Simplify migration_channel_connect() and migration_connect() to not
take an error as input. Move the error handling into the paths that
generate the error.
To achieve this, call migration_connect_error_propagate() from
socket.c and tls.c, which are the async paths.
For the sync paths, the handling is done as normal by returning all
the way to qmp_migrate_finish(), except that now the sync paths don't
pass the error forward into migration_connect() anymore.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-13-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Whenever an error occurs between migrate_init() and the start of
migration_thread, do cleanup immediately.
This allows the special casing for resume to be removed from
migration_connect(), that check is now done at
migration_connect_error_propagate() which already had a case for
resume.
The cleanup at qmp_migrate_finish_cb can also be removed because it
will always be reached either via the error path at
qmp_migrate_finish->migration_connect_error_propagate or via the
migrate_cleanup_bh.
The yank_unregister_instance at qmp_migrate() is now replaced by the
one at migration_cleanup().
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-12-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Move the register_instance call to migrate_prepare() so it can be
paired with the unregister_instance at migration_cleanup(). Otherwise,
the cleanup cannot be run when cpr_state_save() fails because the
instance is registered only after it.
When resuming from a paused postcopy migration, migrate_prepare()
returns early, but migration_cleanup() doesn't run, so the yank will
remain paired.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-11-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Cover the CANCELLING state in migration_connect_error_propagate() and
use it to funnel errors from migrate_prepare() until the end of
migration_connect().
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-10-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
In the next patches migration_cleanup() will be used in qmp_migrate(),
which currently does not show an error message. Move the error
reporting out of migration_cleanup() to avoid duplicate messages.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-9-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Freeing the error at migration_connect() is redundant in the normal
migration case. The freeing already happened at migrate_init():
qmp_migrate()
-> migrate_prepare()
-> migrate_init()
-> qmp_migrate_finish()
-> *_start_outgoing_migration()
-> migration_channel_connect()
-> migration_connect()
For the resume case, migrate_prepare() returns early and doesn't reach
migrate_init(). Move the extra migrate_error_free() call to
migrate_prepare() along with the resume check.
Also change migrate_init() to use migrate_error_free(), so it's easier
to see where are the places the error gets freed.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-8-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
cpr_set_incoming_mode() is only called on the target side, so
migrate_mode() on the source side is the same as s->parameters.mode.
Use the function to reduce explicit access to s->parameters, we have
options.c for that.
Cc: Mark Kanda <mark.kanda@oracle.com>
Cc: Ben Chaney <bchaney@akamai.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-7-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
The postcopy_try_recover() call doesn't need to be duplicated, move it
into migration_incoming_process().
This removes code from migration_fd_process_incoming() so it can be
removed in the near future.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-6-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
The TLS hostname is doing a tour around the world just to be cached
into s->hostname. We're already abusing MigrationState by doing that,
so incorporate the s->hostname into migration_tls_hostname() and stop
passing the string around.
The old route was roughly:
-transport code (socket.c, fd.c, etc):
if (SOCKET_ADDRESS_TYPE_INET)
hostname = saddr->u.inet.host
else
hostname = NULL
migration_channel_connect(..., hostname)
s->hostname = hostname;
migration_tls_client_create(..., hostname)
if (migrate_tls_hostname())
qio_channel_tls_new_client(migrate_tls_hostname())
else
qio_channel_tls_new_client(hostname)
-postcopy_preempt_setup:
postcopy_preempt_send_channel_new
migration_tls_client_create(..., s->hostname)
New route is:
-socket.c only:
if SOCKET_ADDRESS_TYPE_INET
s->hostname = saddr->u.inet.host
migration_channel_connect()
migration_tls_client_create()
qio_channel_tls_new_client(migrate_tls_hostname())
-postcopy_preempt_setup:
postcopy_preempt_send_channel_new
migration_tls_client_create()
qio_channel_tls_new_client(migrate_tls_hostname())
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-5-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
The MigrationState parameter is not used at
migration_tls_channel_process_incoming().
The last usage was removed by commit 3f461a0c0b ("migration: Drop
unused parameter for migration_tls_get_creds()")
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-4-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
When the incoming migration fails during the channel connection phase,
the state transition to FAILED is currently being done in the
MigrationState->state, but the MigrationIncomingState->state is the
one that should be used.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Prasad Pandit <pjp@fedoraproject.org>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-3-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
If local_err is set, migration_connect_error_propagate() will be
called and that function already has a state transition from SETUP to
FAILED.
Reviewed-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260123141656.6765-2-farosas@suse.de
Signed-off-by: Fabiano Rosas <farosas@suse.de>
This feature was deprecated in 9.1. Remove it in this release (11.0).
We also need to remove one unit test (/migration/precopy/fd/file) that
covers the fd: file migration, because it'll stop working now.
Reviewed-by: Prasad Pandit <ppandit@redhat.com>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260115225503.3083355-3-peterx@redhat.com
Signed-off-by: Fabiano Rosas <farosas@suse.de>
It was declared deprecated since 9.2.
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260115225503.3083355-2-peterx@redhat.com
Signed-off-by: Fabiano Rosas <farosas@suse.de>
The parameter can be instead passed into the function to avoid polluting
the global address space of MigrateCommon.
Reviewed-by: Prasad Pandit <ppandit@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260114153751.2427172-3-peterx@redhat.com
Signed-off-by: Fabiano Rosas <farosas@suse.de>
Now postcopy is not the only user of start_hook / end_hook that will pass
in a opaque pointer. It doesn't need to be defined in MigrateCommon as
part of the framework, as all other hook users can pass hook_data around.
Do it too for postcopy.
Reviewed-by: Prasad Pandit <ppandit@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Link: https://lore.kernel.org/qemu-devel/20260114153751.2427172-2-peterx@redhat.com
Signed-off-by: Fabiano Rosas <farosas@suse.de>
In _put() we don't actually allow send a service byte
without fd. So on _get() it's unexpected. Let's be strict.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Link: https://lore.kernel.org/qemu-devel/20260114064710.176268-3-vsementsov@yandex-team.ru
Signed-off-by: Fabiano Rosas <farosas@suse.de>
qemu_file_get_fd() may fail, and return -1 in this case.
We don't support passing "invalid" file descriptors (like -1),
so we should not interpret error here like invalid descriptor.
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Link: https://lore.kernel.org/qemu-devel/20260114064710.176268-2-vsementsov@yandex-team.ru
Signed-off-by: Fabiano Rosas <farosas@suse.de>
-----BEGIN PGP SIGNATURE-----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=u20D
-----END PGP SIGNATURE-----
Merge tag 'pull-trivial-patches' of https://gitlab.com/mjt0k/qemu into staging
trivial patches for 2026-01-20
# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCgAdFiEEZKoqtTHVaQM2a/75gqpKJDselHgFAmlvdRoACgkQgqpKJDse
# lHilOw//fdYjk7Ls+M2F0zgIKHbtQj/ACeIXtYr9mLwgUHCVnYByMbaqVVvyuCfZ
# 6n7unTRC6UyhmAgVz5F9RElUBxrRucqgXHXo9QcbvIBBZoWsnWAXjNWmV6kqBF4s
# L8cDniWRLwwkRaGTf1jjOEHJz8ilVwpXfvbQRkU9KJNelFkoh1rprxj3hkw1hF9C
# TpGda15ho7EaplRdfpN+qjOlHpShgbF2qMPUz1zZAWcpomZK8U5/xiHcuE9Y19/L
# tLC6lEkt1nR+xW/5JRptCAke8jLgLlYUHGsXdZuahpXwfK++ITCtM95X2Ra44gMR
# M3xC8hmcpaO5aF/2JoMkLS7LnuenJ093oBIwyhC9LM+MYt/NWIwLIItu22V25qRj
# bEuD2avDSiR0tWDikzxezj5J7riYsDQCt004Ycw7roBZeEI7yngXD2YTkkurSDjb
# wPvzg+qW8WM9n5K9Mxi66F0C7dFU0lru817nJO3dj4ayPJ7B6Elr47n5NpcfAEpP
# dn7j4+nj2puvAyXcfGwLQyU/o/ayJxkEvAZja1Q5sONwMqf0GHIZ9fDu63njkClQ
# rvzKyIKhRv3inmr65nO8RvnZe/j4ONhJ07IwMEBFjUMqW7aXu0GYPx0no7rtWZvQ
# NXypCGE+ag+PYetcOu2tHjaOLQG5/43eBEWSFCgns5UJcrB6vJ4=
# =u20D
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 20 Jan 2026 11:29:14 PM AEDT
# gpg: using RSA key 64AA2AB531D56903366BFEF982AA4A243B1E9478
# gpg: Good signature from "Michael Tokarev <mjt@debian.org>" [unknown]
# gpg: aka "Michael Tokarev <mjt@corpit.ru>" [unknown]
# gpg: aka "Michael Tokarev <mjt@tls.msk.ru>" [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: 9D8B E14E 3F2A 9DD7 9199 28F1 61AD 3D98 ECDF 2C8E
# Subkey fingerprint: 64AA 2AB5 31D5 6903 366B FEF9 82AA 4A24 3B1E 9478
* tag 'pull-trivial-patches' of https://gitlab.com/mjt0k/qemu:
bsd-user/syscall_defs.h: define STAT_TIME_T_EXT only for 32 bits
linux-user/aarch64/target_fcntl.h: add missing TARGET_O_LARGEFILE definition
configure: add ppc target back to container tests
MAINTAINERS: Add docs/system/i386/ to the general x86 architecture section
tests/qemu-iotests: Use "gsed" for expressions that require GNU sed
configure: Set $PYTHON in the configuration of the optionroms
tests: fix comment declaring runtime in rv64 interruptedmemory test
qemu-options.hx: use KiB, MiB, GiB for power-of-2 units
gdbstub: unlink the unix socket before bind()
Revert "gdbstub: Try unlinking the unix socket before binding"
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
In commit 83d5db95d3 ("meson: Allow system binaries to not
have target-specific units") we allowed targets with empty
target_system_arch[] source set, but missed hw_arch[] could
also be empty when building modules.
Skip such case, otherwise due to commit a1ced48754
("hw/microblaze: Build files once"), building with
--enable-modules triggers:
../meson.build:4034:14: ERROR: Key microblaze is not in the dictionary.
Fixes: a1ced48754 ("hw/microblaze: Build files once")
Reported-by: Frederic Bezies <fredbezies@gmail.com>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/3272
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Message-Id: <20260116131817.38009-1-philmd@linaro.org>
The file hw/virtio/cbor-helpers.c does not include our standard
qemu/osdep.h header, add it.
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Dorjoy Chowdhury <dorjoychy111@gmail.com>
Message-ID: <20260119120402.284985-1-pbonzini@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Add pcapfile property to uevi-vars-* devices, allowing to write out a
capture of the communication traffic between uefi firmware and qemu.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20260114110406.3500357-3-kraxel@redhat.com>
[PMD: Wrap long line to avoid checkpatch.pl warning]
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Allow reusing them elsewhere in qemu.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20260114110406.3500357-2-kraxel@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Daan De Meyer <daan.j.demeyer@gmail.com>
Message-ID: <20251124123521.1058183-5-daan.j.demeyer@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Daan De Meyer <daan.j.demeyer@gmail.com>
Message-ID: <20251124123521.1058183-4-daan.j.demeyer@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
For consistency with LOAD_IMAGE_MAX_DECOMPRESSED_BYTES.
Signed-off-by: Daan De Meyer <daan.j.demeyer@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20251124123521.1058183-3-daan.j.demeyer@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Preparation for adding support for zstd compressed efi zboot kernel
images.
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Daan De Meyer <daan.j.demeyer@gmail.com>
Message-ID: <20251124123521.1058183-2-daan.j.demeyer@gmail.com>
[PMD: Wrap long line to avoid checkpatch.pl warning]
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
"switch" is already part of the function name. No need to repeat it.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-15-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Now that there are dedicated eSDHC device models it is possible to replace
the "vendor" property (which is really a device-specific property) with a
dynamic cast, "fishing" out only the device models which shall have the
quirk applied.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-14-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
The endianness property was only explicitly set in the e500 machine which
now instantiates TYPE_FSL_ESDHC_BE. The property could also not be used
meaningfully in a hypothetical, common TYPE_FSL_ESDHC device model since
sdhci_common_realize() would fail in that case or it would need to know
all MMIO implementations upfront. Remove the property in favor of dedicated
device models.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-12-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Extract an eSDHC (little endian) device model since the uSDHC device model
will get an uSDHC-specific MMIO quirk.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-11-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
TYPE_FSL_ESDHC_BE maches real hardware more closely by reusing code of
TYPE_IMX_USDHC. For example, it fixes Linux to flood the guest console with
"mmc0: Internal clock never stabilised" messages in the QEMU advent calendar
2018 day 19 image.
Reported-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-10-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
For now, TYPE_FSL_ESDHC_BE is basically a big-endian variant of
TYPE_IMX_USDHC. It will be used in the e500 machines in the next step which
prevents Linux to flood the console with "mmc0: Internal clock never
stabilised" messages.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-9-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
The next patches will introduce eSDHC device models (big and little endian).
Prepare for that by renaming the functions. Also, this makes the code have
similar conventions as in Linux and U-Boot.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-8-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Drop the "IMX_" part of the names since these are actually {E,U}SDHC
specific and apply for PowerPC-based SoCs as well. While at it
consolidate all ESDHC constants into the ESDHC section.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-7-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
This reverts commit 1e76667f7a.
Unaware that the constants were named after Linux and U-Boot code,
commit 1e76667f7a ("hw/sd/sdhci: Rename ESDHC_* defines to USDHC_*")
changed the naming to be consistent with the function names. Fix the
constant naming to be consistent with Linux and U-Boot again. In the
next step, the function names will be renamed according to Linux and
U-Boot code as well.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-6-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Just like the Freescale MPC8569E SoC the i.MX25 features an eSDHC but
as little-endian variant. According to the datasheet TYPE_FSL_IMX25 requires
the same reset quirk as the e500 machines. The quirk was introduced in
d060b2789f ("hw/sd/sdhci: Set reset value of interrupt registers") for the
e500 machines, so reuse it now in TYPE_FSL_IMX25.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-5-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
The i.MX 6 SoC features uSDHC controllers which are the successors of eSDHC.
Fix the naming to make this clear.
Fixes: ec46eaa83a ("i.MX: Add i.MX6 SOC implementation.")
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-4-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Now that TYPE_IMX_USDHC sets this property internally it is not needed
on the SoC level any longer.
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-3-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Fixes TYPE_FSL_IMX6UL, TYPE_FSL_IMX7, and TYPE_FSL_IMX8MP to implement
version 3 of the SD specification.
Note that TYPE_FSL_IMX6 already had "sd-spec-version" set accordingly and
that TYPE_FSL_IMX25 correctly sets the same property to version 2 since the
real hardware is an eSDHC which is the uSDHC's predecessor.
Fixes: fd1e5c8179 ("sdhci: Add i.MX specific subtype of SDHCI")
cc: qemu-stable
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-ID: <20260112145418.220506-2-shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Add a qtest suite that validates ARM SMMUv3 translation without guest
firmware or OS. The tests leverage iommu-testdev to trigger DMA
operations and the qos-smmuv3 library to configure IOMMU translation
structures.
This test suite targets the virt machine and covers:
- Stage 1 only translation (VA -> PA via CD page tables)
- Stage 2 only translation (IPA -> PA via STE S2 tables)
- Nested translation (VA -> IPA -> PA, Stage 1 + Stage 2)
- Design to extended to support multiple security spaces
(Non-Secure, Secure, Root, Realm)
Each test case follows this sequence:
1. Initialize SMMUv3 with appropriate command/event queues
2. Build translation tables (STE/CD/PTE) for the target scenario
3. Configure iommu-testdev with IOVA and DMA attributes via MMIO
4. Trigger DMA and validate successful translation
5. Verify data integrity through a deterministic write-read pattern
This bare-metal approach provides deterministic IOMMU testing with
minimal dependencies, making failures directly attributable to the SMMU
translation path.
Signed-off-by: Tao Tang <tangtao1634@phytium.com.cn>
Tested-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Message-ID: <20260119161112.3841386-9-tangtao1634@phytium.com.cn>
[PMD: Cover tests/qtest/iommu-smmuv3-test.c in MAINTAINERS]
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Introduce qos-smmuv3, a reusable library for SMMUv3-related qtest
operations. This module encapsulates common tasks like:
- SMMUv3 initialization (enabling, configuring command/event queues)
- Stream Table Entry (STE) and Context Descriptor (CD) setup
- Multi-level page table construction (L0-L3 for 4KB granules)
- Support for Stage 1, Stage 2, and nested translation modes
- Could be easily extended to support multi-space testing infrastructure
(Non-Secure, Secure, Root, Realm)
The library provides high-level abstractions that allow test code to
focus on IOMMU behavior validation rather than low-level register
manipulation and page table encoding. Key features include:
- Provide memory allocation for translation structures with proper
alignment
- Helper functions to build valid STEs/CDs for different translation
scenarios
- Page table walkers that handle address offset calculations per
security space
This infrastructure is designed to be used by iommu-testdev-based tests
and future SMMUv3 test suites, reducing code duplication and improving
test maintainability.
Signed-off-by: Tao Tang <tangtao1634@phytium.com.cn>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Fabiano Rosas <farosas@suse.de>
Message-ID: <20260119161112.3841386-8-tangtao1634@phytium.com.cn>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Introduce a libqos helper module for the iommu-testdev
device used by qtests. This module provides some common functions to
all IOMMU test cases using iommu-testdev.
Wire the new sources into tests/qtest/libqos/meson.build so
they are built as part of the qtest support library.
Signed-off-by: Tao Tang <tangtao1634@phytium.com.cn>
Message-ID: <20260119161112.3841386-7-tangtao1634@phytium.com.cn>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
This change introduces STE_SET_* and CD_SET_* helpers to centralize and
simplify repeated field setting logic.
Signed-off-by: Tao Tang <tangtao1634@phytium.com.cn>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Message-ID: <20260119161112.3841386-6-tangtao1634@phytium.com.cn>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Fabiano Rosas
Peter Xu
Vladimir Sementsov-Ogievskiy
Markus Armbruster
Richard Henderson
Philippe Mathieu-Daudé
Paolo Bonzini
Gerd Hoffmann
Daan De Meyer
Bernhard Beschow
Tao Tang