Dongwon Kim [Wed, 8 May 2024 17:54:02 +0000 (10:54 -0700)]
ui/console: Use qemu_dmabuf_new() and free() helpers instead
This commit introduces utility functions for the creation and deallocation
of QemuDmaBuf instances. Additionally, it updates all relevant sections
of the codebase to utilize these new utility functions.
v7: remove prefix, "dpy_gl_" from all helpers
qemu_dmabuf_free() returns without doing anything if input is null
(Daniel P. Berrangé <berrange@redhat.com>)
call G_DEFINE_AUTOPTR_CLEANUP_FUNC for qemu_dmabuf_free()
(Daniel P. Berrangé <berrange@redhat.com>)
v8: Introduction of helpers was removed as those were already added
by the previous commit
v9: set dmabuf->allow_fences to 'true' when dmabuf is created in
virtio_gpu_create_dmabuf()/virtio-gpu-udmabuf.c
removed unnecessary spaces were accidently added in the patch,
'ui/console: Use qemu_dmabuf_new() a...'
v11: Calling qemu_dmabuf_close was removed as closing dmabuf->fd will be
done in qemu_dmabuf_free anyway.
(Daniel P. Berrangé <berrange@redhat.com>)
v12: --- Calling qemu_dmabuf_close separately as qemu_dmabuf_free doesn't
do it.
--- 'dmabuf' is now allocated space so it should be freed at the end of
dbus_scanout_texture
v13: --- Immediately free dmabuf after it is released to prevent possible
leaking of the ptr
(Marc-André Lureau <marcandre.lureau@redhat.com>)
--- Use g_autoptr macro to define *dmabuf for auto clean up instead of
calling qemu_dmabuf_free
(Marc-André Lureau <marcandre.lureau@redhat.com>)
v14: --- (vhost-user-gpu) Change qemu_dmabuf_free back to g_clear_pointer
as it was done because of some misunderstanding (v13).
--- (vhost-user-gpu) g->dmabuf[m->scanout_id] needs to be set to NULL
to prevent freed dmabuf to be accessed again in case if(fd==-1)break;
happens (before new dmabuf is allocated). Otherwise, it would cause
invalid memory access when the same function is executed. Also NULL
check should be done before qemu_dmabuf_close (it asserts dmabuf!=NULL.).
(Marc-André Lureau <marcandre.lureau@redhat.com>)
Suggested-by: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Philippe Mathieu-Daudé <philmd@linaro.org> Cc: Daniel P. Berrangé <berrange@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Message-Id: <20240508175403.3399895-6-dongwon.kim@intel.com>
Dongwon Kim [Wed, 8 May 2024 17:54:00 +0000 (10:54 -0700)]
ui/console: Use qemu_dmabuf_get_..() helpers instead
This commit updates all instances where fields within the QemuDmaBuf
struct are directly accessed, replacing them with calls to these new
helper functions.
v6: fix typos in helper names in ui/spice-display.c
v7: removed prefix, "dpy_gl_" from all helpers
v8: Introduction of helpers was removed as those were already added
by the previous commit
v11: -- Use new qemu_dmabuf_close() instead of close(qemu_dmabuf_get_fd()).
(Daniel P. Berrangé <berrange@redhat.com>)
-- Use new qemu_dmabuf_dup_fd() instead of dup(qemu_dmabuf_get_fd()).
(Daniel P. Berrangé <berrange@redhat.com>)
Suggested-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Philippe Mathieu-Daudé <philmd@linaro.org> Cc: Daniel P. Berrangé <berrange@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Message-Id: <20240508175403.3399895-4-dongwon.kim@intel.com>
Dongwon Kim [Wed, 8 May 2024 17:53:59 +0000 (10:53 -0700)]
ui/console: new dmabuf.h and dmabuf.c for QemuDmaBuf struct and helpers
New header and source files are added for containing QemuDmaBuf struct
definition and newly introduced helpers for creating/freeing the struct
and accessing its data.
v10: Change the license type for both dmabuf.h and dmabuf.c from MIT to
GPL to be in line with QEMU's default license
v11: -- Added new helpers, qemu_dmabuf_close for closing dmabuf->fd,
qemu_dmabuf_dup_fd for duplicating dmabuf->fd
(Daniel P. Berrangé <berrange@redhat.com>)
-- Let qemu_dmabuf_fee to call qemu_dmabuf_close before freeing
the struct to make sure fd is closed.
(Daniel P. Berrangé <berrange@redhat.com>)
v12: Not closing fd in qemu_dmabuf_free because there are cases fd
should still be available even after the struct is destroyed
(e.g. virtio-gpu: res->dmabuf_fd).
Suggested-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Philippe Mathieu-Daudé <philmd@linaro.org> Cc: Daniel P. Berrangé <berrange@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Message-Id: <20240508175403.3399895-3-dongwon.kim@intel.com>
Dongwon Kim [Wed, 8 May 2024 17:53:58 +0000 (10:53 -0700)]
ui/gtk: Check if fence_fd is equal to or greater than 0
'fence_fd' needs to be validated always before being referenced
And the passing condition should include '== 0' as 0 is a valid
value for the file descriptor.
Suggested-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Cc: Philippe Mathieu-Daudé <philmd@linaro.org> Cc: Daniel P. Berrangé <berrange@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Message-Id: <20240508175403.3399895-2-dongwon.kim@intel.com>
Dongwon Kim [Fri, 26 Apr 2024 22:50:59 +0000 (15:50 -0700)]
ui/gtk: Draw guest frame at refresh cycle
Draw routine needs to be manually invoked in the next refresh
if there is a scanout blob from the guest. This is to prevent
a situation where there is a scheduled draw event but it won't
happen bacause the window is currently in inactive state
(minimized or tabified). If draw is not done for a long time,
gl_block timeout and/or fence timeout (on the guest) will happen
eventually.
v2: Use gd_gl_area_draw(vc) in gtk-gl-area.c
Suggested-by: Vivek Kasireddy <vivek.kasireddy@intel.com> Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Daniel P. Berrangé <berrange@redhat.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com> Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20240426225059.3871283-1-dongwon.kim@intel.com>
* tag 'for-upstream' of https://gitlab.com/bonzini/qemu: (27 commits)
configs: disable emulators that require it if libfdt is not found
hw/xtensa: require libfdt
kconfig: express dependency of individual boards on libfdt
kconfig: allow compiling out QEMU device tree code per target
meson: move libfdt together with other dependencies
meson: pick libfdt from common_ss when building target-specific files
tests/qtest: arm: fix operation in a build without any boards or devices
i386: select correct components for no-board build
hw/i386: move rtc-reset-reinjection command out of hw/rtc
hw/i386: split x86.c in multiple parts
i386: pc: remove unnecessary MachineClass overrides
i386: correctly select code in hw/i386 that depends on other components
xen: register legacy backends via xen_backend_init
xen: initialize legacy backends from xen_bus_init()
tests/qtest: s390x: fix operation in a build without any boards or devices
s390x: select correct components for no-board build
s390: move css_migration_enabled from machine to css.c
s390_flic: add migration-enabled property
s390x: move s390_cpu_addr2state to target/s390x/sigp.c
sh4: select correct components for no-board build
...
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Paolo Bonzini [Wed, 8 May 2024 07:25:48 +0000 (09:25 +0200)]
configs: disable emulators that require it if libfdt is not found
Since boards can express their dependency on libfdt and
system/device_tree.c, only leave TARGET_NEED_FDT if the target has a
hard dependency.
Those emulators will be skipped if libfdt is disabled, or if it
is "auto" and not found and --disable-download is passed; unless
the target is mentioned explicitly in --target-list, in which case
the build will fail.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Mon, 22 Jan 2024 09:19:17 +0000 (10:19 +0100)]
hw/xtensa: require libfdt
All other boards require libfdt if it can be used (including for example
i386/x86_64), so change the "imply" to "select" and always allow -dtb
in qemu-system-xtensa.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 30 Apr 2024 19:08:15 +0000 (21:08 +0200)]
kconfig: express dependency of individual boards on libfdt
Now that boards are enabled by default and the "CONFIG_FOO=y"
entries are gone from configs/devices/, there cannot be any more
a conflicts between the default contents of configs/devices/
and a failed "depends on" clause.
With this change, each individual board or target can express
whether it needs FDT. It can then include the common code in the
build via "select DEVICE_TREE", which will also as tell meson to link
with libfdt.
This allows building non-microvm x86 emulators without having
libfdt available.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 7 May 2024 12:13:46 +0000 (14:13 +0200)]
kconfig: allow compiling out QEMU device tree code per target
Introduce a new Kconfig symbol, CONFIG_DEVICE_TREE, that specifies whether
to include the common device tree code in system/device_tree.c and to
link to libfdt. For now, include it unconditionally if libfdt is
available.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 25 Jan 2024 11:22:57 +0000 (12:22 +0100)]
meson: move libfdt together with other dependencies
Move the libfdt detection code together with other dependencies instead
of keeping it with subprojects. This has the disadvantage of performing
the detection even if no target requires libfdt; but it has the advantage
that Kconfig will be able to observe the availability of the library.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 7 May 2024 10:22:31 +0000 (12:22 +0200)]
meson: pick libfdt from common_ss when building target-specific files
Avoid having to list dependencies such as libfdt twice, both on common_ss
and specific_ss. Instead, just take all the dependencies in common_ss
and allow the target-specific libqemu-*.fa library to use them.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:44 +0000 (19:00 +0200)]
tests/qtest: arm: fix operation in a build without any boards or devices
ARM/aarch64 are easy to fix because they already have to pass a machine
type by hand. Just guard the tests with a check that the machine actually
exists.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20240509170044.190795-14-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:43 +0000 (19:00 +0200)]
i386: select correct components for no-board build
The local APIC is a part of the CPU and has callbacks that are invoked
from multiple accelerators.
The IOAPIC on the other hand is optional, but ioapic_eoi_broadcast is
used by common x86 code to implement the IOAPIC's implicit EOI mode.
Add a stub in case the IOAPIC device is not included but the APIC is.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Message-ID: <20240509170044.190795-13-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:42 +0000 (19:00 +0200)]
hw/i386: move rtc-reset-reinjection command out of hw/rtc
The rtc-reset-reinjection QMP command is specific to x86, other boards do not
have the ACK tracking functionality that is needed for RTC interrupt
reinjection. Therefore the QMP command is only included in x86, but
qmp_rtc_reset_reinjection() is implemented by hw/rtc/mc146818rtc.c
and requires tracking of all created RTC devices. Move the implementation
to hw/i386, so that 1) it is available even if no RTC device exist
2) the only RTC that exists is easily found in x86ms->rtc.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Message-ID: <20240509170044.190795-12-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:41 +0000 (19:00 +0200)]
hw/i386: split x86.c in multiple parts
Keep the basic X86MachineState definition in x86.c. Move out functions that
are only needed by other files: x86-common.c for the pc and microvm machines,
x86-cpu.c for those used by accelerator code.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Message-ID: <20240509170044.190795-11-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There is no need to override these fields of MachineClass because they are
already set to the right value in the superclass.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Message-ID: <20240509170044.190795-10-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:39 +0000 (19:00 +0200)]
i386: correctly select code in hw/i386 that depends on other components
fw_cfg.c and vapic.c are currently included unconditionally but
depend on other components. vapic.c depends on the local APIC,
while fw_cfg.c includes a piece of AML builder code that depends
on CONFIG_ACPI.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Message-ID: <20240509170044.190795-9-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:38 +0000 (19:00 +0200)]
xen: register legacy backends via xen_backend_init
It is okay to register legacy backends in the middle of xen_bus_init().
All that the registration does is record the existence of the backend
in xenstore.
This makes it possible to remove them from the build without introducing
undefined symbols in xen_be_init(). It also removes the need for the
backend_register callback, whose only purpose is to avoid registering
nonfunctional backends.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240509170044.190795-8-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:37 +0000 (19:00 +0200)]
xen: initialize legacy backends from xen_bus_init()
Prepare for moving the calls to xen_be_register() under the
control of xen_bus_init(), using the normal xen_backend_init()
method that is used by the "modern" backends.
This requires the xenstore global variable to be initialized,
which is done by xen_be_init(). To ensure that everything is
ready at the time the xen_backend_init() functions are called,
remove the xen_be_init() function from all the boards and
place it directly in xen_bus_init().
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240509170044.190795-7-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:36 +0000 (19:00 +0200)]
tests/qtest: s390x: fix operation in a build without any boards or devices
Do the bare minimum to ensure that at least a vanilla
--without-default-devices build works for all targets except i386,
x86_64 and ppc64. In particular this fixes s390x-softmmu; i386 and
x86_64 have about a dozen failing tests that do not pass -M and therefore
require a default machine type; ppc64 has the same issue, though only
with numa-test.
If we can for now ignore the cases where boards and devices are picked
by hand, drive_del-test however can be fixed easily; almost all tests
check for the virtio-blk or virtio-scsi device that they use, and are
already skipped. Only one didn't get the memo; plus another one does
not need a machine at all and can be run with -M none.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20240509170044.190795-6-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:35 +0000 (19:00 +0200)]
s390x: select correct components for no-board build
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20240509170044.190795-5-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:34 +0000 (19:00 +0200)]
s390: move css_migration_enabled from machine to css.c
The CSS subsystem uses global variables, just face the truth and use
a variable also for whether the CSS vmstate is in use; remove the
indirection of fetching it from the machine type, which makes the
TCG code depend unnecessarily on the virtio-ccw machine.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <20240509170044.190795-4-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:33 +0000 (19:00 +0200)]
s390_flic: add migration-enabled property
Instead of mucking with css_migration_enabled(), add a property specific to
the FLIC device, similar to what is done for TYPE_S390_STATTRIB.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20240509170044.190795-3-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Thu, 9 May 2024 17:00:32 +0000 (19:00 +0200)]
s390x: move s390_cpu_addr2state to target/s390x/sigp.c
This function has no dependency on the virtio-ccw machine type, though it
assumes that the CPU address corresponds to the core_id and the index.
If there is any need of something different or more fancy (unlikely)
S390 can include a MachineClass subclass and implement it there. For
now, move it to sigp.c for simplicity.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20240509170044.190795-2-pbonzini@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Tue, 7 May 2024 10:48:22 +0000 (12:48 +0200)]
configure: quote -D options that are passed through to meson
Ensure that they go through unmodified, instead of removing one layer
of quoting.
-D is a pretty specialized option and most options that can have spaces
do not need it (for example, c_args is covered by --extra-cflags).
Therefore it's unlikely that this causes actual trouble. However,
a somewhat realistic failure case would be with -Dpkg_config_path
and a pkg-config directory that contains spaces.
Cc: qemu-stable@nongnu.org Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Wed, 8 May 2024 09:10:54 +0000 (11:10 +0200)]
target/i386: fix feature dependency for WAITPKG
The VMX feature bit depends on general availability of WAITPKG,
not the other way round.
Fixes: 33cc88261c3 ("target/i386: add support for VMX_SECONDARY_EXEC_ENABLE_USER_WAIT_PAUSE", 2023-08-28) Cc: qemu-stable@nongnu.org Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Mon, 6 May 2024 15:34:55 +0000 (17:34 +0200)]
target/i386: move prefetch and multi-byte UD/NOP to new decoder
These are trivial to add, and moving them to the new decoder fixes some
corner cases: raising #UD instead of an instruction fetch page fault for
the undefined opcodes, and incorrectly rejecting 0F 18 prefetches with
register operands (which are treated as reserved NOPs).
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-stable@nongnu.org Reviewed-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Wed, 8 May 2024 15:44:12 +0000 (17:44 +0200)]
target/i386: remove PCOMMIT from TCG, deprecate property
The PCOMMIT instruction was never included in any physical processor.
TCG implements it as a no-op instruction, but its utility is debatable
to say the least. Drop it from the decoder since it is only available
with "-cpu max", which does not guarantee migration compatibility
across versions, and deprecate the property just in case someone is
using it as "pcommit=off".
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Merge tag 'pull-request-2024-05-10' of https://gitlab.com/thuth/qemu into staging
* Attach s390x sclpconsole to a proper parent in the QOM tree
* SCLP related clean-ups
* Report deprecated-props in cpu-model-expansion reply on s390x
* Deprecate "-runas" and introduce "-run-with user=..." instead
* Add some more qtest cases on LoongArch
* tag 'pull-request-2024-05-10' of https://gitlab.com/thuth/qemu:
tests/qtest: Add some test cases support on LoongArch
qemu-options: Deprecate "-runas" and introduce "-run-with user=..." instead
target/s390x: flag te and cte as deprecated
target/s390x: report deprecated-props in cpu-model-expansion reply
s390x/sclp: Simplify get_sclp_device()
s390x/event-facility: Simplify sclp_get_event_facility_bus()
s390x: Introduce a SCLPDevice pointer under the machine
hw/s390x: Attach the sclpconsole to /machine/sclp/s390-sclp-event-facility
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thomas Huth [Mon, 6 May 2024 11:20:58 +0000 (13:20 +0200)]
qemu-options: Deprecate "-runas" and introduce "-run-with user=..." instead
The old "-runas" option has the disadvantage that it is not visible
in the QAPI schema, so it is not available via the normal introspection
mechanisms. We've recently introduced the "-run-with" option for exactly
this purpose, which is meant to handle the options that affect the
runtime behavior. Thus let's introduce a "user=..." parameter here now
and deprecate the old "-runas" option.
Message-ID: <20240506112058.51446-1-thuth@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Thomas Huth <thuth@redhat.com>
target/s390x: report deprecated-props in cpu-model-expansion reply
Retain a list of deprecated features disjoint from any particular
CPU model. A query-cpu-model-expansion reply will now provide a list of
properties (i.e. features) that are flagged as deprecated. Example:
get_sclp_device() scans the whole machine to find a TYPE_SCLP object.
Now that the SCLPDevice instance is available under the machine state,
use it to simplify the lookup. While at it, remove the inline to let
the compiler decide on how to optimize.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Message-ID: <20240502131533.377719-4-clg@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
sclp_get_event_facility_bus() scans the whole machine to find a
TYPE_SCLP_EVENTS_BUS object. The SCLPDevice instance is now available
under the machine state, use it to simplify the lookup and adjust the
creation of the consoles.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Message-ID: <20240502131533.377719-3-clg@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
s390x: Introduce a SCLPDevice pointer under the machine
Initialize directly SCLPDevice from the machine init handler and
remove s390_sclp_init(). We will use the SCLPDevice pointer later to
create the consoles.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Message-ID: <20240502131533.377719-2-clg@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com>
Thomas Huth [Tue, 30 Apr 2024 19:08:43 +0000 (21:08 +0200)]
hw/s390x: Attach the sclpconsole to /machine/sclp/s390-sclp-event-facility
The sclpconsole currently does not have a proper parent in the QOM
tree, so it shows up under /machine/unattached - which is somewhat
ugly. We should rather attach it to /machine/sclp/s390-sclp-event-facility
where the other devices of type TYPE_SCLP_EVENT already reside.
Message-ID: <20240430190843.453903-1-thuth@redhat.com> Reviewed-by: Eric Farman <farman@linux.ibm.com> Reviewed-by: Cédric Le Goater <clg@redhat.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Thomas Huth <thuth@redhat.com>
# -----BEGIN PGP SIGNATURE-----
#
# iLMEAAEKAB0WIQS4/x2g0v3LLaCcbCxAov/yOSY+3wUCZjyDAgAKCRBAov/yOSY+
# 33cfA/4jE0x+eLAT161caSwM3wBOfZRClfUhXdkxLP6GvWbACVQ8l0rEZiw2PuI8
# DFReU2gqs7wAfYKt7Yy62xXlCw1B3aSUzE45gS2TGIP1GqKBwigvpW4i1SgiOoMX
# 4TA+GG16KgR9zaxO48bjjyJ1epc7S3SxdAL09p2U08D9EdSwCA==
# =RLFu
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 09 May 2024 10:02:10 AM CEST
# gpg: using RSA key B8FF1DA0D2FDCB2DA09C6C2C40A2FFF239263EDF
# gpg: Good signature from "Song Gao <m17746591750@163.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: B8FF 1DA0 D2FD CB2D A09C 6C2C 40A2 FFF2 3926 3EDF
* tag 'pull-loongarch-20240509' of https://gitlab.com/gaosong/qemu:
target/loongarch: Put cpucfg operation before CSR register
target/loongarch: Add TCG macro in structure CPUArchState
hw/loongarch: Refine default numa id calculation
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Bibo Mao [Sun, 28 Apr 2024 03:16:51 +0000 (11:16 +0800)]
target/loongarch: Put cpucfg operation before CSR register
On Loongarch, cpucfg is register for cpu feature, some other registers
depend on cpucfg feature such as perf CSR registers. Here put cpucfg
read/write operations before CSR register, so that KVM knows how many
perf CSR registers are valid from pre-set cpucfg feature information.
Signed-off-by: Bibo Mao <maobibo@loongson.cn> Reviewed-by: Song Gao <gaosong@loongson.cn>
Message-Id: <20240428031651.1354587-1-maobibo@loongson.cn> Signed-off-by: Song Gao <gaosong@loongson.cn>
Bibo Mao [Mon, 6 May 2024 01:19:12 +0000 (09:19 +0800)]
target/loongarch: Add TCG macro in structure CPUArchState
In structure CPUArchState some struct elements are only used in TCG
mode, and it is not used in KVM mode. Macro CONFIG_TCG is added to
make it simpiler in KVM mode, also there is the same modification
in c code when these structure elements are used.
When VM runs in KVM mode, TLB entries are not used and do not need
migrate. It is only useful when it runs in TCG mode.
Signed-off-by: Bibo Mao <maobibo@loongson.cn> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20240506011912.2108842-1-maobibo@loongson.cn> Signed-off-by: Song Gao <gaosong@loongson.cn>
Bibo Mao [Tue, 19 Mar 2024 02:26:06 +0000 (10:26 +0800)]
hw/loongarch: Refine default numa id calculation
With numa_test test case, there is subcase named test_def_cpu_split(),
there are 8 sockets and 2 numa nodes. Here is command line:
"-machine smp.cpus=8,smp.sockets=8 -numa node,memdev=ram -numa node"
The required result is:
node 0 cpus: 0 2 4 6
node 1 cpus: 1 3 5 7
Test case numa_test fails on LoongArch, since the actual result is:
node 0 cpus: 0 1 2 3
node 1 cpus: 4 5 6 7
It will be better if all the cpus in one socket share the same numa
node. Here socket id is used to calculate numa id in function
virt_get_default_cpu_node_id().
Signed-off-by: Bibo Mao <maobibo@loongson.cn> Reviewed-by: Song Gao <gaosong@loongson.cn>
Message-Id: <20240319022606.2994565-1-maobibo@loongson.cn> Signed-off-by: Song Gao <gaosong@loongson.cn>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Thomas Huth <thuth@redhat.com> Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20240507175356.281618-1-richard.henderson@linaro.org>
Merge tag 'hw-misc-20240508' of https://github.com/philmd/qemu into staging
Misc HW patches
- Few more g_memdup() replaced by safer g_memdup2() wrapper (Phil)
- Endianness access fixed in vfio-user config space (Mattias)
- Replace qemu_mutex_lock() -> QEMU_LOCK_GUARD in system/physmem (Phil)
- Per-AddressSpace bounce buffering (Mattias)
- Allow to compile x86 PC machines without Floppy Controller (Thomas)
- Cleanups around i386 "isa-bios" memory regions (Bernhard)
- Remove unused usb rndis_config_parameter structure (David)
- Migrate missing clock in STM32L4x5 GPIOs (Inès)
- Deprecate PPC 'ref405ep' machine and 405 CPUs (Cédric)
- Memory leak fixed in Loongarch Virt machine (Song Gao)
- hw/loongarch/ code moved around (Paolo & Bibo Mao)
- Emulate S3 suspend in loongson3_virt machine (Jiaxun)
- Implement IOCSR address space in Loongson IPI (Jiaxun)
- Use QEMU header path relative to include/ directory (Phil)
# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEE+qvnXhKRciHc/Wuy4+MsLN6twN4FAmY7+NgACgkQ4+MsLN6t
# wN4QdhAAnPGMoIGubXiZjrEmCuDU117Hm//g3wFdirn6pSr3XFWk+LGHurpgWzwf
# 6AG2ERE6ouEqbZmhfnkLA9/3cSuS/6YgCDxfQ/2Z5N7hy5+GaG60paIk2iw4wJc+
# z+3EBBNp1MKP7etP/f7XvBbempOGHSENvH83ZEKTXln6oYe/HQWBoTChV/W6Uczv
# Txf45DJ6QPPQl7cBYqtLpg9Nmc4g6d3M5t5YWzkBrecvXiBlKdZnEsI5CBtmXM8D
# h7+wIyNMl0Nc3DgQWN1SsARFDUyHG4bE4K6MS2KkGLkfFF84o8rRtEPV0RvsT+g6
# 0VkhZqLf+Gdq2w1IxCt1k5VCx8HK25+zWZypXlXY1o1dbfssqqbvXEh+3sws67eJ
# 4Hn+8QIXZBsl/zUxBbw5/CVYTHmy4Aosc1TItvDsCPnlicfV2phKSxADlL/WJCCU
# IQBwDMtxNwYM/QMP3bRIN0xNNSpB0+kOzfkRf1hVQmxH+ObqIvvIA+RbI6NhjM3L
# qXw5o0+FltLg7kPGk4Fo/rlhfnydRAE/swrwYd3WRp0sKjDe0lV3r7Bu+FZ4imdB
# 4u24FJK8ygcQOKiyHEX9cEkKruYAXeFmI3WglimYe5Po2LvIonsRLiqX97CGhgDX
# UM9Oo0Sr/0NnV7FnOW75OijNa2PE7pLVgWDjM3Xefmltqui7F/M=
# =uOIs
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 09 May 2024 12:12:40 AM CEST
# gpg: using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.org>" [full]
* tag 'hw-misc-20240508' of https://github.com/philmd/qemu: (28 commits)
misc: Use QEMU header path relative to include/ directory
hw/intc/loongson_ipi: Implement IOCSR address space for MIPS
hw/intc/loongarch_ipi: Rename as loongson_ipi
hw/intc/loongarch_ipi: Remove pointless MAX_CPU check
hw/mips/loongson3_virt: Emulate suspend function
hw/loongarch: Rename LoongArchMachineState with LoongArchVirtMachineState
hw/loongarch: Rename LOONGARCH_MACHINE with LOONGARCH_VIRT_MACHINE
hw/loongarch/virt: Fix memory leak
hw/loongarch: move memory map to boot.c
hw/ppc: Deprecate 'ref405ep' machine and 405 CPUs
hw/gpio: Handle clock migration in STM32L4x5 gpios
hw/usb/dev-network: Remove unused struct 'rndis_config_parameter'
hw/i386/x86: Extract x86_isa_bios_init() from x86_bios_rom_init()
hw/i386/x86: Don't leak "pc.bios" memory region
hw/i386/x86: Don't leak "isa-bios" memory regions
hw/i386: Have x86_bios_rom_init() take X86MachineState rather than MachineState
hw/i386/x86: Eliminate two if statements in x86_bios_rom_init()
hw/i386: Add the possibility to use i440fx and isapc without FDC
hw/i386/Kconfig: Allow to compile Q35 without FDC_ISA
hw/i386/pc: Allow to compile without CONFIG_FDC_ISA
...
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
misc: Use QEMU header path relative to include/ directory
QEMU headers are relative to the include/ directory,
not to the project root directory. Remove "include/".
See also:
https://www.qemu.org/docs/master/devel/style.html#include-directives
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20240507142737.95735-1-philmd@linaro.org>
Jiaxun Yang [Wed, 8 May 2024 13:06:48 +0000 (14:06 +0100)]
hw/intc/loongson_ipi: Implement IOCSR address space for MIPS
Implement IOCSR address space get functions for MIPS/Loongson CPUs.
For MIPS/Loongson without IOCSR (i.e. Loongson-3A1000), get_cpu_iocsr_as
will return as null, and send_ipi_data will fail with MEMTX_DECODE_ERROR,
which matches expected behavior on hardware.
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240508-loongson3-ipi-v1-3-1a7b67704664@flygoat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Jiaxun Yang [Wed, 8 May 2024 13:06:47 +0000 (14:06 +0100)]
hw/intc/loongarch_ipi: Rename as loongson_ipi
This device will be shared among LoongArch and MIPS
based Loongson machine, rename it as loongson_ipi
to reflect this nature.
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240508-loongson3-ipi-v1-2-1a7b67704664@flygoat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Since cpuid will be checked by ipi_getcpu anyway, there is
no point to enforce MAX_CPU here.
This also saved us from including loongarch board header.
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240508-loongson3-ipi-v1-1-1a7b67704664@flygoat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Jiaxun Yang [Wed, 8 May 2024 09:31:09 +0000 (10:31 +0100)]
hw/mips/loongson3_virt: Emulate suspend function
Suspend function is emulated as what hardware actually do.
Doorbell register fields are updates to include suspend value,
suspend vector is encoded in firmware blob and fw_cfg is updated
to include S3 bits as what x86 did.
Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Message-ID: <20240508-loongson3v-suspend-v1-1-186725524a39@flygoat.com>
[PMD: Use g_memdup2(), constify suspend array] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Bibo Mao [Wed, 8 May 2024 03:11:07 +0000 (11:11 +0800)]
hw/loongarch: Rename LoongArchMachineState with LoongArchVirtMachineState
Rename LoongArchMachineState with LoongArchVirtMachineState, and change
variable name LoongArchMachineState *lams with LoongArchVirtMachineState
*lvms.
Rename function specific for virtmachine loongarch_xxx()
with virt_xxx(). However some common functions keep unchanged such as
loongarch_acpi_setup()/loongarch_load_kernel(), since there functions
can be used for real hw boards.
Signed-off-by: Bibo Mao <maobibo@loongson.cn> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240508031110.2507477-3-maobibo@loongson.cn> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Bibo Mao [Wed, 8 May 2024 03:11:06 +0000 (11:11 +0800)]
hw/loongarch: Rename LOONGARCH_MACHINE with LOONGARCH_VIRT_MACHINE
On LoongArch system, there is only virt machine type now, name
LOONGARCH_MACHINE is confused, rename it with LOONGARCH_VIRT_MACHINE.
Machine name about Other real hw boards can be added in future.
Signed-off-by: Bibo Mao <maobibo@loongson.cn> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240508031110.2507477-2-maobibo@loongson.cn> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Fixes: 0cf1478d6 ("hw/loongarch: Add numa support") Signed-off-by: Song Gao <gaosong@loongson.cn> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240507022239.3113987-1-gaosong@loongson.cn> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Paolo Bonzini [Tue, 7 May 2024 14:51:35 +0000 (16:51 +0200)]
hw/loongarch: move memory map to boot.c
Ensure that it can be used even if virt.c is not included in the build, as
is the case for --without-default-devices.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Acked-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20240507145135.270803-1-pbonzini@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
The 'ref405ep' machine and PPC 405 CPU have no known users, firmware
images are not available, OpenWRT dropped support in 2019, U-Boot in
2017, Linux also is dropping support in 2024. It is time to let go of
this ancient hardware and focus on newer CPUs and platforms.
Signed-off-by: Cédric Le Goater <clg@redhat.com> Acked-by: Nicholas Piggin <npiggin@gmail.com>
Message-ID: <20240507123332.641708-1-clg@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Dr. David Alan Gilbert <dave@treblig.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240505171444.333302-5-dave@treblig.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/i386/x86: Extract x86_isa_bios_init() from x86_bios_rom_init()
The function is inspired by pc_isa_bios_init() and should eventually replace it.
Using x86_isa_bios_init() rather than pc_isa_bios_init() fixes pflash commands
to work in the isa-bios region.
While at it convert the magic number 0x100000 (== 1MiB) to increase readability.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Message-ID: <20240508175507.22270-6-shentey@gmail.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Fix the leaking in x86_bios_rom_init() by adding a "bios" attribute to
X86MachineState. Note that it is only used in the -bios case.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Message-ID: <20240508175507.22270-5-shentey@gmail.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Fix the leaking in x86_bios_rom_init() and pc_isa_bios_init() by adding an
"isa_bios" attribute to X86MachineState.
Suggested-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Message-ID: <20240508175507.22270-4-shentey@gmail.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Bernhard Beschow [Tue, 30 Apr 2024 15:06:39 +0000 (17:06 +0200)]
hw/i386: Have x86_bios_rom_init() take X86MachineState rather than MachineState
The function creates and leaks two MemoryRegion objects regarding the BIOS which
will be moved into X86MachineState in the next steps to avoid the leakage.
Signed-off-by: Bernhard Beschow <shentey@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240430150643.111976-3-shentey@gmail.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Bernhard Beschow [Tue, 30 Apr 2024 15:06:38 +0000 (17:06 +0200)]
hw/i386/x86: Eliminate two if statements in x86_bios_rom_init()
Given that memory_region_set_readonly() is a no-op when the readonlyness is
already as requested it is possible to simplify the pattern
if (condition) {
foo(true);
}
to
foo(condition);
which is shorter and allows to see the invariant of the code more easily.
Signed-off-by: Bernhard Beschow <shentey@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240430150643.111976-2-shentey@gmail.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Thomas Huth [Thu, 25 Apr 2024 18:43:15 +0000 (20:43 +0200)]
hw/i386: Add the possibility to use i440fx and isapc without FDC
The i440fx and the isapc machines can be used in binaries without
FDC, too. We just have to make sure that they don't try to instantiate
the FDC when it is not available.
Signed-off-by: Thomas Huth <thuth@redhat.com> Acked-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240425184315.553329-4-thuth@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Thomas Huth [Thu, 25 Apr 2024 18:43:14 +0000 (20:43 +0200)]
hw/i386/Kconfig: Allow to compile Q35 without FDC_ISA
The q35 machine can be used without floppy disk controller (FDC),
but due to our current Kconfig setup, the FDC code is still always
included in the binary. To fix this, the "PC" config option should
only imply the "FDC_ISA" instead of always selecting it.
The i440fx and the isa-pc machine currently always instantiate
the FDC, so we have to add the select statements now there instead.
Signed-off-by: Thomas Huth <thuth@redhat.com> Acked-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240425184315.553329-3-thuth@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Thomas Huth [Thu, 25 Apr 2024 18:43:13 +0000 (20:43 +0200)]
hw/i386/pc: Allow to compile without CONFIG_FDC_ISA
The q35 machine can work without FDC. But to be able to also link
a QEMU binary that does not include the FDC code, we have to make
it possible to disable the spots that call into the FDC code.
Signed-off-by: Thomas Huth <thuth@redhat.com> Acked-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20240425184315.553329-2-thuth@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Instead of using a single global bounce buffer, give each AddressSpace
its own bounce buffer. The MapClient callback mechanism moves to
AddressSpace accordingly.
This is in preparation for generalizing bounce buffer handling further
to allow multiple bounce buffers, with a total allocation limit
configured per AddressSpace.
Reviewed-by: Peter Xu <peterx@redhat.com> Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Mattias Nissler <mnissler@rivosinc.com>
Message-ID: <20240507094210.300566-2-mnissler@rivosinc.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
[PMD: Split patch, part 2/2] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
system/physmem: Replace qemu_mutex_lock() calls with QEMU_LOCK_GUARD
Simplify cpu_[un]register_map_client() and cpu_notify_map_clients()
by replacing the pair of qemu_mutex_lock/qemu_mutex_unlock calls by
the WITH_QEMU_LOCK_GUARD() macro.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Mattias Nissler <mnissler@rivosinc.com> Reviewed-by: Peter Xu <peterx@redhat.com>
Message-Id: <20240507123025.93391-2-philmd@linaro.org>
Mattias Nissler [Wed, 23 Aug 2023 09:29:30 +0000 (02:29 -0700)]
hw/remote/vfio-user: Fix config space access byte order
PCI config space is little-endian, so on a big-endian host we need to
perform byte swaps for values as they are passed to and received from
the generic PCI config space access machinery.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Jagannathan Raman <jag.raman@oracle.com> Signed-off-by: Mattias Nissler <mnissler@rivosinc.com>
Message-ID: <20240507094210.300566-6-mnissler@rivosinc.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/ppc/spapr_pci: Replace g_memdup() by g_memdup2()
Per https://discourse.gnome.org/t/port-your-module-from-g-memdup-to-g-memdup2-now/5538
The old API took the size of the memory to duplicate as a guint,
whereas most memory functions take memory sizes as a gsize. This
made it easy to accidentally pass a gsize to g_memdup(). For large
values, that would lead to a silent truncation of the size from 64
to 32 bits, and result in a heap area being returned which is
significantly smaller than what the caller expects. This can likely
be exploited in various modules to cause a heap buffer overflow.
Replace g_memdup() by the safer g_memdup2() wrapper.
Trivially safe because the argument was directly from sizeof.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Acked-by: David Gibson <david@gibson.dropber.id.au>
Message-Id: <20210903174510.751630-17-philmd@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
hw/hppa/machine: Replace g_memdup() by g_memdup2()
Per https://discourse.gnome.org/t/port-your-module-from-g-memdup-to-g-memdup2-now/5538
The old API took the size of the memory to duplicate as a guint,
whereas most memory functions take memory sizes as a gsize. This
made it easy to accidentally pass a gsize to g_memdup(). For large
values, that would lead to a silent truncation of the size from 64
to 32 bits, and result in a heap area being returned which is
significantly smaller than what the caller expects. This can likely
be exploited in various modules to cause a heap buffer overflow.
Replace g_memdup() by the safer g_memdup2() wrapper.
Trivially safe because the argument was directly from sizeof.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20210903174510.751630-12-philmd@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Per https://discourse.gnome.org/t/port-your-module-from-g-memdup-to-g-memdup2-now/5538
The old API took the size of the memory to duplicate as a guint,
whereas most memory functions take memory sizes as a gsize. This
made it easy to accidentally pass a gsize to g_memdup(). For large
values, that would lead to a silent truncation of the size from 64
to 32 bits, and result in a heap area being returned which is
significantly smaller than what the caller expects. This can likely
be exploited in various modules to cause a heap buffer overflow.
Replace g_memdup() by the safer g_memdup2() wrapper.
Trivially safe because the argument was directly from sizeof.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Acked-by: David Gibson <david@gibson.dropbear.id.au>
Message-Id: <20210903174510.751630-27-philmd@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
block/qcow2-bitmap: Replace g_memdup() by g_memdup2()
Per https://discourse.gnome.org/t/port-your-module-from-g-memdup-to-g-memdup2-now/5538
The old API took the size of the memory to duplicate as a guint,
whereas most memory functions take memory sizes as a gsize. This
made it easy to accidentally pass a gsize to g_memdup(). For large
values, that would lead to a silent truncation of the size from 64
to 32 bits, and result in a heap area being returned which is
significantly smaller than what the caller expects. This can likely
be exploited in various modules to cause a heap buffer overflow.
Replace g_memdup() by the safer g_memdup2() wrapper.
Trivially safe because the argument was directly from sizeof.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210903174510.751630-6-philmd@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Peter missed the Sphinx HMP document for the "resume/-r" flag in commit 7a4da28b26 ("qmp: hmp: add migrate "resume" option"). Add it.
When at it, slightly cleanup the lines around:
- Move "detach/-d" to a separate section rather than appending it at the
end of the command description. Add a hint for how to query the migration
results in detached mode.
- Add "postcopy" keyword to "resume/-r" help messages, as it only applies
to postcopy.
Cc: Dr. David Alan Gilbert <dave@treblig.org> Cc: Fabiano Rosas <farosas@suse.de> Fixes: 7a4da28b26 ("qmp: hmp: add migrate "resume" option") Reported-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com> Signed-off-by: Fabiano Rosas <farosas@suse.de>
The fd: URI can currently trigger two different types of migration, a
TCP migration using sockets and a file migration using a plain
file. This is in conflict with the recently introduced (8.2) QMP
migrate API that takes structured data as JSON-like format. We cannot
keep the same backend for both types of migration because with the new
API the code is more tightly coupled to the type of transport. This
means a TCP migration must use the 'socket' transport and a file
migration must use the 'file' transport.
If we keep allowing fd: when using a file, this creates an issue when
the user converts the old-style (fd:) to the new style ("transport":
"socket") invocation because the file descriptor in question has
previously been allowed to be either a plain file or a socket.
To avoid creating too much confusion, we can simply deprecate the fd:
+ file usage, which is thought to be rarely used currently and instead
establish a 1:1 correspondence between fd: URI and socket transport,
and file: URI and file transport.
Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Fabiano Rosas <farosas@suse.de>
The 'compress' migration capability enables the old compression code
which has shown issues over the years and is thought to be less stable
and tested than the more recent multifd-based compression. The old
compression code has been deprecated in 8.2 and now is time to remove
it.
Deprecation commit 864128df46 ("migration: Deprecate old compression
method").
Acked-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Fabiano Rosas <farosas@suse.de>
The block migration has been considered obsolete since QEMU 8.2 in
favor of the more flexible storage migration provided by the
blockdev-mirror driver. Two releases have passed so now it's time to
remove it.
migration: Remove 'blk/-b' option from migrate commands
The block migration is considered obsolete and has been deprecated in
8.2. Remove the migrate command option that enables it. This only
affects the QMP and HMP commands, the feature can still be accessed by
setting the migration 'block' capability. The whole feature will be
removed in a future patch.
Deprecation commit 8846b5bfca ("migration: migrate 'blk' command
option is deprecated.").
Reviewed-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Fabiano Rosas <farosas@suse.de>
qapi: introduce exit-on-error parameter for migrate-incoming
Now we do set MIGRATION_FAILED state, but don't give a chance to
orchestrator to query migration state and get the error.
Let's provide a possibility for QMP-based orchestrators to get an error
like with outgoing migration.
For hmp_migrate_incoming(), let's enable the new behavior: HMP is not
and ABI, it's mostly intended to use by developer and it makes sense
not to stop the process.
For x-exit-preconfig, let's keep the old behavior:
- it's called from init(), so here we want to keep current behavior by
default
- it does exit on error by itself as well
So, if we want to change the behavior of x-exit-preconfig, it should be
another patch.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru> Acked-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Fabiano Rosas <farosas@suse.de> Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Fabiano Rosas <farosas@suse.de>
It's bad idea to leave critical section with error object freed, but
s->error still set, this theoretically may lead to use-after-free
crash. Let's avoid it.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru> Reviewed-by: Fabiano Rosas <farosas@suse.de> Reviewed-by: Peter Xu <peterx@redhat.com> Signed-off-by: Fabiano Rosas <farosas@suse.de>
* tag 'for-upstream' of https://gitlab.com/bonzini/qemu: (26 commits)
target/i386: remove duplicate prefix decoding
target/i386: split legacy decoder into a separate function
target/i386: decode x87 instructions in a separate function
target/i386: remove now-converted opcodes from old decoder
target/i386: port extensions of one-byte opcodes to new decoder
target/i386: move BSWAP to new decoder
target/i386: move remaining conditional operations to new decoder
target/i386: merge and enlarge a few ranges for call to disas_insn_new
target/i386: move C0-FF opcodes to new decoder (except for x87)
target/i386: generalize gen_movl_seg_T0
target/i386: move 60-BF opcodes to new decoder
target/i386: allow instructions with more than one immediate
target/i386: extract gen_far_call/jmp, reordering temporaries
target/i386: move 00-5F opcodes to new decoder
target/i386: reintroduce debugging mechanism
target/i386: cleanup *gen_eob*
target/i386: clarify the "reg" argument of functions returning CCPrepare
target/i386: do not use s->T0 and s->T1 as scratch registers for CCPrepare
target/i386: extend cc_* when using them to compute flags
target/i386: pull cc_op update to callers of gen_jmp_rel{,_csize}
...
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Merge tag 'pull-tcg-20240507' of https://gitlab.com/rth7680/qemu into staging
tcg: Add write_aofs to GVecGen3i
tcg/i386: Simplify immediate 8-bit logical vector shifts
tcg/i386: Optimize setcond of TST{EQ,NE} with 0xffffffff
tcg/optimize: Optimize setcond with zmask
accel/tcg: Introduce CF_BP_PAGE
target/sh4: Update DisasContextBase.insn_start
gitlab: Drop --static from s390x linux-user build
gitlab: Streamline ubuntu-22.04-s390x
# -----BEGIN PGP SIGNATURE-----
#
# iQFRBAABCgA7FiEEekgeeIaLTbaoWgXAZN846K9+IV8FAmY6OoAdHHJpY2hhcmQu
# aGVuZGVyc29uQGxpbmFyby5vcmcACgkQZN846K9+IV8FEwf7Bhs9bV2Kp4LxUzGq
# +dSHHc/WuCyIILLDQ4kZyXvILuI59wYhrWBUUTzBnAZ/tEf0oMG2y57F/lIcxz9w
# VvsFicMOhtjQ8iBEfl/rkkaYs9BLcxqMTAA3PxNBE6l3bzjcHSTkhey4MoPGRibn
# CkwaLzb2ebNjfgzC1IsNf/tyiMXl0tBQM7JVV4EztaOGEmqw8X0/PyVZDiC3WUNC
# tf9yqiNIlgGkn7rj3sT/rNdi4xlzQybgrb1MCFT6z5cqsW2bwqivRpxHi4yulHKI
# VhYA3kud+TX2ASukpibsSkA+9SbcH/qwOugPhPIu+KANsFUcVKL6Anzv6Ysl9kZ0
# +Wnbow==
# =FJCW
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 07 May 2024 07:28:16 AM PDT
# gpg: using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F
# gpg: issuer "richard.henderson@linaro.org"
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [ultimate]
* tag 'pull-tcg-20240507' of https://gitlab.com/rth7680/qemu:
gitlab: Streamline ubuntu-22.04-s390x
gitlab: Drop --static from s390x linux-user build
gitlab: Drop --disable-libssh from ubuntu-22.04-s390x.yml
target/sh4: Update DisasContextBase.insn_start
accel/tcg: Introduce CF_BP_PAGE
tcg/optimize: Optimize setcond with zmask
tcg/i386: Optimize setcond of TST{EQ,NE} with 0xffffffff
tcg/i386: Simplify immediate 8-bit logical vector shifts
tcg: Add write_aofs to GVecGen3i
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Paolo Bonzini [Tue, 9 Apr 2024 15:31:23 +0000 (17:31 +0200)]
target/i386: remove duplicate prefix decoding
Now that a bulk of opcodes go through the new decoder, it is sensible
to do some cleanup. Go immediately through disas_insn_new and only jump
back after parsing the prefixes.
disas_insn() now only contains the three sigsetjmp cases, and they
are more easily managed if they are inlined into i386_tr_translate_insn.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Fri, 22 Dec 2023 16:30:06 +0000 (17:30 +0100)]
target/i386: split legacy decoder into a separate function
Split the bits that have some duplication with disas_insn_new, from
those that should be the main topic of the conversion. This is the
first step towards removing duplicate decoding of prefixes between
disas_insn and disas_insn_new.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Fri, 13 Oct 2023 06:42:58 +0000 (08:42 +0200)]
target/i386: decode x87 instructions in a separate function
These are unlikely to be converted to the table-based decoding
soon (perhaps there could be generic ESC decoding in decode-new.c.inc
for the Mod/RM byte, but not operand decoding), so keep them separate
from the remaining legacy-decoded instructions.
Acked-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Paolo Bonzini [Sat, 21 Oct 2023 15:51:23 +0000 (17:51 +0200)]
target/i386: port extensions of one-byte opcodes to new decoder
A few two-byte opcodes are simple extensions of existing one-byte opcodes;
they are easy to decode and need no change to emit.c.inc. Port them to
the new decoder.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
projects / mirror_qemu.git / log