu <wuhaotsh@google.com>
Message-id: 20220208155911.3408455-2-peter.maydell@linaro.org
Currently we don't allow guests under hvf to use the PAuth extension,
because we didn't have any special code to handle that, and therefore
in arm_cpu_pauth_finalize() we will sanitize the ID_AA64ISAR1 value
the guest sees to clear the PAuth related fields.
Add support for this in the same way that KVM does it, by defaulting
to "PAuth enabled" if the host CPU has it and allowing the user to
disable it via '-cpu pauth=no' on the command line.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Alexander Graf <agraf@csgraf.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220204165506.2846058-7-peter.maydell@linaro.org
Currently when using hvf we mishandle '-cpu max': we fall through to
the TCG version of its initfn, which then sets a lot of feature bits
that the real host CPU doesn't have. The hvf accelerator code then
exposes these bogus ID register values to the guest because it
doesn't check that the host really has the features.
Make '-cpu host' be like '-cpu max' for hvf, as we do with kvm.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Alexander Graf <agraf@csgraf.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220204165506.2846058-6-peter.maydell@linaro.org
Warner Losh