# -----BEGIN PGP SIGNATURE-----
#
# iQEzBAABCAAdFiEEIV1G9IJGaJ7HfzVi7wSWWzmNYhEFAmdEEtgACgkQ7wSWWzmN
# YhH5qAgAlKdcx/gFt4EBXtjVq/qbPluEGOQxvcRYLlN90rPLHPgCjAoT5ly3fIv1
# 4kCgcVZyG8SdGu1n0TzTTS9kg5tL7weQ9xEWwF0oyyuZABgAB7w/wpC8MHSkJFOn
# 2Tv+2Iab0dJ+e1pw71OMpE/YR5X2xq5vopsSHRtnyGWfRPGswJFwka+f8FS5DSiq
# 2CeNxADgTkPxJgDmOrNSsAPz8Rns77FAZdvDMqFjx1Lrqm8kPv9jzwOMO+a/2LpC
# t6OkpFzGjiiskPjSnSn/tzo4TfWYoABjJaI7b3vEqmNEJSTAaxltZNtSXZucctEt
# 1ihnFdjr/wPwGK/5Wu+qGnfDbFNxBw==
# =W4y1
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 25 Nov 2024 06:02:00 GMT
# gpg: using RSA key 215D46F48246689EC77F3562EF04965B398D6211
# gpg: Good signature from "Jason Wang (Jason Wang on RedHat) <jasowang@redhat.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg: It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 215D 46F4 8246 689E C77F 3562 EF04 965B 398D 6211
* tag 'net-pull-request' of https://github.com/jasowang/qemu:
virtio-net: Copy received header to buffer
virtio-net: Initialize hash reporting values
virtio-net: Fix hash reporting when the queue changes
virtio-net: Do not check for the queue before RSS
virtio-net: Fix size check in dhclient workaround
net: checksum: Convert data to void *
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
These were introduced in the avocado tests to workaround read issues
when interacting with console. They are no longer necessary and we can
use the expected "login:" string or the command prompt now. Drop the
last use of exec_command.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20241122090322.1934697-4-clg@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Drop the SSH connection which was introduced in the avocado tests to
workaround read issues when interacting with console.
EXTRA_BOOTARGS was introduced to reduce the console output at Linux
boot time. This didn't have the desired effect as we still had issues
when trying to match patterns on the console and we had to use the ssh
connection as a workaround.
While at it, remove the U-Boot EXTRA_BOOTARGS variable which has
become useless.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20241122090322.1934697-3-clg@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
The docs for submitting a patch describe using your "Real Name" with
the Signed-off-by line. Although somewhat ambiguous, this has often
been interpreted to mean someone's legal name.
In recent times, there's been a general push back[1] against the notion
that use of Signed-off-by in a project automatically requires / implies
the use of legal ("real") names and greater awareness of the downsides.
Full discussion of the problems of such policies is beyond the scope of
this commit message, but at a high level they are liable to marginalize,
disadvantage, and potentially result in harm, to contributors.
TL;DR: there are compelling reasons for a person to choose distinct
identities in different contexts & a decision to override that choice
should not be taken lightly.
A number of key projects have responded to the issues raised by making
it clear that a contributor is free to determine the identity used in
SoB lines:
* Linux has clarified[2] that they merely expect use of the
contributor's "known identity", removing the previous explicit
rejection of pseudonyms.
* CNCF has clarified[3] that the real name is simply the identity
the contributor chooses to use in the context of the community
and does not have to be a legal name, nor birth name, nor appear
on any government ID.
Since we have no intention of ever routinely checking any form of ID
documents for contributors[4], realistically we have no way of knowing
anything about the name they are using, except through chance, or
through the contributor volunteering the information. IOW, we almost
certainly already have people using pseudonyms for contributions.
This proposes to accept that reality and eliminate unnecessary friction,
by following Linux & the CNCF in merely asking that a contributors'
commonly known identity, of their choosing, be used with the SoB line.
[1] Raised in many contexts at many times, but a decent overall summary
can be read at https://drewdevault.com/2023/10/31/On-real-names.html
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d4563201f33a022fc0353033d9dfeb1606a88330
[3] https://github.com/cncf/foundation/blob/659fd32c86dc/dco-guidelines.md
[4] Excluding the rare GPG key signing parties for regular maintainers
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241021190939.1482466-1-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-40-alex.bennee@linaro.org>
The peripheral and PrimeCell identification registers of pl011 are located at
offset 0xFE0 - 0xFFC. To check if a read falls to such registers, the C
implementation checks if the offset-shifted-by-2 (not the offset itself) is in
the range 0x3F8 - 0x3FF.
Use the same check in the Rust implementation.
This fixes the timeout of the following avocado tests:
* tests/avocado/boot_linux_console.py:BootLinuxConsole.test_arm_virt
* tests/avocado/replay_kernel.py:ReplayKernelNormal.test_arm_virt
* tests/avocado/replay_kernel.py:ReplayKernelNormal.test_arm_vexpressa9
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Junjie Mao <junjie.mao@hotmail.com>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <SY0P300MB102644C4AC34A3AAD75DC4D5955C2@SY0P300MB1026.AUSP300.PROD.OUTLOOK.COM>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-39-alex.bennee@linaro.org>
Move the tests to a new file so that they can be run via
qemu-system-aarch64 in the functional framework.
Since these were the last tests in tests/avocado/tuxrun_baselines.py,
we can now remove that file, too.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-20-alex.bennee@linaro.org>
If a failure occurs early in the QemuBaseTest constructor, the
'log_filename' object atttribute may not exist yet. This happens
most notably if the QEMU_TEST_QEMU_BINARY is not set. We can't
initialize 'log_filename' earlier as we use the binary to identify
the architecture which is then used to build the path in which the
logs are stored.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241121154218.1423005-19-berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-19-alex.bennee@linaro.org>
Support the QEMU_TEST_QMP_BACKDOOR=backdoor.sock env variable as a
way to get a QMP backdoor for debugging a stalled QEMU test. Most
typically this would be used if running the tests directly:
$ QEMU_TEST_QMP_BACKDOOR=backdoor.sock \
QEMU_TEST_QEMU_BINARY=./build/qemu-system-arm \
PYTHONPATH=./python \
./tests/functional/test_arm_tuxrun.py
And then, when the test stalls, in a second shell run:
$ ./scripts/qmp/qmp-shell backdoor.sock
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241121154218.1423005-18-berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-18-alex.bennee@linaro.org>
The tuxrun tests send a series of strings to the guest to login
and then run commands. Since we have been unable to match on
console output that isn't followed by a newline, the test used
many time.sleep() statements to pretend to synchronize with
the guest.
This has proved to be unreliable for the aarch64be instance of
the tuxrun tests, with the test often hanging. The hang is a
very subtle timing problem, and it is suspected that some
(otherwise apparently harmless) I/O error messages could be
resulting in full FIFO buffers, stalling interaction with
the guest.
With the newly rewritten console interaction able to match
strings that don't have a following newline, the tux run
tests can now match directly on the login prompt, and/or
shell PS1 prompt.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2689
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241121154218.1423005-17-berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-17-alex.bennee@linaro.org>
The console interaction that waits for predicted strings uses
readline(), and thus is only capable of waiting for strings
that are followed by a newline.
This is inconvenient when needing to match on some things,
particularly login prompts, or shell prompts, causing tests
to use time.sleep(...) instead, which is unreliable.
Switch to reading the console 1 byte at a time, comparing
against the success/failure messages until we see a match,
regardless of whether a newline is encountered.
The success/failure comparisons are done with the python bytes
type, rather than strings, to avoid the problem of needing to
decode partially received multibyte utf8 characters.
Heavily inspired by a patch proposed by Cédric, but written
again to work in bytes, rather than strings.
Co-developed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241121154218.1423005-16-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-16-alex.bennee@linaro.org>
When waiting for expected output, the 'success_message' is a mandatory
parameter, with 'failure_message' defaulting to None.
The code has logic which indicates it was trying to cope with
'success_message' being None and 'failure_message' being non-None but
it does not appear able to actually do anything useful. The check for
'success_message is None' will break out of the loop before any check
for 'failure_message' has been performed.
IOW, for practcal purposes 'success_message' must be non-None unless
'send_string' is set. Assert this expectation and simplify the loop
logic.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241121154218.1423005-15-berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-15-alex.bennee@linaro.org>
The build/tests/functional subdirectories are consuming huge amounts
of disk space.
Split the location for scratch files into a 'scratch' sub-directory,
separate from log files, and delete it upon completion of each test.
The new env variable QEMU_TEST_KEEP_SCRATCH can be set to preserve
this scratch dir for debugging access if required.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20241121154218.1423005-3-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241121165806.476008-3-alex.bennee@linaro.org>
The libssh does not handle non-blocking mode in SFTP correctly. The
driver code already changes the mode to blocking for the SFTP
initialization, but for some reason changes to non-blocking mode.
This used to work accidentally until libssh in 0.11 branch merged
the patch to avoid infinite looping in case of network errors:
https://gitlab.com/libssh/libssh-mirror/-/merge_requests/498
Since then, the ssh driver in qemu fails to read files over SFTP
as the first SFTP messages exchanged after switching the session
to non-blocking mode return SSH_AGAIN, but that message is lost
int the SFTP internals and interpretted as SSH_ERROR, which is
returned to the caller:
https://gitlab.com/libssh/libssh-mirror/-/issues/280
This is indeed an issue in libssh that we should address in the
long term, but it will require more work on the internals. For
now, the SFTP is not supported in non-blocking mode.
Fixes: https://gitlab.com/libssh/libssh-mirror/-/issues/280
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
Message-ID: <20241113125526.2495731-1-rjones@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
qemu_create_cli_devices() should use qmp_device_add() to match the
behavior of the QMP monitor. A comment explained that libvirt changes
implementing strict CLI syntax were needed.
Peter Krempa <pkrempa@redhat.com> has confirmed that modern libvirt uses
the same JSON for -device (CLI) and device_add (QMP). Go ahead and use
qmp_device_add().
Cc: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20240827192751.948633-3-stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
The QMP device_add monitor command converts the QDict arguments to
QemuOpts and then back again to QDict. This process only supports scalar
types. Device properties like virtio-blk-pci's iothread-vq-mapping (an
array of objects) are silently dropped by qemu_opts_from_qdict() during
the QemuOpts conversion even though QAPI is capable of validating them.
As a result, hotplugging virtio-blk-pci devices with the
iothread-vq-mapping property does not work as expected (the property is
ignored).
Get rid of the QemuOpts conversion in qmp_device_add() and call
qdev_device_add_from_qdict() with from_json=true. Using the QMP
command's QDict arguments directly allows non-scalar properties.
The HMP is also adjusted since qmp_device_add()'s now expects properly
typed JSON arguments and cannot be used from HMP anymore. Move the code
that was previously in qmp_device_add() (with QemuOpts conversion and
from_json=false) into hmp_device_add() so that its behavior is
unchanged.
This patch changes the behavior of QMP device_add but not HMP
device_add. QMP clients that sent incorrectly typed device_add QMP
commands no longer work. This is a breaking change but clients should be
using the correct types already. See the netdev_add QAPIfication in
commit db2a380c84 for similar reasoning and object-add in commit
9151e59a8b. Unlike those commits, we continue to rely on 'gen': false
for the time being.
Markus helped me figure this out and even provided a draft patch. The
code ended up very close to what he suggested.
Suggested-by: Markus Armbruster <armbru@redhat.com>
Cc: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20240827192751.948633-2-stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
pci_devfn properties accept both integer and string values, but
integer 1 and string '1' have different meanings: The integer value
means device 0, function 1 whereas the string value '1' is short for
'1.0' and means device 1, function 0.
This test wants the string version so that the device actually becomes
visible for the guest. device_add hides the problem because it goes
through QemuOpts, which turns all properties into strings - this is a
QEMU bug that we want to fix, but that cancelled out the bug in this
test.
Fix the test first so that device_add can be fixed afterwards.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20241122224042.149258-1-kwolf@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
pci_devfn properties accept either a string or an integer as input. To
implement this, set_pci_devfn() first tries to visit the option as a
string, and if that fails, it visits it as an integer instead. While the
QemuOpts visitor happens to accept this, it is invalid according to the
visitor interface. QObject input visitors run into an assertion failure
when this is done.
QObject input visitors are used with the JSON syntax version of -device
on the command line:
$ ./qemu-system-x86_64 -enable-kvm -M q35 -device pcie-pci-bridge,id=pci.1,bus=pcie.0 -blockdev null-co,node-name=disk -device '{ "driver": "virtio-blk-pci", "drive": "disk", "id": "virtio-disk0", "bus": "pci.1", "addr": 1 }'
qemu-system-x86_64: ../qapi/qobject-input-visitor.c:143: QObject *qobject_input_try_get_object(QObjectInputVisitor *, const char *, _Bool): Assertion `removed' failed.
The proper way to accept both strings and integers is using the
alternate mechanism, which tells us the type of the input before it's
visited. With this information, we can directly visit it as the right
type.
This fixes set_pci_devfn() by using the alternate mechanism.
Cc: qemu-stable@nongnu.org
Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20241119120353.57812-1-kwolf@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
As of (at least) pylint 3.3.1, this code trips pylint up into believing
we are raising something other than an Exception. We are not: the first
two values may indeed be "None", but the last and final value must by
definition be a SystemExit exception.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-ID: <20241101173700.965776-5-jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Newest versions of pylint complain about specifically positional
arguments in addition to too many in general. We already disable the
general case, so silence this new warning too.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-ID: <20241101173700.965776-4-jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
I have a vague memory that I suggested this base class to Vladimir and
said "Maybe someday it will break, and I'll just fix it then." Guess
that's today.
Fixes various mypy errors in the "make check-tox" python test for at
least Python3.8; seemingly requires a fairly modern mypy and/or Python
base version to trigger.
Signed-off-by: John Snow <jsnow@redhat.com>
Message-ID: <20241101173700.965776-3-jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
receive_header() used to cast the const qualifier of the pointer to the
received packet away to modify the header. Avoid this by copying the
received header to buffer.
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
The specification says hash_report should be set to
VIRTIO_NET_HASH_REPORT_NONE if VIRTIO_NET_F_HASH_REPORT is negotiated
but not configured with VIRTIO_NET_CTRL_MQ_RSS_CONFIG. However,
virtio_net_receive_rcu() instead wrote out the content of the extra_hdr
variable, which is not uninitialized in such a case.
Fix this by zeroing the extra_hdr.
Fixes: e22f0603fb ("virtio-net: reference implementation of hash report")
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
virtio_net_process_rss() fills the values used for hash reporting, but
the values used to be thrown away with a recursive function call if
the queue changes after RSS. Avoid the function call to keep the values.
Fixes: a4c960eedc ("virtio-net: Do not write hashes to peer buffer")
Buglink: https://issues.redhat.com/browse/RHEL-59572
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
virtio_net_can_receive() checks if the queue is ready, but RSS will
change the queue to use so, strictly speaking, we may still be able to
receive the packet even if the queue initially provided is not ready.
Perform RSS before virtio_net_can_receive() to cover such a case.
Fixes: 4474e37a5b ("virtio-net: implement RX RSS processing")
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
work_around_broken_dhclient() accesses IP and UDP headers to detect
relevant packets and to calculate checksums, but it didn't check if
the packet has size sufficient to accommodate them, causing out-of-bound
access hazards. Fix this by correcting the size requirement.
Fixes: 1d41b0c1ec ("Work around dhclient brokenness")
Cc: qemu-stable@nongnu.org
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Convert the data parameter of net_checksum_calculate() to void * to
save unnecessary casts for callers.
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Peter Maydell