dump: Add arch cleanup function

Some architectures (s390x) need to cleanup after a failed dump to be
able to continue to run the vm. Add a cleanup function pointer and
call it if it's set.

Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-ID: <20231109120443.185979-3-frankja@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x/dump: Remove unneeded dump info function pointer init

dump_state_prepare() now sets the function pointers to NULL so we only
need to touch them if we're going to use them.

Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20231109120443.185979-2-frankja@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

Merge tag 'pull-request-2023-11-13' of https://gitlab.com/thuth/qemu into staging

* Fix compilation with Clang 17 on s390x hosts
* Two small s390x PCI fixes
* Update MAINTAINERS file with more entries
* Fix NetBSD VM test
* Clean up some bad wordings

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEJ7iIR+7gJQEY8+q5LtnXdP5wLbUFAmVSAoIRHHRodXRoQHJl
# ZGhhdC5jb20ACgkQLtnXdP5wLbVXBg//VVZS5CXEfOFV91I1kqQnLCvgwmuAyqEg
# PI2/HBxuhzeBx+F1t7uR0n15tUPi1zkFFBOpyBDBubvWcp4vGvFwLQoiBCUvNzBA
# +b1vMySP+K0OO1X5yT3cFHXF9q0o0V5WADwemf5RglIPjlTOIiz9qhD4EYqd2QHC
# EUd9Y45DP4Y0V5raHLjY990f/zr3PuSAB6MASFTUnKdgGkRqonLWdLDdIZNDrZuL
# oGwx1ALXgBOMV3yNyQx9jZBT24git/ai1vd9AU/d3JRKDPsd+4vC39+PTI9NH4h6
# oQglvo399f64cir1f1JJ3MN4ZtwXZpwUkjeTMcR9XZxk7GibU7P2arG5M3TERdmE
# VLqylYsnbJojWOeCH+TViJapRhg1CzUveVlQofr7GHvf2N3oy3BrKaV715gauEyW
# zpjbhSPpIQu9WFXt8+tSquqbvpAP/VlLrOV73D4LzJ7WdTa9CHmSek8D0zoRQDZR
# 8OixrgoBKS+pmBDmTve5gFsIKhZIz9CrmaAKKYdskC8blENxCng8LOFp7sg2PK3M
# U0lWYoDS7qZ85761Bl+QaBdFocdahQqkO/LUQuhoSt2OvA1EGAz2FdVSKkmPDdSS
# P/homr4hOXIqJFSsZj0YNUTIXsXwLBvKjvcJPAWYgbXZhim0LtPQTQO3+ignwGyu
# RXjaVkvkf/s=
# =+2rp
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 13 Nov 2023 06:03:30 EST
# 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-2023-11-13' of https://gitlab.com/thuth/qemu:
  hw/audio/es1370: Clean up comment
  tests/tsan: Rename the file with the entries that should be ignored
  test-resv-mem: Fix CID 1523911
  tests/vm/netbsd: Use Python v3.11
  MAINTAINERS: Add a general architecture section for x86
  MAINTAINERS: Extend the Stellaris section
  MAINTAINERS: Add hw/display/sii9022.c to the Versatile Express section
  MAINTAINERS: Add hw/input/ads7846.c to the PXA2XX section
  MAINTAINERS: Add include/hw/input/pl050.h to the PrimeCell/CMSDK section
  s390x/pci: only limit DMA aperture if vfio DMA limit reported
  s390x/pci: bypass vfio DMA counting when using cdev
  host/include/generic/host/atomic128: Fix compilation problem with Clang 17

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'pull-qapi-2023-11-13' of https://repo.or.cz/qemu/armbru into staging

QAPI patches patches for 2023-11-13

# -----BEGIN PGP SIGNATURE-----
#
# iQJGBAABCAAwFiEENUvIs9frKmtoZ05fOHC0AOuRhlMFAmVR7j4SHGFybWJydUBy
# ZWRoYXQuY29tAAoJEDhwtADrkYZTmT0P/imN9c6xnkx8Kff0iEaT88N8YUHEu5gt
# ECiQtfpbU8y8itIrWpuXWk7HUC1Jp397vKgUhmmr5Im32vx1waUGz/zxdBiitw0a
# 2McXfD6Lbz3LN+q002Vq0QjP3tMeoc8DoP/CagmdXniGrxyJSs/oIsUAdqZnymug
# ZREYRUlbrgJ8XA06n9qIaCGpiihBLl2XBbx4SV9JWle/9vKpII9EBrWNXTcCkde3
# W6gHEKgGAKa1NGYMJtOncanpnzVP0qoLmb3jmiR9tpj3YBB/5LP9l4O11DTC+PQ5
# nn+OmtmjuKE2a2C79Y8VCbRPR6MPSDDbuWwu7Q+A7vXi/1gnkMeXH+A6jE0d3y+p
# PWKqBK65U+yS0IwiUGpMpcGo2WtZiBLzp3ragt+gzi3gV5GvEvPv0uvcEd66in95
# +M2RoVhkbi+e03mF+vDldvbwBYPwlNdgAX/ly7SR76YZybSMK0Jkp7yfvxN35LJc
# kgCo0CY9kAYYUgPUJMiPhaxSfRVk738Qex7QCvFwDiZE8l4XEwoT1PRO0fs8G5aI
# /Yz7hhjkaLw+i+k/tCVCVe+hOgMxxVUz/gZzV4+yrfQNxz41v4WqCfE8a+m6RROo
# TaceVJPKqsyoJ2zmS+CpEPFWkWXEQSQ14yu2wdF8wMyKEWDS4yhRG78rnYC2/ZhC
# zullCL7QfFbf
# =bGOD
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 13 Nov 2023 04:37:02 EST
# gpg:                using RSA key 354BC8B3D7EB2A6B68674E5F3870B400EB918653
# gpg:                issuer "armbru@redhat.com"
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" [full]
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>" [full]
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* tag 'pull-qapi-2023-11-13' of https://repo.or.cz/qemu/armbru:
  tests/qapi-schema: Tidy up pylint warnings and advice
  sphinx/qapidoc: Tidy up pylint warning raise-missing-from
  qapi: Fix QAPISchemaEntity.__repr__()

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'pull-shadow-2023-11-13' of https://repo.or.cz/qemu/armbru into staging

-Wshadow=local patches for 2023-11-13

# -----BEGIN PGP SIGNATURE-----
#
# iQJGBAABCAAwFiEENUvIs9frKmtoZ05fOHC0AOuRhlMFAmVR7ZISHGFybWJydUBy
# ZWRoYXQuY29tAAoJEDhwtADrkYZT7YkP/RUaHzka7vJhcAJSFgCviZ7NYZFbDYkT
# MT9Hi4XYYkOqS0BQ9xZPZDy47oB4pqZlNfMCUD3vmD8LLdVsUyRIbZxu59SdgTOa
# dnNvXIsC16y7Hk6VRfWB4bifOa1rmbF2el5WevA1UpXMRnjkMbtahbXXlSudQpeQ
# Y0e5mILKcbLD1Q6RuTounkfZ/C5pJZKycxVrN8YvFBQ2pILR/JtVwCU9rQ525Sgx
# tgScQ+Z+S4Pw+hyD8kpnw+xqk/iXprgX7GmOhCiy4yLdndq1nZgXOXnZvuIvG3n2
# cvM+dNTVH32qlLMqNdwAaOjAVStobdt85oiyMTnkBfi1+6B+5I6FzyH31/XhV86t
# siCqZ1vsL8N8cuA9mhbKS2eaLrf4ubAyh07SycaKNWfxVMgIAN0rpqaUzKghGcT3
# y9TfFI48zKgCS8AmMuUYy9G6xPlQy726ZHi4yb+t+j2dDx80eAQyXZRPQBiEomND
# cIotxEAPdAdl+52lCwMpYPt+rnxGg4EW05de/CrjKag9Fizfz4cOuwCsSp32XuNe
# TmE1Gi3pKaX/goLO2oTesZ6h8rbMoGZ3XG7YWoIkDpjb+XKuUXXxg+b1cv/0dntk
# Qn6izIFdSscYXakxAkq6ci/3xS0ork3FXjDHihOKj9IWR8YrhqaDZ//XHnIL1Ld7
# QcXs7wElHeJM
# =66br
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 13 Nov 2023 04:34:10 EST
# gpg:                using RSA key 354BC8B3D7EB2A6B68674E5F3870B400EB918653
# gpg:                issuer "armbru@redhat.com"
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" [full]
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>" [full]
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* tag 'pull-shadow-2023-11-13' of https://repo.or.cz/qemu/armbru:
  meson: Enable -Wshadow=local
  block/snapshot: Fix compiler warning with -Wshadow=local

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'mem-2023-11-13' of https://github.com/davidhildenbrand/qemu into staging

Hi,

"Host Memory Backends" and "Memory devices" queue ("mem"):
- One virtio-mem fix leading to a QEMU crash in QEMU debug builds

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEG9nKrXNcTDpGDfzKTd4Q9wD/g1oFAmVR4DsRHGRhdmlkQHJl
# ZGhhdC5jb20ACgkQTd4Q9wD/g1qKMQ//fe/4mJOXQ8l5OZ3ScpC2K7yoB9dowJiQ
# vobja0X0UhyMIOEH4V5RDtMrW3WcYzD2rVwehpLel3QbwcGa7TTB8NtkTx/t4L8P
# tRQe3epGvz+0Kkx4kBFcNBYNR5Skl1rg9kcDhYxNmoOLngWjJcDqRBryfc3V9pEs
# dl9sWXaQn82MGNQGuWFnTOUeOgg1LIdKMRcU2AzhAhrA/e4BqOof/JW+PVdQfzDq
# 4Jhq74pDmKiuH9GmRZgbNlNFX+GxRk63jJrRw4HDAbSD5dBmVnLAjgFZ0sBcKxe0
# HyiGrZOZNIMhMl/GwwQ7NilN03Hl6Hqlx03nz96/2DbiEKr6sOAErIclkUOVlr7k
# YeJvFv+iijqyC4XF43OqoIOz8mtkxan8CuiZW/6/FV9mS/Rb3r8of/BnrK2a8/Kh
# RJLX3tsmrxFdFDxVXWPw+UYrJy8g0xQP2Ils3OReO8QO9qqCytPqJFQsSHDlK3T3
# 2K5FiDpMu7cjFezLyRF0LkPSWg1CV7D6Vc8mp+amc2K4Ltiyhp4xZ2TBKrEC8HHE
# zs+EyEIfsna4SaKwVUVRimWF3+B4GojoAcAD0zju+uhD8Zw+z553zXpr5TSx0Une
# cbMs1n5MTzE6pQo1MmL3hu1xaf6Xdx7hnJPlcnjlKXGFol8ghv6tBkHbOQA5B1/H
# 7hVX43f3epM=
# =7M1K
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 13 Nov 2023 03:37:15 EST
# gpg:                using RSA key 1BD9CAAD735C4C3A460DFCCA4DDE10F700FF835A
# gpg:                issuer "david@redhat.com"
# gpg: Good signature from "David Hildenbrand <david@redhat.com>" [unknown]
# gpg:                 aka "David Hildenbrand <davidhildenbrand@gmail.com>" [full]
# gpg:                 aka "David Hildenbrand <hildenbr@in.tum.de>" [unknown]
# gpg: WARNING: The key's User ID is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 1BD9 CAAD 735C 4C3A 460D  FCCA 4DDE 10F7 00FF 835A

* tag 'mem-2023-11-13' of https://github.com/davidhildenbrand/qemu:
  virtio-mem: fix division by zero in virtio_mem_activate_memslots_to_plug()

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

MAINTAINERS: update virtio-fs mailing list address

The old virtio-fs mailing list address is no longer in use. Switch to
the new mailing list address.

Cc: Philippe Mathieu-Daudé <philmd@linaro.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: German Maglione <gmaglione@redhat.com>
Cc: Hanna Czenczek <hreitz@redhat.com>
Reviewed-by: German Maglione <gmaglione@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20231111004920.148348-1-stefanha@redhat.com>

hw/audio/es1370: Clean up comment

Replace a sweary comment with one that's a bit more helpful to
future readers of the code.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Volker Rümelin <vr_qemu@t-online.de>
Message-ID: <20231110164318.2197569-1-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/tsan: Rename the file with the entries that should be ignored

Let's use a better file name here.

Message-ID: <20231109174720.375873-1-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

test-resv-mem: Fix CID 1523911

Coverity complains about passing "&expected" to "run_range_inverse_array",
which dereferences null "expected". I guess the problem is that the
compare_ranges() loop dereferences 'e' without testing it. However the
loop condition is based on 'ranges' which is garanteed to have
the same length as 'expected' given the g_assert_cmpint() just
before the loop. So the code looks safe to me.

Nevertheless adding a test on expected before the loop to get rid of the
warning.

Fixes: CID 1523901
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reported-by: Coverity (CID 1523901)
Message-ID: <20231110083654.277345-1-eric.auger@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/vm/netbsd: Use Python v3.11

We requiere the 'ninja-build', which depends on 'python311':

  $ pkgin show-deps ninja-build
  direct dependencies for ninja-build-1.11.1nb1
          python311>=3.11.0

So we end up installing both Python v3.10 and v3.11:

  [31/76] installing python311-3.11.5...
  [54/76] installing python310-3.10.13...
  [74/76] installing py310-expat-3.10.13nb1...

Then the build system picks Python v3.11, and doesn't find
py-expat because we only installed the 3.10 version:

  python determined to be '/usr/pkg/bin/python3.11'
  python version: Python 3.11.5

  *** Ouch! ***

  Python's pyexpat module is not found.
  It's normally part of the Python standard library, maybe your distribution packages it separately?
  Either install pyexpat, or alleviate the need for it in the first place by installing pip and setuptools for '/usr/pkg/bin/python3.11'.

  (Hint: NetBSD's pkgsrc debundles this to e.g. 'py310-expat'.)

  ERROR: python venv creation failed

Fix by installing py-expat for v3.11. Remove the v3.10
packages since we aren't using them anymore.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20231109150900.91186-1-philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

MAINTAINERS: Add a general architecture section for x86

It's a little bit weird that the files in target/i386/ which
are not in a subfolder there do not have any associated
maintainer (and thus nobody might be CC:-ed on changes to
these files). We should have a general x86 section for these
files, similar to what we already have for s390x and mips.
Since Paolo is already listed as maintainer for both, the
x86 KVM and TCG CPUs, I'd like to suggest him as maintainer
for the general files, too.

Message-ID: <20230929134551.395438-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

MAINTAINERS: Extend the Stellaris section

This header include/hw/timer/stellaris-gptm.h obviously belongs to the
Stellaris machines, so let's add it to the corresponding section.

And hw/display/ssd0303.c and hw/display/ssd0323.c are only used
by hw/arm/stellaris.c, so add them to the corresponding section
in the MAINTAINERS file, too.

Message-ID: <20231020060936.524988-5-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

MAINTAINERS: Add hw/display/sii9022.c to the Versatile Express section

This graphics adapter is only used by the Versatile Express machine,
so add it to the corresponding section in MAINTAINERS.

Message-ID: <20231020060936.524988-4-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

MAINTAINERS: Add hw/input/ads7846.c to the PXA2XX section

The code from hw/input/ads7846.c is only used by hw/arm/spitz.c,
so add this file to the same section where hw/arm/spitz.c is
listed.

Message-ID: <20231020060936.524988-3-thuth@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

MAINTAINERS: Add include/hw/input/pl050.h to the PrimeCell/CMSDK section

The corresponding pl050.c file is already listed here, so we should
mention the header here, too.

Message-ID: <20231020060936.524988-2-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

s390x/pci: only limit DMA aperture if vfio DMA limit reported

If the host kernel lacks vfio DMA limit reporting, do not attempt
to shrink the guest DMA aperture.

Fixes: df202e3ff3 ("s390x/pci: shrink DMA aperture to be bound by vfio DMA limit")
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Message-ID: <20231110175108.465851-3-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

s390x/pci: bypass vfio DMA counting when using cdev

The current code assumes that there is always a vfio group, but
that's no longer guaranteed with the iommufd backend when using
cdev.  In this case, we don't need to track the vfio dma limit
anyway.

Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Message-ID: <20231110175108.465851-2-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

host/include/generic/host/atomic128: Fix compilation problem with Clang 17

When compiling QEMU with Clang 17 on a s390x, the compilation fails:

In file included from ../accel/tcg/cputlb.c:32:
In file included from /root/qemu/include/exec/helper-proto-common.h:10:
In file included from /root/qemu/include/qemu/atomic128.h:62:
/root/qemu/host/include/generic/host/atomic128-ldst.h:68:15: error:
 __sync builtin operation MUST have natural alignment (consider using __
atomic). [-Werror,-Wsync-alignment]
   68 |     } while (!__sync_bool_compare_and_swap_16(ptr_align, old, new.i));
      |               ^
In file included from ../accel/tcg/cputlb.c:32:
In file included from /root/qemu/include/exec/helper-proto-common.h:10:
In file included from /root/qemu/include/qemu/atomic128.h:61:
/root/qemu/host/include/generic/host/atomic128-cas.h:36:11: error:
 __sync builtin operation MUST have natural alignment (consider using __a
tomic). [-Werror,-Wsync-alignment]
   36 |     r.i = __sync_val_compare_and_swap_16(ptr_align, c.i, n.i);
      |           ^
2 errors generated.

It's arguably a bug in Clang since we already use __builtin_assume_aligned()
to tell the compiler that the pointer is properly aligned. But according to
https://github.com/llvm/llvm-project/issues/69146 it seems like the Clang
folks don't see an easy fix on their side and recommend to use a type
declared with __attribute__((aligned(16))) to work around this problem.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1934
Message-ID: <20231108085954.313071-1-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/qapi-schema: Tidy up pylint warnings and advice

Pylint warns:

    tests/qapi-schema/test-qapi.py:139:13: W1514: Using open without explicitly specifying an encoding (unspecified-encoding)
    tests/qapi-schema/test-qapi.py:143:13: W1514: Using open without explicitly specifying an encoding (unspecified-encoding)

Add encoding='utf-8'.

Pylint advises:

    tests/qapi-schema/test-qapi.py:143:13: R1732: Consider using 'with' for resource-allocating operations (consider-using-with)

Silence this by returning the value directly.

Pylint advises:

    tests/qapi-schema/test-qapi.py:221:4: R1722: Consider using sys.exit() (consider-using-sys-exit)
    tests/qapi-schema/test-qapi.py:226:4: R1722: Consider using sys.exit() (consider-using-sys-exit)

Sure, why not.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20231025092925.1785934-1-armbru@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>

sphinx/qapidoc: Tidy up pylint warning raise-missing-from

Pylint advises:

    docs/sphinx/qapidoc.py:518:12: W0707: Consider explicitly re-raising using 'raise ExtensionError(str(err)) from err' (raise-missing-from)

>From its manual:

    Python's exception chaining shows the traceback of the current
    exception, but also of the original exception.  When you raise a
    new exception after another exception was caught it's likely that
    the second exception is a friendly re-wrapping of the first
    exception.  In such cases `raise from` provides a better link
    between the two tracebacks in the final error.

Makes sense, so do it.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20231025092159.1782638-2-armbru@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>

qapi: Fix QAPISchemaEntity.__repr__()

I messed it up on merge.  It's a debugging aid, so no impact on build.

Fixes: e307a8174bb8 (qapi: provide a friendly string representation of QAPI classes)
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20231024104841.1569250-1-armbru@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>

meson: Enable -Wshadow=local

Local variables shadowing other local variables or parameters make the
code needlessly hard to understand.  Bugs love to hide in such code.
Evidence: commit bbde656263d (migration/rdma: Fix save_page method to
fail on polling error).

Enable -Wshadow=local to prevent such issues.  Possible thanks to
recent cleanups.  Enabling -Wshadow would prevent more issues, but
we're not yet ready for that.

As usual, the warning is only enabled when the compiler recognizes it.
GCC does, Clang doesn't.

Some shadowed locals remain in bsd-user.  Since BSD prefers Clang,
let's not wait for its cleanup.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20231026053115.2066744-2-armbru@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>

virtio-mem: fix division by zero in virtio_mem_activate_memslots_to_plug()

When running with "dynamic-memslots=off", we enter
virtio_mem_activate_memslots_to_plug() to return immediately again
because "vmem->dynamic_memslots == false". However, the compiler might
not optimize out calculating start_idx+end_idx, where we divide by
vmem->memslot_size. In such a configuration, the memslot size is 0 and
we'll get a division by zero:

    (qemu) qom-set vmem0 requested-size 3G
    (qemu) q35.sh: line 38: 622940 Floating point exception(core dumped)

The same is true for virtio_mem_deactivate_unplugged_memslots(), however
we never really reach that code without a prior
virtio_mem_activate_memslots_to_plug() call.

Let's fix it by simply calling these functions only with
"dynamic-memslots=on".

This was found when using a debug build of QEMU.

Message-ID: <20231023111341.219317-1-david@redhat.com>
Reprted-by: Mario Casquero <mcasquer@redhat.com>
Fixes: 177f9b1ee464 ("virtio-mem: Expose device memory dynamically via multiple memslots if enabled")
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Maciej S. Szmigiero <maciej.szmigiero@oracle.com>
Tested-by: Mario Casquero <mcasquer@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>

block/snapshot: Fix compiler warning with -Wshadow=local

No need to declare a new variable in the the inner code block
here, we can re-use the "ret" variable that has been declared
at the beginning of the function. With this change, the code
can now be successfully compiled with -Wshadow=local again.

Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-ID: <20231023175038.111607-1-thuth@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
[Commit message tweaked]
Signed-off-by: Markus Armbruster <armbru@redhat.com>

Merge tag 'qdev-array-prop' of https://repo.or.cz/qemu/kevin into staging

qdev: Make array properties user accessible again

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmVOZicRHGt3b2xmQHJl
# ZGhhdC5jb20ACgkQfwmycsiPL9arpw/+NKGRhSMrSq9Az+z5+ANUfw5SNLJYf1hH
# jm5ITA1Gr9htqHtBfEOdkms2wef6m7onF72rHVUlBKdqCPNMGLme5B0oQ8PZ1X1t
# OxAZ8KYwlO98QvOYl617SA/8wxc0U4/zi192kJpbRkKF6KdbbMGtLKjHyEitA/Yv
# izx1vkKOgQyMFGF1JgIyG4R3WmsKQW1XLqb3emVNRzCqmJpkvMJZQG8tnyEAXlIS
# gkY69cTpaKVaM1OxdB45gjlKTGzLWC/3tTGH+u8q356fvgm/QIgrokCirCZFPIl0
# C8hvzPm/L8hkvWtUb3EZx0DLiunWcAGvoLgBNODHojKRtQ6X9TRTrjJ41ZCLXVqv
# tVJm+XGKC0CZ/WW5yqVOmnzfPH4z8ubzSoRv5ryz3xDb5B/Zr10+ScE+/Ee24wJ2
# HIehxc1LgVGGpikP88/Ns/nAlIVUQxxYvSJ23R5D1+UpP6FCy6Y1pKyRtZGzPCIe
# N4Y+52GtelBR8gOjay5INn/Yf8Fh6sFxX556BW0XKYcbQgvl2bxASe/KVnAVZ1NB
# 8DsaAWlK+hPGopwyp2lDRuGd4kusNbzQvIUZ0mr1g9HQ/iSnT/9RFdExsj+K6QTr
# pX42QCe4mWHPAKx38cez+Bhx4TEOw+GmHuTp/oLdBRuY8DPu/I0Ny364uiW+At/R
# 8jF+jt5uVZc=
# =MV6O
# -----END PGP SIGNATURE-----
# gpg: Signature made Sat 11 Nov 2023 01:19:35 HKT
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* tag 'qdev-array-prop' of https://repo.or.cz/qemu/kevin:
  qdev: Rework array properties based on list visitor
  qdev: Make netdev properties work as list elements
  qom: Add object_property_set_default_list()
  hw/rx/rx62n: Use qdev_prop_set_array()
  hw/arm/xlnx-versal: Use qdev_prop_set_array()
  hw/arm/virt: Use qdev_prop_set_array()
  hw/arm/vexpress: Use qdev_prop_set_array()
  hw/arm/sbsa-ref: Use qdev_prop_set_array()
  hw/arm/mps2: Use qdev_prop_set_array()
  hw/arm/mps2-tz: Use qdev_prop_set_array()
  hw/i386/pc: Use qdev_prop_set_array()

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

qdev: Rework array properties based on list visitor

Until now, array properties are actually implemented with a hack that
uses multiple properties on the QOM level: a static "foo-len" property
and after it is set, dynamically created "foo[i]" properties.

In external interfaces (-device on the command line and device_add in
QMP), this interface was broken by commit f3558b1b ('qdev: Base object
creation on QDict rather than QemuOpts') because QDicts are unordered
and therefore it could happen that QEMU tried to set the indexed
properties before setting the length, which fails and effectively makes
array properties inaccessible. In particular, this affects the 'ports'
property of the 'rocker' device, which used to be configured like this:

-device rocker,len-ports=2,ports[0]=dev0,ports[1]=dev1

This patch reworks the external interface so that instead of using a
separate top-level property for the length and for each element, we use
a single true array property that accepts a list value. In the external
interfaces, this is naturally expressed as a JSON list and makes array
properties accessible again. The new syntax looks like this:

-device '{"driver":"rocker","ports":["dev0","dev1"]}'

Creating an array property on the command line without using JSON format
is currently not possible. This could be fixed by switching from
QemuOpts to a keyval parser, which however requires consideration of the
compatibility implications.

All internal users of devices with array properties go through
qdev_prop_set_array() at this point, so updating it takes care of all of
them.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1090
Fixes: f3558b1b763683bb877f7dd5b282469cdadc65c3
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231109174240.72376-12-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qdev: Make netdev properties work as list elements

The 'name' parameter of QOM setters is primarily used to specify the name
of the currently parsed input element in the visitor interface. For
top-level qdev properties, this is always set and matches 'prop->name'.

However, for list elements it is NULL, because each element of a list
doesn't have a separate name. Passing a non-NULL value runs into
assertion failures in the visitor code.

Therefore, using 'name' in error messages is not right for property
types that are used in lists, because "(null)" (or even a segfault)
isn't very helpful to identify what QEMU is complaining about.

Change netdev properties to use 'prop->name' instead, which will contain
the name of the array property after switching array properties to lists
in the external interface. (This is still not perfect, as it doesn't
identify which element in the list caused the error, but strictly better
than before.)

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231109174240.72376-11-kwolf@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qom: Add object_property_set_default_list()

This function provides a default for properties that are accessed using
the list visitor interface. The default is always an empty list.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-10-kwolf@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/rx/rx62n: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231109174240.72376-9-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/arm/xlnx-versal: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-8-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/arm/virt: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-7-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/arm/vexpress: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-6-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/arm/sbsa-ref: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-5-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/arm/mps2: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-4-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/arm/mps2-tz: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-3-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

hw/i386/pc: Use qdev_prop_set_array()

Instead of manually setting "foo-len" and "foo[i]" properties, build a
QList and use the new qdev_prop_set_array() helper to set the whole
array property with a single call.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <20231109174240.72376-2-kwolf@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

Merge tag 'tracing-pull-request' of https://gitlab.com/stefanha/qemu into staging

Pull request

# -----BEGIN PGP SIGNATURE-----
#
# iQEzBAABCAAdFiEEhpWov9P5fNqsNXdanKSrs4Grc8gFAmVMhGAACgkQnKSrs4Gr
# c8iBoQgAgWjGzp++3Iik0kdMObRlXVpTWwrlYohPg+5NanQB1LifYMt/2MTB70OA
# TUcMcZwzHCUSrtqp+IwU9m5YBZlk+NTQIUn3j5r6Rdv3TnHElBJgpiNWRqpQoDo6
# ZPqSymoyOh2pWtyKQtPQWhx1SsUeF5dEsbOy7fqZi7wLEUqBM1+9LV9C/Y6pIAa7
# KRSnbyNPviyV5B4qX1gaH17DKhj6QyRF/secv8TkKFlfr/6GelNySxEj9+9Om7qO
# R1NT09ZjjMXnNqA9cRzz00fP7bTP36WF065jkYc1oGAo9VwhWAfysju1qbSN2whe
# 4CwucP9lSPuJ+g5tv+7nfX+JfY8+Gw==
# =jAvl
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 09 Nov 2023 15:04:00 HKT
# gpg:                using RSA key 8695A8BFD3F97CDAAC35775A9CA4ABB381AB73C8
# gpg: Good signature from "Stefan Hajnoczi <stefanha@redhat.com>" [ultimate]
# gpg:                 aka "Stefan Hajnoczi <stefanha@gmail.com>" [ultimate]
# Primary key fingerprint: 8695 A8BF D3F9 7CDA AC35  775A 9CA4 ABB3 81AB 73C8

* tag 'tracing-pull-request' of https://gitlab.com/stefanha/qemu:
  tracetool: avoid invalid escape in Python string

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'xen-virtio-fix-1-tag' of https://gitlab.com/sstabellini/qemu into staging

xen-virtio-fix-1

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCgAdFiEE0E4zq6UfZ7oH0wrqiU+PSHDhrpAFAmVKp0cACgkQiU+PSHDh
# rpA0Wg//RUJeBg/Ohjw/ETMjr7+zpsxm0JEjqE5uDYF9+wbX5uAcPOjYLpCbWdKY
# Q9nXmJCbnoqPYzjUrPpx0i8iiVnjbKY28sIqKzPDl4b2s8iPuGrjyPBbcZO1qvx9
# j3UOxiFqcIOIV/Ypf6rdo8hiIwZlHSYR3ZzWLFh2sFGmMee9TjDHiCwRXqZiIzE3
# ks39nV6+Ob1xVTLNZpHjSAsafd/SAusQv66OiGWPixM3DEMKCXUNWtFIfXd5tnVW
# P/4uyRFHzDH+hBZZx2NMiWalTNpcWb4LYhgv/zFiSlnUuw7g0fUBWM1qHWaLcTfY
# BEzm6578BVunFDJe5PEtc5bC1vSZarv6ndRKo7e/t5S/1bzyWzuI8nWCrUKscBSu
# ZgFeJyO9bLRae5QDIiDG6id2dafP9x4zIrn07enuMQVYkjQynA6oS49KL1I+RP38
# Cae+dfadqBksxDbFkh0t02tWjVNeR5MzmkcktlWEn8h136VzZ+fPn93TSTewPxsf
# 3ii6dhiu8+P7A5tfXuRhqlgS8QgaJt6EXDkttjIsC5tWuBo28jZQKw77cZ4DWq1a
# eTy6hn9F0vX+8rNb1dq5Ct6BT6i2svmAloXKBvkaSPLwGLn2lKiUA3ky+9n0ZWEe
# GiyTwkvdaZndrjup+B3MbtQWmJhN+niJZveTJgCdG6or4tvGOuI=
# =1At5
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 08 Nov 2023 05:08:23 HKT
# gpg:                using RSA key D04E33ABA51F67BA07D30AEA894F8F4870E1AE90
# gpg: Good signature from "Stefano Stabellini <sstabellini@kernel.org>" [expired]
# gpg:                 aka "Stefano Stabellini <stefano.stabellini@eu.citrix.com>" [expired]
# gpg: Note: This key has expired!
# Primary key fingerprint: D04E 33AB A51F 67BA 07D3  0AEA 894F 8F48 70E1 AE90

* tag 'xen-virtio-fix-1-tag' of https://gitlab.com/sstabellini/qemu:
  Xen: Fix xen_set_irq() and xendevicemodel_set_irq_level()

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

tracetool: avoid invalid escape in Python string

This is an error in Python 3.12; fix it by using a raw string literal.

Cc: <qemu-stable@nongnu.org>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20231108105649.60453-1-marcandre.lureau@redhat.com>

Merge tag 'for-upstream' of https://repo.or.cz/qemu/kevin into staging

Block layer patches

- Graph locking part 6 (bs->file/backing)
- ahci: trigger either error IRQ or regular IRQ, not both

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmVLvccRHGt3b2xmQHJl
# ZGhhdC5jb20ACgkQfwmycsiPL9ZkFg//awQoPiGnYzHpqcx2tGCM2AqBV+mFkbZr
# BKI5vp8FYfJtgMuHjC8jabL24NRMPpT+HbCzoxwjJU+nnnr85qr7R5iGwG6kfgX6
# HJlAXYXdY6e7l+FV44PBJ52vOoGCsh1GHg8HlKsHMaxSdXi9C1axHJ6rCAjnWXE0
# FQ4znCBVs/9HiKsvu4Wdm5muX2ShftFRM/toAwA+fLEOealX8WEXoRFJXI40bYbR
# OR7aJXWMDQrljlqdKk2FXvK337/tpofXmXf3NIE1R2pmY4x5Fg8bfChZn4UaaCdN
# n+0AhmE4ScI0rXuaXXYOvTO9vdTzXeBROG6tX03t9rrQfB6wPcGVeXRo/uusslAW
# sDH8NLz7uHFOooV02Fs8CKDdVrNNw5qjziclSGa0Po7vqOV1TKI8OTiNpsDLmdI5
# +DQvC6N+IU1pSOXImATSHkheGWggsegrsgN6PdrlzHEXJwWoAaRD0T06MRn74/pL
# gCegK2ez4RJYsci7C5b0gaqY/QBsMj8EUfEGVHvVyuVSoPRwiq4ehPqSQ+siA3xP
# KxYR0e4+QIfRmxqCzaJhiQ3DDGdt8UcO3yF0XcKXEqWwgFAGQKNeUG314jginvmA
# iaJzC0dHbiGcagAk7Ey8iyzfxQDWM6ixzJtGv7VLILepzCuu8vaJXy5qeEtTM/ZI
# EXoDGceNSvw=
# =ikBW
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 09 Nov 2023 00:56:39 HKT
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* tag 'for-upstream' of https://repo.or.cz/qemu/kevin: (25 commits)
  hw/ide/ahci: trigger either error IRQ or regular IRQ, not both
  block: Protect bs->file with graph_lock
  block: Take graph lock for most of .bdrv_open
  vhdx: Take locks for accessing bs->file
  qcow2: Take locks for accessing bs->file
  block: Add missing GRAPH_RDLOCK annotations
  block: Introduce bdrv_co_change_backing_file()
  blkverify: Add locking for request_fn
  block: Protect bs->backing with graph_lock
  block: Mark bdrv_replace_node() GRAPH_WRLOCK
  block: Mark bdrv_replace_node_common() GRAPH_WRLOCK
  block: Inline bdrv_set_backing_noperm()
  block: Mark bdrv_set_backing_hd_drained() GRAPH_WRLOCK
  block: Mark bdrv_cow_child() and callers GRAPH_RDLOCK
  block: Mark bdrv_filter_child() and callers GRAPH_RDLOCK
  block: Mark bdrv_chain_contains() and callers GRAPH_RDLOCK
  block: Mark bdrv_(un)freeze_backing_chain() and callers GRAPH_RDLOCK
  block: Mark bdrv_skip_filters() and callers GRAPH_RDLOCK
  block: Mark bdrv_skip_implicit_filters() and callers GRAPH_RDLOCK
  block: Mark bdrv_filter_or_cow_bs() and callers GRAPH_RDLOCK
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'pull-halloween-omnibus-081123-1' of https://gitlab.com/stsquad/qemu into staging

Final test, gdbstub, plugin and gitdm updates for 8.2

  - fix duplicate register in arm xml
  - hide various duplicate system registers from gdbstub
  - add new gdb register test to the CI (skipping s390x/ppc64 for now)
  - introduce GDBFeatureBuilder
  - move plugin initialisation to after vCPU init completes
  - enable building TCG plugins on Windows platform
  - various gitdm updates
  - some mailmap fixes
  - disable testing for nios2 signals which have regressed

# -----BEGIN PGP SIGNATURE-----
#
# iQEzBAABCgAdFiEEZoWumedRZ7yvyN81+9DbCVqeKkQFAmVLpk4ACgkQ+9DbCVqe
# KkT7Zwf+LgNS2T8Gd6UBMk50Zwew3DSzK3HRRkAlxSV9vN9TCprnVDGJn7ObRpfq
# QCwiTmh20JRPFFBEsPGy/ozNPZsuWbt1/vyh3fnU4KD3aMySuyc/Hb9/mONPC9VE
# zh1mUxLCx10uwG5qF8jupIp22BQYD7B9i4YSF1gAUGsQNU7BPvcBDeDzyhCItJen
# 73oG9RQm7vDbjTOcGDkAMAG8iwLt07oMgFrDSgD8x7RWOxG8aiM3ninAW6S5GcO3
# s49t0rTqJIu+pOncYYzmPvFxyZ/6W82tsJYtfxlVML02qj24HOmLWywRWgL5b10y
# TyXsDba3Ru8ez/kEaVVX6u9N1G/Ktg==
# =or8W
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 08 Nov 2023 23:16:30 HKT
# gpg:                using RSA key 6685AE99E75167BCAFC8DF35FBD0DB095A9E2A44
# gpg: Good signature from "Alex Bennée (Master Work Key) <alex.bennee@linaro.org>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 6685 AE99 E751 67BC AFC8  DF35 FBD0 DB09 5A9E 2A44

* tag 'pull-halloween-omnibus-081123-1' of https://gitlab.com/stsquad/qemu: (23 commits)
  Revert "tests/tcg/nios2: Re-enable linux-user tests"
  mailmap: fixup some more corrupted author fields
  contrib/gitdm: add Daynix to domain-map
  contrib/gitdm: map HiSilicon to Huawei
  contrib/gitdm: add domain-map for Cestc
  contrib/gitdm: Add Rivos Inc to the domain map
  plugins: allow plugins to be enabled on windows
  gitlab: add dlltool to Windows CI
  plugins: disable lockstep plugin on windows
  plugins: make test/example plugins work on windows
  plugins: add dllexport and dllimport to api funcs
  configure: tell meson and contrib_plugins about DLLTOOL
  cpu: Call plugin hooks only when ready
  gdbstub: Introduce GDBFeatureBuilder
  gdbstub: Introduce gdb_find_static_feature()
  gdbstub: Add num_regs member to GDBFeature
  tests/avocado: update the tcg_plugins test
  tests/tcg: add an explicit gdbstub register tester
  target/arm: hide aliased MIDR from gdbstub
  target/arm: hide all versions of DBGD[RS]AR from gdbstub
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

hw/ide/ahci: trigger either error IRQ or regular IRQ, not both

According to AHCI 1.3.1, 5.3.8.1 RegFIS:Entry, if ERR_STAT is set,
we jump to state ERR:FatalTaskfile, which will raise a TFES IRQ
unconditionally, regardless if the I bit is set in the FIS or not.

Thus, we should never raise a normal IRQ after having sent an error
IRQ.

NOTE: for QEMU platforms that use SeaBIOS, this patch depends on QEMU
commit 784155cdcb02 ("seabios: update submodule to git snapshot"), and
QEMU commit 14f5a7bae4cb ("seabios: update binaries to git snapshot"),
which update SeaBIOS to a version that contains SeaBIOS commit 1281e340
("ahci: handle TFES irq correctly").

Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Message-ID: <20231011131220.1992064-1-nks@flawful.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Protect bs->file with graph_lock

Almost all functions that access bs->file already take the graph
lock now. Add locking to the remaining users and finally annotate the
struct field itself as protected by the graph lock.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-25-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Take graph lock for most of .bdrv_open

Most implementations of .bdrv_open first open their file child (which is
an operation that internally takes the write lock and therefore we
shouldn't hold the graph lock while calling it), and afterwards many
operations that require holding the graph lock, e.g. for accessing
bs->file.

This changes block drivers that follow this pattern to take the graph
lock after opening the child node.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-24-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

vhdx: Take locks for accessing bs->file

This updates the vhdx code to add GRAPH_RDLOCK annotations for all
places that read bs->file.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-23-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

qcow2: Take locks for accessing bs->file

This updates the qcow2 code to add GRAPH_RDLOCK annotations for all
places that read bs->file.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-22-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Add missing GRAPH_RDLOCK annotations

This adds GRAPH_RDLOCK to some driver callbacks that are already called
with the graph lock held, and which will need the annotation because
they access bs->file, but don't have it yet.

This also covers a few callbacks that were not marked GRAPH_RDLOCK
before, but where updating BlockDriver is trivially possible.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-21-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Introduce bdrv_co_change_backing_file()

bdrv_change_backing_file() is called both inside and outside coroutine
context. This makes it difficult for it to take the graph lock
internally. It also means that driver implementations need to be able to
run outside of coroutines, too. Switch it to the usual model with a
coroutine based implementation and a co_wrapper instead. The new
function is marked GRAPH_RDLOCK.

As the co_wrapper now runs the function in the AioContext of the node
(as it should always have done), this is not GLOBAL_STATE_CODE() any
more.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-20-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

blkverify: Add locking for request_fn

This is either bdrv_co_preadv() or bdrv_co_pwritev() which both need to
have the graph locked. Annotate the function pointer accordingly and add
locking to its callers.

This shouldn't actually have resulted in a bug because the graph lock is
already held by blkverify_co_prwv(), which waits for the coroutines to
terminate. Annotate with GRAPH_RDLOCK as well to make this clearer.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-19-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Protect bs->backing with graph_lock

Almost all functions that access bs->backing already take the graph
lock now. Add locking to the remaining users and finally annotate the
struct field itself as protected by the graph lock.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-18-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

Revert "tests/tcg/nios2: Re-enable linux-user tests"

nios2 signal tests are broken again:

  retry.py -n 10 -c -- ./qemu-nios2 ./tests/tcg/nios2-linux-user/signals
  Results summary:
  0: 8 times (80.00%), avg time 2.254 (0.00 varience/0.00 deviation)
  -11: 2 times (20.00%), avg time 0.253 (0.00 varience/0.00 deviation)
  Ran command 10 times, 8 passes

This wasn't picked up by CI as we don't have a docker container that
can build QEMU with the nios2 compiler. I don't have time to bisect
the breakage and the target is orphaned anyway so take the easy route
and revert it.

This reverts commit 20e7524ff9f0cab4c9a0306014d6f3d7b467ae1e.

Cc: Chris Wulff <crwulff@gmail.com>
Cc: Marek Vasut <marex@denx.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-23-alex.bennee@linaro.org>

mailmap: fixup some more corrupted author fields

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Cc: Timothée Cocault <timothee.cocault@gmail.com>
Cc: fanwenjie <fanwj@mail.ustc.edu.cn>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-22-alex.bennee@linaro.org>

contrib/gitdm: add Daynix to domain-map

Daynix describes itself as a cloud technology company so I assume
employee contributions should count as such.

Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-21-alex.bennee@linaro.org>

contrib/gitdm: map HiSilicon to Huawei

HiSilicon is a wholly owned subsidiary of Huawei so map the domain to
the same company to avoid splitting the contributions.

Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-20-alex.bennee@linaro.org>

contrib/gitdm: add domain-map for Cestc

Signed-off-by: luzhipeng <luzhipeng@cestc.cn>
Message-Id: <20230628072236.1925-1-luzhipeng@cestc.cn>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-19-alex.bennee@linaro.org>

contrib/gitdm: Add Rivos Inc to the domain map

Whatever they are up to a number of people for the company are
contributing to QEMU so lets group them together.

Reviewed-by: Palmer Dabbelt <palmer@rivosinc.com>
Acked-by: Palmer Dabbelt <palmer@rivosinc.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-18-alex.bennee@linaro.org>

plugins: allow plugins to be enabled on windows

allow plugins to be enabled in the configure script on windows. Also,
add the qemu_plugin_api.lib to the installer.

Signed-off-by: Greg Manning <gmanning@rapitasystems.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231102172053.17692-5-gmanning@rapitasystems.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
[AJB: add check for dlltool to configure]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-17-alex.bennee@linaro.org>

gitlab: add dlltool to Windows CI

We also --disable-plugins for the two mingw based cross builds as
although they have dlltool they seem to be unhappy linking.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>

plugins: disable lockstep plugin on windows

The lockstep plugin uses unix sockets and would require a different
communication mechanism to work on Windows.

Signed-off-by: Greg Manning <gmanning@rapitasystems.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231102172053.17692-4-gmanning@rapitasystems.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-16-alex.bennee@linaro.org>

plugins: make test/example plugins work on windows

Generate a qemu_plugin_api.lib delay import lib on windows, for
windows qemu plugins to link against.

Implement an example dll load fail hook to link up the API functions
correctly when a plugin is loaded on windows.

Update the build scripts for the test and example plugins to use these
things.

Signed-off-by: Greg Manning <gmanning@rapitasystems.com>
Acked-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231102172053.17692-3-gmanning@rapitasystems.com>
[AJB: use find_program for dlltool, s/Windows/windows/]
Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-15-alex.bennee@linaro.org>

plugins: add dllexport and dllimport to api funcs

In qemu-plugin.h, mark all API functions as __declspec(dllexport) when
compiling the executables, and as __declspec(dllimport) when being used
to compile plugins against.

Signed-off-by: Greg Manning <gmanning@rapitasystems.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231102172053.17692-2-gmanning@rapitasystems.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20231106185112.2755262-14-alex.bennee@linaro.org>

configure: tell meson and contrib_plugins about DLLTOOL

To cleanly handle cross-building we need to export the details of
dlltool into meson's list of cross binaries and into the
contrib/plugins/ make configuration.

Cc: Greg Manning <gmanning@rapitasystems.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-13-alex.bennee@linaro.org>

cpu: Call plugin hooks only when ready

The initialization and exit hooks will not affect the state of vCPU
outside TCG context, but they may depend on the state of vCPU.
Therefore, it's better to call plugin hooks after the vCPU state is
fully initialized and before it gets uninitialized.

Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231025093128.33116-16-akihiko.odaki@daynix.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-12-alex.bennee@linaro.org>

gdbstub: Introduce GDBFeatureBuilder

GDBFeatureBuilder unifies the logic to generate dynamic GDBFeature.

Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231025093128.33116-4-akihiko.odaki@daynix.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-11-alex.bennee@linaro.org>

gdbstub: Introduce gdb_find_static_feature()

This function is useful to determine the number of registers exposed to
GDB from the XML name.

Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231025093128.33116-3-akihiko.odaki@daynix.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-10-alex.bennee@linaro.org>

gdbstub: Add num_regs member to GDBFeature

Currently the number of registers exposed to GDB is written as magic
numbers in code. Derive the number of registers GDB actually see from
XML files to replace the magic numbers in code later.

Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231025093128.33116-2-akihiko.odaki@daynix.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-9-alex.bennee@linaro.org>

tests/avocado: update the tcg_plugins test

There are a number of things that are broken on the test currently so
lets fix that up:

  - replace retired Debian kernel for tuxrun_baseline one
  - remove "detected repeat instructions test" since ea185a55
  - log total counted instructions/memory accesses

Acked-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-8-alex.bennee@linaro.org>

tests/tcg: add an explicit gdbstub register tester

We already do a couple of "info registers" for specific tests but this
is a more comprehensive multiarch test. It also has some output
helpful for debugging the gdbstub by showing which XML features are
advertised and what the underlying register numbers are.

My initial motivation was to see if there are any duplicate register
names exposed via the gdbstub while I was reviewing the proposed
register interface for TCG plugins.

Mismatches between the xml and remote-desc are reported for debugging
but do not fail the test.

We also skip the tests for the following arches for now until we can
investigate and fix any issues:

  - s390x (fails to read v0l->v15l, not seen in remote-registers)
  - ppc64 (fails to read vs0h->vs31h, not seen in remote-registers)

Cc: Akihiko Odaki <akihiko.odaki@daynix.com>
Cc: Luis Machado <luis.machado@linaro.org>
Cc: Ilya Leoshkevich <iii@linux.ibm.com>
Cc: qemu-s390x@nongnu.org
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Daniel Henrique Barboza <danielhb413@gmail.com>
Cc: qemu-ppc@nongnu.org
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-7-alex.bennee@linaro.org>

target/arm: hide aliased MIDR from gdbstub

This is just a constant alias register with the same value as the
"other" MIDR so it serves no purpose being presented to gdbstub.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20231106185112.2755262-6-alex.bennee@linaro.org>

target/arm: hide all versions of DBGD[RS]AR from gdbstub

This avoids two duplicates being presented to gdbstub. As the
registers are RAZ anyway it is unlikely their value would be of use to
someone using gdbstub anyway.

Acked-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-5-alex.bennee@linaro.org>

target/arm: mark the 32bit alias of PAR when LPAE enabled

We also mark it ARM_CP_NO_GDB so we avoid duplicate PAR's in the
system register XML we send to gdb.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20231107105145.2916124-1-alex.bennee@linaro.org>

gdb-xml: fix duplicate register in arm-neon.xml

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Fixes: 56aebc8916 ("Add GDB XML register description support")
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-3-alex.bennee@linaro.org>

default-configs: Add TARGET_XML_FILES definition

loongarch64-linux-user has references to XML files so include them.

Fixes: d32688ecdb ("default-configs: Add loongarch linux-user support")
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-Id: <20231030054834.39145-6-akihiko.odaki@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
[AJB: remove base32 from list]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231106185112.2755262-2-alex.bennee@linaro.org>

Merge tag 'pull-ppc-20231107' of https://gitlab.com/danielhb/qemu into staging

ppc patch queue for 2023-11-07:

This queue, the last one before the 8.2 feature freeze, has miscellanous
changes that includes new PowerNV features and the new AmigaONE XE
board.

# -----BEGIN PGP SIGNATURE-----
#
# iIwEABYKADQWIQQX6/+ZI9AYAK8oOBk82cqW3gMxZAUCZUqiORYcZGFuaWVsaGI0
# MTNAZ21haWwuY29tAAoJEDzZypbeAzFkBSUA/2qm8CyrRqY5+tsjtWQqZmPZ3L1F
# CgnXFNqtY2tzbTe5AQCi6FeQBEmXbZYVfryZyA+CQ4DUERc+18pe6hV3bBR9Cg==
# =cnHS
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 08 Nov 2023 04:46:49 HKT
# gpg:                using EDDSA key 17EBFF9923D01800AF2838193CD9CA96DE033164
# gpg:                issuer "danielhb413@gmail.com"
# gpg: Good signature from "Daniel Henrique Barboza <danielhb413@gmail.com>" [unknown]
# gpg: WARNING: The key's User ID is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 17EB FF99 23D0 1800 AF28  3819 3CD9 CA96 DE03 3164

* tag 'pull-ppc-20231107' of https://gitlab.com/danielhb/qemu:
  ppc: qtest already exports qtest_rtas_call()
  hw/pci-host: Update PHB5 XSCOM registers
  ppc/pnv: Fix number of I2C engines and ports for power9/10
  ppc/pnv: Connect PNV I2C controller to powernv10
  ppc/pnv: Connect I2C controller model to powernv9 chip
  ppc/pnv: Add an I2C controller model
  tests/avocado: Add test for amigaone board
  hw/ppc: Add emulation of AmigaOne XE board
  hw/pci-host: Add emulation of Mai Logic Articia S

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'misc-fixes-pull-request' of https://gitlab.com/berrange/qemu into staging

 * Print error with f32 audio format in wav backend instead of abort
 * Drop redundant listener name suffix
 * Update libvirt mailing list address
 * Cleanup RSA key error paths
 * Fix CPU uarch ABI script QMP command calling

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEE2vOm/bJrYpEtDo4/vobrtBUQT98FAmVKiwoACgkQvobrtBUQ
# T9+SjxAAmDBWE9WI/eKZEb3QVNi3W90bVNIKMSdu7DMaqSPGMe6dcNeJy6kHL2ud
# EgdvtKgWq5wwFDojrYiUB75tSFGP7InK2a7EN7t60kE/XxtqR/fvLo11d+gUI7Z6
# vetsHLtDYEs3Hr/75qg1/ock/G1TxzlJknzww1zDjQrwFSaJnOWDdY+RSRGTwEho
# PRfFRkIDZqs3LAKyjRdiYWms90vogG2Bwygmo/mirgta9Lmkz9WC9SN/EqOvOBOE
# NyaO84xu38U3JwN5d9w4MFdgayFNRYTcvTu8J0NdHLp16CYBViGecIGb4deDYN0U
# jhmfKChyr0tJucPsRndXQU2KFusePjiSWEsG9FdbpXdjFGIwinJUDKOaHI4Kk9pc
# stGCwb8+9Ec0e31JqtMBd5VeoToK8UsthVqAx1brnXv9K5liSXS7nyFPwNp/SMi4
# UNpjawSItAHEgc6VzqDNZY+dGtueMZmM0lf7cMLho3BgAeyTbr1NuwOIJx28JP2E
# NnkAwDHjj9u1mkhRM7BbgBzGq006naWP8V6dCGQSH+/D0C9P3yzs7XLtvwtUDO3P
# DalGRsazT+o+pyuyw0uyV2z6GuLAl9ZhI0MJCgIxsmASdxANxMcIaCrs1yc0IG2A
# y16e4/Ic/FzqkbwkkBy3tkDF7vI0++1MuqmcWki+RFz1hzqmY+Y=
# =KHzn
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 08 Nov 2023 03:07:54 HKT
# gpg:                using RSA key DAF3A6FDB26B62912D0E8E3FBE86EBB415104FDF
# gpg: Good signature from "Daniel P. Berrange <dan@berrange.com>" [full]
# gpg:                 aka "Daniel P. Berrange <berrange@redhat.com>" [full]
# Primary key fingerprint: DAF3 A6FD B26B 6291 2D0E  8E3F BE86 EBB4 1510 4FDF

* tag 'misc-fixes-pull-request' of https://gitlab.com/berrange/qemu:
  scripts/cpu-x86-uarch-abi.py: Fix parameter error of cmd
  MAINTAINERS: update libvirt devel mailing list address
  audio: don't abort on f32 audio format in wav backend
  io: Stop appending -listen to net listeners
  crypto/rsakey-builtin.c.inc: Clean up two error paths

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'pull-request-2023-11-07' of https://gitlab.com/thuth/qemu into staging

* Fix s390x CPU reconfiguration information in the SCLP facility map
* Fix condition code problem in the CLC and LAALG instruction
* Fix ordering of the new s390x topology list entries
* Add some more files to the MAINTAINERS file
* Allow newer versions of Tesseract in the m68k nextcube test

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEJ7iIR+7gJQEY8+q5LtnXdP5wLbUFAmVKgksRHHRodXRoQHJl
# ZGhhdC5jb20ACgkQLtnXdP5wLbWIHg//TM3JOpsMEqHKlUKqOJH02mFQrK6H7LG0
# BC56FG7T+/mpYs1NTG92t8nCK03C2ZCweQWD7ZulRJAjPhZv+TF5bJEForivU7+k
# PKEshz9xKCWn2YGyNnf2LA06J1JkF215+KlReOoxwSgj1cPlHfBLQ0DtxmpJJZ1G
# h5p4d26BbSlwR58HrFWTlhgJMPenl59BETUGIK1FklBxunmZeeijddfniAhOT44y
# i0u9/H9KCg3tkwBROUy+42QV+ef32kz/yvi5RmYQI5W7PixO4sxH6MYduOjshsu9
# wK70f8EOwiZV6lFxqmbV7vxFeNnp5IuaVU7PMBoAkwZqLw99mSFy1+1BabCuL5b+
# 3iUTiD4UW48MYwE2Ua6Lit4kpfjhwcp/UYz6pIk6TCBQX6LfzO+nj+rod0GdIpyZ
# 4Lwm7jBtpTlYkGrsMvpA/qcidOtqPA1lmBTNlY1hFodQF6KWtyObn0w5AM80xeeU
# /mGxQDz97Bpz7LKZvhu+k38jaWvnJFnl3jF1zet88CYL9YL+YI/k1KjhFafCXb0V
# 38Xpt5JTWxyLSh2B3gx0OpokX5bftvW9GlLix0HqL7c23uYwR2Bq+Rd6I8SAlk4C
# uJq6gqP8IFBFHfgbmyqf/fyd/eHxm7J1voIdy9PZyxZ1JYT9A7yu56qV6SJYwCpr
# aARwui/Dm4o=
# =y+cC
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 08 Nov 2023 02:30:35 HKT
# 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-2023-11-07' of https://gitlab.com/thuth/qemu:
  target/s390x/cpu topology: Fix ordering and creation of TLEs
  tests/tcg/s390x: Test ADD LOGICAL WITH CARRY
  tests/tcg/s390x: Test LAALG with negative cc_src
  target/s390x: Fix LAALG not updating cc_src
  tests/tcg/s390x: Test CLC with inaccessible second operand
  target/s390x: Fix CLC corrupting cc_src
  target/s390x/cpu_models: Use 'first_cpu' in s390_get_feat_block()
  s390/sclp: fix SCLP facility map
  tests/avocado: Allow newer versions of tesseract in the nextcube test
  MAINTAINERS: Add artist.c to the hppa machine section
  MAINTAINERS: Add the virtio-gpu documentation to the corresponding section

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'misc-cpus-20231107' of https://github.com/philmd/qemu into staging

Misc hardware patch queue

HW emulation:
- PMBus fixes and tests (Titus)
- IDE fixes and tests (Fiona)
- New ADM1266 sensor (Titus)
- Better error propagation in PCI-ISA i82378 (Philippe)
- Declare SD model QOM types using DEFINE_TYPES macro (Philippe)

Topology:
- Fix CPUState::nr_cores calculation (Zhuocheng Ding and Zhao Liu)

Monitor:
- Synchronize CPU state in 'info lapic' (Dongli Zhang)

QOM:
- Have 'cpu-qom.h' target-agnostic (Philippe)
- Move ArchCPUClass definition to each target's cpu.h (Philippe)
- Call object_class_is_abstract once in cpu_class_by_name (Philippe)

UI:
- Use correct key names in titles on MacOS / SDL2 (Adrian)

MIPS:
- Fix MSA BZ/BNZ and TX79 LQ/SQ opcodes (Philippe)

Nios2:
- Create IRQs *after* vCPU is realized (Philippe)

PPC:
- Restrict KVM objects to system emulation (Philippe)
- Move target-specific definitions out of 'cpu-qom.h' (Philippe)

S390X:
- Make hw/s390x/css.h and hw/s390x/sclp.h headers target agnostic (Philippe)

X86:
- HVF & KVM cleanups (Philippe)

Various targets:
- Use env_archcpu() to optimize (Philippe)

Misc:
- Few global variable shadowing removed (Philippe)
- Introduce cpu_exec_reset_hold and factor tcg_cpu_reset_hold out (Philippe)
- Remove few more 'softmmu' mentions (Philippe)
- Fix and cleanup in vl.c (Akihiko & Marc-André)
- Resource leak fix in dump (Zongmin Zhou)
- MAINTAINERS updates (Thomas, Daniel)

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEE+qvnXhKRciHc/Wuy4+MsLN6twN4FAmVKKmEACgkQ4+MsLN6t
# wN4xHQ//X/enH4C7K3VP/tSinDiwmXN2o61L9rjqSDQkBaCtktZx4c8qKSDL7V4S
# vwzmvvBn3biMXQwZNVJo9d0oz2qoaF9tI6Ao0XDHAan9ziagfG9YMqWhkCfj077Q
# jLdCqkUuMJBvQgXGB1a6UgCme8PQx7h0oqjbCNfB0ZBls24b5DiEjO87LE4OTbTi
# zKRhYEpZpGwIVcy+1dAsbaBpGFP06sr1doB9Wz4c06eSx7t0kFSPk6U4CyOPrGXh
# ynyCxPwngxIXmarY8gqPs3SBs7oXsH8Q/ZOHr1LbuXhwSuw/0zBQU9aF7Ir8RPan
# DB79JjPrtxTAhICKredWT79v9M18D2/1MpONgg4vtx5K2FzGYoAJULCHyfkHMRSM
# L6/H0ZQPHvf7w72k9EcSQIhd0wPlMqRmfy37/8xcLiw1h4l/USx48QeKaeFWeSEu
# DgwSk+R61HbrKvQz/U0tF98zUEyBaQXNrKmyzht0YE4peAtpbPNBeRHkd0GMae/Z
# HOmkt8QlFQ0T14qSK7mSHaSJTUzRvFGD01cbuCDxVsyCWWsesEikXBACZLG5RCRY
# Rn1WeX1H9eE3kKi9iueLnhzcF9yM5XqFE3f6RnDzY8nkg91lsTMSQgFcIpv6uGyp
# 3WOTNSC9SoFyI3x8pCWiKOGytPUb8xk+PnOA85wYvVmT+7j6wus=
# =OVdQ
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 07 Nov 2023 20:15:29 HKT
# gpg:                using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.org>" [full]
# Primary key fingerprint: FAAB E75E 1291 7221 DCFD  6BB2 E3E3 2C2C DEAD C0DE

* tag 'misc-cpus-20231107' of https://github.com/philmd/qemu: (75 commits)
  dump: Add close fd on error return to avoid resource leak
  ui/sdl2: use correct key names in win title on mac
  MAINTAINERS: Add more guest-agent related files to the corresponding section
  MAINTAINERS: Add include/hw/xtensa/mx_pic.h to the XTFPGA machine section
  MAINTAINERS: update libvirt devel mailing list address
  MAINTAINERS: Add the CAN documentation file to the CAN section
  MAINTAINERS: Add include/hw/timer/tmu012.h to the SH4 R2D section
  hw/sd: Declare QOM types using DEFINE_TYPES() macro
  hw/i2c: pmbus: reset page register for out of range reads
  hw/i2c: pmbus: immediately clear faults on request
  tests/qtest: add tests for ADM1266
  hw/sensor: add ADM1266 device model
  hw/i2c: pmbus: add VCAP register
  hw/i2c: pmbus: add fan support
  hw/i2c: pmbus: add vout mode bitfields
  hw/i2c: pmbus add support for block receive
  tests/qtest: ahci-test: add test exposing reset issue with pending callback
  hw/ide: reset: cancel async DMA operation before resetting state
  hw/cpu: Update the comments of nr_cores and nr_dies
  system/cpus: Fix CPUState.nr_cores' calculation
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Xen: Fix xen_set_irq() and xendevicemodel_set_irq_level()

Remove '=' from 'if CONFIG_XEN_CTRL_INTERFACE_VERSION <= 41500'.
Because xendevicemodel_set_irq_level() was introduced in 4.15 version.

Also, update xendevicemodel_set_irq_level() to return -1 for older versions.

Signed-off-by: Vikram Garhwal <vikram.garhwal@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>

ppc: qtest already exports qtest_rtas_call()

Having two functions with the same name is a bad idea.  As spapr only
uses the function locally, made it static.

When you compile with clang, you get this compilation error:

/usr/bin/ld: tests/qtest/libqos/libqos.fa.p/.._libqtest.c.o: in function `qtest_rtas_call':
/scratch/qemu/clang/full/all/../../../../../mnt/code/qemu/full/tests/qtest/libqtest.c:1195: multiple definition of `qtest_rtas_call'; libqemu-ppc64-softmmu.fa.p/hw_ppc_spapr_rtas.c.o:/scratch/qemu/clang/full/all/../../../../../mnt/code/qemu/full/hw/ppc/spapr_rtas.c:536: first defined here
clang-16: error: linker command failed with exit code 1 (use -v to see invocation)
ninja: build stopped: subcommand failed.
make: *** [Makefile:162: run-ninja] Error 1

Signed-off-by: Juan Quintela <quintela@redhat.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Message-ID: <20231030163834.4638-1-quintela@redhat.com>
[dhb: remove 'spapr_rtas.h' include from spapr_rtas.c]
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

hw/pci-host: Update PHB5 XSCOM registers

Add new XSCOM registers introduced in PHB5.
Apply bit-masks within xscom-write methods.
Bit-masks specified using PPC_BITMASK macro.

Signed-off-by: Saif Abrar <saif.abrar@linux.vnet.ibm.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Harsh Prateek Bora <harshpb@linux.ibm.com>
Message-ID: <20231016175948.10869-1-saif.abrar@linux.vnet.ibm.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

ppc/pnv: Fix number of I2C engines and ports for power9/10

Power9 is supposed to have 4 PIB-connected I2C engines with the
following number of ports on each engine:

    0: 2
    1: 13
    2: 2
    3: 2

Power10 also has 4 engines but has the following number of ports
on each engine:

    0: 14
    1: 14
    2: 2
    3: 16

Current code assumes that they all have the same (maximum) number.
This can be a problem if software expects to see a certain number
of ports present (Power Hypervisor seems to care).

Fixed this by adding separate tables for power9 and power10 that
map the I2C controller number to the number of I2C buses that should
be attached for that engine.

Reviewed-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Glenn Miles <milesg@linux.vnet.ibm.com>
Message-ID: <20231025152714.956664-1-milesg@linux.vnet.ibm.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

ppc/pnv: Connect PNV I2C controller to powernv10

Wires up four I2C controller instances to the powernv10 chip
XSCOM address space.

Each controller instance is wired up to two I2C buses of
its own.  No other I2C devices are connected to the buses
at this time.

Signed-off-by: Glenn Miles <milesg@linux.vnet.ibm.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Message-ID: <20231017221434.810363-1-milesg@linux.vnet.ibm.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

ppc/pnv: Connect I2C controller model to powernv9 chip

Wires up three I2C controller instances to the powernv9 chip
XSCOM address space.

Each controller instance is wired up to a single I2C bus of
its own.  No other I2C devices are connected to the buses
at this time.

Signed-off-by: Cédric Le Goater <clg@kaod.org>
[milesg: Split wiring from addition of model itself]
[milesg: Added new commit message]
[milesg: Moved hardcoded attributes into PnvChipClass]
[milesg: Removed TODO comment for I2C]
Signed-off-by: Glenn Miles <milesg@linux.vnet.ibm.com>
Acked-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-ID: <20231016222013.3739530-3-milesg@linux.vnet.ibm.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

ppc/pnv: Add an I2C controller model

The more recent IBM power processors have an embedded I2C
controller that is accessible by software via the XSCOM
address space.

Each instance of the I2C controller is capable of controlling
multiple I2C buses (one at a time).  Prior to beginning a
transaction on an I2C bus, the bus must be selected by writing
the port number associated with the bus into the PORT_NUM
field of the MODE register.  Once an I2C bus is selected,
the status of the bus can be determined by reading the
Status and Extended Status registers.

I2C bus transactions can be started by writing a command to
the Command register and reading/writing data from/to the
FIFO register.

Not supported :

 . 10 bit I2C addresses
 . Multimaster
 . Slave

Signed-off-by: Cédric Le Goater <clg@kaod.org>
[milesg: Split wiring to powernv9 into its own commit]
[milesg: Added more detail to commit message]
[milesg: Added SPDX Licensed Identifier to new files]
[milesg: updated copyright dates]
[milesg: Added use of g_autofree]
[milesg: Added NULL check after pnv_i2c_get_bus]
Signed-off-by: Glenn Miles <milesg@linux.vnet.ibm.com>
Acked-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-ID: <20231016222013.3739530-2-milesg@linux.vnet.ibm.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

tests/avocado: Add test for amigaone board

Add an avocado test for the amigaone board that tests it with the
firmware.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-ID: <b1a0246840fcff1fe6bbd8685e2474a9231b34c5.1698406922.git.balaton@eik.bme.hu>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

hw/ppc: Add emulation of AmigaOne XE board

The AmigaOne is a rebranded MAI Teron board that uses U-Boot firmware
with patches to support AmigaOS and is very similar to pegasos2 so can
be easily emulated sharing most code with pegasos2. The reason to
emulate it is that AmigaOS comes in different versions for AmigaOne
and PegasosII which only have drivers for one machine and firmware so
these only run on the specific machine. Adding this board allows
another AmigaOS version to be used reusing already existing peagasos2
emulation. (The AmigaOne was the first of these boards so likely most
widespread which then inspired Pegasos that was later replaced with
PegasosII due to problems with Articia S, so these have a lot of
similarity. Pegasos mainly ran MorphOS while the PegasosII version of
AmigaOS was added later and therefore less common than the AmigaOne
version.)

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Tested-by: Rene Engel <ReneEngel80@emailn.de>
Acked-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-ID: <804935e7a5921548d630576159ae2c758fe6e275.1699382232.git.balaton@eik.bme.hu>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>

target/s390x/cpu topology: Fix ordering and creation of TLEs

In case of horizontal polarization entitlement has no effect on
ordering.
Moreover, since the comparison is used to insert CPUs at the correct
position in the TLE list, this affects the creation of TLEs and now
correctly collapses horizontally polarized CPUs into one TLE.

Fixes: f4f54b582f ("target/s390x/cpu topology: handle STSI(15) and build the SYSIB")
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Message-ID: <20231027163637.3060537-1-nsg@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/tcg/s390x: Test ADD LOGICAL WITH CARRY

Add a test that tries different combinations of ADD LOGICAL WITH
CARRY instructions.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Message-ID: <20231106093605.1349201-6-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/tcg/s390x: Test LAALG with negative cc_src

Add a small test to prevent regressions.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20231106093605.1349201-5-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Fix LAALG not updating cc_src

LAALG uses op_laa() and wout_addu64(). The latter expects cc_src to be
set, but the former does not do it. This can lead to assertion failures
if something sets cc_src to neither 0 nor 1 before.

Fix by introducing op_laa_addu64(), which sets cc_src, and using it for
LAALG.

Fixes: 4dba4d6fef61 ("target/s390x: Use atomic operations for LOAD AND OP")
Cc: qemu-stable@nongnu.org
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20231106093605.1349201-4-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/tcg/s390x: Test CLC with inaccessible second operand

Add a small test to prevent regressions.

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-ID: <20231106093605.1349201-3-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x: Fix CLC corrupting cc_src

CLC updates cc_src before accessing the second operand; if the latter
is inaccessible, the former ends up containing a bogus value.

Fix by reading cc_src into a temporary first.

Fixes: 4f7403d52b1c ("target-s390: Convert CLC")
Closes: https://gitlab.com/qemu-project/qemu/-/issues/1865
Cc: qemu-stable@nongnu.org
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-ID: <20231106093605.1349201-2-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

target/s390x/cpu_models: Use 'first_cpu' in s390_get_feat_block()

We already have a global 'first_cpu' variable storing a pointer
to the first CPU, no need to use a static one.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20231030093150.65297-1-philmd@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

s390/sclp: fix SCLP facility map

Qemu's SCLP implementation incorrectly reports that it supports CPU
reconfiguration. If a guest issues a CPU reconfiguration request it
is rejected as invalid command.

Fix the SCLP_HAS_CPU_INFO mask, and remove the unused
SCLP_CMDW_CONFIGURE_CPU and SCLP_CMDW_DECONFIGURE_CPU defines.

Reviewed-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Halil Pasic <pasic@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Message-ID: <20231024100703.929679-1-hca@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/avocado: Allow newer versions of tesseract in the nextcube test

Current Linux distros ship version 5 of the tesseract OCR software,
so the nextcube screen test is ignored there. Let's make the check
more flexible to allow newer versions, too, and remove the old v3
test since most Linux distros don't ship this version anymore.

Message-ID: <20231101204323.35533-1-huth@tuxfamily.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>

block: Mark bdrv_replace_node() GRAPH_WRLOCK

Instead of taking the writer lock internally, require callers to already
hold it when calling bdrv_replace_node(). Its callers may already want
to hold the graph lock and so wouldn't be able to call functions that
take it internally.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-17-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Mark bdrv_replace_node_common() GRAPH_WRLOCK

Instead of taking the writer lock internally, require callers to already
hold it when calling bdrv_replace_node_common(). Basically everthing in
the function needs the lock and its callers may already want to hold the
graph lock and so wouldn't be able to call functions that take it
internally.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-16-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Inline bdrv_set_backing_noperm()

It's only a single line and has a single caller. Inlining makes things
a bit easier to follow.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-15-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Mark bdrv_set_backing_hd_drained() GRAPH_WRLOCK

Instead of taking the writer lock internally, require callers to already
hold it when calling bdrv_set_backing_hd_drained(). Basically everthing
in the function needs the lock and its callers may already want to hold
the graph lock and so wouldn't be able to call functions that take it
internally.

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20231027155333.420094-14-kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>