Xiaoyao Li [Thu, 19 Dec 2024 11:01:24 +0000 (06:01 -0500)]
i386/cpu: Set up CPUID_HT in x86_cpu_expand_features() instead of cpu_x86_cpuid()
Currently CPUID_HT is evaluated in cpu_x86_cpuid() each time. It's not a
correct usage of how feature bit is maintained and evaluated. The
expected practice is that features are tracked in env->features[] and
cpu_x86_cpuid() should be the consumer of env->features[].
Track CPUID_HT in env->features[FEAT_1_EDX] instead and evaluate it in
cpu's realizefn().
Xiaoyao Li [Thu, 19 Dec 2024 11:01:22 +0000 (06:01 -0500)]
i386/cpu: Hoist check of CPUID_EXT3_TOPOEXT against threads_per_core
Now it changes to use env->topo_info.threads_per_core and doesn't depend
on qemu_init_vcpu() anymore. Put it together with other feature checks
before qemu_init_vcpu()
Xiaoyao Li [Thu, 19 Dec 2024 11:01:21 +0000 (06:01 -0500)]
i386/cpu: Track a X86CPUTopoInfo directly in CPUX86State
The name of nr_modules/nr_dies are ambiguous and they mislead people.
The purpose of them is to record and form the topology information. So
just maintain a X86CPUTopoInfo member in CPUX86State instead. Then
nr_modules and nr_dies can be dropped.
As the benefit, x86 can switch to use information in
CPUX86State::topo_info and get rid of the nr_cores and nr_threads in
CPUState. This helps remove the dependency on qemu_init_vcpu(), so that
x86 can get and use topology info earlier in x86_cpu_realizefn(); drop
the comment that highlighted the depedency.
Xiaoyao Li [Thu, 19 Dec 2024 11:01:20 +0000 (06:01 -0500)]
i386/topology: Introduce helpers for various topology info of different level
Introduce various helpers for getting the topology info of different
semantics. Using the helper is more self-explanatory.
Besides, the semantic of the helper will stay unchanged even when new
topology is added in the future. At that time, updating the
implementation of the helper without affecting the callers.
Xiaoyao Li [Thu, 19 Dec 2024 11:01:17 +0000 (06:01 -0500)]
i386/cpu: Drop the variable smp_cores and smp_threads in x86_cpu_pre_plug()
No need to define smp_cores and smp_threads, just using ms->smp.cores
and ms->smp.threads is straightforward. It's also consistent with other
checks of socket/die/module.
Paolo Bonzini [Tue, 24 Dec 2024 15:59:12 +0000 (16:59 +0100)]
target/i386/kvm: Replace ARRAY_SIZE(msr_handlers) with KVM_MSR_FILTER_MAX_RANGES
kvm_install_msr_filters() uses KVM_MSR_FILTER_MAX_RANGES as the bound
when traversing msr_handlers[], while other places still compute the
size by ARRAY_SIZE(msr_handlers).
In fact, msr_handlers[] is an array with the fixed size
KVM_MSR_FILTER_MAX_RANGES, and this has to be true because
kvm_install_msr_filters copies from one array to the other.
For code consistency, assert that they match and use
ARRAY_SIZE(msr_handlers) everywehere.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Zhao Liu [Wed, 6 Nov 2024 03:07:27 +0000 (11:07 +0800)]
target/i386/kvm: Clean up error handling in kvm_arch_init()
Currently, there're following incorrect error handling cases in
kvm_arch_init():
* Missed to handle failure of kvm_get_supported_feature_msrs().
* Missed to return when kvm_vm_enable_disable_exits() fails.
* MSR filter related cases called exit() directly instead of returning
to kvm_init(). (The caller of kvm_arch_init() - kvm_init() - needs to
know if kvm_arch_init() fails in order to perform cleanup).
Zhao Liu [Wed, 6 Nov 2024 03:07:25 +0000 (11:07 +0800)]
target/i386/kvm: Clean up return values of MSR filter related functions
Before commit 0cc42e63bb54 ("kvm/i386: refactor kvm_arch_init and split
it into smaller functions"), error_report() attempts to print the error
code from kvm_filter_msr(). However, printing error code does not work
due to kvm_filter_msr() returns bool instead int.
0cc42e63bb54 fixed the error by removing error code printing, but this
lost useful error messages. Bring it back by making kvm_filter_msr()
return int.
This also makes the function call chain processing clearer, allowing for
better handling of error result propagation from kvm_filter_msr() to
kvm_arch_init(), preparing for the subsequent cleanup work of error
handling in kvm_arch_init().
Zhao Liu [Wed, 6 Nov 2024 03:07:18 +0000 (11:07 +0800)]
i386/cpu: Mark avx10_version filtered when prefix is NULL
In x86_cpu_filter_features(), if host doesn't support AVX10, the
configured avx10_version should be marked as filtered regardless of
whether prefix is NULL or not.
Check prefix before warn_report() instead of checking for
have_filtered_features.
Cc: qemu-stable@nongnu.org Fixes: commit bccfb846fd52 ("target/i386: add AVX10 feature and AVX10 version property") Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Tao Su <tao1.su@linux.intel.com> Link: https://lore.kernel.org/r/20241106030728.553238-2-zhao1.liu@intel.com Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 17 Oct 2024 10:10:39 +0000 (12:10 +0200)]
target/i386: improve code generation for BT
Because BT does not write back to the source operand, it can modify it to
ensure that one of the operands of TSTNE is a constant (after either gen_BT
or the optimizer's constant propagation). This produces better and more
optimizable TCG ops. For example, the sequence
Paolo Bonzini [Thu, 19 Dec 2024 10:24:13 +0000 (11:24 +0100)]
make-release: only leave tarball of wrap-file subprojects
The QEMU source archive is including the sources downloaded from crates.io
in both tarball form (in subprojects/packagecache) and expanded/patched
form (in the subprojects directory). The former is the more authoritative
form, as it has a hash that can be verified in the wrap file and checked
against the download URL, so keep that one only. This works also with
--disable-download; when building QEMU for the first time from the
tarball, Meson will print something like
Using proc-macro2-1-rs source from cache.
for each subproject, and then go on to extract the tarball and apply the
overlay or the patches in subprojects/packagefiles.
Reported-by: Michael Tokarev <mjt@tls.msk.ru>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2719 Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 7 Jan 2025 10:42:49 +0000 (11:42 +0100)]
qom: remove unused field
The "concrete_class" field of InterfaceClass is only ever written, and as far
as I can tell is not particularly useful when debugging either; remove it.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Fri, 20 Dec 2024 12:10:03 +0000 (13:10 +0100)]
rust: hide warnings for subprojects
This matches cargo's own usage of "--cap-lints allow" when building
dependencies. The dummy changes to the .wrap files help Meson notice
that the subproject is out of date.
Also remove an unnecessary subprojects/unicode-ident-1-rs/meson.build file.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Fri, 29 Nov 2024 09:46:44 +0000 (10:46 +0100)]
rust: qdev: expose inherited methods to subclasses of SysBusDevice
The ObjectDeref trait now provides all the magic that is required to fake
inheritance. Replace the "impl SysBusDevice" block of qemu_api::sysbus
with a trait, so that sysbus_init_irq() can be invoked as "self.init_irq()"
without any intermediate upcast.
Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This is going to be fairly common. Using a custom procedural macro
provides better error messages and automatically finds the right
type.
Note that this is different from the same-named macro in the
derive_more crate. That one provides conversion from e.g. tuples
to enums with tuple variants, not from integers to enums.
Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Mon, 2 Dec 2024 12:16:19 +0000 (13:16 +0100)]
rust: pl011: only leave embedded object initialization in instance_init
Leave IRQ and MMIO initialization to instance_post_init. In Rust the
two callbacks are more distinct, because only instance_post_init has a
fully initialized object available.
While at it, add a wrapper for sysbus_init_mmio so that accesses to
the SysBusDevice correctly use shared references.
Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Fri, 29 Nov 2024 10:38:59 +0000 (11:38 +0100)]
rust: qom: move device_id to PL011 class side
There is no need to monkeypatch DeviceId::Luminary into the already-initialized
PL011State. Instead, now that we can define a class hierarchy, we can define
PL011Class and make device_id a field in there.
There is also no need anymore to have "Arm" as zero, so change DeviceId into a
wrapper for the array; all it does is provide an Index<hwaddr> implementation
because arrays can only be indexed by usize.
Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 24 Oct 2024 09:57:02 +0000 (11:57 +0200)]
rust: macros: check that #[derive(Object)] requires #[repr(C)]
Convert derive_object to the same pattern of first making a
Result<proc_macro2::TokenStream, CompileError>, and then doing
.unwrap_or_else(Into::into) to support checking the validity of
the input. Add is_c_repr to check that all QOM structs include
a #[repr(C)] attribute.
Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Wed, 11 Dec 2024 11:18:06 +0000 (12:18 +0100)]
rust: add a utility module for compile-time type checks
It is relatively common in the low-level qemu_api code to assert that
a field of a struct has a specific type; for example, it can be used
to ensure that the fields match what the qemu_api and C code expects
for safety.
Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Wed, 11 Dec 2024 10:38:20 +0000 (11:38 +0100)]
rust: qom: add ParentField
Add a type that, together with the C function object_deinit, ensures the
correct drop order for QOM objects relative to their superclasses.
Right now it is not possible to implement the Drop trait for QOM classes
that are defined in Rust, as the drop() function would not be called when
the object goes away; instead what is called is ObjectImpl::INSTANCE_FINALIZE.
It would be nice for INSTANCE_FINALIZE to just drop the object, but this has
a problem: suppose you have
When instance_finalize is called for MySubclass, it will walk the struct's
list of fields and call the drop method for MySuperclass. Then, object_deinit
recurses to the superclass and calls the same drop method again. This
will cause double-freeing of the Box<Data>.
What's happening here is that QOM wants to control the drop order of
MySuperclass and MySubclass's fields. To do so, the parent field must
be marked ManuallyDrop<>, which is quite ugly. Instead, add a wrapper
type ParentField<> that is specific to QOM. This hides the implementation
detail of *what* is special about the ParentField, and will also be easy
to check in the #[derive(Object)] macro.
Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 7 Jan 2025 15:52:25 +0000 (16:52 +0100)]
rust: add --check-cfg test to rustc arguments
rustc will check that every reachable #[cfg] matches a list of
the expected config names and values. Recent versions of rustc are
also complaining about #[cfg(test)], even if it is basically a standard
part of the language. So, always allow it.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stefan Hajnoczi [Mon, 6 Jan 2025 14:39:02 +0000 (09:39 -0500)]
Merge tag 'qga-pull-2025-01-06' of https://github.com/kostyanf14/qemu into staging
qga-pull-2025-01-06
# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEEwsLBCepDxjwUI+uE711egWG6hOcFAmd7vqMACgkQ711egWG6
# hOffRxAAotgBsE+o8fsZ2tfOKNPekW0hlw/hceDMJRA2UwOSPfw1fXfw59w4Pnfr
# 4xwMC6O8Lu9ohBCBWHUvh3261gJgXQkLASbbzmF2oewfXZyvPXQI8nz78Ol3LBTG
# gL8lwaBci3YuFtc+2/55VdQsWUqtrRMvBW9WSXTEC+0dQJv+VzblXlEF7hQkKppT
# oGiHQL7pEA1UP7bRo4TyaoDnc8a+xz1J+vtEZUZghtreT7I3ELai/PFdo0U99fkf
# HZfjyj2sHCZto+tAokjBcqf2RXDRqUVRsn3GgC1MQbh1LRdfShmhCTbgYYk/1MmD
# 0xwiqAsw814W25299LM3xP2LHPm1jKtkZyCyuSXme9QtN9mC3F0TipR+HMRErAj0
# GQTBOJ0LinZsx5U/+ih4/qPj7RRov+SFzpVxBV3NUkpneVFp5FQgOo4n8l+h57ap
# fmkZ6/hb8itn2oux7S9v/LkcmWE3FqThKO6qMXOhBhQDCKpICz8liYO/tPdB4x1Q
# /HHQ9oon0A2eQw/53AYqz0SoazOqNtadg/hsQ11OHDExUjdp4M6hyxtmrJEQz4Et
# AFvIby98lJZCZ1u65dv/Prb+gW0E8AQ5Ib0jJllAm7tL/GjVyhbRlUl8S9R2uTcZ
# Gsb6e3DMBOny/lR9+2M4rCyCqXM58gTohuqtcXvAe8l2a3h23B4=
# =uk2Q
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 06 Jan 2025 06:29:39 EST
# gpg: using RSA key C2C2C109EA43C63C1423EB84EF5D5E8161BA84E7
# gpg: Good signature from "Kostiantyn Kostiuk (Upstream PR sign) <kkostiuk@redhat.com>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg: There is no indication that the signature belongs to the owner.
# Primary key fingerprint: C2C2 C109 EA43 C63C 1423 EB84 EF5D 5E81 61BA 84E7
* tag 'qga-pull-2025-01-06' of https://github.com/kostyanf14/qemu:
qemu-ga: Optimize freeze-hook script logic of logging error
qga: implement a 'guest-get-load' command
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Dehan Meng [Wed, 25 Dec 2024 08:37:44 +0000 (16:37 +0800)]
qemu-ga: Optimize freeze-hook script logic of logging error
Make sure the error log of fsfreeze hooks
when freeze/thaw/snapshot could be logged
to system logs if the default logfile of
qga can't be written or other situations
Signed-off-by: Dehan Meng <demeng@redhat.com> Reviewed-by: Yan Vugenfirer <yvugenfi@redhat.com> Reviewed-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Message-ID: <20241225083744.277374-1-demeng@redhat.com> Signed-off-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Windows has no native equivalent API, but it would be possible to
simulate it as illustrated here (BSD-3-Clause):
https://github.com/giampaolo/psutil/pull/1485
This is left as an exercise for future contributors.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com> Reviewed-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Message-ID: <20241202121927.864335-1-berrange@redhat.com> Signed-off-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Stefan Hajnoczi [Fri, 3 Jan 2025 14:14:11 +0000 (09:14 -0500)]
Merge tag 'qtest-20250102-pull-request' of https://gitlab.com/farosas/qemu into staging
Qtest pull request
- migration-test: fix UADK build, reinstate postcopy tests and other cleanups
- a couple of memory leak fixes for bios-tables-test and virtio-iommu-test
- drop dead code from fw_cfg-test
# -----BEGIN PGP SIGNATURE-----
#
# iQJEBAABCAAuFiEEqhtIsKIjJqWkw2TPx5jcdBvsMZ0FAmd3DSMQHGZhcm9zYXNA
# c3VzZS5kZQAKCRDHmNx0G+wxnUy5EACPnvy3U2jen2Cc5Hke69byCQtMkDTAJHH5
# xlwTLoTEF25AXMjLyee9vM4MoDHl+EUaGd3zmmAu2/P5/zOHWrBh9VbkFKDDvJJS
# XTVSZ0Iv9a0LDGKsv74Pk7LUZCEanr9a0qpfYTFPuwUh9yu7pd4QeYpbEdwzKzX6
# TMpJTYOE5gUTgP6X8Ir6u3po8DChfE9Eb+Z3LhRU7S+Xi3FoUWRFgRzuvXhXLkf3
# SPNEA27XCaqWQjG8SfQ9yhafrEz9OIGq39Q2unSJ8Df8riYurrdhxYrWNZ/6URQi
# /sYoBJZ2IMBzLg4cVOWXXQwdFkmU+0LUYPOANy6MrYRXl1QJtWway3eHOj+ZlkIH
# r2ulAJWAXqIQ3Ki6Dv3TYCg0/snF8arFzmdsaAbn0M8YYQB7XAlcqHus5axEMIVO
# 0VG7rXctRyH2h7DFfKeW/G1aaK5ox+/0aJoVhjmFJSlYZ2DUx+htfsPlIJmwhRRF
# i61VaOw7NsrHsW0fEgTRurQBVVuZ2/nJT8saJwvXpFiJu0X1GQ6lmKQNlWqm5h6C
# JyLc096+vOG0Cqih7p+M3NvKTHNQZQzUPSLJU2UXItPDPv1oAmFrHpVlBQiUD37I
# h/2R6vxZIY5ei2ECFITnl+LaV4MNjGExpeOEi020KSLEz9Y8nk+i5E+4oDUct93z
# wAsWLadmxg==
# =sLkH
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 02 Jan 2025 17:03:15 EST
# gpg: using RSA key AA1B48B0A22326A5A4C364CFC798DC741BEC319D
# gpg: issuer "farosas@suse.de"
# gpg: Good signature from "Fabiano Rosas <farosas@suse.de>" [unknown]
# gpg: aka "Fabiano Almeida Rosas <fabiano.rosas@suse.com>" [unknown]
# gpg: WARNING: The key's User ID is not certified with a trusted signature!
# gpg: There is no indication that the signature belongs to the owner.
# Primary key fingerprint: AA1B 48B0 A223 26A5 A4C3 64CF C798 DC74 1BEC 319D
* tag 'qtest-20250102-pull-request' of https://gitlab.com/farosas/qemu:
qtest/fw-cfg: remove compiled out code
tests/qtest/migration: Re-enable postcopy tests
tests/migration: Drop arch_[source|target]
tests/qtest/virtio-iommu-test: Don't pass uninitialized data into qtest_memwrite
tests/qtest/bios-tables-test: Free tables at dump_aml_files
tests/qtest/migration: Initialize buffer in probe_o_direct_support
tests/qtest/migration: Do proper cleanup in the dirty_limit test
tests/qtest/migration: Fix compile errors when CONFIG_UADK is set
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Fabiano Rosas [Wed, 18 Dec 2024 19:22:22 +0000 (16:22 -0300)]
tests/qtest/migration: Re-enable postcopy tests
Postcopy tests have been inadvertently disabled since commit 124a3c58b8 ("tests/qtest/migration: Move ufd_version_check to
utils"). That commit moved the ufd_version_check() function to another
file but failed to make sense of the ifdefs and includes:
The <sys/syscall> include was incorrectly dropped. It is needed to
pull in <asm/unistd.h> for __NR_userfaultfd.
The <sys/ioctl.h> was moved under the wrong ifdef.
Fixes: 124a3c58b8 ("tests/qtest/migration: Move ufd_version_check to utils") Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Fabiano Rosas <farosas@suse.de>
Message-Id: <20241218192223.10551-2-farosas@suse.de>
Peter Xu [Mon, 16 Dec 2024 16:14:13 +0000 (11:14 -0500)]
tests/migration: Drop arch_[source|target]
Coverity complained about them. These two variables are never used now
after commit 832c732c5d ("migration-test: Create arch_opts"), and/or commit 34cc54fb35 ("tests/qtest/migration-test: Use custom asm bios for ppc64").
Fabiano Rosas [Mon, 9 Dec 2024 20:44:25 +0000 (17:44 -0300)]
tests/qtest/virtio-iommu-test: Don't pass uninitialized data into qtest_memwrite
Valgrind complains about:
Use of uninitialised value of size 8
&
Conditional jump or move depends on uninitialised value(s)
both at:
at 0x5265931: _itoa_word (_itoa.c:180)
by 0x527EEC7: __vfprintf_internal (vfprintf-internal.c:1687)
by 0x528C8B0: __vsprintf_internal (iovsprintf.c:96)
by 0x526B920: sprintf (sprintf.c:30)
by 0x1296C7: qtest_memwrite (libqtest.c:1273)
by 0x193C04: send_map (virtio-iommu-test.c:125)
by 0x194392: test_attach_detach (virtio-iommu-test.c:214)
by 0x17BDE7: run_one_test (qos-test.c:181)
by 0x4B0699D: test_case_run (gtestutils.c:2900)
by 0x4B0699D: g_test_run_suite_internal (gtestutils.c:2988)
by 0x4B068B2: g_test_run_suite_internal (gtestutils.c:3005)
by 0x4B068B2: g_test_run_suite_internal (gtestutils.c:3005)
by 0x4B068B2: g_test_run_suite_internal (gtestutils.c:3005)
Uninitialised value was created by a stack allocation
at 0x193AFD: send_map (virtio-iommu-test.c:103)
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20241209204427.17763-5-farosas@suse.de> Signed-off-by: Fabiano Rosas <farosas@suse.de>
Fabiano Rosas [Mon, 9 Dec 2024 20:44:23 +0000 (17:44 -0300)]
tests/qtest/migration: Initialize buffer in probe_o_direct_support
Valgrind complains about the probe_o_direct_support() function reading
from an uninitialized buffer. For probing O_DIRECT support we don't
actually need to write to the file, just make sure the pwrite call
doesn't reject the write. Still, write zeroes to the buffer to
suppress the warning.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20241209204427.17763-3-farosas@suse.de> Signed-off-by: Fabiano Rosas <farosas@suse.de>
Stefan Hajnoczi [Thu, 2 Jan 2025 14:35:06 +0000 (09:35 -0500)]
Merge tag 'pull-request-2025-01-02' of https://gitlab.com/thuth/qemu into staging
* Update year in copyright statements
* Convert the VNC test to the functional framework
* Improve and update the ppc64_hv functional test
* Fix broken rx_gdbsim and arm_quanta_gsj functional tests
* tag 'pull-request-2025-01-02' of https://gitlab.com/thuth/qemu:
tests/functional/test_arm_quanta_gsj: Fix broken test
tests/functional/test_rx_gdbsim: Use stable URL for test_linux_sash
tests/functional/test_ppc64_hv: Update to Alpine 3.21.0
tests/functional/test_ppc64_hv: Update repo management
tests/functional/test_ppc64_hv: Simplify console handling
tests/functional: Extract the find_free_ports() function into a helper file
tests/functional/test_vnc: Remove the test_no_vnc test
tests/functional/test_vnc: Do not use a hard-coded VNC port
tests/functional: Convert the vnc test
docs: update copyright date to the year 2025
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Thomas Huth [Thu, 2 Jan 2025 07:30:35 +0000 (08:30 +0100)]
tests/functional/test_arm_quanta_gsj: Fix broken test
ASSET_IMAGE needs to be prefixed with "self." ... this bug
apparently went in unnoticed because the test is not run by
default.
Message-ID: <20250102073403.36328-1-thuth@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Thomas Huth <thuth@redhat.com>
tests/functional/test_rx_gdbsim: Use stable URL for test_linux_sash
Yoshinori said [*] URL references on OSDN were stable, but they
appear not to be. Mirror the artifacts on GitHub to avoid failures
while testing on CI.
Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Reported-by: Alex Bennée <alex.bennee@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-ID: <20200630202631.7345-1-f4bug@amsat.org>
[huth: Adapt the patch to the new version in the functional framework]
Message-ID: <20241229083419.180423-1-huth@tuxfamily.org> Signed-off-by: Thomas Huth <thuth@redhat.com>
Nicholas Piggin [Fri, 20 Dec 2024 02:46:17 +0000 (12:46 +1000)]
tests/functional/test_ppc64_hv: Update to Alpine 3.21.0
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Message-ID: <20241220024617.1968556-5-npiggin@gmail.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
`setup-apkrepos` can be used to set repos rather than open-coding URLs.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Message-ID: <20241220024617.1968556-4-npiggin@gmail.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
Since functional tests have character-based console output parsing,
there is no need for strange hacks to work around old line-based.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Message-ID: <20241220024617.1968556-3-npiggin@gmail.com> Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
Thomas Huth [Wed, 18 Dec 2024 13:14:38 +0000 (14:14 +0100)]
tests/functional: Extract the find_free_ports() function into a helper file
We'll need this functionality in other functional tests, too, so
let's extract it into the qemu_test module.
Also add an __enter__ and __exit__ function that can be used for
using this functionality in a locked context, so that tests that
are running in parallel don't try to compete for the same ports
later.
Also make sure to only use ports in the "Dynamic Ports" range
(see https://www.rfc-editor.org/rfc/rfc6335) and "randomize" the
start of the probed range with the PID of the test process to
further avoid possible clashes with other competing processes.
Message-ID: <20241218131439.255841-5-thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
Thomas Huth [Wed, 18 Dec 2024 13:14:36 +0000 (14:14 +0100)]
tests/functional/test_vnc: Remove the test_no_vnc test
This test matches exactly the first three lines of the following
test_no_vnc_change_password test, so there is exactly zero additional
test coverage in here.
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241218131439.255841-3-thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
Thomas Huth [Wed, 18 Dec 2024 13:14:37 +0000 (14:14 +0100)]
tests/functional/test_vnc: Do not use a hard-coded VNC port
Two tests here are using the hard-coded VNC port :0 ... if there
is already a QEMU or other program running that is using this
port, the tests will be failing. Fortunately, QEMU can also
auto-detect a free port with the "to=..." parameter, so let's
use that for the tests to avoid the problem.
Message-ID: <20241218131439.255841-4-thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
Thomas Huth [Wed, 18 Dec 2024 13:14:35 +0000 (14:14 +0100)]
tests/functional: Convert the vnc test
Nothing thrilling in here, it's just a straight forward conversion.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241218131439.255841-2-thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
Stefan Hajnoczi [Wed, 1 Jan 2025 20:17:07 +0000 (15:17 -0500)]
Merge tag 'hw-misc-20241231' of https://github.com/philmd/qemu into staging
Misc HW patches queue
- Allow more than 4 legacy IRQs on Generic PCI Express Bridge (Alexander)
- Add MMIO-based Inter-VM shared memory device 'ivshmem-flat' (Gustavo)
- Use UHCI register definitions (Guenter)
- Propagate CPU endianness to microblaze_load_kernel (Philippe)
- Mark x86/TriCore devices as little-endian, OpenRISC/SPARC as big (Philippe)
- Don't set callback_opaque NULL in fw_cfg_modify_bytes_read (Shameer)
- Simplify non-KVM checks on AMD IOMMU XTSup feature (Philippe)
- Trivial cleanups on xilinx_ethlite, vmcoreinfo, qxl (Philippe, Hyman)
- Move USB-HCD-XHCI msi/msix properties from NEC to superclass (Phil)
- Redesign of main thread event handling due to macOS Cocoa (Phil)
- Introduce ParavirtualizedGraphics.Framework support 'apple-gfx' (Phil)
- Pad short Ethernet frames on macOS vmnet (William)
* tag 'hw-misc-20241231' of https://github.com/philmd/qemu: (29 commits)
hw/display/qxl: Do not use C99 // comments
net/vmnet: Pad short Ethernet frames
MAINTAINERS: Add myself as maintainer for apple-gfx, reviewer for HVF
hw/display/apple-gfx: Adds configurable mode list
hw/display/apple-gfx: Adds PCI implementation
hw/display/apple-gfx: Introduce ParavirtualizedGraphics.Framework support
ui & main loop: Redesign of system-specific main thread event handling
hw/usb/hcd-xhci: Unimplemented/guest error logging for port MMIO
hw/usb/hcd-xhci-pci: Move msi/msix properties from NEC to superclass
hw/block/virtio-blk: Replaces request free function with g_free
hw/i386/amd_iommu: Simplify non-KVM checks on XTSup feature
hw/misc/vmcoreinfo: Rename opaque pointer as 'opaque'
hw/misc/vmcoreinfo: Declare QOM type using DEFINE_TYPES macro
fw_cfg: Don't set callback_opaque NULL in fw_cfg_modify_bytes_read()
hw/net/xilinx_ethlite: Rename rxbuf -> port_index
hw/net/xilinx_ethlite: Correct maximum RX buffer size
hw/net/xilinx_ethlite: Update QOM style
hw/net/xilinx_ethlite: Remove unuseful debug logs
hw/net/xilinx_ethlite: Convert some debug logs to trace events
hw/sparc: Mark devices as big-endian
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
William Hooper [Sat, 6 Jan 2024 22:35:46 +0000 (14:35 -0800)]
net/vmnet: Pad short Ethernet frames
At least on macOS 12.7.2, vmnet doesn't pad Ethernet frames, such as the
host's ARP replies, to the minimum size (60 bytes before the frame check
sequence) defined in IEEE Std 802.3-2022, so guests' Ethernet device
drivers may drop them with "frame too short" errors.
This patch calls eth_pad_short_frame() to add padding, as in net/tap.c
and net/slirp.c. Thanks to Bin Meng, Philippe Mathieu-Daudé, and Phil
Dennis-Jordan for reviewing earlier versions.
Signed-off-by: William Hooper <wsh@wshooper.org>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2058 Reviewed-by: Phil Dennis-Jordan <phil@philjordan.eu>
Message-ID: <20241102205653.30476-1-wsh@wshooper.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
MAINTAINERS: Add myself as maintainer for apple-gfx, reviewer for HVF
I'm happy to take responsibility for the macOS PV graphics code. As
HVF patches don't seem to get much attention at the moment, I'm also
adding myself as designated reviewer for HVF and x86 HVF to try and
improve that.
Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu> Reviewed-by: Roman Bolshakov <rbolshakov@ddn.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241223221645.29911-6-phil@philjordan.eu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
This change adds a property 'display_modes' on the graphics device
which permits specifying a list of display modes. (screen resolution
and refresh rate)
The property is an array of a custom type to make the syntax slightly
less awkward to use, for example:
This change wires up the PCI variant of the paravirtualised
graphics device, mainly useful for x86-64 macOS guests, implemented
by macOS's ParavirtualizedGraphics.framework. It builds on code
shared with the vmapple/mmio variant of the PVG device.
Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-ID: <20241223221645.29911-4-phil@philjordan.eu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/display/apple-gfx: Introduce ParavirtualizedGraphics.Framework support
MacOS provides a framework (library) that allows any vmm to implement a
paravirtualized 3d graphics passthrough to the host metal stack called
ParavirtualizedGraphics.Framework (PVG). The library abstracts away
almost every aspect of the paravirtualized device model and only provides
and receives callbacks on MMIO access as well as to share memory address
space between the VM and PVG.
This patch implements a QEMU device that drives PVG for the VMApple
variant of it.
Signed-off-by: Alexander Graf <graf@amazon.com> Co-authored-by: Alexander Graf <graf@amazon.com>
Subsequent changes:
* Cherry-pick/rebase conflict fixes, API use updates.
* Moved from hw/vmapple/ (useful outside that machine type)
* Overhaul of threading model, many thread safety improvements.
* Asynchronous rendering.
* Memory and object lifetime fixes.
* Refactoring to split generic and (vmapple) MMIO variant specific
code.
Implementation wise, most of the complexity lies in the differing threading
models of ParavirtualizedGraphics.framework, which uses libdispatch and
internal locks, versus QEMU, which heavily uses the BQL, especially during
memory-mapped device I/O. Great care has therefore been taken to prevent
deadlocks by never calling into PVG methods while holding the BQL, and
similarly never acquiring the BQL in a callback from PVG. Different strategies
have been used (libdispatch, blocking and non-blocking BHs, RCU, etc.)
depending on the specific requirements at each framework entry and exit point.
ui & main loop: Redesign of system-specific main thread event handling
macOS's Cocoa event handling must be done on the initial (main) thread
of the process. Furthermore, if library or application code uses
libdispatch, the main dispatch queue must be handling events on the main
thread as well.
So far, this has affected Qemu in both the Cocoa and SDL UIs, although
in different ways: the Cocoa UI replaces the default qemu_main function
with one that spins Qemu's internal main event loop off onto a
background thread. SDL (which uses Cocoa internally) on the other hand
uses a polling approach within Qemu's main event loop. Events are
polled during the SDL UI's dpy_refresh callback, which happens to run
on the main thread by default.
As UIs are mutually exclusive, this works OK as long as nothing else
needs platform-native event handling. In the next patch, a new device is
introduced based on the ParavirtualizedGraphics.framework in macOS.
This uses libdispatch internally, and only works when events are being
handled on the main runloop. With the current system, it works when
using either the Cocoa or the SDL UI. However, it does not when running
headless. Moreover, any attempt to install a similar scheme to the
Cocoa UI's main thread replacement fails when combined with the SDL
UI.
This change tidies up main thread management to be more flexible.
* The qemu_main global function pointer is a custom function for the
main thread, and it may now be NULL. When it is, the main thread
runs the main Qemu loop. This represents the traditional setup.
* When non-null, spawning the main Qemu event loop on a separate
thread is now done centrally rather than inside the Cocoa UI code.
* For most platforms, qemu_main is indeed NULL by default, but on
Darwin, it defaults to a function that runs the CFRunLoop.
* The Cocoa UI sets qemu_main to a function which runs the
NSApplication event handling runloop, as is usual for a Cocoa app.
* The SDL UI overrides the qemu_main function to NULL, thus
specifying that Qemu's main loop must run on the main
thread.
* The GTK UI also overrides the qemu_main function to NULL.
* For other UIs, or in the absence of UIs, the platform's default
behaviour is followed.
This means that on macOS, the platform's runloop events are always
handled, regardless of chosen UI. The new PV graphics device will
thus work in all configurations. There is no functional change on other
operating systems.
Implementing this via a global function pointer variable is a bit
ugly, but it's probably worth investigating the existing UI thread rule
violations in the SDL (e.g. #2537) and GTK+ back-ends. Fixing those
issues might precipitate requirements similar but not identical to those
of the Cocoa UI; hopefully we'll see some kind of pattern emerge, which
can then be used as a basis for an overhaul. (In fact, it may turn
out to be simplest to split the UI/native platform event thread from the
QEMU main event loop on all platforms, with any UI or even none at all.)
Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Tested-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-ID: <20241223221645.29911-2-phil@philjordan.eu>
[PMD: Declare 'qemu_main' symbol in tests/qtest/fuzz/fuzz.c,
add missing g_assert_not_reached() call in main()] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/usb/hcd-xhci: Unimplemented/guest error logging for port MMIO
The XHCI device code uses tracing rather than logging on various code
paths that are so far unimplemented. In some cases, these code paths
actually indicate faulty guest software. This patch switches instances
in the read and write handlers for the port MMIO region to use
qemu_log_mask() with LOG_UNIMP or LOG_GUEST_ERROR, as appropriate in
each case.
Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241227121336.25838-5-phil@philjordan.eu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/usb/hcd-xhci-pci: Move msi/msix properties from NEC to superclass
The NEC XHCI controller exposes the underlying PCI device's msi and
msix properties, but the superclass and thus the qemu-xhci device do
not. There does not seem to be any obvious reason for this limitation.
This change moves these properties to the superclass so they are
exposed by both PCI XHCI device variants.
Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241227121336.25838-3-phil@philjordan.eu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/block/virtio-blk: Replaces request free function with g_free
The virtio_blk_free_request() function has been a 1-liner forwarding
to g_free() for a while now. We may as well call g_free on the request
pointer directly.
Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Tested-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-ID: <20241223221645.29911-14-phil@philjordan.eu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/i386/amd_iommu: Simplify non-KVM checks on XTSup feature
Generic code wanting to access KVM specific methods should
do so being protected by the 'kvm_enabled()' helper.
Doing so avoid link failures when optimization is disabled
(using --enable-debug), see for example commits c04cfb4596a
("hw/i386: fix short-circuit logic with non-optimizing builds")
and 0266aef8cd6 ("amd_iommu: Fix kvm_enable_x2apic link error
with clang in non-KVM builds").
XTSup feature depends on KVM, so protect the whole block
checking the XTSup feature with a check on whether KVM is
enabled.
Since x86_cpus_init() already checks APIC ID > 255 imply
kernel support for irqchip and X2APIC, remove the confuse
and unlikely reachable "AMD IOMMU xtsup=on requires support
on the KVM side" message.
Fix a type in "configuration" in error message.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Vasant Hegde <vasant.hegde@amd.com>
Message-Id: <20241129155802.35534-1-philmd@linaro.org>
hw/misc/vmcoreinfo: Rename opaque pointer as 'opaque'
Both QEMUResetHandler and FWCfgWriteCallback take an opaque
pointer argument, no need to cast.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241219153857.57450-3-philmd@linaro.org>
hw/misc/vmcoreinfo: Declare QOM type using DEFINE_TYPES macro
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241219153857.57450-2-philmd@linaro.org>
fw_cfg: Don't set callback_opaque NULL in fw_cfg_modify_bytes_read()
On arm/virt platform, Chen Xiang reported a Guest crash while
attempting the below steps,
1. Launch the Guest with nvdimm=on
2. Hot-add a NVDIMM dev
3. Reboot
4. Guest boots fine.
5. Reboot again.
6. Guest boot fails.
QEMU_EFI reports the below error:
ProcessCmdAddPointer: invalid pointer value in "etc/acpi/tables"
OnRootBridgesConnected: InstallAcpiTables: Protocol Error
Debugging shows that on first reboot(after hot adding NVDIMM),
Qemu updates the etc/table-loader len,
And in fw_cfg_modify_bytes_read() we set the "callback_opaque" for
the key entry to NULL. Because of this, on the second reboot,
virt_acpi_build_update() is called with a NULL "build_state" and
returns without updating the ACPI tables. This seems to be
upsetting the firmware.
To fix this, don't change the callback_opaque in fw_cfg_modify_bytes_read().
'rxbuf' is the index of the dual port RAM used.
Rename it as 'port_index'.
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20241112181044.92193-8-philmd@linaro.org>
hw/net/xilinx_ethlite: Correct maximum RX buffer size
The current max RX bufsize is set to 0x800. This is
invalid, since it contains the MMIO registers region.
Add the correct definition (valid for both TX & RX,
see datasheet p. 20, Table 11 "XPS Ethernet Lite MAC
Memory Map") and use it.
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20241112181044.92193-6-philmd@linaro.org>
Use XlnxXpsEthLite typedef, OBJECT_DECLARE_SIMPLE_TYPE macro;
convert type_init() to DEFINE_TYPES().
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Message-Id: <20241112181044.92193-5-philmd@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Message-Id: <20241112181044.92193-4-philmd@linaro.org>
hw/net/xilinx_ethlite: Convert some debug logs to trace events
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com>
Message-Id: <20241112181044.92193-3-philmd@linaro.org>
These devices are only used by the SPARC targets, which are
only built as big-endian. Therefore the DEVICE_NATIVE_ENDIAN
definition expand to DEVICE_BIG_ENDIAN (besides, the
DEVICE_LITTLE_ENDIAN case isn't tested). Simplify directly
using DEVICE_BIG_ENDIAN.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241106184612.71897-6-philmd@linaro.org>
The openrisc little-endian control is in a control register:
SR[LEE] (which we do not implement at present).
These devices are only used by the OpenRISC target, which is
only built as big-endian. Therefore the DEVICE_NATIVE_ENDIAN
definition expand to DEVICE_BIG_ENDIAN (besides, the
DEVICE_LITTLE_ENDIAN case isn't tested). Simplify directly
using DEVICE_BIG_ENDIAN.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Message-Id: <20241106184612.71897-5-philmd@linaro.org>
These devices are only used by the TriCore target, which is
only built as little-endian. Therefore the DEVICE_NATIVE_ENDIAN
definition expand to DEVICE_LITTLE_ENDIAN (besides, the
DEVICE_BIG_ENDIAN case isn't tested). Simplify directly using
DEVICE_LITTLE_ENDIAN.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Message-Id: <20241106184612.71897-3-philmd@linaro.org>
These devices are only used by the X86 targets, which are only
built as little-endian. Therefore the DEVICE_NATIVE_ENDIAN
definition expand to DEVICE_LITTLE_ENDIAN (besides, the
DEVICE_BIG_ENDIAN case isn't tested). Simplify directly using
DEVICE_LITTLE_ENDIAN.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Message-Id: <20241106184612.71897-2-philmd@linaro.org>
hw/microblaze: Propagate CPU endianness to microblaze_load_kernel()
Pass vCPU endianness as argument so we can load kernels
with different endianness (different from the qemu-system-binary
builtin one).
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Anton Johansson <anjo@rev.ng> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@amd.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241107012223.94337-3-philmd@linaro.org>
Gustavo Romero [Mon, 27 Nov 2023 05:20:20 +0000 (05:20 +0000)]
hw/misc/ivshmem-flat: Add ivshmem-flat device
Add a new device, ivshmem-flat, which is similar to the ivshmem PCI but
does not require a PCI bus. It's meant to be used on machines like those
with Cortex-M MCUs, which usually lack a PCI/PCIe bus, e.g. lm3s6965evb
and mps2-an385.
The device currently only supports the sysbus bus.
The new device, just like the ivshmem PCI device, supports both peer
notification via hardware interrupts and shared memory.
The device shared memory size can be set using the 'shmem-size' option
and it defaults to 4 MiB, which is the default size of shmem allocated
by the ivshmem server.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1134 Signed-off-by: Gustavo Romero <gustavo.romero@linaro.org>
[PMD: Rebased updating Property and using DEFINE_TYPES macro] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241216141818.111255-2-gustavo.romero@linaro.org>
The original code was right in that long name in LFN directory
entry uses other parts of the entry for the name too, not just
the original "name" field. So it is wrong to limit the offset
to be within the name field. Some other mechanism is needed
to fix the ubsan report and whole messy usage of bytes past the
given field.
Reported-by: Volker Rümelin <vr_qemu@t-online.de> Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Alexander Graf [Wed, 14 Jun 2023 22:56:24 +0000 (22:56 +0000)]
hw/pci-host/gpex: Allow more than 4 legacy IRQs
Some boards such as vmapple don't do real legacy PCI IRQ swizzling.
Instead, they just keep allocating more board IRQ lines for each new
legacy IRQ. Let's support that mode by giving instantiators a new
"nr_irqs" property they can use to support more than 4 legacy IRQ lines.
In this mode, GPEX will export more IRQ lines, one for each device.
Signed-off-by: Alexander Graf <graf@amazon.com> Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu> Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com> Tested-by: Akihiko Odaki <akihiko.odaki@daynix.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241223221645.29911-9-phil@philjordan.eu> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Stefan Hajnoczi [Sun, 29 Dec 2024 08:25:40 +0000 (03:25 -0500)]
Merge tag 'pull-request-2024-12-29' of https://gitlab.com/huth/qemu into staging
* Rework "next-cube" to bring it up-to-date with current coding standards/APIs
* Remove overlapping memory regions of the "next-cube" machine
* Encapsulate the next-cube SCSI related code into a separate device
* QOM-ify the next-rtc device
* Declare m68k devices as big endian instead of native endian
* Disable the CD-ROM drive for the next-cube machine
* tag 'pull-request-2024-12-29' of https://gitlab.com/huth/qemu: (35 commits)
next-cube: add my copyright to the top of the file
next-cube: rename old_scr2 and scr2_2 in next_scr2_rtc_update()
next-cube: move next_rtc_cmd_is_write() and next_rtc_data_in_irq() functions
next-cube: add rtc-power-out named gpio to trigger the NEXT_PWR_I interrupt
next-cube: add rtc-cmd-reset named gpio to reset the rtc state machine
next-cube: use named gpio output for next-rtc data
next-cube: move rtc-data-in gpio from next-pc to next-rtc device
next-cube: move reset of next-rtc fields from next-pc to next-rtc
next-cube: QOMify NeXTRTC
next-cube: don't use rtc phase value of -1
next-cube: use named gpio to read RTC data bit in scr2
next-cube: use named gpio to set RTC data bit in scr2
next-cube: always use retval to return rtc read values
next-cube: separate rtc read and write shift logic
next-cube: use qemu_irq to drive int_status in next_scr2_rtc_update()
next-cube: rename typedef struct NextRtc to NeXTRTC
next-cube: convert next-pc device to use Resettable interface
next-cube: rearrange NeXTState declarations to improve readability
next-cube: remove unused next.scr memory region
next-cube: add empty slots for unknown accesses to next.scr memory region
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Mark Cave-Ayland [Sun, 22 Dec 2024 13:00:12 +0000 (13:00 +0000)]
next-cube: add my copyright to the top of the file
This series has involved rewriting and/or updating a considerable part of the
next-cube emulation so update the copyright in next-cube.c to reflect this.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Thomas Huth <huth@tuxfamily.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241222130012.1013374-34-mark.cave-ayland@ilande.co.uk> Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Mark Cave-Ayland [Sun, 22 Dec 2024 13:00:11 +0000 (13:00 +0000)]
next-cube: rename old_scr2 and scr2_2 in next_scr2_rtc_update()
Rename them to old_scr2_rtc and scr2_rtc to reflect that they contain the previous
and current values of the SCR2 RTC bits.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Message-ID: <20241222130012.1013374-33-mark.cave-ayland@ilande.co.uk> Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Mark Cave-Ayland [Sun, 22 Dec 2024 13:00:10 +0000 (13:00 +0000)]
next-cube: move next_rtc_cmd_is_write() and next_rtc_data_in_irq() functions
Move these functions in next-cube.c so that they are with the rest of the
next-rtc functions.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Message-ID: <20241222130012.1013374-32-mark.cave-ayland@ilande.co.uk> Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Mark Cave-Ayland [Sun, 22 Dec 2024 13:00:09 +0000 (13:00 +0000)]
next-cube: add rtc-power-out named gpio to trigger the NEXT_PWR_I interrupt
This allows us to decouple the next-pc and next-rtc devices from each
other in next_rtc_data_in_irq().
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-ID: <20241222130012.1013374-31-mark.cave-ayland@ilande.co.uk> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Mark Cave-Ayland [Sun, 22 Dec 2024 13:00:08 +0000 (13:00 +0000)]
next-cube: add rtc-cmd-reset named gpio to reset the rtc state machine
This allows us to decouple the next-pc and next-rtc devices from each
other in next_scr2_rtc_update().
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-ID: <20241222130012.1013374-30-mark.cave-ayland@ilande.co.uk> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Mark Cave-Ayland [Sun, 22 Dec 2024 13:00:07 +0000 (13:00 +0000)]
next-cube: use named gpio output for next-rtc data
Add a named gpio output for the next-rtc data and then update
next_rtc_data_in_irq() to drive the IRQ directly. This enables the next-rtc to
next-pc data to be wired up using the standard qdev gpio APIs.
At the same time rename the pc-rtc-data-in gpio to rtc-data-in which is possible
now that the previous rtc-data-in gpio has been moved to the next-rtc device.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-ID: <20241222130012.1013374-29-mark.cave-ayland@ilande.co.uk> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Mark Cave-Ayland [Sun, 22 Dec 2024 13:00:06 +0000 (13:00 +0000)]
next-cube: move rtc-data-in gpio from next-pc to next-rtc device
Add a new rtc-data-out gpio to the next-pc device and wire it up to the next-rtc
rtc-data-in gpio using the standard qdev gpio APIs.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Message-ID: <20241222130012.1013374-28-mark.cave-ayland@ilande.co.uk> Signed-off-by: Thomas Huth <huth@tuxfamily.org>
projects / thirdparty / qemu.git / log
