Peter Maydell [Fri, 5 Jan 2024 16:08:58 +0000 (16:08 +0000)]
Merge tag 'hw-cpus-20240105' of https://github.com/philmd/qemu into staging
HW core patch queue
- Unify CPU QOM type checks (Gavin)
- Simplify uses of some CPU related property (Philippe)
(start-powered-off, ARM reset-cbar and mp-affinity)
- Header and documentation cleanups (Zhao, Philippe)
- Have Memory API return boolean indicating possible error
- Fix frame filter mask in CAN sja1000 model (Pavel)
- QOM embed MCF5206 timer into SoC (Thomas)
- Simplify LEON3 qemu_irq_ack handler (Clément)
# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEE+qvnXhKRciHc/Wuy4+MsLN6twN4FAmWYIxwACgkQ4+MsLN6t
# wN66fA//UBwgYqcdpg6Wz17qzgq1TWeZHHzYh7HbZRUCxhdSgS6TSQOH9Fi8VNYq
# Ed5a5l4ovP/2NRN1/S5PPBydyKXTU7wintHm2+suQbLSmplIE6yr0Ca6o8FLEeJ3
# hnE0dAoQCLS7eDpoeOEpGjzmJFiBSWLvyqAZLa/rZkCnCiZRHB6g/nAEM8I3I9bl
# //H20d3a/fektZxGnpEAeoMxrl4iA9hkFYVW8lbu6EhNFBPUkkj5Y8w47Kq/BIvD
# NmLTPgu4d7oahwlfsM6jWdRDG9zlEkXQor817PHwl00o45yAfeITsy40GvJeEYaI
# BcDLFfWrSm9SQb7/suXGeyU/SLmx7rsmJWfNYUoMr6807QcSH4ScPCfgzEQ4j8IV
# PmeVsxxLxT9CSzfxhMx5cXt33H2l+tEzwJ5UJCLQvmvTu+aDkt46Q09X/7j0z89m
# zSk/HBtdACIzwEWBAJsKuzarRTZNUvyXEsOxZ5l7xOxJpzpsNV2YVuChClVGtHOJ
# kr1PE2hxEMPY1vDyKU6ckDvW+XXgYhOXrPAxdx8gIwwd4oyDC5vVlIajvlqbOAsp
# Es7zq40b/is3ZnByEDbZ+yYvdYRLtVf/lDPK3KIv7IhrTNzH/HT1egshOQAVirY1
# Gw8f3fXqL3/84w383VI4efrSlKBJeb0i2SJ50y2N1clrF1qnlx0=
# =an4B
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 05 Jan 2024 15:41:16 GMT
# gpg: using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.org>" [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: FAAB E75E 1291 7221 DCFD 6BB2 E3E3 2C2C DEAD C0DE
* tag 'hw-cpus-20240105' of https://github.com/philmd/qemu: (71 commits)
target/sparc: Simplify qemu_irq_ack
hw/net/can/sja1000: fix bug for single acceptance filter and standard frame
hw/m68k/mcf5206: Embed m5206_timer_state in m5206_mbar_state
hw/pci-host/raven: Propagate error in raven_realize()
hw/nvram: Simplify memory_region_init_rom_device() calls
hw/misc: Simplify memory_region_init_ram_from_fd() calls
hw/sparc: Simplify memory_region_init_ram_nomigrate() calls
hw/arm: Simplify memory_region_init_rom() calls
hw: Simplify memory_region_init_ram() calls
misc: Simplify qemu_prealloc_mem() calls
util/oslib: Have qemu_prealloc_mem() handler return a boolean
backends: Reduce variable scope in host_memory_backend_memory_complete
backends: Have HostMemoryBackendClass::alloc() handler return a boolean
backends: Simplify host_memory_backend_memory_complete()
backends: Use g_autofree in HostMemoryBackendClass::alloc() handlers
memory: Have memory_region_init_ram_from_fd() handler return a boolean
memory: Have memory_region_init_ram_from_file() handler return a boolean
memory: Have memory_region_init_resizeable_ram() return a boolean
memory: Have memory_region_init_rom_device() handler return a boolean
memory: Simplify memory_region_init_rom_device_nomigrate() calls
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Pavel Pisa [Wed, 3 Jan 2024 23:14:26 +0000 (00:14 +0100)]
hw/net/can/sja1000: fix bug for single acceptance filter and standard frame
A CAN sja1000 standard frame filter mask has been computed and applied
incorrectly for standard frames when single Acceptance Filter Mode
(MOD_AFM = 1) has been selected. The problem has not been found
by Linux kernel testing because it uses dual filter mode (MOD_AFM = 0)
and leaves falters fully open.
The problem has been noticed by Grant Ramsay when testing with Zephyr
RTOS which uses single filter mode.
Signed-off-by: Pavel Pisa <pisa@cmp.felk.cvut.cz> Reported-by: Grant Ramsay <gramsay@enphaseenergy.com>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2028 Fixes: 733210e754 ("hw/net/can: SJA1000 chip register level emulation")
Message-ID: <20240103231426.5685-1-pisa@fel.cvut.cz>
Thomas Huth [Thu, 21 Dec 2023 12:29:39 +0000 (13:29 +0100)]
hw/m68k/mcf5206: Embed m5206_timer_state in m5206_mbar_state
There's no need to explicitely allocate the memory here, we can
simply embed it into the m5206_mbar_state instead.
Signed-off-by: Thomas Huth <huth@tuxfamily.org> Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231221122939.11001-1-huth@tuxfamily.org>
hw/pci-host/raven: Propagate error in raven_realize()
When an Error** reference is available, it is better to
propagate local errors, rather then using generic ones,
which might terminate the whole QEMU process.
Since qemu_prealloc_mem() returns whether or not an error
occured, we don't need to check the @errp pointer. Remove
local_err uses when we can return directly.
util/oslib: Have qemu_prealloc_mem() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have qemu_prealloc_mem()
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-19-philmd@linaro.org>
backends: Have HostMemoryBackendClass::alloc() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have HostMemoryBackendClass::alloc
return a boolean indicating whether an error is set or not.
memory: Have memory_region_init_ram_from_fd() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_ram_from_fd
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-14-philmd@linaro.org>
memory: Have memory_region_init_ram_from_file() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_ram_from_file
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-13-philmd@linaro.org>
memory: Have memory_region_init_resizeable_ram() return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_resizeable_ram
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-12-philmd@linaro.org>
memory: Have memory_region_init_rom_device() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_rom_device
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-11-philmd@linaro.org>
memory: Have memory_region_init_rom_device_nomigrate() return a boolean
Following the example documented since commit e3fe3988d7
("error: Document Error API usage rules"), have
memory_region_init_rom_device_nomigrate() return a boolean
indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-9-philmd@linaro.org>
memory: Have memory_region_init_rom() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_rom()
return a boolean indicating whether an error is set or not.
memory: Have memory_region_init_ram() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_ram()
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-7-philmd@linaro.org>
memory: Have memory_region_init_rom_nomigrate() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_rom_nomigrate
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-4-philmd@linaro.org>
[PMD: Only update 'readonly' field on success (Manos Pitsidianakis)]
Message-Id: <af352e7d-3346-4705-be77-6eed86858d18@linaro.org>
memory: Have memory_region_init_ram_nomigrate() handler return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_ram_nomigrate
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-3-philmd@linaro.org>
memory: Have memory_region_init_ram_flags_nomigrate() return a boolean
Following the example documented since commit e3fe3988d7 ("error:
Document Error API usage rules"), have memory_region_init_ram_nomigrate
return a boolean indicating whether an error is set or not.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Gavin Shan <gshan@redhat.com>
Message-Id: <20231120213301.24349-2-philmd@linaro.org>
hw: Simplify accesses to the CPUState::'start-powered-off' property
The 'start-powered-off' property has been added to ARM CPUs in
commit 5de164304a ("arm: Allow secondary KVM CPUs to be booted
via PSCI"), then eventually got generalized to all CPUs in commit c1b701587e ("target/arm: Move start-powered-off property to generic
CPUState"). Since all CPUs have it, no need to check whether it is
available. Updating this property can't fail, so use &error_abort.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20231123143813.42632-5-philmd@linaro.org>
hw/ppc/spapr_cpu_core: Access QDev properties with proper API
CPUState::start_powered_off field is part of the internal
implementation of a QDev CPU. It is exposed as the QDev
"start-powered-off" property. External components should
use the qdev properties API to access it.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Harsh Prateek Bora <harshpb@linux.ibm.com>
Message-Id: <20231123143813.42632-2-philmd@linaro.org>
hw/arm/bcm2836: Use ARM_CPU 'mp-affinity' property
The 'mp-affinity' property is present since commit 15a21fe028
("target-arm: Add mp-affinity property for ARM CPU class").
Use it and remove a /* TODO */ comment. Since all ARM CPUs
have this property, use &error_abort, because this call can
not fail.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20231123143813.42632-4-philmd@linaro.org>
hw/arm/bcm2836: Simplify use of 'reset-cbar' property
bcm2836_realize() is called by
- bcm2836_class_init() which sets:
bc->cpu_type = ARM_CPU_TYPE_NAME("cortex-a7")
- bcm2837_class_init() which sets:
bc->cpu_type = ARM_CPU_TYPE_NAME("cortex-a53")
Both Cortex-A7 / A53 have the ARM_FEATURE_CBAR set. If it isn't,
then this is a programming error: use &error_abort.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20231123143813.42632-3-philmd@linaro.org>
Zhao Liu [Mon, 27 Nov 2023 14:56:11 +0000 (22:56 +0800)]
hw/cpu/cluster: Cleanup unused included header in cluster.c
Remove unused header (qemu/module.h and qemu/cutils.h) in cluster.c,
and reorder the remaining header files (except qemu/osdep.h) in
alphabetical order.
Tested by "./configure" and then "make".
Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231127145611.925817-3-zhao1.liu@linux.intel.com>
Zhao Liu [Mon, 27 Nov 2023 14:56:10 +0000 (22:56 +0800)]
hw/cpu/core: Cleanup unused included header in core.c
Remove unused header (qemu/module.h and sysemu/cpus.h) in core.c,
and reorder the remaining header files (except qemu/osdep.h) in
alphabetical order.
Tested by "./configure" and then "make".
Signed-off-by: Zhao Liu <zhao1.liu@intel.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231127145611.925817-2-zhao1.liu@linux.intel.com>
'next_cpu' was converted to 'node' in commit bdc44640cb
("cpu: Use QTAILQ for CPU list").
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231129183243.15859-1-philmd@linaro.org>
hw/core/cpu: Remove final vestiges of dynamic state tracing
The dynamic state tracing was removed in commit d0aaf08bb9.
Fixes: d0aaf08bb9 ("tcg: remove the final vestiges of dstate") Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231129182734.15565-1-philmd@linaro.org>
Gavin Shan [Mon, 4 Dec 2023 00:47:22 +0000 (10:47 +1000)]
hw/arm/virt: Hide host CPU model for tcg
The 'host' CPU model isn't available until KVM or HVF is enabled.
For example, the following error messages are seen when the guest
is started with option '-cpu cortex-a8' on tcg after the next commit
is applied to check the CPU type in machine_run_board_init().
Gavin Shan [Mon, 4 Dec 2023 00:47:21 +0000 (10:47 +1000)]
machine: Print CPU model name instead of CPU type
The names of supported CPU models instead of CPU types should be
printed when the user specified CPU type isn't supported, to be
consistent with the output from '-cpu ?'.
Correct the error messages to print CPU model names instead of CPU
type names.
Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231204004726.483558-5-gshan@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Gavin Shan [Mon, 4 Dec 2023 00:47:20 +0000 (10:47 +1000)]
machine: Improve is_cpu_type_supported()
It's no sense to check the CPU type when mc->valid_cpu_types[0] is
NULL, which is a program error. Raise an assert on this.
A precise hint for the error message is given when mc->valid_cpu_types[0]
is the only valid entry. Besides, enumeration on mc->valid_cpu_types[0]
when we have mutiple valid entries there is avoided to increase the code
readability, as suggested by Philippe Mathieu-Daudé.
Besides, @cc comes from machine->cpu_type or mc->default_cpu_type. For
the later case, it can be NULL and it's also a program error. We should
use assert() in this case.
Signed-off-by: Gavin Shan <gshan@redhat.com>
Message-ID: <20231204004726.483558-4-gshan@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Gavin Shan [Mon, 4 Dec 2023 00:47:19 +0000 (10:47 +1000)]
machine: Introduce helper is_cpu_type_supported()
The logic, to check if the specified CPU type is supported in
machine_run_board_init(), is independent enough. Factor it out into
helper is_cpu_type_supported(). machine_run_board_init() looks a bit
clean with this. Since we're here, @machine_class is renamed to @mc to
avoid multiple line spanning of code. The comments are tweaked a bit
either.
No functional change intended.
Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231204004726.483558-3-gshan@redhat.com>
[PMD: Only call new helper if machine->cpu_type is not NULL] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Gavin Shan [Mon, 4 Dec 2023 00:47:18 +0000 (10:47 +1000)]
machine: Use error handling when CPU type is checked
Functions that use an Error **errp parameter to return errors should
not also report them to the user, because reporting is the caller's
job. The principle is violated by machine_run_board_init() because
it calls error_report(), error_printf(), and exit(1) when the machine
doesn't support the requested CPU type.
Clean this up by using error_setg() and error_append_hint() instead.
No functional change, as the only caller passes &error_fatal.
Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20231204004726.483558-2-gshan@redhat.com>
[PMD: Correct error_append_hint() argument] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Add a helper to return a machine default CPU type.
If this machine is restricted to a single CPU type,
use it as default, obviously.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231116163726.28952-1-philmd@linaro.org>
Signed-off-by: Gavin Shan <gshan@redhat.com>
Message-ID: <20231114235628.534334-22-gshan@redhat.com>
[PMD: Split patch in 2, only include the "Use generic cpu_list" change]
Message-ID: <51ffd060-b2f8-405c-83e1-a0663c0183f5@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Gavin Shan [Tue, 14 Nov 2023 23:56:03 +0000 (09:56 +1000)]
cpu: Add generic cpu_list()
Add generic cpu_list() to replace the individual target's implementation
in the subsequent commits. Currently, there are 3 targets with no cpu_list()
implementation: microblaze and nios2. With this applied, those two targets
switch to the generic cpu_list().
[gshan@gshan q]$ ./build/qemu-system-microblaze -cpu ?
Available CPUs:
microblaze-cpu
[gshan@gshan q]$ ./build/qemu-system-nios2 -cpu ?
Available CPUs:
nios2-cpu
Suggested-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231114235628.534334-7-gshan@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Gavin Shan [Tue, 14 Nov 2023 23:56:02 +0000 (09:56 +1000)]
cpu: Add helper cpu_model_from_type()
Add helper cpu_model_from_type() to extract the CPU model name from
the CPU type name in two circumstances: (1) The CPU type name is the
combination of the CPU model name and suffix. (2) The CPU type name
is same to the CPU model name.
The helper will be used in the subsequent commits to conver the
CPU type name to the CPU model name.
Suggested-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20231114235628.534334-6-gshan@redhat.com>
[PMD: Mention returned string must be released with g_free()] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
cpu: Call object_class_dynamic_cast() once in cpu_class_by_name()
For all targets, the CPU class returned from CPUClass::class_by_name()
and object_class_dynamic_cast(oc, CPU_RESOLVING_TYPE) need to be
compatible. Lets apply the check in cpu_class_by_name() for once,
instead of having the check in CPUClass::class_by_name() for individual
target.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Gavin Shan <gshan@redhat.com>
Message-ID: <20231114235628.534334-4-gshan@redhat.com>
Gavin Shan [Tue, 14 Nov 2023 23:55:59 +0000 (09:55 +1000)]
target/hppa: Remove object_class_is_abstract()
Since commit 3a9d0d7b64 ("hw/cpu: Call object_class_is_abstract()
once in cpu_class_by_name()"), there is no need to check if @oc is
abstract because it has been covered by cpu_class_by_name().
Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231114235628.534334-3-gshan@redhat.com>
[PMD: Mention commit 3a9d0d7b64] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Gavin Shan [Tue, 14 Nov 2023 23:55:58 +0000 (09:55 +1000)]
target/alpha: Remove fallback to ev67 cpu class
'ev67' CPU class will be returned to match everything, which makes
no sense as mentioned in the comments. Remove the logic to fall
back to 'ev67' CPU class to match everything.
Signed-off-by: Gavin Shan <gshan@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231114235628.534334-2-gshan@redhat.com>
[PMD: Reword subject, replace 'any' -> 'ev67' on linux-user] Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Peter Maydell [Fri, 5 Jan 2024 13:35:25 +0000 (13:35 +0000)]
Merge tag 'migration-20240104-pull-request' of https://gitlab.com/peterx/qemu into staging
migration 1st pull for 9.0
- We lost Juan and Leo in the maintainers file
- Steven's suspend state fix
- Steven's fix for coverity on migrate_mode
- Avihai's migration cleanup series
# -----BEGIN PGP SIGNATURE-----
#
# iIgEABYKADAWIQS5GE3CDMRX2s990ak7X8zN86vXBgUCZZY0TxIccGV0ZXJ4QHJl
# ZGhhdC5jb20ACgkQO1/MzfOr1wbSxgEAoM5g3wkc22lpAlRpU+hJUqT9NVOVQSK+
# Fk7XJYTdSgABAKzykA6hAmU5Kj+yVI6jI874SVZbs2FWpFs4osvsKk4D
# =sfuM
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 04 Jan 2024 04:30:07 GMT
# gpg: using EDDSA key B9184DC20CC457DACF7DD1A93B5FCCCDF3ABD706
# gpg: issuer "peterx@redhat.com"
# gpg: Good signature from "Peter Xu <xzpeter@gmail.com>" [unknown]
# gpg: aka "Peter Xu <peterx@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: B918 4DC2 0CC4 57DA CF7D D1A9 3B5F CCCD F3AB D706
* tag 'migration-20240104-pull-request' of https://gitlab.com/peterx/qemu: (26 commits)
migration: fix coverity migrate_mode finding
migration/multifd: Remove unnecessary usage of local Error
migration: Remove unnecessary usage of local Error
migration: Fix migration_channel_read_peek() error path
migration/multifd: Remove error_setg() in migration_ioc_process_incoming()
migration/multifd: Fix leaking of Error in TLS error flow
migration/multifd: Simplify multifd_channel_connect() if else statement
migration/multifd: Fix error message in multifd_recv_initial_packet()
migration: Remove errp parameter in migration_fd_process_incoming()
migration: Refactor migration_incoming_setup()
migration: Remove nulling of hostname in migrate_init()
migration: Remove migrate_max_downtime() declaration
tests/qtest: postcopy migration with suspend
tests/qtest: precopy migration with suspend
tests/qtest: option to suspend during migration
tests/qtest: migration events
migration: preserve suspended for bg_migration
migration: preserve suspended for snapshot
migration: preserve suspended runstate
migration: propagate suspended runstate
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Peter Maydell [Thu, 4 Jan 2024 19:55:19 +0000 (19:55 +0000)]
Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging
* configure: use a native non-cross compiler for linux-user
* meson: cleanups
* target/i386: miscellaneous cleanups and optimizations
* target/i386: implement CMPccXADD
* target/i386: the sgx_epc_get_section stub is reachable
* esp: check for NULL result from scsi_device_find()
* tag 'for-upstream' of https://gitlab.com/bonzini/qemu: (46 commits)
meson.build: report graphics backends separately
configure, meson: rename targetos to host_os
meson: rename config_all
meson: remove CONFIG_ALL
meson: remove config_targetos
meson: remove CONFIG_POSIX and CONFIG_WIN32 from config_targetos
meson: remove OS definitions from config_targetos
meson: always probe u2f and canokey if the option is enabled
meson: move subdirs to "Collect sources" section
meson: move config-host.h definitions together
meson: move CFI detection code with other compiler flags
meson: keep subprojects together
meson: move accelerator dependency checks together
meson: move option validation together
meson: move program checks together
meson: add more sections to main meson.build
configure: unify again the case arms in probe_target_compiler
configure: remove unnecessary subshell
Makefile: clean qemu-iotests output
meson: use version_compare() to compare version
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Avihai Horon [Sun, 31 Dec 2023 09:30:16 +0000 (11:30 +0200)]
migration/multifd: Remove unnecessary usage of local Error
According to Error API, usage of ERRP_GUARD() or a local Error instead
of errp is needed if errp is passed to void functions, where it is later
dereferenced to see if an error occurred.
There are several places in multifd.c that use local Error although it
is not needed. Change these places to use errp directly.
Avihai Horon [Sun, 31 Dec 2023 09:30:15 +0000 (11:30 +0200)]
migration: Remove unnecessary usage of local Error
According to Error API, usage of ERRP_GUARD() or a local Error instead
of errp is needed if errp is passed to void functions, where it is later
dereferenced to see if an error occurred.
There are several places in migration.c that use local Error although it
is not needed. Change these places to use errp directly.
migration_channel_read_peek() calls qio_channel_readv_full() and handles
both cases of return value == 0 and return value < 0 the same way, by
calling error_setg() with errp. However, if return value < 0, errp is
already set, so calling error_setg() with errp will lead to an assert.
Fix it by handling these cases separately, calling error_setg() with
errp only in return value == 0 case.
Fixes: 6720c2b32725 ("migration: check magic value for deciding the mapping of channels") Signed-off-by: Avihai Horon <avihaih@nvidia.com> Reviewed-by: Fabiano Rosas <farosas@suse.de> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Link: https://lore.kernel.org/r/20231231093016.14204-10-avihaih@nvidia.com Signed-off-by: Peter Xu <peterx@redhat.com>
Avihai Horon [Sun, 31 Dec 2023 09:30:13 +0000 (11:30 +0200)]
migration/multifd: Remove error_setg() in migration_ioc_process_incoming()
If multifd_load_setup() fails in migration_ioc_process_incoming(),
error_setg() is called with errp. This will lead to an assert because in
that case errp already contains an error.
Fix it by removing the redundant error_setg().
Fixes: 6720c2b32725 ("migration: check magic value for deciding the mapping of channels") Signed-off-by: Avihai Horon <avihaih@nvidia.com> Reviewed-by: Fabiano Rosas <farosas@suse.de> Link: https://lore.kernel.org/r/20231231093016.14204-9-avihaih@nvidia.com Signed-off-by: Peter Xu <peterx@redhat.com>
Avihai Horon [Sun, 31 Dec 2023 09:30:12 +0000 (11:30 +0200)]
migration/multifd: Fix leaking of Error in TLS error flow
If there is an error in multifd TLS handshake task,
multifd_tls_outgoing_handshake() retrieves the error with
qio_task_propagate_error() but never frees it.
Fix it by freeing the obtained Error.
In addition, the error is not reported at all, so report it with
migrate_set_error().
Fixes: 29647140157a ("migration/tls: add support for multifd tls-handshake") Signed-off-by: Avihai Horon <avihaih@nvidia.com> Reviewed-by: Fabiano Rosas <farosas@suse.de> Link: https://lore.kernel.org/r/20231231093016.14204-8-avihaih@nvidia.com Signed-off-by: Peter Xu <peterx@redhat.com>
Avihai Horon [Sun, 31 Dec 2023 09:30:10 +0000 (11:30 +0200)]
migration/multifd: Fix error message in multifd_recv_initial_packet()
In multifd_recv_initial_packet(), if MultiFDInit_t->id is greater than
the configured number of multifd channels, an irrelevant error message
about multifd version is printed.
Change the error message to a relevant one about the channel id.
Avihai Horon [Sun, 31 Dec 2023 09:30:08 +0000 (11:30 +0200)]
migration: Refactor migration_incoming_setup()
Commit 6720c2b32725 ("migration: check magic value for deciding the
mapping of channels") extracted the only code that could fail in
migration_incoming_setup().
Now migration_incoming_setup() can't fail, so refactor it to return void
and remove errp parameter.
Avihai Horon [Sun, 31 Dec 2023 09:30:07 +0000 (11:30 +0200)]
migration: Remove nulling of hostname in migrate_init()
MigrationState->hostname is set to NULL in migrate_init(). This is
redundant because it is already freed and set to NULL in
migrade_fd_cleanup(). Remove it.
Steve Sistare [Wed, 3 Jan 2024 20:05:41 +0000 (12:05 -0800)]
tests/qtest: postcopy migration with suspend
Add a test case to verify that the suspended state is handled correctly by
live migration postcopy. The test suspends the src, migrates, then wakes
the dest.
Steve Sistare [Wed, 3 Jan 2024 20:05:40 +0000 (12:05 -0800)]
tests/qtest: precopy migration with suspend
Add a test case to verify that the suspended state is handled correctly
during live migration precopy. The test suspends the src, migrates, then
wakes the dest.
Steve Sistare [Wed, 3 Jan 2024 20:05:39 +0000 (12:05 -0800)]
tests/qtest: option to suspend during migration
Add an option to suspend the src in a-b-bootblock.S, which puts the guest
in S3 state after one round of writing to memory. The option is enabled by
poking a 1 into the suspend_me word in the boot block prior to starting the
src vm. Generate symbol offsets in a-b-bootblock.h so that the suspend_me
offset is known. Generate the bootblock for each test, because suspend_me
may differ for each.
Steve Sistare [Wed, 3 Jan 2024 20:05:38 +0000 (12:05 -0800)]
tests/qtest: migration events
Define a state object to capture events seen by migration tests, to allow
more events to be captured in a subsequent patch, and simplify event
checking in wait_for_migration_pass. No functional change.
Steve Sistare [Wed, 3 Jan 2024 20:05:37 +0000 (12:05 -0800)]
migration: preserve suspended for bg_migration
Do not wake a suspended guest during bg_migration, and restore the prior
state at finish rather than unconditionally running. Allow the additional
state transitions that occur.
Steve Sistare [Wed, 3 Jan 2024 20:05:36 +0000 (12:05 -0800)]
migration: preserve suspended for snapshot
Restoring a snapshot can break a suspended guest. Snapshots suffer from
the same suspended-state issues that affect live migration, plus they must
handle an additional problematic scenario, which is that a running vm must
remain running if it loads a suspended snapshot.
To save, the existing vm_stop call now completely stops the suspended
state. Finish with vm_resume to leave the vm in the state it had prior
to the save, correctly restoring the suspended state.
To load, if the snapshot is not suspended, then vm_stop + vm_resume
correctly handles all states, and leaves the vm in the state it had prior
to the load. However, if the snapshot is suspended, restoration is
trickier. First, call vm_resume to restore the state to suspended so the
current state matches the saved state. Then, if the pre-load state is
running, call wakeup to resume running.
Prior to these changes, the vm_stop to RUN_STATE_SAVE_VM and
RUN_STATE_RESTORE_VM did not change runstate if the current state was
suspended, but now it does, so allow these transitions.
Steve Sistare [Wed, 3 Jan 2024 20:05:35 +0000 (12:05 -0800)]
migration: preserve suspended runstate
A guest that is migrated in the suspended state automaticaly wakes and
continues execution. This is wrong; the guest should end migration in
the same state it started. The root cause is that the outgoing migration
code automatically wakes the guest, then saves the RUNNING runstate in
global_state_store(), hence the incoming migration code thinks the guest is
running and continues the guest if autostart is true.
On the outgoing side, delete the call to qemu_system_wakeup_request().
Now that vm_stop completely stops a vm in the suspended state (from the
preceding patches), the existing call to vm_stop_force_state is sufficient
to correctly migrate all vmstate.
On the incoming side, call vm_start if the pre-migration state was running
or suspended. For the latter, vm_start correctly restores the suspended
state, and a future system_wakeup monitor request will cause the vm to
resume running.
Steve Sistare [Wed, 3 Jan 2024 20:05:34 +0000 (12:05 -0800)]
migration: propagate suspended runstate
If the outgoing machine was previously suspended, propagate that to the
incoming side via global_state, so a subsequent vm_start restores the
suspended state. To maintain backward and forward compatibility, reclaim
some space from the runstate member.
Steve Sistare [Wed, 3 Jan 2024 20:05:32 +0000 (12:05 -0800)]
cpus: check running not RUN_STATE_RUNNING
When a vm transitions from running to suspended, runstate notifiers are
not called, so the notifiers still think the vm is running. Hence, when
we call vm_start to restore the suspended state, we call vm_state_notify
with running=1. However, some notifiers check for RUN_STATE_RUNNING.
They must check the running boolean instead.
Steve Sistare [Wed, 3 Jan 2024 20:05:31 +0000 (12:05 -0800)]
cpus: stop vm in suspended runstate
Currently, a vm in the suspended state is not completely stopped. The VCPUs
have been paused, but the cpu clock still runs, and runstate notifiers for
the transition to stopped have not been called. This causes problems for
live migration. Stale cpu timers_state is saved to the migration stream,
causing time errors in the guest when it wakes from suspend, and state that
would have been modified by runstate notifiers is wrong.
Modify vm_stop to completely stop the vm if the current state is suspended,
transition to RUN_STATE_PAUSED, and remember that the machine was suspended.
Modify vm_start to restore the suspended state.
This affects all callers of vm_stop and vm_start, notably, the qapi stop and
cont commands:
old behavior:
RUN_STATE_SUSPENDED --> stop --> RUN_STATE_SUSPENDED
projects / mirror_qemu.git / log