When we unrealize a CPU object (which happens on vCPU hot-unplug), we
should destroy all the AddressSpace objects we created via calls to
cpu_address_space_init() when the CPU was realized.
Commit 24bec42f3d added a function to do this for a specific
AddressSpace, but did not add any places where the function was
called.
Since we always want to destroy all the AddressSpaces on unrealize,
regardless of the target architecture, we don't need to try to keep
track of how many are still undestroyed, or make the target
architecture code manually call a destroy function for each AS it
created. Instead we can adjust the function to always completely
destroy the whole cpu->ases array, and arrange for it to be called
during CPU unrealize as part of the common code.
Without this fix, AddressSanitizer will report a leak like this
from a run where we hot-plugged and then hot-unplugged an x86 KVM
vCPU:
Direct leak of 416 byte(s) in 1 object(s) allocated from:
#0 0x5b638565053d in calloc (/data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/qemu-system-x86_64+0x1ee153d) (BuildId: c1cd6022b195142106e1bffeca23498c2b752bca)
#1 0x7c28083f77b1 in g_malloc0 (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x637b1) (BuildId: 1eb6131419edb83b2178b682829a6913cf682d75)
#2 0x5b6386999c7c in cpu_address_space_init /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../system/physmem.c:797:25
#3 0x5b638727f049 in kvm_cpu_realizefn /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../target/i386/kvm/kvm-cpu.c:102:5
#4 0x5b6385745f40 in accel_cpu_common_realize /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../accel/accel-common.c:101:13
#5 0x5b638568fe3c in cpu_exec_realizefn /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../hw/core/cpu-common.c:232:10
#6 0x5b63874a2cd5 in x86_cpu_realizefn /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../target/i386/cpu.c:9321:5
#7 0x5b6387a0469a in device_set_realized /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../hw/core/qdev.c:494:13
#8 0x5b6387a27d9e in property_set_bool /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../qom/object.c:2375:5
#9 0x5b6387a2090b in object_property_set /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../qom/object.c:1450:5
#10 0x5b6387a35b05 in object_property_set_qobject /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../qom/qom-qobject.c:28:10
#11 0x5b6387a21739 in object_property_set_bool /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../qom/object.c:1520:15
#12 0x5b63879fe510 in qdev_realize /data_nvme1n1/linaro/qemu-from-laptop/qemu/build/x86-tgts-asan/../../hw/core/qdev.c:276:12
Cc: qemu-stable@nongnu.org
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2517
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20250929144228.1994037-4-peter.maydell@linaro.org
Signed-off-by: Peter Xu <peterx@redhat.com>
(cherry picked from commit 300a87c502)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
If an AddressSpace has been created in its own allocated
memory, cleaning it up requires first destroying the AS
and then freeing the memory. Doing this doesn't work:
address_space_destroy(as);
g_free_rcu(as, rcu);
because both address_space_destroy() and g_free_rcu()
try to use the same 'rcu' node in the AddressSpace struct
and the address_space_destroy hook gets overwritten.
Provide a new address_space_destroy_free() function which
will destroy the AS and then free the memory it uses, all
in one RCU callback.
(CC to stable because the next commit needs this function.)
Cc: qemu-stable@nongnu.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20250929144228.1994037-3-peter.maydell@linaro.org
Signed-off-by: Peter Xu <peterx@redhat.com>
(cherry picked from commit 041600e23f)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
address_space_destroy() doesn't actually immediately destroy the AS;
it queues it to be destroyed via RCU. This means you can't g_free()
the memory the AS struct is in until that has happened.
Clarify this in the documentation.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20250929144228.1994037-2-peter.maydell@linaro.org
Signed-off-by: Peter Xu <peterx@redhat.com>
(cherry picked from commit 9e7bfda490)
(Mjt: this is just a comment fix, but it makes subsequent changes to apply cleanly)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Both qemu_minrampagesize() and qemu_maxrampagesize() are
related to host memory backends, having the following call
stack:
qemu_minrampagesize()
-> find_min_backend_pagesize()
-> object_dynamic_cast(obj, TYPE_MEMORY_BACKEND)
qemu_maxrampagesize()
-> find_max_backend_pagesize()
-> object_dynamic_cast(obj, TYPE_MEMORY_BACKEND)
Having TYPE_MEMORY_BACKEND defined in "system/hostmem.h":
include/system/hostmem.h:23:#define TYPE_MEMORY_BACKEND "memory-backend"
Move their prototype declaration to "system/hostmem.h".
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Message-Id: <20250308230917.18907-7-philmd@linaro.org>
Acked-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20250311085743.21724-2-philmd@linaro.org
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Unlike cpr-reboot mode, cpr-transfer mode cannot save volatile ram blocks
in the migration stream file and recreate them later, because the physical
memory for the blocks is pinned and registered for vfio. Add a blocker
for volatile ram blocks.
Also add a blocker for RAM_GUEST_MEMFD. Preserving guest_memfd may be
sufficient for CPR, but it has not been tested yet.
Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-ID: <1740667681-257312-1-git-send-email-steven.sistare@oracle.com>
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Some of these bits are actually common to all cpus; while the
reset have common reservations for target-specific usage.
While generic code cannot know what the target-specific usage is,
common code can know what to do with the bits, e.g. single-step.
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Re-use the TARGET_PAGE_BITS_VARY mechanism to define
TARGET_PAGE_SIZE and friends when not compiling per-target.
Inline qemu_target_page_{size,mask,bits} as they are now trivial.
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Since we no longer support 64-bit guests on 32-bit hosts,
we can use a 32-bit type on a 32-bit host. This shrinks
the size of the structure to 16 bytes on a 32-bit host.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Since we no longer support 64-bit guests on 32-bit hosts,
we can use a 32-bit type on a 32-bit host.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Right now, we only allow for writing to memory regions that allow direct
access using memcpy etc; all other writes are simply ignored. This
implies that debugging guests will not work as expected when writing
to MMIO device regions.
Let's extend cpu_memory_rw_debug() to write to more memory regions,
including MMIO device regions. Reshuffle the condition in
memory_access_is_direct() to make it easier to read and add a comment.
While this change implies that debug access can now also write to MMIO
devices, we now are also permit ELF image loads and similar users of
cpu_memory_rw_debug() to write to MMIO devices; currently we ignore
these writes.
Peter assumes [1] that there's probably a class of guest images, which
will start writing junk (likely zeroes) into device model registers; we
previously would silently ignore any such bogus ELF sections. Likely
these images are of questionable correctness and this can be ignored. If
ever a problem, we could make these cases use address_space_write_rom()
instead, which is left unchanged for now.
This patch is based on previous work by Stefan Zabka.
[1] https://lore.kernel.org/all/CAFEAcA_2CEJKFyjvbwmpt=on=GgMVamQ5hiiVt+zUr6AY3X=Xg@mail.gmail.com/
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/213
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20250210084648.33798-8-david@redhat.com
Signed-off-by: Peter Xu <peterx@redhat.com>
Let's factor the complete "directly accessible" check independent of
the "write" condition out so we can reuse it next.
We can now split up the checks RAM and ROMD check, so we really only check
for RAM DEVICE in case of RAM -- ROM DEVICE is neither RAM not RAM DEVICE.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20250210084648.33798-4-david@redhat.com
Signed-off-by: Peter Xu <peterx@redhat.com>
Let's factor more of the generic "is this directly accessible" check,
independent of the "write" condition out.
Note that the "!mr->rom_device" check in the write case essentially
disallows the memory_region_is_romd() condition again. Further note that
RAM DEVICE regions are also RAM regions, so we can check for RAM+ROMD
first.
This is a preparation for further changes.
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20250210084648.33798-3-david@redhat.com
Signed-off-by: Peter Xu <peterx@redhat.com>
Migration pull request
- Purge of ram_save_target_page_legacy
- Cleanups to postcopy, json writer, migration states
- New migration mode cpr-transfer
- Fix for a -Werror=maybe-uninitialized instance in savevm
# -----BEGIN PGP SIGNATURE-----
#
# iQJEBAABCAAuFiEEqhtIsKIjJqWkw2TPx5jcdBvsMZ0FAmeaT8EQHGZhcm9zYXNA
# c3VzZS5kZQAKCRDHmNx0G+wxndXrEACTT+rdoEvOsNs4nM2a67GjxUoQZVTAWn+8
# lYhhNZLA4E+qHwpHTDCwyfyvCe615r72+bF7QO1KTrYeXGJg4SPk5kbEhCDqqjEu
# SGqrlPwkC1x3WkTvb228iDddDQ8dccko3Sy6wAyz0o8dtp5p4iK+57qzB/84u94L
# y3zQ+owOo9OLnXgdfMpN99HGQSvPR7CbP/2L293IrMCuPDUo9XhI7ARNS/phbT3Z
# aDl10WGHKz1SJWOkPj137E6+xMKuCmOZDTufTcTaHfyliD04JRWgEZVnKJxKJDxd
# 9e+lzHvXuYfO7YO11fr7DttPRnLEfjipELVTxrudM92eZ95XwdL4+ggfBTGHt76P
# yFUrp7G8qsUjWd+DHPmoo6Gx71zPbE6v9J2NMn2/1k4WdPOYy7HTmDgCkRirRTvV
# irYkHtdSFFsj3c0g4P4mhOzXnvUkGXzgrjteM5hlLy3bjSeZz9VMZADjiGqFGVPb
# 6euPcLLa9oynkoP5UXmFd/9PjWcgnfIbQu2MVlIyWhjvTGZKSGecVZmH5pWTJuBV
# xCbab1jYprRFpUIAMo94rgvRQRosomS1+GjGndFkX5++dTTlFSqpDLSGcEnPSGRx
# o9n+IldNiqh2vjN1bj60pLfmrHN/F+hsGTsDJlW+kfeyBXBkGArg1rDjN5ae7GvD
# UZK0N+OG0g==
# =jwOI
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 29 Jan 2025 10:56:49 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 'migration-20250129-pull-request' of https://gitlab.com/farosas/qemu: (42 commits)
migration: refactor ram_save_target_page functions
migration: Trivial cleanup on JSON writer of vmstate_save()
migration: Merge precopy/postcopy on switchover start
migration: Always set DEVICE state
migration: Cleanup qemu_savevm_state_complete_precopy()
migration: Unwrap qemu_savevm_state_complete_precopy() in postcopy
migration: Notify COMPLETE once for postcopy
migration: Take BQL slightly longer in postcopy_start()
migration: Drop cached migration state in migration_maybe_pause()
migration: Adjust locking in migration_maybe_pause()
migration: Adjust postcopy bandwidth during switchover
migration: Synchronize all CPU states only for non-iterable dump
migration: Drop inactivate_disk param in qemu_savevm_state_complete*
migration: Avoid two src-downtime-end tracepoints for postcopy
migration: Optimize postcopy on downtime by avoiding JSON writer
migration: Do not construct JSON description if suppressed
migration: Remove postcopy implications in should_send_vmdesc()
migration: cpr-transfer documentation
migration-test: cpr-transfer
tests/qtest: assert qmp connected
...
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Define the RAM_PRIVATE flag.
In RAMBlock creation functions, if MAP_SHARED is 0 in the flags parameter,
in a subsequent patch the implementation may still create a shared mapping
if other conditions require it. Callers who specifically want a private
mapping, eg for objects specified by the user, must pass RAM_PRIVATE.
After RAMBlock creation, MAP_SHARED in the block's flags indicates whether
the block is shared or private, and MAP_PRIVATE is omitted.
Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/r/1736967650-129648-6-git-send-email-steven.sistare@oracle.com
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Convert `unspecified` member of MemTxAttrs from bit field to bool, so
that bindgen could generate more ergonomic Rust binding with bool type.
As a result, MemTxAttrs needs to be expanded from 4 bytes to 8 bytes.
Therefore, move `unspecified` to after the bit fields and add reserved
members to ensure that the whole structure is packed into 8 bytes.
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Zhao Liu <zhao1.liu@intel.com>
Link: https://lore.kernel.org/r/20250121151322.171832-2-zhao1.liu@intel.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This started as a clean-up to properly pass a Error handler to the
gdbserver_start so we could do the right thing for command line and
HMP invocations.
Now that we have cleaned up foreach_device_config_or_exit() in earlier
patches we can further simplify by it by passing &error_fatal instead
of checking the return value. Having a return value is still useful
for HMP though so tweak the return to use a simple bool instead.
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20250116160306.1709518-11-alex.bennee@linaro.org>
Convert all targets simultaneously, as the gen_intermediate_code
function disappears from the target. While there are possible
workarounds, they're larger than simply performing the conversion.
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Remove the single target-specific definition used in
"exec/translator.h" (TARGET_PAGE_MASK) by un-inlining
is_same_page().
Rename the method as translator_is_same_page() and
improve its documentation.
Use it in translator_use_goto_tb().
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241218154145.71353-1-philmd@linaro.org>
When moving tb_cflags() in commit 88d4b5138a ("tcg: Make
tb_cflags() usable from target-agnostic code") we forgot to
include "qemu/atomic.h", which declares qatomic_read().
Explicitly include it now to avoid issue when refactoring
unrelated headers.
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Message-Id: <20241217141326.98947-2-philmd@linaro.org>
Include "cpu.h" earlier to get the MMU_USER_IDX definition soon
enough and avoid when refactoring unrelated headers:
In file included from include/exec/translator.h:271,
from ../../accel/tcg/translator.c:13:
include/exec/cpu-all.h: In function ‘cpu_mmu_index’:
include/exec/cpu-all.h:274:12: error: ‘MMU_USER_IDX’ undeclared (first use in this function)
274 | return MMU_USER_IDX;
| ^~~~~~~~~~~~
include/exec/cpu-all.h:274:12: note: each undeclared identifier is reported only once for each function it appears in
ninja: build stopped: subcommand failed.
We need to forward-declare cpu_mmu_index() to avoid on user emulation:
In file included from include/exec/cpu-all.h:263,
from include/exec/translator.h:271,
from ../../accel/tcg/translator.c:13:
../../target/sparc/cpu.h: In function ‘cpu_get_tb_cpu_state’:
../../target/sparc/cpu.h:757:13: error: implicit declaration of function ‘cpu_mmu_index’ [-Werror=implicit-function-declaration]
757 | flags = cpu_mmu_index(env_cpu(env), false);
| ^~~~~~~~~~~~~
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241218155202.71931-5-philmd@linaro.org>
