Hexagon (target/hexagon) Merge arguments to probe_pkt_scalar_hvx_stores

Reducing the number of arguments reduces the overhead of the helper
call

Signed-off-by: Taylor Simpson <tsimpson@quicinc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230405164211.30015-2-tsimpson@quicinc.com>

Hexagon (target/hexagon) Remove redundant/unused macros

Remove the following macros (remnants of the old generator design)
    READ_REG
    READ_PREG
    WRITE_RREG
    WRITE_PREG
Modify macros that rely on the above

The following are unused
    READ_IREG
    fGET_FIELD
    fSET_FIELD
    fREAD_P3
    fREAD_NPC
    fWRITE_LC0
    fWRITE_LC1

Signed-off-by: Taylor Simpson <tsimpson@quicinc.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230405183048.147767-1-tsimpson@quicinc.com>

Use black code style for python scripts

Signed-off-by: Marco Liebel <quic_mliebel@quicinc.com>
Signed-off-by: Taylor Simpson <tsimpson@quicinc.com>
Acked-by: Taylor Simpson <tsimpson@quicinc.com>
Tested-by: Taylor Simpson <tsimpson@quicinc.com>
Message-Id: <20230320092533.2859433-3-quic_mliebel@quicinc.com>

Use f-strings in python scripts

Replace python 2 format string with f-strings

Signed-off-by: Marco Liebel <quic_mliebel@quicinc.com>
Signed-off-by: Taylor Simpson <tsimpson@quicinc.com>
Reviewed-by: Taylor Simpson <tsimpson@quicinc.com>
Tested-by: Taylor Simpson <tsimpson@quicinc.com>
Message-Id: <20230320092533.2859433-2-quic_mliebel@quicinc.com>

Hexagon (translate.c): avoid redundant PC updates on COF

When there is a conditional change of flow or an endloop instruction, we
preload HEX_REG_PC with ctx->next_PC at gen_start_packet(). Nonetheless,
we still generate TCG code to do this update again at gen_goto_tb() when
the condition for the COF is not met, thus producing redundant
instructions. This can be seen with the following packet:

 0x004002e4:  0x5c20d000 {       if (!P0) jump:t PC+0 }

Which generates this TCG code:

   ---- 004002e4
-> mov_i32 pc,$0x4002e8
   and_i32 loc9,p0,$0x1
   mov_i32 branch_taken,loc9
   add_i32 pkt_cnt,pkt_cnt,$0x2
   add_i32 insn_cnt,insn_cnt,$0x2
   brcond_i32 branch_taken,$0x0,ne,$L1
   goto_tb $0x0
   mov_i32 pc,$0x4002e4
   exit_tb $0x7fb0c36e5200
   set_label $L1
   goto_tb $0x1
-> mov_i32 pc,$0x4002e8
   exit_tb $0x7fb0c36e5201
   set_label $L0
   exit_tb $0x7fb0c36e5203

Note that even after optimizations, the redundant PC update is still
present:

   ---- 004002e4
-> mov_i32 pc,$0x4002e8                     sync: 0  dead: 0 1  pref=0xffff
   mov_i32 branch_taken,$0x1                sync: 0  dead: 0 1  pref=0xffff
   add_i32 pkt_cnt,pkt_cnt,$0x2             sync: 0  dead: 0 1  pref=0xffff
   add_i32 insn_cnt,insn_cnt,$0x2           sync: 0  dead: 0 1 2  pref=0xffff
   goto_tb $0x1
-> mov_i32 pc,$0x4002e8                     sync: 0  dead: 0 1  pref=0xffff
   exit_tb $0x7fb0c36e5201
   set_label $L0
   exit_tb $0x7fb0c36e5203

With this patch, the second redundant update is properly discarded.

Note that we need the additional "move_to_pc" flag instead of just
avoiding the update whenever `dest == ctx->next_PC`, as that could
potentially skip updates from a COF with met condition, whose
ctx->branch_dest just happens to be equal to ctx->next_PC.

Signed-off-by: Matheus Tavares Bernardino <quic_mathbern@quicinc.com>
Signed-off-by: Taylor Simpson <tsimpson@quicinc.com>
Reviewed-by: Anton Johansson <anjo@rev.ng>
Reviewed-by: Taylor Simpson <tsimpson@quicinc.com>
Message-Id: <fc059153c3f0526d97b7f13450c02b276b0908e1.1679519341.git.quic_mathbern@quicinc.com>

Merge tag 'pull-for-8.0-220323-1' of https://gitlab.com/stsquad/qemu into staging

Misc fixes for 8.0 (testing, plugins, gitdm)

  - update Alpine image used for testing images
  - include libslirp in custom runner build env
  - update gitlab-runner recipe for CentOS
  - update docker calls for better caching behaviour
  - document some plugin callbacks
  - don't use tags to define drives for lkft baseline tests
  - fix missing clear of plugin_mem_cbs
  - fix iotests to report individual results
  - update the gitdm metadata for contributors
  - avoid printing comments before g_test_init()
  - probe for multiprocess support before running avocado test
  - refactor igb.py into netdev-ethtool.py avocado test
  - rebuild openbsd to have more space space for iotests

# -----BEGIN PGP SIGNATURE-----
#
# iQEzBAABCgAdFiEEZoWumedRZ7yvyN81+9DbCVqeKkQFAmQbGkoACgkQ+9DbCVqe
# KkT2kQgAif70dGyn2bcv7OYv2LgAcSzDGTIsOfMM6gYc9bm3nU/R/cj9cy8Qgbnu
# v6BSVeig5AVBWI1UTuMNeKW1BWoQYfxg2kdduSyAzMZS44r09ch9iabSbxbadC1e
# L1RrRWlzs/MwWX8IclAyEj1jr+DB+/DwoG61IP3215XXSy84e/XV4j+JAyBEzXQ9
# LdznGyqyItg3S6rnVpRP/wjR0P3VlrYOLOjFfCw7gB8JrlW7KIr8hWkHXYuS2mF5
# UyXTBwbXwYB5BAx0zXC3SVgl0Gs1qymaRUX77dlotlpVVo0Ql8a06dHPtYrjMNgA
# /Nyat3Dbbu7Rai+IzZIJl3tGx850wg==
# =XR0M
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 22 Mar 2023 15:10:02 GMT
# gpg:                using RSA key 6685AE99E75167BCAFC8DF35FBD0DB095A9E2A44
# gpg: Good signature from "Alex Bennée (Master Work Key) <alex.bennee@linaro.org>" [full]
# Primary key fingerprint: 6685 AE99 E751 67BC AFC8  DF35 FBD0 DB09 5A9E 2A44

* tag 'pull-for-8.0-220323-1' of https://gitlab.com/stsquad/qemu: (35 commits)
  qtests: avoid printing comments before g_test_init()
  contrib/gitdm: add group map for AMD
  contrib/gitdm: add more individual contributors
  contrib/gitdm: add revng to domain map
  contrib/gitdm: add Alibaba to the domain-map
  contrib/gitdm: add Amazon to the domain map
  contrib/gitdm: Add SYRMIA to the domain map
  contrib/gitdm: Add ASPEED Technology to the domain map
  iotests: remove the check-block.sh script
  iotests: register each I/O test separately with meson
  iotests: always use a unique sub-directory per test
  iotests: connect stdin to /dev/null when running tests
  iotests: print TAP protocol version when reporting tests
  iotests: strip subdir path when listing tests
  iotests: allow test discovery before building
  iotests: explicitly pass source/build dir to 'check' command
  tests/vm: custom openbsd partitioning to increase /home space
  tests/vm: skip X11 in openbsd installation
  include/qemu/plugin: Inline qemu_plugin_disable_mem_helpers
  include/qemu: Split out plugin-event.h
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

qtests: avoid printing comments before g_test_init()

The TAP protocol version line must be the first thing printed on
stdout. The migration test failed that requirement in certain
scenarios:

  # Skipping test: Userfault not available (builtdtime)
  TAP version 13
  # random seed: R02Sc120c807f11053eb90bfea845ba1e368
  1..32
  # Start of x86_64 tests
  # Start of migration tests
  ....

The TAP version is printed by g_test_init(), so we need to make
sure that any methods which print are run after that.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20230317170553.592707-1-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>

contrib/gitdm: add group map for AMD

AMD recently acquired Xilinx and contributors have been transitioning
their emails across.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Vikram Garhwal <vikram.garhwal@amd.com>
Cc: Stefano Stabellini <stefano.stabellini@amd.com>
Cc: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com>
Cc: Tong Ho <tong.ho@xilinx.com>
Reviewed-by: Francisco Iglesias <francisco.iglesias@amd.com>
Message-Id: <20230315174331.2959-33-alex.bennee@linaro.org>

contrib/gitdm: add more individual contributors

I've only added the names explicitly acked.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Bin Meng <bmeng@tinylab.org>
Acked-by: Jason A. Donenfeld <Jason@zx2c4.com>
Acked-by: Strahinja Jankovic <strahinja.p.jankovic@gmail.com>
Acked-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230315174331.2959-32-alex.bennee@linaro.org>

contrib/gitdm: add revng to domain map

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Anton Johansson <anjo@rev.ng>
Cc: Niccolò Izzo <nizzo@rev.ng>
Cc: Paolo Montesel <babush@rev.ng>
Reviewed-by: Alessandro Di Federico <ale@rev.ng>
Message-Id: <20230315174331.2959-31-alex.bennee@linaro.org>

contrib/gitdm: add Alibaba to the domain-map

This replaces the previous attempt to add c-sky.com. Group everything
under Alibaba now.

Added as requested by LIU Zhiwei.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Reviewed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
Message-Id: <20230315174331.2959-30-alex.bennee@linaro.org>

contrib/gitdm: add Amazon to the domain map

We have multiple contributors from both .co.uk and .com versions of
the address. Also add .de for completeness sake.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Alexander Graf <graf@amazon.com>
Cc: Paul Durrant <pdurrant@amazon.com>
Cc: David Wooodhouse <dwmw@amazon.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Paul Durrant <pdurrant@amazon.com>
Reviewed-by: Alexander Graf <graf@amazon.com>
Message-Id: <20230315174331.2959-29-alex.bennee@linaro.org>

contrib/gitdm: Add SYRMIA to the domain map

The company website lists QEMU amongst the things they work on so I
assume these are corporate contributions.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Milica Lazarevic <milica.lazarevic@syrmia.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230315174331.2959-28-alex.bennee@linaro.org>

contrib/gitdm: Add ASPEED Technology to the domain map

We have a number of contributors from this domain which is a corporate
endeavour.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Steven Lee <steven_lee@aspeedtech.com>
Cc: Troy Lee <troy_lee@aspeedtech.com>
Cc: Howard Chiu <howard_chiu@aspeedtech.com>
Cc: Jamin Lin <jamin_lin@aspeedtech.com>
Reviewed-by: Troy Lee <troy_lee@aspeedtech.com>
Message-Id: <20230315174331.2959-27-alex.bennee@linaro.org>

iotests: remove the check-block.sh script

Now that meson directly invokes the individual I/O tests, the
check-block.sh wrapper script is no longer required.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-9-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-26-alex.bennee@linaro.org>

iotests: register each I/O test separately with meson

Currently meson registers a single test that invokes an entire group of
I/O tests, hiding the test granularity from meson. There are various
downsides of doing this

 * You cannot ask 'meson test' to invoke a single I/O test
 * The meson test timeout can't be applied to the individual
   tests
 * Meson only gets a pass/fail for the overall I/O test group
   not individual tests
 * If a CI job gets killed by the GitLab timeout, we don't
   get visibility into how far through the I/O tests
   execution got.

This switches meson to perform test discovery by invoking 'check' in
dry-run mode. It then registers one meson test case for each I/O
test. Parallel execution remains disabled since the I/O tests do not
use self contained execution environments and thus conflict with
each other.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-8-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-25-alex.bennee@linaro.org>

iotests: always use a unique sub-directory per test

The current test runner is only safe against parallel execution within
a single instance of the 'check' process, and only if -j is given a
value greater than 2. This prevents running multiple copies of the
'check' process for different test scenarios.

This change switches the output / socket directories to always include
the test name, image format and image protocol. This should allow full
parallelism of all distinct test scenarios. eg running both qcow2 and
raw tests at the same time, or both file and nbd tests at the same
time.

It would be possible to allow for parallelism of the same test scenario
by including the pid, but that would potentially let many directories
accumulate over time on failures, so is not done.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-7-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-24-alex.bennee@linaro.org>

iotests: connect stdin to /dev/null when running tests

Currently the tests have their stdin inherited from the test harness,
meaning they are connected to a TTY. The QEMU processes spawned by
certain tests, however, modify TTY settings and if the test exits
abnormally the settings might not be restored.

The python test harness thus has some logic which will capture the
initial TTY settings and restore them once all tests are finished.

This does not, however, take into account the possibility of many
copies of the 'check' program running in parallel. With parallel
execution, a later invokation may save the TTY state that QEMU has
already modified, and thus restore bad state leaving the TTY
non-functional.

None of the I/O tests shnould actually be interactive requiring
user input and so they should not require a TTY at all. To avoid
this while TTY save/restore complexity we can connect the test
stdin to /dev/null instead.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-6-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-23-alex.bennee@linaro.org>

iotests: print TAP protocol version when reporting tests

Recently meson started complaining that TAP test reports don't include
the TAP protocol version. While this warning is bogus and has since been
removed from Meson, it looks like good practice to include this header
going forward. The GLib library test harness has started unconditionally
printing the version, so this brings the I/O tests into line.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-5-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-22-alex.bennee@linaro.org>

iotests: strip subdir path when listing tests

When asking 'check' to list individual tests by invoking it in dry run
mode, it prints the paths to the tests relative to the base of the
I/O test directory.

When asking 'check' to run an individual test, however, it mandates that
only the unqualified test name is given, without any path prefix. This
inconsistency makes it harder to ask for a list of tests and then invoke
each one.

Thus the test listing code is change to flatten the test names, by
printing only the base name, which can be directly invoked.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-4-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-21-alex.bennee@linaro.org>

iotests: allow test discovery before building

The 'check' script can be invoked in "dry run" mode, in which case it
merely does test discovery and prints out all their names. Despite only
doing test discovery it still validates that the various QEMU binaries
can be found. This makes it impossible todo test discovery prior to
building QEMU. This is a desirable feature to support, because it will
let meson discover tests.

Fortunately the code in the TestEnv constructor is ordered in a way
that makes this fairly trivial to achieve. We can just short circuit
the constructor after the basic directory paths have been set.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-3-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-20-alex.bennee@linaro.org>

iotests: explicitly pass source/build dir to 'check' command

The 'check' script has some rather dubious logic whereby it assumes
that if invoked as a symlink, then it is running from a separate
source tree and build tree, otherwise it assumes the current working
directory is a combined source and build tree.

This doesn't work if you want to invoke the 'check' script using
its full source tree path while still using a split source and build
tree layout. This would be a typical situation with meson if you ask
it to find the 'check' script path using files('check').

Rather than trying to make the logic more magical, add support for
explicitly passing the dirs using --source-dir and --build-dir. If
either is omitted the current logic is maintained.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Acked-by: Hanna Czenczek <hreitz@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230303160727.3977246-2-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-19-alex.bennee@linaro.org>

tests/vm: custom openbsd partitioning to increase /home space

The openbsd image is 20GB in size, but the automatic partitioning
done by the installer leaves /home with a mere ~3.5 GB of space,
wasting free space across many other partitions that are not
used by our build process:

openbsd$ df
Filesystem  512-blocks      Used     Avail Capacity  Mounted on
/dev/sd0a      1229692    213592    954616    18%    /
/dev/sd0k      7672220        40   7288572     0%    /home
/dev/sd0d      1736604        24   1649752     0%    /tmp
/dev/sd0f      4847676   2505124   2100172    54%    /usr
/dev/sd0g      1326684    555656    704696    44%    /usr/X11R6
/dev/sd0h      4845436   1445932   3157236    31%    /usr/local
/dev/sd0j     10898972         4  10354020     0%    /usr/obj
/dev/sd0i      3343644         4   3176460     0%    /usr/src
/dev/sd0e      2601212     19840   2451312     1%    /var

This change tells the installer todo custom partitioning with
4 GB on /, 256 MB swap, and the remaining ~15GB for /home

openbsd$ df
Filesystem  512-blocks      Used     Avail Capacity  Mounted on
/dev/sd0a      7932412   4740204   2795588    63%    /
/dev/sd0d     32164636        40  30556368     0%    /home

This will avoid ENOSPC failures when tests that need to create
big files (disk images) run in parallel.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230322123639.836104-3-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>

tests/vm: skip X11 in openbsd installation

As a VM used only for automated testing there is no need to
install the X11 stack.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230322123639.836104-2-berrange@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>

include/qemu/plugin: Inline qemu_plugin_disable_mem_helpers

Now that we've broken the include loop with cpu.h,
we can bring this inline.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230310195252.210956-8-richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-18-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>

include/qemu: Split out plugin-event.h

The usage in hw/core/cpu.h only requires QEMU_PLUGIN_EV_MAX.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230310195252.210956-7-richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-17-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>

*: Add missing includes of qemu/plugin.h

This had been pulled in from hw/core/cpu.h,
but that will be removed.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230310195252.210956-6-richard.henderson@linaro.org>
[AJB: also syscall-trace.h]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-16-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>

*: Add missing includes of qemu/error-report.h

This had been pulled in via qemu/plugin.h from hw/core/cpu.h,
but that will be removed.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230310195252.210956-5-richard.henderson@linaro.org>
[AJB: add various additional cases shown by CI]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-15-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>

include/qemu/plugin: Remove QEMU_PLUGIN_ASSERT

This macro is no longer used.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230310195252.210956-4-richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-14-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>

tcg: Drop plugin_gen_disable_mem_helpers from tcg_gen_exit_tb

Now that we call qemu_plugin_disable_mem_helpers in cpu_tb_exec,
we don't need to do this in generated code as well.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230310195252.210956-3-richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-13-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>

tcg: Clear plugin_mem_cbs on TB exit

Do this in cpu_tb_exec (normal exit) and cpu_loop_exit (exception),
adjacent to where we reset can_do_io.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1381
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230310195252.210956-2-richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-12-alex.bennee@linaro.org>
Reviewed-by: Emilio Cota <cota@braap.org>

tests/avocado: don't use tags to define drive

We are abusing the avocado tags which are intended to provide test
selection metadata to provide parameters to our test. This works OK up
until the point you need to have ,'s in the field as this is the tag
separator character which is the case for a number of the drive
parameters. Fix this by making drive a parameter to the common helper
function.

Fixes: 267fe57c23 (tests: add tuxrun baseline test to avocado)
Reviewed-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-11-alex.bennee@linaro.org>

tests/tcg: add some help output for running individual tests

So you can do:

  cd tests/tcg/aarch64-linux-user
  make -f ../Makefile.target help

To see the list of tests. You can then run each one individually.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230315174331.2959-8-alex.bennee@linaro.org>

include/qemu: add documentation for memory callbacks

Some API documentation was missed, rectify that.

Fixes: https://gitlab.com/qemu-project/qemu/-/issues/1497
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230315174331.2959-7-alex.bennee@linaro.org>

gitlab: update centos-8-stream job

A couple of clean-ups here:

  - inherit from the custom runners job for artefacts
  - call check-avocado directly
  - add some comments to the top about setup

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230315174331.2959-6-alex.bennee@linaro.org>

scripts/ci: update gitlab-runner playbook to handle CentOS

This was broken when we moved to using the pre-built packages as we
didn't take care to ensure we used RPMs where required.

NB: I could never get this to complete on my test setup but I suspect
this was down to network connectivity and timeouts while downloading.

Fixes: 69c4befba1 (scripts/ci: update gitlab-runner playbook to use latest runner)
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-5-alex.bennee@linaro.org>

scripts/ci: add libslirp-devel to build-environment

Without libslip enabled we won't have user networking which means the
KVM tests won't run.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230315174331.2959-4-alex.bennee@linaro.org>

tests/docker: all add DOCKER_BUILDKIT to RUNC environment

It seems we also need to pass DOCKER_BUILDKIT as an argument to docker
itself to get the full benefit of caching.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Suggested-by: Fabiano Rosas <farosas@suse.de>
Tested-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230315174331.2959-3-alex.bennee@linaro.org>

tests/avocado: re-factor igb test to avoid timeouts

The core of the test was utilising "ethtool -t eth1 offline" to run
through a test sequence. For reasons unknown the test hangs under some
configurations of the build on centos8-stream. Fundamentally running
the old fedora-31 cloud-init is just too much for something that is
directed at testing one device. So we:

  - replace fedora with a custom kernel + buildroot rootfs
  - rename the test from IGB to NetDevEthtool
  - re-factor the common code, add (currently skipped) tests for other
     devices which support ethtool
  - remove the KVM limitation as its fast enough to run in KVM or TCG

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Cc: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-Id: <20230322145529.4079753-1-alex.bennee@linaro.org>

tests/avocado: probe for multi-process support before running test

A recent attempt to let avocado run more tests on the CentOS stream
build failed because there was no gating on the multiprocess feature.
Like missing accelerators avocado should gracefully skip when the
feature is not enabled.

In this case we use the existence of the proxy device as a proxy for
multi-process support.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Cc: Elena Ufimtseva <elena.ufimtseva@oracle.com>
Cc: Jagannathan Raman <jag.raman@oracle.com>
Cc: John G Johnson <john.g.johnson@oracle.com>
Message-Id: <20230321111752.2681128-1-alex.bennee@linaro.org>

tests/avocado: update AArch64 tests to Alpine 3.17.2

To test Alpine boot on SBSA-Ref target we need Alpine Linux
'standard' image as 'virt' one lacks kernel modules.

So to minimalize Avocado cache I move test to 'standard' image.

Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20230302191146.1790560-1-marcin.juszkiewicz@linaro.org>
Message-Id: <20230315174331.2959-2-alex.bennee@linaro.org>

Update version for v8.0.0-rc1 release

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Merge tag 'pull-target-arm-20230321' of https://git.linaro.org/people/pmaydell/qemu-arm into staging

target-arm queue:
 * contrib/elf2dmp: Support Windows Server 2022
 * hw/char/cadence_uart: Fix guards on invalid BRGR/BDIV settings
 * target/arm: Add Neoverse-N1 IMPDEF registers
 * hw/usb/imx: Fix out of bounds access in imx_usbphy_read()
 * docs/system/arm/cpu-features.rst: Fix formatting
 * target/arm: Don't advertise aarch64-pauth.xml to gdb

# -----BEGIN PGP SIGNATURE-----
#
# iQJNBAABCAA3FiEE4aXFk81BneKOgxXPPCUl7RQ2DN4FAmQZrwQZHHBldGVyLm1h
# eWRlbGxAbGluYXJvLm9yZwAKCRA8JSXtFDYM3gmFD/9Ib/G7f21IQkhi0d0MoJeQ
# 529QbzHbXH272OvO2zFdev98o6EVbbeGzGqgaa0lv6OASwvNUIFVJAwZUX6Bb756
# dJ9k5aS2249SGQ8AzM65bCL4HxSVFan5+t9P890SyQk3zIzzQtSVjci/K2P2cFx1
# bKzbCZys/qjZgncPaPeuc9irkmAKlqc9UwqgUV3xvhBAfq1eFHk/bVIhcTVxNwUy
# quCYOt1GwtsOKn+nUcKclOcmBb7diCu6iFCGlO7XF9Rjaa+egW3OhUnGqUFROsdu
# j4drjeQT8gWY92m8PlnsZb0YUeefAwD7iVZGIAEp3G+9GEXdOvotrQVKtMLMZkq0
# /YInUjYAFu1w7DqhelvSYGVoVioP13HxsFWpmKNYNSJIHtS7QCfmHfUBPQnWjHD5
# XUO/K7vbsp69yi/rDDoHvQ3sqxJUuiF1Wuyj+hRK1JXRhLkRL+tBE7urlqqoJ1wH
# 0vL6oNj5GdvNJssIkb7yXx72irgAUu8XTC7bEvGCVfaylmei3SsS35qQmGePzO/z
# ok7WePQ/tM/FJ8JLVTXur9YsG7EqMROdszQRE4Yla3NE6BOr7HCCj7ZdCfy5SXL4
# IlZ69UELcYghcfIDRrRLXDSdfs98voRxIRDHy0rz64hUHlLBOnfqw/dcHvZBAB09
# CV7QPcDOR87jY228DT4EzA==
# =D7pq
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 21 Mar 2023 13:20:04 GMT
# gpg:                using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg:                issuer "peter.maydell@linaro.org"
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>" [ultimate]
# gpg:                 aka "Peter Maydell <pmaydell@gmail.com>" [ultimate]
# gpg:                 aka "Peter Maydell <pmaydell@chiark.greenend.org.uk>" [ultimate]
# gpg:                 aka "Peter Maydell <peter@archaic.org.uk>" [ultimate]
# Primary key fingerprint: E1A5 C593 CD41 9DE2 8E83  15CF 3C25 25ED 1436 0CDE

* tag 'pull-target-arm-20230321' of https://git.linaro.org/people/pmaydell/qemu-arm:
  target/arm: Don't advertise aarch64-pauth.xml to gdb
  docs/system/arm/cpu-features.rst: Fix formatting
  hw/usb/imx: Fix out of bounds access in imx_usbphy_read()
  contrib/elf2dmp: add PE name check and Windows Server 2022 support
  contrib/elf2dmp: move PE dir search to pe_get_data_dir_entry
  contrib/elf2dmp: fix code style
  hw/char/cadence_uart: Fix guards on invalid BRGR/BDIV settings
  target/arm: Add Neoverse-N1 registers

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

target/arm: Don't advertise aarch64-pauth.xml to gdb

Unfortunately a bug in older versions of gdb means that they will
crash if QEMU sends them the aarch64-pauth.xml.  This bug is fixed in
gdb commit 1ba3a3222039eb25, and there are plans to backport that to
affected gdb release branches, but since the bug affects gdb 9
through 12 it is very widely deployed (for instance by distros).

It is not currently clear what the best way to deal with this is; it
has been proposed to define a new XML feature name that old gdb will
ignore but newer gdb can handle.  Since QEMU's 8.0 release is
imminent and at least one of our CI runners is now falling over this,
disable the pauth XML for the moment.  We can follow up with a more
considered fix either in time for 8.0 or else for the 8.1 release.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

docs/system/arm/cpu-features.rst: Fix formatting

The markup for the Arm CPU feature documentation is incorrect,
and results in the HTML not rendering correctly -- the first
line of each description is rendered in boldface as if it
were part of the option name.

Reformat to match the styling used in cpu-models-x86.rst.inc.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1479
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20230316105808.1414003-1-peter.maydell@linaro.org
Reviewed-by: Cornelia Huck <cohuck@redhat.com>

hw/usb/imx: Fix out of bounds access in imx_usbphy_read()

The i.MX USB Phy driver does not check register ranges, resulting in out of
bounds accesses if an attempt is made to access non-existing PHY registers.
Add range check and conditionally report bad accesses to fix the problem.

While at it, also conditionally log attempted writes to non-existing or
read-only registers.

Reported-by: Qiang Liu <cyruscyliu@gmail.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Tested-by: Qiang Liu <cyruscyliu@gmail.com>
Message-id: 20230316234926.208874-1-linux@roeck-us.net
Link: https://gitlab.com/qemu-project/qemu/-/issues/1408
Fixes: 0701a5efa015 ("hw/usb: Add basic i.MX USB Phy support")
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

contrib/elf2dmp: add PE name check and Windows Server 2022 support

Since its inception elf2dmp has checked MZ signatures within an
address space above IDT[0] interrupt vector and took first PE image
found as Windows Kernel.
But in Windows Server 2022 memory dump this address space range is
full of invalid PE fragments and the tool must check that PE image
is 'ntoskrnl.exe' actually.
So, introduce additional validation by checking image name from
Export Directory against 'ntoskrnl.exe'.

Signed-off-by: Viktor Prutyanov <viktor@daynix.com>
Tested-by: Yuri Benditovich <yuri.benditovich@daynix.com>
Reviewed-by: Annie Li <annie.li@oracle.com>
Message-id: 20230222211246.883679-4-viktor@daynix.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

contrib/elf2dmp: move PE dir search to pe_get_data_dir_entry

Move out PE directory search functionality to be reused not only
for Debug Directory processing but for arbitrary PE directory.

Signed-off-by: Viktor Prutyanov <viktor@daynix.com>
Reviewed-by: Annie Li <annie.li@oracle.com>
Message-id: 20230222211246.883679-3-viktor@daynix.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

contrib/elf2dmp: fix code style

Originally elf2dmp were added with some code style issues,
especially in pe.h header, and some were introduced by
2d0fc797faaa73fbc1d30f5f9e90407bf3dd93f0. Fix them now.

Signed-off-by: Viktor Prutyanov <viktor@daynix.com>
Reviewed-by: Annie Li <annie.li@oracle.com>
Message-id: 20230222211246.883679-2-viktor@daynix.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

hw/char/cadence_uart: Fix guards on invalid BRGR/BDIV settings

The cadence UART attempts to avoid allowing the guest to set invalid
baud rate register values in the uart_write() function.  However it
does the "mask to the size of the register field" and "check for
invalid values" in the wrong order, which means that a malicious
guest can get a bogus value into the register by setting also some
high bits in the value, and cause QEMU to crash by division-by-zero.

Do the mask before the bounds check instead of afterwards.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1493
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Edgar E. Iglesias <edgar@zeroasic.com>
Reviewed-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Qiang Liu <cyruscyliu@gmail.com>
Message-id: 20230314170804.1196232-1-peter.maydell@linaro.org

target/arm: Add Neoverse-N1 registers

Add implementation defined registers for neoverse-n1 which
would be accessed by TF-A. Since there is no DSU in Qemu,
CPUCFR_EL1.SCU bit is set to 1 to avoid DSU registers definition.

Signed-off-by: Chen Baozi <chenbaozi@phytium.com.cn>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Tested-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Message-id: 20230313033936.585669-1-chenbaozi@phytium.com.cn
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Merge tag 'ui-pull-request' of https://gitlab.com/marcandre.lureau/qemu into staging

ui/ fixes for 8.0

# -----BEGIN PGP SIGNATURE-----
#
# iQJQBAABCAA6FiEEh6m9kz+HxgbSdvYt2ujhCXWWnOUFAmQZcpUcHG1hcmNhbmRy
# ZS5sdXJlYXVAcmVkaGF0LmNvbQAKCRDa6OEJdZac5VI/D/48GzUBrNuA8lep7/K4
# Mw1Ec7X3nn6KZbHsbvCASJuMH3wKYKZJS4JQuXvpHLxpDnzycYKpbrdEG8WrNo50
# 7tmahsZfA7rs5U3GToBgSdsTUlruvUoU4ycWLKfOJD0A47jaLprAFsBWvY3cYVO6
# 37EzCdzgV+FA/btaceNwdANItIGGUNPyXKSAdOebyDRD/3J4zTgSh3eG4ux86hzf
# P5Vo/m/hsqLc3S+M840XBABPitrmo+P7kS8NP8a5uifyE+bAbTBVV8WypMXK3lcX
# GMZh4LTfUPaABqBAhGZh/Bf3wOJ7P4w8+AXiMLBbxTYQmmSwwLZ0kwVcEIhrw7tK
# 8TLLz1hrVC88KQLCJ5m99w4xa0DEFZIwh6qEaStA+/TeJR02k49HpWiWkwa/yn55
# qRjaPV9mMhGaNCD9+E2ipX0krx6f03/TqE70IpyhDKuvQbjPv/4Q48gvF7R9IcuI
# mK45CAakudRjGE+2ZygGRvJQnZ/rRQX/spCbipOtd9Ay9yWFyrXj9zbkb97OUe6G
# 5HhhcfWAKXmRKh/V+xFIyBpN30XJwMxn6UJVacQwDRZamJMMmVi0rvZ7L1zhjbq5
# ZBjY6mZhufDk8YUTZiLz3BzDPcw+PWjkaGetFeoVkY7YDHajo0P5c3o23XJvberI
# 8Gaz61t7YB5uXq9WxKjyoRE23A==
# =npHn
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 21 Mar 2023 09:02:13 GMT
# gpg:                using RSA key 87A9BD933F87C606D276F62DDAE8E10975969CE5
# gpg:                issuer "marcandre.lureau@redhat.com"
# gpg: Good signature from "Marc-André Lureau <marcandre.lureau@redhat.com>" [full]
# gpg:                 aka "Marc-André Lureau <marcandre.lureau@gmail.com>" [full]
# Primary key fingerprint: 87A9 BD93 3F87 C606 D276  F62D DAE8 E109 7596 9CE5

* tag 'ui-pull-request' of https://gitlab.com/marcandre.lureau/qemu:
  ui: fix crash on serial reset, during init
  ui/sdl2: remove workaround forcing x11
  ui: return the default console cursor when con == NULL
  ui/gtk: fix cursor moved to left corner
  ui/dbus: fix passing SOCKET to GSocket API & leak
  ui/spice: fix SOCKET handling regression
  win32: add qemu_close_socket_osfhandle()

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

ui: fix crash on serial reset, during init

For ex, when resetting the xlnx-zcu102 machine:

(lldb) bt
* thread #1, queue = 'com.apple.main-thread', stop reason =
EXC_BAD_ACCESS (code=1, address=0x50)
   * frame #0: 0x10020a740 gd_vc_send_chars(vc=0x000000000) at
gtk.c:1759:41 [opt]
     frame #1: 0x100636264 qemu_chr_fe_accept_input(be=<unavailable>) at
char-fe.c:159:9 [opt]
     frame #2: 0x1000608e0 cadence_uart_reset_hold [inlined]
uart_rx_reset(s=0x10810a960) at cadence_uart.c:158:5 [opt]
     frame #3: 0x1000608d4 cadence_uart_reset_hold(obj=0x10810a960) at
cadence_uart.c:530:5 [opt]
     frame #4: 0x100580ab4 resettable_phase_hold(obj=0x10810a960,
opaque=0x000000000, type=<unavailable>) at resettable.c:0 [opt]
     frame #5: 0x10057d1b0 bus_reset_child_foreach(obj=<unavailable>,
cb=(resettable_phase_hold at resettable.c:162), opaque=0x000000000,
type=RESET_TYPE_COLD) at bus.c:97:13 [opt]
     frame #6: 0x1005809f8 resettable_phase_hold [inlined]
resettable_child_foreach(rc=0x000060000332d2c0, obj=0x0000600002c1c180,
cb=<unavailable>, opaque=0x000000000, type=RESET_TYPE_COLD) at
resettable.c:96:9 [opt]
     frame #7: 0x1005809d8 resettable_phase_hold(obj=0x0000600002c1c180,
opaque=0x000000000, type=RESET_TYPE_COLD) at resettable.c:173:5 [opt]
     frame #8: 0x1005803a0
resettable_assert_reset(obj=0x0000600002c1c180, type=<unavailable>) at
resettable.c:60:5 [opt]
     frame #9: 0x10058027c resettable_reset(obj=0x0000600002c1c180,
type=RESET_TYPE_COLD) at resettable.c:45:5 [opt]

While the chardev is created early, the VirtualConsole is associated
after, during qemu_init_displays().

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230220072251.3385878-1-marcandre.lureau@redhat.com>

ui/sdl2: remove workaround forcing x11

This workaround was put in place in the original implementation almost
10 years ago, considering a very old SDL2 version. Currently it prevents
users to run in a wayland-only environment without manually forcing the
backend.
The SDL2 wayland backend has been supported by distributions for a very
long time (e.g. in Fedora, first available 8 years ago), and is now
considered stable and becoming the default for new SDL2 releases.
Instead of requiring the x11 backend to exist by default, let new qemu
releases run with the default chosen by the installed SDL2 version.

Signed-off-by: Erico Nunes <ernunes@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20230301141205.514338-1-ernunes@redhat.com>

ui: return the default console cursor when con == NULL

VNC code relies on con==NULL to mean the default console.

Fixes:
https://gitlab.com/qemu-project/qemu/-/issues/1548

Fixes: commit 385ac97f8 ("ui: keep current cursor with QemuConsole")
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reported-by: Helge Konetzka <hk@zapateado.de>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230319111017.1319880-1-marcandre.lureau@redhat.com>

ui/gtk: fix cursor moved to left corner

Do not attempt to move the pointer if the widget is not yet realized.
The mouse cursor is placed to the corner of the screen, on X11 at least,
as x_root and y_root are then miscalculated. (this is not reproducible
on Wayland, because Gtk doesn't implement device warping there)

This also fixes the following warning at start:
qemu: Gdk: gdk_window_get_root_coords: assertion 'GDK_IS_WINDOW (window)' failed

Fixes: 6effaa16ac98 ("ui: set cursor position upon listener
registration")
Reported-by: Bernhard Beschow <shentey@gmail.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Tested-by: Bernhard Beschow <shentey@gmail.com>
Message-Id: <20230320132624.1612464-1-marcandre.lureau@redhat.com>

ui/dbus: fix passing SOCKET to GSocket API & leak

-display dbus is not currently available to win32 users, so it's not
considered a regression.

Note also the close() leak fix in case of error.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230320133643.1618437-4-marcandre.lureau@redhat.com>

ui/spice: fix SOCKET handling regression

Spice uses SOCKET on win32, but QEMU now uses file-descriptors.

Fixes "8.0.0rc0 Regression: spicy windows doesn't open":
https://gitlab.com/qemu-project/qemu/-/issues/1549

Fixes: commit abe34282b ("win32: avoid mixing SOCKET and file descriptor space")
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230320133643.1618437-3-marcandre.lureau@redhat.com>

win32: add qemu_close_socket_osfhandle()

Close the given file descriptor, but returns the underlying SOCKET.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20230320133643.1618437-2-marcandre.lureau@redhat.com>

Merge tag 'edk2-stable202302-20230320-pull-request' of https://gitlab.com/kraxel/qemu into staging

update edk2 to 202302 stable tag

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCgAdFiEEoDKM/7k6F6eZAf59TLbY7tPocTgFAmQYKWwACgkQTLbY7tPo
# cTinlA/9EVWaAyb7UV56My5Zbo41okXev8xmZV8R0G2xAbGaM22v/mPZaifA9Iqn
# JG19Go4xzTmX12Ttg/Def30Bu2RcjHfHccYycUQ7brd40rS8Gc2QYZX4p9NltUda
# dJOMB4e+7MKkavCrqfiHB6LWGMk27cT5puvMkmt65+/WsoBV1gcc0SmFK8tWIM3N
# dZqGdPr9ml72+fv7Cf7crShGEfwV/d3LkJ30Zccq0mbS08QcLm1SbyIB2+SdCyrE
# 35okXRJePACdh7rOWQgWrYAy1CTb54Vup7Fraq0a9NjfknKoQBxWt0w1L95d3p0d
# i8j0Yl1G5rD+Ybqvz/KgDLVNpKddttpkD1CECfZynOioHvwXWODrcJCKF9xsXcRR
# 9Lmeq090ottEbrYXln/clAjwNnF5ALNMEdvmOBZnFBG1H1bxTMI7VwWZEwRBQ5nu
# qtJf11oxuF9Lsi8OzIFcWx0hHkUmIrjCvnyNrVdxoMVPHcXXz1yD8MVV8vgg4jcL
# Fs/s9h62ymv0n3L0OYbgCfXgVWgomcgAkWfHcv+k8sEyzRa2v2wQ9zyVMnjelNkJ
# i/YEn+dupaZ20a6cLoYGQGBOACJ3iHyd1wZ0XzwSYYMq50X/ESRXAo5XofCAhzuz
# 5LQ3VgzQevFOuRwwqjnAwaZHt5M6Ysj0V28iYpMBJDsk+sMP5nM=
# =7oxT
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 20 Mar 2023 09:37:48 GMT
# gpg:                using RSA key A0328CFFB93A17A79901FE7D4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>" [full]
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>" [full]
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>" [full]
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* tag 'edk2-stable202302-20230320-pull-request' of https://gitlab.com/kraxel/qemu:
  edk2: update firmware binaries
  edk2: replace build scripts
  edk2: update submodule to edk2-stable202302
  acpi: disable tests/data/acpi updates
  bios-tables-test: use 128M numa nodes on aarch64
  acpi: enable tests/data/acpi updates

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Merge tag 'pull-request-2023-03-20' of https://gitlab.com/thuth/qemu into staging

* Mark Nios II as orphan
* Many s390x emulation fixes
* Disable flaky complete_in_standby blockjob unit test
* White space cleanups in various files

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEJ7iIR+7gJQEY8+q5LtnXdP5wLbUFAmQYVEoRHHRodXRoQHJl
# ZGhhdC5jb20ACgkQLtnXdP5wLbWu/A//QdVG7wS66lhdkPPu/GN+eDNaNsTzPPZd
# YwH0bpy2YoopInwa1ggZ1zPatmbSOVUcKjUfNYLcQzUBQnmeFtpuAF+4IGDhYfa+
# agkjWvmrCrIww4Wn+OLdLReW1+GtRlKZKu7xbYKkeLyqr9SHUD2lNUZ5RXj2t9Lw
# BXkPDXkMFqfVDzHtAxyXL7Z7yk3UTnZKBjDxiwE3tZ4S+N4r7m/TvNHiopI2bTVg
# JeDTxN40ylbOlBfOLC/qhlLUVnyBsDalIfKffIdLZ5Qn+FnKfZhkt9km4i7OL+ZX
# 1wKnTh/J8OCYqHOyhJdZGT8QxZH79qDm12/nKXzIQkxtJoKuz6Fm5FY3ZJNrb+IJ
# ybxSpAFBJB/8sUjbzl8ZjGxbZecIyEBKT518+oBoqjNcy3n8/m2BhBFr6f6F4cJC
# kdTnRS0XYKGLbJLz0+VBARE7hoHNckYsup/DGscppyYKNA6rFXbq/MI3+oMGAnPw
# Ua/+gXU/hwq8TPg97A7O0hS4TKSKMtdZFEDmAGzkejVHvm/3BvcYrVoKDljnUAQV
# SaERp/Elefbt3EufkuZp77AwLU8RcEFscitEIACmLbmwm3aKpqS+R2u1EJ4ZcoFT
# dHUBkokXW1/i+FtNRmjEKV9z398cNYXGEDLPfMnFbknHwbS53pqeQqqlvUidltGk
# LWviyiqKihk=
# =tYCc
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 20 Mar 2023 12:40:42 GMT
# gpg:                using RSA key 27B88847EEE0250118F3EAB92ED9D774FE702DB5
# gpg:                issuer "thuth@redhat.com"
# gpg: Good signature from "Thomas Huth <th.huth@gmx.de>" [full]
# gpg:                 aka "Thomas Huth <thuth@redhat.com>" [full]
# gpg:                 aka "Thomas Huth <huth@tuxfamily.org>" [full]
# gpg:                 aka "Thomas Huth <th.huth@posteo.de>" [unknown]
# Primary key fingerprint: 27B8 8847 EEE0 2501 18F3  EAB9 2ED9 D774 FE70 2DB5

* tag 'pull-request-2023-03-20' of https://gitlab.com/thuth/qemu: (24 commits)
  replace TABs with spaces
  qemu/osdep: Switch position of "extern" and "G_NORETURN"
  tests/unit/test-blockjob: Disable complete_in_standby test
  target/s390x/tcg/mem_helper: Remove bad assert() statement
  tests/tcg/s390x: Test unaligned accesses
  target/s390x: Update do_unaligned_access() comment
  target/s390x: Handle STGRL to non-aligned addresses
  target/s390x: Handle STRL to non-aligned addresses
  target/s390x: Handle CLRL and CLGFRL with non-aligned addresses
  target/s390x: Handle CGRL and CLGRL with non-aligned addresses
  target/s390x: Handle CRL and CGFRL with non-aligned addresses
  target/s390x: Handle LLGFRL from non-aligned addresses
  target/s390x: Handle LRL and LGFRL from non-aligned addresses
  target/s390x: Handle LGRL from non-aligned addresses
  target/s390x: Handle EXECUTE of odd addresses
  target/s390x: Handle branching to odd addresses
  tests/tcg/s390x: Add ex-relative-long.c
  target/s390x: Fix EXECUTE of relative long instructions
  tests/tcg/s390x: Add rxsbg.c
  target/s390x: Fix R[NOX]SBG with T=1
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

replace TABs with spaces

Bring the files in line with the QEMU coding style, with spaces
for indentation.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/378
Signed-off-by: Yeqi Fu <fufuyqqqqqq@gmail.com>
Message-Id: <20230315032649.57568-1-fufuyqqqqqq@gmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

qemu/osdep: Switch position of "extern" and "G_NORETURN"

Fixes the Windows build under msys2 using GCC 12 which fails with the following
error:

  [184/579] Compiling C++ object qga/vss-win32/qga-vss.dll.p/install.cpp.obj
  FAILED: qga/vss-win32/qga-vss.dll.p/install.cpp.obj
  "c++" "-m64" "-mcx16" "-Iqga/vss-win32/qga-vss.dll.p" "-Iqga/vss-win32" "-I../src/qga/vss-win32" "-I." "-Iqapi" "-Itrace" "-Iui" "-Iui/shader" "-IC:/msys64/mingw64/include/glib-2.0" "-IC:/msys64/mingw64/lib/glib-2.0/include" "-fdiagnostics-color=auto" "-Wall" "-Winvalid-pch" "-Wnon-virtual-dtor" "-Werror" "-std=gnu++11" "-g" "-iquote" "." "-iquote" "C:/msys64/home/shentey/Projects/qemu/src" "-iquote" "C:/msys64/home/shentey/Projects/qemu/src/include" "-iquote" "C:/msys64/home/shentey/Projects/qemu/src/tcg/i386" "-D__STDC_LIMIT_MACROS" "-D__STDC_CONSTANT_MACROS" "-D__STDC_FORMAT_MACROS" "-fno-pie" "-no-pie" "-D_GNU_SOURCE" "-D_FILE_OFFSET_BITS=64" "-D_LARGEFILE_SOURCE" "-fno-strict-aliasing" "-fno-common" "-fwrapv" "-Wundef" "-Wwrite-strings" "-Wtype-limits" "-Wformat-security" "-Wformat-y2k" "-Winit-self" "-Wignored-qualifiers" "-Wempty-body" "-Wendif-labels" "-Wexpansion-to-defined" "-Wimplicit-fallthrough=2" "-Wmissing-format-attribute" "-Wno-missing-include-dirs" "-Wno-shift-negative-value" "-Wno-psabi" "-fstack-protector-strong" "-Wno-unknown-pragmas" "-Wno-delete-non-virtual-dtor" "-Wno-non-virtual-dtor" -MD -MQ qga/vss-win32/qga-vss.dll.p/install.cpp.obj -MF "qga/vss-win32/qga-vss.dll.p/install.cpp.obj.d" -o qga/vss-win32/qga-vss.dll.p/install.cpp.obj "-c" ../src/qga/vss-win32/install.cpp
  In file included from C:/msys64/mingw64/lib/glib-2.0/include/glibconfig.h:9,
              from C:/msys64/mingw64/include/glib-2.0/glib/gtypes.h:34,
              from C:/msys64/mingw64/include/glib-2.0/glib/galloca.h:34,
              from C:/msys64/mingw64/include/glib-2.0/glib.h:32,
              from C:/msys64/home/shentey/Projects/qemu/src/include/glib-compat.h:32,
              from C:/msys64/home/shentey/Projects/qemu/src/include/qemu/osdep.h:144,
              from ../src/qga/vss-win32/install.cpp:13:
  C:/msys64/mingw64/include/glib-2.0/glib/gmacros.h:1075:21: error: standard attributes in middle of decl-specifiers
  1075 | # define G_NORETURN [[noreturn]]
        |                     ^
  C:/msys64/home/shentey/Projects/qemu/src/include/qemu/osdep.h:240:8: note: in expansion of macro 'G_NORETURN'
  240 | extern G_NORETURN
        |        ^~~~~~~~~~
  C:/msys64/mingw64/include/glib-2.0/glib/gmacros.h:1075:21: note: standard attributes must precede the decl-specifiers to apply to the declaration, or follow them to apply to the type
  1075 | # define G_NORETURN [[noreturn]]
        |                     ^
  C:/msys64/home/shentey/Projects/qemu/src/include/qemu/osdep.h:240:8: note: in expansion of macro 'G_NORETURN'
  240 | extern G_NORETURN
        |        ^~~~~~~~~~
  C:/msys64/mingw64/include/glib-2.0/glib/gmacros.h:1075:21: error: attribute ignored [-Werror=attributes]
  1075 | # define G_NORETURN [[noreturn]]
        |                     ^
  C:/msys64/home/shentey/Projects/qemu/src/include/qemu/osdep.h:240:8: note: in expansion of macro 'G_NORETURN'
  240 | extern G_NORETURN
        |        ^~~~~~~~~~
  C:/msys64/mingw64/include/glib-2.0/glib/gmacros.h:1075:21: note: an attribute that appertains to a type-specifier is ignored
  1075 | # define G_NORETURN [[noreturn]]
        |                     ^
  C:/msys64/home/shentey/Projects/qemu/src/include/qemu/osdep.h:240:8: note: in expansion of macro 'G_NORETURN'
  240 | extern G_NORETURN
        |        ^~~~~~~~~~
  cc1plus.exe: all warnings being treated as errors

Apparently it also fixes the compilation with Clang 15 (see
https://gitlab.com/qemu-project/qemu/-/issues/1541 ).

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1541
Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Message-Id: <20230318185931.181659-1-shentey@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/unit/test-blockjob: Disable complete_in_standby test

The blockjob/complete_in_standby test is flaky and fails
intermittently in CI:

172/621 qemu:unit / test-blockjob
           ERROR           0.26s   killed by signal 6 SIGABRT
11:03:46 MALLOC_PERTURB_=176
G_TEST_SRCDIR=/Users/pm215/src/qemu-for-merges/tests/unit
G_TEST_BUILDDIR=/Users/pm215/src/qemu-for-merges/build/all/tests/unit
/Users/pm215/src/qemu-for-merges/build/all/tests/unit/test-blockjob
--tap -k
----------------------------------- output -----------------------------------
stdout:
# random seed: R02S8c79d6e1c01ce0b25475b2210a253242
1..9
# Start of blockjob tests
ok 1 /blockjob/ids
stderr:
Assertion failed: (job->status == JOB_STATUS_STANDBY), function
test_complete_in_standby, file ../../tests/unit/test-blockjob.c, line
499.

Seen on macOS/x86_64, FreeBSD 13/x86_64, msys2-64bit, eg:

https://gitlab.com/qemu-project/qemu/-/jobs/3872508803
https://gitlab.com/qemu-project/qemu/-/jobs/3950667240

Disable this subtest until somebody has time to investigate.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20230317143534.1481947-1-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x/tcg/mem_helper: Remove bad assert() statement

The "assert(!nonfault)" statement can be triggered by running the
"mvpg" s390x kvm-unit-test with TCG. According to Richard: "... the
assert looks backward. We should only arrive there if nonfault was
true for the probe (otherwise the probe would have raised the
exception directly).  I would think we could just remove the assert."

Fixes: 4049431478 ("target/s390x: Fix s390_probe_access for user-only")
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230317135737.597570-1-thuth@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

edk2: update firmware binaries

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

edk2: replace build scripts

Remove Makefile.edk2 and the edk2*.sh scripts and replace them
with a python script (which already handles fedora rpm builds)
and a config file for it.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

edk2: update submodule to edk2-stable202302

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>

acpi: disable tests/data/acpi updates

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

bios-tables-test: use 128M numa nodes on aarch64

Recent edk2 versions don't boot with very small numa nodes.
Bump the size from 64M to 128M.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

acpi: enable tests/data/acpi updates

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

tests/tcg/s390x: Test unaligned accesses

Add a number of small test that check whether accessing unaligned
addresses in various ways leads to a specification exception.

Run these test both in softmmu and user configurations; expect a PGM
in one case and SIGILL in the other.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230316164428.275147-13-iii@linux.ibm.com>
[thuth: Added -Wl,--build-id=none to LDFLAGS]
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Update do_unaligned_access() comment

Relative long instructions now depend on do_unaligned_access() too.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-12-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle STGRL to non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-11-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle STRL to non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-10-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle CLRL and CLGFRL with non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-9-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle CGRL and CLGRL with non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-8-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle CRL and CGFRL with non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-7-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle LLGFRL from non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-6-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle LRL and LGFRL from non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-5-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle LGRL from non-aligned addresses

Use MO_ALIGN and let do_unaligned_access() generate a specification
exception.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Suggested-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-4-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle EXECUTE of odd addresses

Generate a specification exception in the helper before trying to fetch
the instruction.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-3-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Handle branching to odd addresses

Let branching happen and try to generate a new translation block with
an odd address. Generate a specification exception in
cpu_get_tb_cpu_state().

Reported-by: Harold Grovesteen <h.grovsteen@tx.rr.com>
Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230316164428.275147-2-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/tcg/s390x: Add ex-relative-long.c

Test EXECUTE and EXECUTE RELATIVE LONG with relative long instructions
as targets.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-Id: <20230316210751.302423-3-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Fix EXECUTE of relative long instructions

The code uses the wrong base for relative addressing: it should use the
target instruction address and not the EXECUTE's address.

Fix by storing the target instruction address in the new CPUS390XState
member and loading it from the code generated by gen_ri2().

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20230316210751.302423-2-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/tcg/s390x: Add rxsbg.c

Add a small test for RXSBG with T=1 to prevent regressions.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Message-Id: <20230316172205.281369-3-iii@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Fix R[NOX]SBG with T=1

RXSBG usage in the "filetests" test from the wasmtime testsuite makes
tcg_reg_alloc_op() attempt to temp_load() a TEMP_VAL_DEAD temporary,
causing an assertion failure:

    0x01000a70:  ec14 b040 3057  rxsbg    %r1, %r4, 0xb0, 0x40, 0x30

    OP after optimization and liveness analysis:
     ---- 0000000001000a70 0000000000000004 0000000000000006
     rotl_i64 tmp2,r4,$0x30                   dead: 1 2  pref=0xffff
     and_i64 tmp2,tmp2,$0x800000000000ffff    dead: 1  pref=0xffff
    [xor_i64 tmp3,tmp3,tmp2                   dead: 1 2  pref=0xffff]
     and_i64 cc_dst,tmp3,$0x800000000000ffff  sync: 0  dead: 0 1 2  pref=0xffff
     mov_i64 psw_addr,$0x1000a76              sync: 0  dead: 0 1  pref=0xffff
     mov_i32 cc_op,$0x6                       sync: 0  dead: 0 1  pref=0xffff
     call lookup_tb_ptr,$0x6,$1,tmp8,env      dead: 1  pref=none
     goto_ptr tmp8                            dead: 0
     set_label $L0
     exit_tb $0x7fffe809d183

    ../tcg/tcg.c:3865: tcg fatal error

The reason is that tmp3 does not have an initial value, which confuses
the register allocator. This also affects the correctness of the
results.

Fix by assigning R1 to it.

Exposed by commit e2e641fa3d5 ("tcg: Change default temp lifetime to
TEMP_TB").

Fixes: d6c6372e186e ("target-s390: Implement R[NOX]SBG")
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20230316172205.281369-2-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/tcg/s390x: Add PSW modification tests

Add several small tests that check the PSW modification instructions:

* lpsw.S checks whether LPSW works correctly in the "happy" case.

* lpswe-early.S checks whether early exceptions are recognized and
  whether the correct ILC and old PSW are stored when they happen.

* ssm-early.S, stosm-early.S and exrl-ssm-early.S check the special
  handling of SSM and STOSM with respect to early exceptions.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Message-Id: <20230315020408.384766-4-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Implement Early Exception Recognition

Generate a specification exception if a reserved bit is set in the PSW
mask or if the PSW address is out of bounds dictated by the addressing
mode.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20230315020408.384766-3-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Fix LPSW

Currently LPSW does not invert the mask bit 12 and incorrectly copies
the BA bit into the address.

Fix by generating code similar to what s390_cpu_load_normal() does.

Reported-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Co-developed-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20230315020408.384766-2-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

MAINTAINERS: Mark the Nios II CPU as orphan

Marek and Chris haven't been active for Nios II since years
(the last time seems to have been in 2017), and we've got
unhandled severe Nios II bug tickets in the bug tracker since
a long time, so to avoid wrong expectations of people who are
looking at the MAINTAINERS file, it's maybe best to mark the
Nios II entry as orphan nowadays.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230313183352.274744-1-thuth@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

Merge tag 'trivial-branch-for-8.0-pull-request' of https://gitlab.com/laurent_vivier/qemu into staging

Trivial branch pull request 20230317

Fix doc
Fix sh4 cpu log output

# -----BEGIN PGP SIGNATURE-----
#
# iQJGBAABCAAwFiEEzS913cjjpNwuT1Fz8ww4vT8vvjwFAmQUdowSHGxhdXJlbnRA
# dml2aWVyLmV1AAoJEPMMOL0/L748ylwP/RisLo3XrvNBVuDW/cLh7vhFcaeSqEzH
# /RfdJ0DeqreXQYiwqsUS+YrtBARpBwuvBk1RGg46chx3IZJp7fmfLjs0sSSiESsR
# kL8tJqRFFdCXmpmUWQqmluiCEBLllq8uDHMaAyXh95V2VRt4vrR8K2x3GW0yrarS
# 4BvMdTAJBmjjpXRZ+/NY88EEEqCHgliWTHm+JKLj7II5duUYZO+r0XpCQELFdHig
# uEYOupSUm/H4X8VuKuqNG1YGUz9c8X7nYZ+lgKLNH/i8vco+dPSoL4fZvG1ts/NH
# kcA+EdGNDJZdvhOUDlvebSIZ9uuVjxy4WD8CIJQ5A2LGSdmv4KUJjQWGBw/r5B/B
# e+uhrFnXJPrRSytSohB0iEzlrETBHAzq1yzoc7TYy70SzHtBylR0Xxwxnaq2N9ux
# k4wyeXF6toX49RiZ0UmoYO82vtPqg1LOsmYDYWKNzgWbRsTIVJAXuAwBAVIzOgJV
# hs93YO3gCaIjh8n/V43VvQzborNhp0xf6Ch7F0cgD1klAAKX1ICX4Mn2Qj7tPmgh
# GUsdTpleaFTpsKxgpNP1eU5QYtI9ZcGBhAhsSV9SSnrRsWo1dlppZJkUWEndIHGr
# iUnP5OPLtWU5tM2WbVNb3pio/4xrzft3lu4DKH2IecZbBHF11FpK6f9mrlOagAz0
# V/3JSBg10dZQ
# =3s21
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 17 Mar 2023 14:17:48 GMT
# gpg:                using RSA key CD2F75DDC8E3A4DC2E4F5173F30C38BD3F2FBE3C
# gpg:                issuer "laurent@vivier.eu"
# gpg: Good signature from "Laurent Vivier <lvivier@redhat.com>" [full]
# gpg:                 aka "Laurent Vivier <laurent@vivier.eu>" [full]
# gpg:                 aka "Laurent Vivier (Red Hat) <lvivier@redhat.com>" [full]
# Primary key fingerprint: CD2F 75DD C8E3 A4DC 2E4F  5173 F30C 38BD 3F2F BE3C

* tag 'trivial-branch-for-8.0-pull-request' of https://gitlab.com/laurent_vivier/qemu:
  docs/sphinx/kerneldoc.py: Honour --enable-werror
  target/sh4: Honor QEMU_LOG_FILENAME with QEMU_LOG=cpu
  exec/memory: Fix kernel-doc warning

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Merge tag '20230316-xtensa' of https://github.com/OSLL/qemu-xtensa into staging

target/xtensa updates for v8.0:

- enable testing big-endian xtensa cores

# -----BEGIN PGP SIGNATURE-----
#
# iQJHBAABCgAxFiEEK2eFS5jlMn3N6xfYUfnMkfg/oEQFAmQTd6gTHGpjbXZia2Jj
# QGdtYWlsLmNvbQAKCRBR+cyR+D+gRCowEACpQULqjxkpd23bd5uYQYsuJRzbSblF
# 6xux3IK5kD2y8vjzVegTs1XdZNhNzLRfEOijy+DxP1+5S6jRoACRkv0zKaBigbfA
# C7pLTxKGXk4OLA1sEdAXZ8jTBxDNQG542kF/PSPUHLxbJyb+lH+4707vb/kgQQdz
# nhCVfK75hqW2leohWsEES1jPxazyqu5PLMr9zPa0yfhbUU3g1AgTzFHAVrocP5mN
# QKsaSwP5fwnF2ozTRM7MhHJj9iLlX0IYxeAcA22kUhr6kaAmL4SQdpwiREbCgOts
# XE2Hpo+W5WQRw1B94au0spMZwYGDhFeApLhHRAI6hpLlOyTzC1bHV0yetHEWxLI4
# yPO4gCrkM+bIC5ErywTaCrTfCbaJkmMsz7epFLpawAfEtBTysfOzQEdVYcDKA7DO
# uP1qvW3moIOFLJl+s8c5RQzuSP2kB7Ziox92p6bG9o4XM5Wp0SQU201xKuWkwoC2
# VLsd04IopaDQZdMkErqv9+moebI3w3s9zebbzjDicBA+JfgeOltEWp8tIMAXDTx6
# Rk+A6qh7VW73AACwP10GOtC2bslFCokgsuDuabFDC1w6bhWxgBv2bNk+8WJgR8Oe
# BQKZgtH5XmnpTVZCrpgLQhJBWO+shVLm6sOy+wkTKuOwksn/ADd+q2QwAvKKWU6J
# o+E0K5KnN3p4Qw==
# =v5XD
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 16 Mar 2023 20:10:16 GMT
# gpg:                using RSA key 2B67854B98E5327DCDEB17D851F9CC91F83FA044
# gpg:                issuer "jcmvbkbc@gmail.com"
# gpg: Good signature from "Max Filippov <filippov@cadence.com>" [unknown]
# gpg:                 aka "Max Filippov <max.filippov@cogentembedded.com>" [full]
# gpg:                 aka "Max Filippov <jcmvbkbc@gmail.com>" [full]
# Primary key fingerprint: 2B67 854B 98E5 327D CDEB  17D8 51F9 CC91 F83F A044

* tag '20230316-xtensa' of https://github.com/OSLL/qemu-xtensa:
  tests/tcg/xtensa: allow testing big-endian cores
  tests/tcg/xtensa: add linker.ld to CLEANFILES

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Merge tag 'migration-20230316-pull-request' of https://gitlab.com/juan.quintela/qemu into staging

Migration Pull request

Hi

This is just fixes for migration.
- Fix rdma (dave)
- Remove unused variable (Zhijian)
- Fix AVX512 and XBZRLE (Matheus)
- Fix migration preempt (Peter)
- Fix populate_vfio_info (Steve)
- Fix multifd send trace (Wei)

Please apply.

Later, Juan.

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEEGJn/jt6/WMzuA0uC9IfvGFhy1yMFAmQTMa0ACgkQ9IfvGFhy
# 1yMKCBAAyHcE4TO3NrFpG87nGbLhr1zjXkLEKGuAA/E6NKKrf+OrSV2Wq0aBEO/B
# +yAAiXEPsjheh7gwS8FkgAgIj6PwRTtlKae/Tr0JtraukxYT4qzubkaw3X2VjEwf
# uJ8/I+BZOLBUxlZHYoyDP4CWX7elEMlzKdcShcpJCLTgLjxFXjQ9lZwQBgu51cF2
# YitYRDMsVMr77y20eTqCUrLPcSgTM+8h3PtdyDRH2D+KF1IEkjGmSuF8lBc2WiN+
# BiuW6RlaLjhQiBuWNbkrjdulArnim9qhIT5didJuE7p9L50WFSqZs3EV7a4fJjUl
# 1TWJ4k/fzcYf5GF1ypVBrZ1WWCsyJe/fd8ZddGyz9Xpu6Ml7BevVA+0oGeBDYaDB
# T7o2TrHzz2vYCl20jwPbUeafjN8MppBdtHaJ2pP0nA3cPZHPziU2BE9D2/AUxMib
# ICrNe+nVqXUKqx8qZrRPzIBsnoAgmhkNC/k8T7VEhfmm+s6C+xEE0eyhE299TYJn
# ETpQCqiDPCDlkC8iL3OAd8Nn8kucT7EL7qcV2PcpWxgKPc7mCCNIHdpfpM2aZduJ
# 0S/m19ICS8PSWSjFpSK2/vKLOyjA1PKA869iXFD+OOPb8jY5piTYOrj1gowkRjMx
# JOOVDX04lBxbgUVaZnC+iLuh/ZlMRFv2GFAHIk66XDsmUOzRddI=
# =Lwbm
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 16 Mar 2023 15:11:41 GMT
# gpg:                using RSA key 1899FF8EDEBF58CCEE034B82F487EF185872D723
# gpg: Good signature from "Juan Quintela <quintela@redhat.com>" [full]
# gpg:                 aka "Juan Quintela <quintela@trasno.org>" [full]
# Primary key fingerprint: 1899 FF8E DEBF 58CC EE03  4B82 F487 EF18 5872 D723

* tag 'migration-20230316-pull-request' of https://gitlab.com/juan.quintela/qemu:
  migration: fix populate_vfio_info
  migration/multifd: correct multifd_send_thread to trace the flags
  migration/rdma: Remove deprecated variable rdma_return_path
  migration/xbzrle: fix out-of-bounds write with axv512
  migration/xbzrle: use ctz64 to avoid undefined result
  migration/rdma: Fix return-path case
  migration: Wait on preempt channel in preempt thread

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Merge tag 'seabios-1.16.2-20230316-pull-request' of https://gitlab.com/kraxel/qemu into staging

update seabios to 1.16.2

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCgAdFiEEoDKM/7k6F6eZAf59TLbY7tPocTgFAmQS8cMACgkQTLbY7tPo
# cTjbEBAAkM1vhtNYOSKEHVBzSvJjkdV5wr3+WBaglLOpH9H30pRoGLGP8RpF9pUA
# 6fib7/gKNvTtQSCCp31aiK9M6zi3KV51VFcPFoiZfsAwGwuWHDnYGAa2GXssFya+
# kCzErlMFUfJM64rX+/VRktMMfMIV+dNPyWaJyD44Hzds+EpoHfdyZlxlOyBqJ3ju
# eV2CVutXtfUQ4H5eG7feWjPfv+ih4TeC9dPlQuU/nAd8HBt4gDjBhf7zy4pz5De5
# teUJYNuLYLZZxiD4sUw/OqZ9C9j5JBaIz4wZBsiHRSa6d7M+su+QQPGg3CIqFtqo
# Q9m/ITD+B1tgB6k928q6IPZkMVSqKeflEQHPDcDiEYJ4TNQ2h5r/QdCGQoGXdPJ4
# /fmyovZ/9pxCs3BecRXytrYyT4R4HY7DsjxnTMyMgnRiuj9BlJgYJDiJCQyN9DDW
# mHO1AWD+UrKy1k9MzAx9t3GFDeUga997g9jPfLjbJRJwagbHWLbTAB1QEK8E6lZ1
# V5LKX/DYfrw9A1XWFuDCx0L4FJ43dtEGzBBwcoZd01Kpe9va9QS56cdKadGlrPv+
# WYeX+js+1ufSFYagoH6QfFgaShgq+CaXEzDccBJRdFqDqWh8ah6zaHTQRRvjn/33
# brT/XAVBGsLCjZ2Soa2oTd9OBL+pFDbyU+uVcY2ahGGMeSJZH8k=
# =0SRX
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 16 Mar 2023 10:38:59 GMT
# gpg:                using RSA key A0328CFFB93A17A79901FE7D4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>" [full]
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>" [full]
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>" [full]
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* tag 'seabios-1.16.2-20230316-pull-request' of https://gitlab.com/kraxel/qemu:
  update seabios binaries to 1.16.2
  update seabios submodule to 1.16.2

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

migration: fix populate_vfio_info

Include CONFIG_DEVICES so that populate_vfio_info is instantiated for
CONFIG_VFIO.  Without it, the 'info migrate' command never returns
info about vfio.

Fixes: 43bd0bf30f ("migration: Move populate_vfio_info() into a separate file")
Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

migration/multifd: correct multifd_send_thread to trace the flags

The p->flags could be updated via the send_prepare callback, e.g. OR-ed
with MULTIFD_FLAG_ZLIB via zlib_send_prepare. Assign p->flags to the
local "flags" before the send_prepare callback could only get partial of
p->flags. Fix it by moving the assignment of p->flags to the local flags
after the callback, so that the correct flags can be traced.

Fixes: ab7cbb0b9a3b ("multifd: Make no compression operations into its own structure")
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

migration/rdma: Remove deprecated variable rdma_return_path

It's no longer needed since commit
44bcfd45e98 ("migration/rdma: destination: create the return patch after the first accept")

Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

migration/xbzrle: fix out-of-bounds write with axv512

xbzrle_encode_buffer_avx512() checks for overflows too scarcely in its
outer loop, causing out-of-bounds writes:

$ ../configure --target-list=aarch64-softmmu --enable-sanitizers --enable-avx512bw
$ make tests/unit/test-xbzrle && ./tests/unit/test-xbzrle

==5518==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x62100000b100 at pc 0x561109a7714d bp 0x7ffed712a440 sp 0x7ffed712a430
WRITE of size 1 at 0x62100000b100 thread T0
    #0 0x561109a7714c in uleb128_encode_small ../util/cutils.c:831
    #1 0x561109b67f6a in xbzrle_encode_buffer_avx512 ../migration/xbzrle.c:275
    #2 0x5611099a7428 in test_encode_decode_overflow ../tests/unit/test-xbzrle.c:153
    #3 0x7fb2fb65a58d  (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x7a58d)
    #4 0x7fb2fb65a333  (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x7a333)
    #5 0x7fb2fb65aa79 in g_test_run_suite (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x7aa79)
    #6 0x7fb2fb65aa94 in g_test_run (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x7aa94)
    #7 0x5611099a3a23 in main ../tests/unit/test-xbzrle.c:218
    #8 0x7fb2fa78c082 in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x24082)
    #9 0x5611099a608d in _start (/qemu/build/tests/unit/test-xbzrle+0x28408d)

0x62100000b100 is located 0 bytes to the right of 4096-byte region [0x62100000a100,0x62100000b100)
allocated by thread T0 here:
    #0 0x7fb2fb823a06 in __interceptor_calloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cc:153
    #1 0x7fb2fb637ef0 in g_malloc0 (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x57ef0)

Fix that by performing the overflow check in the inner loop, instead.

Signed-off-by: Matheus Tavares Bernardino <quic_mathbern@quicinc.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

migration/xbzrle: use ctz64 to avoid undefined result

__builtin_ctzll() produces undefined results when the argument is 0.
This can be seen through test-xbzrle, which produces the following
warning:

../migration/xbzrle.c:265: runtime error: passing zero to ctz(), which is not a valid argument

Replace __builtin_ctzll() with our ctz64() wrapper which properly
handles 0.

Signed-off-by: Matheus Tavares Bernardino <quic_mathbern@quicinc.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>