OSDN Git Service

qmiga/qemu.git
16 months agoMerge tag 'pull-request-2023-03-07' of https://gitlab.com/thuth/qemu into staging
Peter Maydell [Thu, 9 Mar 2023 15:19:30 +0000 (15:19 +0000)]
Merge tag 'pull-request-2023-03-07' of https://gitlab.com/thuth/qemu into staging

* Refine the distro support policy
* Deprecate 32-bit x86 and arm hosts for system emulation
* Check bison version to be >= 3.0
* Compile vnc test only if vnc is really enabled
* Check docs/config/ich9-ehci-uhci.cfg via the readconfig-test
* s390x: Add support for list-directed IPL from ECKD DASD

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEJ7iIR+7gJQEY8+q5LtnXdP5wLbUFAmQIQD0RHHRodXRoQHJl
# ZGhhdC5jb20ACgkQLtnXdP5wLbX+EA/5ATc/Rz9Y8TZF36/sUKrmjOxQnX91hgRW
# zovgZejoHtek1AC4pLd9kcpo1JonXkOfmbkNRYTw6+1wY6Ipt2apkWAc+MqfvQ80
# dGznbLXnX/E2QHW7h5XqvAhRGvjog9b38zAQcR37PKZ9DkSfWBwMgTT2SXYpLUjR
# YWgqPGBYRq1ViG4q7hv2dHqi/idzh+DOWFGU5NAQC1fEBOmPYN9u6vfbX86dwZsy
# S9pqw16yN2MZKL2bgIViFIjdhPPMO3RFcv8NTPipzk3g/pHNIv58+xCE8r/Tp8LW
# JRQzdh6JbGvj1BXH68igXWGUFW6dhOXZWhgmFe4QEDaiCj91pHGvQCYgo2agx2bl
# rXaSQXSMhIcm+t9C9kO6UNxZf+1HohcM1ZlQXhvX6FcBPVMUx/52YEpiK+OtBKyq
# 43AlKp4fp1xXLFSOyMNgSMR90sksng6CEoQqTr0jCZtF8H18wb+eHfb6dME3XRCw
# SKQeNGhkP1FxaQnlRAkeBxW7GkDE5YBjf9pSIaLZ/8VsLdyo3SK/DLkre5+qnu/V
# lksZlt1K7xDLxRFopBuVJ3MHeFSRazBDbWfrRFOALkdbGEEeZnnmEPFjD3mag+P4
# 1y9WmBJHRLlvhlm195WA8babenUZoOfLkPrL0k7mNLs9sZyx1eIkZfOUgCWjFE2V
# 9VBOWL8oi/w=
# =p13b
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 08 Mar 2023 07:58:53 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-2023-03-07' of https://gitlab.com/thuth/qemu:
  pc-bios/s390-ccw: Update s390-ccw.img with the list-directed IPL fix
  pc-bios: Add support for List-Directed IPL from ECKD DASD
  docs/config: Set the "kvm" accelerator via "[accel]" section
  tests/qtest/readconfig: Test docs/config/ich9-ehci-uhci.cfg
  tests/qtest/readconfig: Rework test_object_rng_resp into a generic function
  gitlab-ci.d/crossbuilds: Drop the 32-bit arm system emulation jobs
  docs/about/deprecated: Deprecate 32-bit arm hosts for system emulation
  gitlab-ci.d/crossbuilds: Drop the i386 system emulation job
  docs/about/deprecated: Deprecate 32-bit x86 hosts for system emulation
  include/hw/i386: Clean up includes in x86.h
  test: Check vnc enable before compiling vnc test
  Hexagon (meson.build): define min bison version
  docs/about/build-platforms: Refine the distro support policy

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
16 months agoMerge tag 'xenfv-2' of git://git.infradead.org/users/dwmw2/qemu into staging
Peter Maydell [Thu, 9 Mar 2023 13:22:05 +0000 (13:22 +0000)]
Merge tag 'xenfv-2' of git://git.infradead.org/users/dwmw2/qemu into staging

Enable PV backends with Xen/KVM emulation

This is phase 2, following on from the basic platform support which was
already merged.

 • Add a simple single-tenant internal XenStore implementation
 • Indirect Xen gnttab/evtchn/foreignmem/xenstore through operations table
 • Provide emulated back ends for Xen operations
 • Header cleanups to allow PV back ends to build without Xen itself
 • Enable PV back ends in emulated mode
 • Documentation update

Tested-by: Paul Durrant <paul@xen.org>
... on real Xen (master branch, 4.18) with a Debian guest.

# -----BEGIN PGP SIGNATURE-----
#
# iQJGBAABCgAwFiEEMUsIrNDeSBEzpfKGm+mA/QrAFUQFAmQHu3wSHGR3bXdAYW1h
# em9uLmNvLnVrAAoJEJvpgP0KwBVE5LYP/0VodDsQdP7Z4L+/IzgBSgEec7qmyQFB
# KlBZS/PmvCZKb0DHLI3GhXIyzD+/fnLtGSRl0rYObnKP7im+MpEDGmn97f6nIITk
# AzkdsVhNEBQFXCkLgQ9y8kTrTmsod9O4sqn0+naa2TX4FPcRN0MaNmpuLEubvaRS
# +JuyHmwy9ZeeAnsU31uJ0nx4F1hW9IDaatNoDeFcFnKCXQp36rtdZUViMowUJvwu
# Q+Xyg6dybusznaoiXd485tTPrTt+FK/wEARse3q2gRh9QblLu0r5BFb0rOfhYCTQ
# jw+5lBsOX+UlffmB9IDakRpVe4RKhvvRQSkRvYkPCshsqud9zMGhaquKg1vKBgca
# I31XSN0LCcon/ahHGtmVAxyZUpWdEnfzO1TbTNpz9oacROklgVgEYdw5Vwca71VD
# SURl6uCt9Jb9WmsR4twus4i4qDjQIDOtOF0hcxpl7HGktkxlGxUVI4qVLXARtVCS
# OTB6N0LlhJ2woj2wYK5BRTiOj03T2MkJEWaYhDdIrQREKWe2Sn4xTOH5kGbQQnOr
# km93odjBZFRHsAUnzXHXW3+yHjMefH7KrHePbmvsO4foGF77bBxosuC2ehFfvNJ0
# VM/H04NDtPYCBwdAr545PSN/q+WzEPQaquLZ0UuTBuPpMMOYd+Ff8YvQWJPyCM18
# 1mq9v6Xe9RQZ
# =JGLX
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 07 Mar 2023 22:32:28 GMT
# gpg:                using RSA key 314B08ACD0DE481133A5F2869BE980FD0AC01544
# gpg:                issuer "dwmw@amazon.co.uk"
# gpg: Good signature from "David Woodhouse <dwmw@amazon.co.uk>" [unknown]
# gpg:                 aka "David Woodhouse <dwmw@amazon.com>" [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: 314B 08AC D0DE 4811 33A5  F286 9BE9 80FD 0AC0 1544

* tag 'xenfv-2' of git://git.infradead.org/users/dwmw2/qemu: (27 commits)
  docs: Update Xen-on-KVM documentation for PV disk support
  MAINTAINERS: Add entry for Xen on KVM emulation
  i386/xen: Initialize Xen backends from pc_basic_device_init() for emulation
  hw/xen: Implement soft reset for emulated gnttab
  hw/xen: Map guest XENSTORE_PFN grant in emulated Xenstore
  hw/xen: Add emulated implementation of XenStore operations
  hw/xen: Add emulated implementation of grant table operations
  hw/xen: Hook up emulated implementation for event channel operations
  hw/xen: Only advertise ring-page-order for xen-block if gnttab supports it
  hw/xen: Avoid crash when backend watch fires too early
  hw/xen: Build PV backend drivers for CONFIG_XEN_BUS
  hw/xen: Rename xen_common.h to xen_native.h
  hw/xen: Use XEN_PAGE_SIZE in PV backend drivers
  hw/xen: Move xenstore_store_pv_console_info to xen_console.c
  hw/xen: Add xenstore operations to allow redirection to internal emulation
  hw/xen: Add foreignmem operations to allow redirection to internal emulation
  hw/xen: Pass grant ref to gnttab unmap operation
  hw/xen: Add gnttab operations to allow redirection to internal emulation
  hw/xen: Add evtchn operations to allow redirection to internal emulation
  hw/xen: Create initial XenStore nodes
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agoMerge tag 'pull-aspeed-20230307' of https://github.com/legoater/qemu into staging
Peter Maydell [Thu, 9 Mar 2023 10:23:05 +0000 (10:23 +0000)]
Merge tag 'pull-aspeed-20230307' of https://github.com/legoater/qemu into staging

aspeed queue:

* Small adjustments for the newest Meta machines
* blk_pread_nonzeroes() fix required for pflash and m25p80 devices
* Improve error reporting on file size for m25p80 devices

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEEoPZlSPBIlev+awtgUaNDx8/77KEFAmQHXi8ACgkQUaNDx8/7
# 7KHB7w//SQzcTNAUgMc6zbe3bULS9YU7csg0zX896DwQ9Ub4wtbnIps0/JwOsx6E
# ss52YbbRVlHKcbahX6WW8JHtc6Wepcmnp4vC5dXdqgRryLTK5GgBu/GKcBb6qnHZ
# uXwJpf7IwygOXQFg9Wy4zl3ipmrjgzMy6ZEbc3oqQUto+8E7HVMJiMXnrT5Y0yLt
# Y76PywBxkYWcd522lWAXziXFxHaEJGMGrDCI/op4TuNAk0QLSRHp+x851JBF0Uyx
# Lkp1cEjgADoH1FUfvo7aCNtAWGP4IFDprEupI3idPqjZ8KnnwPk2cnBMGeKJ5lWQ
# XfSTwG5B3hYwGM0uYwHzHXUsq5/JGa0iDhSvjZVD8rf/w9BTek/GjypbJpL2XW6b
# 4QaGGKGFWWMHbAaNVP6oEGqmkoC3QRKoQuaP6epaFyuSjNRWXkOhf5S5V10akjSc
# xis0Z8gIpyLrtFT2TFc5BN9+z+PLgyaYyh7wZ2pFy1i6SOgh4FIFBOzQJDb05mmf
# aFMD+6FOGx0wo8OXfy5EQyElWhVSYio+Hx9DaDP3d3cd/twHJx8nZ8QlXh2ThRd8
# jNLdFReJBImCJHSmYzbc/tGDOjg6+TWkZOzdjqTra08UwisK4XyOKj8E+sodWr0y
# e/hI9nmIoarLCytMN1zElQFvj8xjpZ2lZOxr2EiX5ugYg0iuJhU=
# =Xs26
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 07 Mar 2023 15:54:23 GMT
# gpg:                using RSA key A0F66548F04895EBFE6B0B6051A343C7CFFBECA1
# gpg: Good signature from "Cédric Le Goater <clg@kaod.org>" [undefined]
# 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: A0F6 6548 F048 95EB FE6B  0B60 51A3 43C7 CFFB ECA1

* tag 'pull-aspeed-20230307' of https://github.com/legoater/qemu:
  hw/arm/aspeed: Modified BMC FRU byte data in yosemitev2
  hw/arm/aspeed: Added TMP421 type sensor's support in tiogapass
  hw/arm/aspeed: Added TMP421 type sensor's support in yosemitev2
  pflash: Fix blk_pread_nonzeroes()
  m25p80: Improve error when the backend file size does not match the device

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agoMerge tag 'mips-misc-20230308' of https://github.com/philmd/qemu into staging
Peter Maydell [Thu, 9 Mar 2023 10:22:49 +0000 (10:22 +0000)]
Merge tag 'mips-misc-20230308' of https://github.com/philmd/qemu into staging

MIPS (and few misc) patches

- MIPS
  - Remove obsolete "mips" board from target-mips.rst
  - Fix JALS32/J32/SWM32 instructions for microMIPS
  - Fix CP0.Config7.WII handling on pre-R6 cores

- HW
  - Revert "Remove intermediate IRQ forwarder" commits
  - Implement legacy LTIM Edge/Level Bank Select in Intel 8259 INTC
  - Improve PCI IRQ routing in VT82C686 / Pegasos II
  - Basic implementation of VIA AC97 audio playback
  - Implement 'resume on connection status change' in USB OHCI

- UI
  - Override windowDidResignKey

- memory
  - Dump HPA and access type in HMP 'info ramblock'

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEE+qvnXhKRciHc/Wuy4+MsLN6twN4FAmQHzH8ACgkQ4+MsLN6t
# wN4RbxAAtbsbJKHikHevCiE9Fi0E4HHI4su00m3anImogfU3CKIHA7WHgeUVCdVH
# aLoFKjvE3d45FA4YwMs13wIo89tv6btn1y8C9iy+yjktdABPUr8OJphuaDxU+yNf
# XhPm4WsS0tEg5KzzTHF7qotJGw7Zd0Aca8oezFVBSL8b73lqiJiWBEouFTK9j5Oi
# s1uvAOPG3oxSlT7IIbnLRIEff4hi5FZh+LxRDgE3ChcOyY2W/DhrpdVIazv9Cpki
# facQ0ozMG4uuZ+HvviuTkK1vLX1+BcS0P1fnDPkXEPAxqx9jdqsMqWHbbnseQPN3
# xcVhw+GOZ67x8qAWIBKDG7nfChbcXgJ2sHxQmvb2XlxnOYw1oO5aRlrnn7ZPEWYA
# NbqUHB8G88wDcrms+Y+xCfO8idnr7Kzf4/1R1J1+5yEjg8Y1wu4t0asqZvhXA2HL
# F1yhHDCRY8w9pLYmPFGBrINBCoosiDn61g+JTngPffq1zJttmWjSLe9BYOF8Kiw+
# 4YjkCx43wK6RLTZNhU8g7iuqoYbHCQcXx5ZnGEadk+UJcfGrLnOrQbtAhvysS2wo
# msyum0FNWhnx/IZ6bmhmbFC8F/hASgyiV9CDwU2oOZ2oAkRiFXYBfXruUAt+6uLT
# UnAihAEsyUjyg5YNb4r8ZNkdeCPN6p3s2xY8OHphqu717K6uJXw=
# =D/0W
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 07 Mar 2023 23:45:03 GMT
# gpg:                using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.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: FAAB E75E 1291 7221 DCFD  6BB2 E3E3 2C2C DEAD C0DE

* tag 'mips-misc-20230308' of https://github.com/philmd/qemu:
  log: Remove unneeded new line
  memory: Dump HPA and access type of ramblocks
  ui/cocoa: Override windowDidResignKey
  hw/usb/ohci: Implement resume on connection status change
  hw/audio/via-ac97: Basic implementation of audio playback
  hw/usb/vt82c686-uhci-pci: Use PCI IRQ routing
  hw/ppc/pegasos2: Fix PCI interrupt routing
  hw/isa/vt82c686: Implement PCI IRQ routing
  hw/intc/i8259: Implement legacy LTIM Edge/Level Bank Select
  hw/display/sm501: Add debug property to control pixman usage
  Revert "hw/isa/vt82c686: Remove intermediate IRQ forwarder"
  Revert "hw/isa/i82378: Remove intermediate IRQ forwarder"
  hw/mips/itu: Pass SAAR using QOM link property
  hw/mips: Declare all length properties as unsigned
  target/mips: Set correct CP0.Config[4, 5] values for M14K(c)
  target/mips: Implement CP0.Config7.WII bit support
  target/mips: Fix SWM32 handling for microMIPS
  target/mips: Fix JALS32/J32 instruction handling for microMIPS
  target/mips: Replace [g_]assert(0) -> g_assert_not_reached()
  docs/system: Remove "mips" board from target-mips.rst

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agopc-bios/s390-ccw: Update s390-ccw.img with the list-directed IPL fix
Thomas Huth [Mon, 6 Mar 2023 11:05:10 +0000 (12:05 +0100)]
pc-bios/s390-ccw: Update s390-ccw.img with the list-directed IPL fix

This update includes the list-directed IPL fix from Jared Rossi.

Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agopc-bios: Add support for List-Directed IPL from ECKD DASD
Jared Rossi [Tue, 21 Feb 2023 17:45:48 +0000 (12:45 -0500)]
pc-bios: Add support for List-Directed IPL from ECKD DASD

Check for a List Directed IPL Boot Record, which would supersede the CCW type
entries.  If the record is valid, proceed to use the new style pointers
and perform LD-IPL. Each block pointer is interpreted as either an LD-IPL
pointer or a legacy CCW pointer depending on the type of IPL initiated.

In either case CCW- or LD-IPL is transparent to the user and will boot the same
image regardless of which set of pointers is used. Because the interactive boot
menu is only written with the old style pointers, the menu will be disabled for
List Directed IPL from ECKD DASD.

If the LD-IPL fails, retry the IPL using the CCW type pointers.

If no LD-IPL boot record is found, simply perform CCW type IPL as usual.

Signed-off-by: Jared Rossi <jrossi@linux.ibm.com>
Message-Id: <20230221174548.1866861-2-jrossi@linux.ibm.com>
[thuth: Drop some superfluous parantheses]
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agodocs/config: Set the "kvm" accelerator via "[accel]" section
Thomas Huth [Tue, 28 Feb 2023 21:15:30 +0000 (22:15 +0100)]
docs/config: Set the "kvm" accelerator via "[accel]" section

Configuring the accelerator should nowadays be done via the "-accel"
command line parameter, and thus via the "[accel]" section in config
files. We also need this change for the upcoming qtests that will
use these config files, since the qtests are already using "-accel"
for setting the "qtest" accelerator and QEMU does not like mixing
"-accel ..." and "-machine accel=...".

Message-Id: <20230228211533.201837-4-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agotests/qtest/readconfig: Test docs/config/ich9-ehci-uhci.cfg
Thomas Huth [Tue, 28 Feb 2023 21:15:29 +0000 (22:15 +0100)]
tests/qtest/readconfig: Test docs/config/ich9-ehci-uhci.cfg

We've got some sample config files in docs/config/ but no means
of regression checking them. Thus let's test them in our readconfig
qtest, starting with ich9-ehci-uhci.cfg. Note: To enable the test
to read the config files from the build folder, we have to install
a symlink for docs/config in the build directory.

Message-Id: <20230228211533.201837-3-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agotests/qtest/readconfig: Rework test_object_rng_resp into a generic function
Thomas Huth [Tue, 28 Feb 2023 21:15:28 +0000 (22:15 +0100)]
tests/qtest/readconfig: Rework test_object_rng_resp into a generic function

test_object_rng_resp() can be reworked quite easily to allow
testing for arbitrary objects in the qom-list response.

Message-Id: <20230228211533.201837-2-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agogitlab-ci.d/crossbuilds: Drop the 32-bit arm system emulation jobs
Thomas Huth [Mon, 6 Mar 2023 08:46:58 +0000 (09:46 +0100)]
gitlab-ci.d/crossbuilds: Drop the 32-bit arm system emulation jobs

Hardly anybody still uses 32-bit arm environments for running QEMU,
so let's stop wasting our scarce CI minutes with these jobs.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
Message-Id: <20230306084658.29709-6-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agodocs/about/deprecated: Deprecate 32-bit arm hosts for system emulation
Thomas Huth [Mon, 6 Mar 2023 08:46:57 +0000 (09:46 +0100)]
docs/about/deprecated: Deprecate 32-bit arm hosts for system emulation

For running QEMU in system emulation mode, the user needs a rather
strong host system, i.e. not only an embedded low-frequency controller.
All recent beefy arm host machines should support 64-bit now, it's
unlikely that anybody is still seriously using QEMU on a 32-bit arm
CPU, so we deprecate the 32-bit arm hosts here to finally save use
some time and precious CI minutes.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
Message-Id: <20230306084658.29709-5-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agolog: Remove unneeded new line
BALATON Zoltan [Thu, 19 Jan 2023 21:40:33 +0000 (22:40 +0100)]
log: Remove unneeded new line

The help text of the -d plugin option has a new line at the end which
is not needed as one is added automatically. Fixing it removes the
unexpected empty line in -d help output.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230119214033.600FB74645F@zero.eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agomemory: Dump HPA and access type of ramblocks
Ted Chen [Mon, 5 Dec 2022 12:07:12 +0000 (20:07 +0800)]
memory: Dump HPA and access type of ramblocks

It's convenient to dump HVA and RW/RO status of a ramblock in "info ramblock"
for debug purpose.

Before:
            Offset               Used              Total
0x0000000000000000 0x0000000400000000 0x0000000400000000

After:
            Offset               Used              Total                HVA  RO
0x0000000000000000 0x0000000400000000 0x0000000400000000 0x00007f12ebe00000  rw

Signed-off-by: Ted Chen <znscnchen@gmail.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20221205120712.269013-1-znscnchen@gmail.com>
[PMD: Add uintptr_t cast for 32-bit hosts]
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agoui/cocoa: Override windowDidResignKey
Akihiko Odaki [Tue, 28 Feb 2023 07:09:46 +0000 (16:09 +0900)]
ui/cocoa: Override windowDidResignKey

This fixes pressed keys being stuck when the deck is clicked and the
window loses focus.

In the past, Gustavo Noronha Silva also had a patch to fix this issue
though it only ungrabs mouse and does not release keys, and depends on
another out-of-tree patch:
https://github.com/akihikodaki/qemu/pull/3/commits/e906a80147b1dc6d4f31b6a08064ef9871a2b76c

Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-Id: <20230228070946.12370-1-akihiko.odaki@daynix.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/usb/ohci: Implement resume on connection status change
BALATON Zoltan [Mon, 20 Feb 2023 18:15:09 +0000 (19:15 +0100)]
hw/usb/ohci: Implement resume on connection status change

If certain bit is set remote wake up should change state from
suspended to resume and generate interrupt. There was a todo comment
for this, implement that by moving existing resume logic to a function
and call that.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Message-Id: <35c4d4ccf2f73e6a87cdbd28fb6a1b33de72ed74.1676916640.git.balaton@eik.bme.hu>
[PMD: Have ohci_resume() return a boolean]
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/audio/via-ac97: Basic implementation of audio playback
BALATON Zoltan [Sun, 23 Jan 2022 20:40:42 +0000 (21:40 +0100)]
hw/audio/via-ac97: Basic implementation of audio playback

Add basic implementation of the AC'97 sound part used in VIA south
bridge chips. Not all features of the device is emulated, only one
playback channel is supported for now but this is enough to get sound
output from some guests using this device on pegasos2.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Volker Rümelin <vr_qemu@t-online.de>
Tested-by: Rene Engel <ReneEngel80@emailn.de>
Message-Id: <63b99410895312f40e7be479f581da0805e605a1.1678188711.git.balaton@eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/usb/vt82c686-uhci-pci: Use PCI IRQ routing
Bernhard Beschow [Sat, 25 Feb 2023 17:08:59 +0000 (18:08 +0100)]
hw/usb/vt82c686-uhci-pci: Use PCI IRQ routing

According to the PCI specification, PCI_INTERRUPT_LINE shall have no
effect on hardware operations. Now that the VIA south bridges implement
the internal PCI interrupt router let's be more conformant to the PCI
specification.

Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Tested-by: Rene Engel <ReneEngel80@emailn.de>
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Message-Id: <9fb86a74d16db65e3aafbb154238d55e123053eb.1678188711.git.balaton@eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/ppc/pegasos2: Fix PCI interrupt routing
BALATON Zoltan [Thu, 16 Feb 2023 20:27:32 +0000 (21:27 +0100)]
hw/ppc/pegasos2: Fix PCI interrupt routing

According to the PegasosII schematics the PCI interrupt lines are
connected to both the gpp pins of the Mv64361 north bridge and the
PINT pins of the VT8231 south bridge so guests can get interrupts from
either of these. So far we only had the MV64361 connections which
worked for on board devices but for additional PCI devices (such as
network or sound card added with -device) guest OSes expect interrupt
from the ISA IRQ 9 where the firmware routes these PCI interrupts in
VT8231 ISA bridge. After the previous patches we can now model this
and also remove the board specific connection from mv64361. Also
configure routing of these lines when using Virtual Open Firmware to
match board firmware for guests that expect this.

This fixes PCI interrupts on pegasos2 under Linux, MorphOS and AmigaOS.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Tested-by: Rene Engel <ReneEngel80@emailn.de>
Message-Id: <520ff9e6eeef600ee14a4116c0c7b11940cc499c.1678188711.git.balaton@eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/isa/vt82c686: Implement PCI IRQ routing
BALATON Zoltan [Thu, 16 Feb 2023 20:21:35 +0000 (21:21 +0100)]
hw/isa/vt82c686: Implement PCI IRQ routing

The real VIA south bridges implement a PCI IRQ router which is configured
by the BIOS or the OS. In order to respect these configurations, QEMU
needs to implement it as well. The real chip may allow routing IRQs from
internal functions independently of PCI interrupts but since guests
usually configute it to a single shared interrupt we don't model that
here for simplicity.

Note: The implementation was taken from piix4_set_irq() in hw/isa/piix4.

Suggested-by: Bernhard Beschow <shentey@gmail.com>
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Bernhard Beschow <shentey@gmail.com>
Tested-by: Rene Engel <ReneEngel80@emailn.de>
Message-Id: <fbb016c7d0e19093335c237e15f5f6c62c4393b4.1678188711.git.balaton@eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/intc/i8259: Implement legacy LTIM Edge/Level Bank Select
David Woodhouse [Thu, 2 Mar 2023 09:06:26 +0000 (10:06 +0100)]
hw/intc/i8259: Implement legacy LTIM Edge/Level Bank Select

Back in the mists of time, before EISA came along and required per-pin
level control in the ELCR register, the i8259 had a single chip-wide
level-mode control in bit 3 of ICW1.

Even in the PIIX3 datasheet from 1996 this is documented as 'This bit is
disabled', but apparently MorphOS is using it in the version of the
i8259 which is in the Pegasos2 board as part of the VT8231 chipset.

It's easy enough to implement, and I think it's harmless enough to do so
unconditionally.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
[balaton: updated commit message as asked by author]
Tested-by: BALATON Zoltan <balaton@eik.bme.hu>
Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-Id: <3f09b2dd109d19851d786047ad5c2ff459c90cd7.1678188711.git.balaton@eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/display/sm501: Add debug property to control pixman usage
BALATON Zoltan [Sat, 25 Feb 2023 22:46:16 +0000 (23:46 +0100)]
hw/display/sm501: Add debug property to control pixman usage

Add a property to allow disabling pixman and always use the fallbacks
for different operations which is useful for testing different drawing
methods or debugging pixman related issues.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Tested-by: Rene Engel <ReneEngel80@emailn.de>
Message-Id: <61768ffaefa71b65a657d1365823bd43c7ee9354.1678188711.git.balaton@eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agoRevert "hw/isa/vt82c686: Remove intermediate IRQ forwarder"
BALATON Zoltan [Wed, 1 Mar 2023 00:17:08 +0000 (01:17 +0100)]
Revert "hw/isa/vt82c686: Remove intermediate IRQ forwarder"

To be 'usable', QDev objects (which are QOM objects) must be
1/ initialized (at this point their properties can be modified), then
2/ realized (properties are consumed).
Some devices (objects) might depend on other devices. When creating
the 'QOM composition tree', parent objects can't be 'realized' until
all their children are. We might also have circular dependencies.
A common circular dependency occurs with IRQs. Device (A) has an
output IRQ wired to device (B), and device (B) has one to device (A).
When (A) is realized and connects its IRQ to an unrealized (B), the
IRQ handler on (B) is not yet created. QEMU pass IRQ between objects
as pointer. When (A) poll (B)'s IRQ, it is NULL. Later (B) is realized
and its IRQ pointers are populated, but (A) keeps a reference to a
NULL pointer.
A common pattern to bypass this circular limitation is to use 'proxy'
objects. Proxy (P) is created (and realized) before (A) and (B). Then
(A) and (B) can be created in different order, it doesn't matter: (P)
pointers are already populated.

Commit bb98e0f59cde ("hw/isa/vt82c686: Remove intermediate IRQ
forwarder") neglected the QOM/QDev circular dependency issue, and
removed the 'proxy' between the southbridge, its PCI functions and the
interrupt controller, resulting in PCI functions wiring output IRQs to
'NULL', leading to guest failures (IRQ never delivered) [1] [2].

Since we are entering feature freeze, it is safer to revert the
offending patch until we figure a way to strengthen our APIs.

[1] https://lore.kernel.org/qemu-devel/928a8552-ab62-9e6c-a492-d6453e338b9d@redhat.com/
[2] https://lore.kernel.org/qemu-devel/cover.1677628524.git.balaton@eik.bme.hu/

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Tested-by: Rene Engel <ReneEngel80@emailn.de>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Message-Id: <cdfb3c5a42e505450f6803124f27856434c5b298.1677628524.git.balaton@eik.bme.hu>
[PMD: Reworded description]
Inspired-by: Bernhard Beschow <shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agoRevert "hw/isa/i82378: Remove intermediate IRQ forwarder"
Philippe Mathieu-Daudé [Tue, 7 Mar 2023 11:10:08 +0000 (12:10 +0100)]
Revert "hw/isa/i82378: Remove intermediate IRQ forwarder"

To be 'usable', QDev objects (which are QOM objects) must be
1/ initialized (at this point their properties can be modified), then
2/ realized (properties are consumed).
Some devices (objects) might depend on other devices. When creating
the 'QOM composition tree', parent objects can't be 'realized' until
all their children are. We might also have circular dependencies.
A common circular dependency occurs with IRQs. Device (A) has an
output IRQ wired to device (B), and device (B) has one to device (A).
When (A) is realized and connects its IRQ to an unrealized (B), the
IRQ handler on (B) is not yet created. QEMU pass IRQ between objects
as pointer. When (A) poll (B)'s IRQ, it is NULL. Later (B) is realized
and its IRQ pointers are populated, but (A) keeps a reference to a
NULL pointer.
A common pattern to bypass this circular limitation is to use 'proxy'
objects. Proxy (P) is created (and realized) before (A) and (B). Then
(A) and (B) can be created in different order, it doesn't matter: (P)
pointers are already populated.

Commit cef2e7148e ("hw/isa/i82378: Remove intermediate IRQ forwarder")
neglected the QOM/QDev circular dependency issue, and removed the
'proxy' between the southbridge, its PCI functions and the interrupt
controller, resulting in PCI functions wiring output IRQs to
'NULL', leading to guest failures (IRQ never delivered) [1] [2].

Since we are entering feature freeze, it is safer to revert the
offending patch until we figure a way to strengthen our APIs.

[1] https://lore.kernel.org/qemu-devel/928a8552-ab62-9e6c-a492-d6453e338b9d@redhat.com/
[2] https://lore.kernel.org/qemu-devel/cover.1677628524.git.balaton@eik.bme.hu/

This reverts commit cef2e7148e32d61338de0220619d308bf42af770.

Reported-by: Thomas Huth <thuth@redhat.com>
Inspired-by: Bernhard Beschow <shentey@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agohw/mips/itu: Pass SAAR using QOM link property
Philippe Mathieu-Daudé [Thu, 2 Feb 2023 13:22:42 +0000 (14:22 +0100)]
hw/mips/itu: Pass SAAR using QOM link property

QOM objects shouldn't access each other internals fields
except using the QOM API.

mips_cps_realize() instantiates a TYPE_MIPS_ITU object, and
directly sets the 'saar' pointer:

   if (saar_present) {
       s->itu.saar = &env->CP0_SAAR;
   }

In order to avoid that, pass the MIPS_CPU object via a QOM
link property, and set the 'saar' pointer in mips_itu_realize().

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Tested-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Reviewed-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Message-Id: <20230203113650.78146-10-philmd@linaro.org>

17 months agohw/mips: Declare all length properties as unsigned
Philippe Mathieu-Daudé [Thu, 2 Feb 2023 09:46:45 +0000 (10:46 +0100)]
hw/mips: Declare all length properties as unsigned

Some length properties are signed, other unsigned:

  hw/mips/cps.c:183:    DEFINE_PROP_UINT32("num-vp", MIPSCPSState, num_vp, 1),
  hw/mips/cps.c:184:    DEFINE_PROP_UINT32("num-irq", MIPSCPSState, num_irq, 256),
  hw/misc/mips_cmgcr.c:215:    DEFINE_PROP_INT32("num-vp", MIPSGCRState, num_vps, 1),
  hw/misc/mips_cpc.c:167:    DEFINE_PROP_UINT32("num-vp", MIPSCPCState, num_vp, 0x1),
  hw/misc/mips_itu.c:552:    DEFINE_PROP_INT32("num-fifo", MIPSITUState, num_fifo,
  hw/misc/mips_itu.c:554:    DEFINE_PROP_INT32("num-semaphores", MIPSITUState,

Since negative values are not used (the minimum is '0'),
unify by declaring all properties as unsigned.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230203113650.78146-9-philmd@linaro.org>

17 months agotarget/mips: Set correct CP0.Config[4, 5] values for M14K(c)
Marcin Nowakowski [Thu, 16 Feb 2023 05:17:17 +0000 (06:17 +0100)]
target/mips: Set correct CP0.Config[4, 5] values for M14K(c)

Signed-off-by: Marcin Nowakowski <marcin.nowakowski@fungible.com>
Suggested-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230216051717.3911212-5-marcin.nowakowski@fungible.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agotarget/mips: Implement CP0.Config7.WII bit support
Marcin Nowakowski [Thu, 16 Feb 2023 05:17:16 +0000 (06:17 +0100)]
target/mips: Implement CP0.Config7.WII bit support

Some pre-release 6 cores use CP0.Config7.WII bit to indicate that a
disabled interrupt should wake up a sleeping CPU.
Enable this bit by default for M14K(c) and P5600. There are potentially
other cores that support this feature, but I do not have a complete
list.

Signed-off-by: Marcin Nowakowski <marcin.nowakowski@fungible.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230216051717.3911212-4-marcin.nowakowski@fungible.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agotarget/mips: Fix SWM32 handling for microMIPS
Marcin Nowakowski [Thu, 16 Feb 2023 05:17:15 +0000 (06:17 +0100)]
target/mips: Fix SWM32 handling for microMIPS

SWM32 should store a sequence of 32-bit words from the GPRs, but it was
incorrectly coded to store 16-bit words only. As a result, an LWM32 that
usually follows would restore invalid register values.

Fixes: 7dd547e5ab ("target/mips: Use cpu_*_mmuidx_ra instead of
MMU_MODE*_SUFFIX")

Signed-off-by: Marcin Nowakowski <marcin.nowakowski@fungible.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230216051717.3911212-3-marcin.nowakowski@fungible.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agotarget/mips: Fix JALS32/J32 instruction handling for microMIPS
Marcin Nowakowski [Thu, 16 Feb 2023 05:17:14 +0000 (06:17 +0100)]
target/mips: Fix JALS32/J32 instruction handling for microMIPS

microMIPS J & JAL instructions perform a jump in a 128MB region and 5
top bits of the address need to be preserved. This is different behavior
compared to standard mips systems, where the jump is executed within a
256MB region.
Note that microMIPS32 instruction set documentation appears to have
inconsistent information regarding JALX32 instruction - it is written in
the doc that:

"To execute a procedure call within the current 256 MB-aligned region
(...)
The low 26 bits of the target address is the target field shifted left
2 bits."

But the target address is already 26 bits. Moreover, the operation
description indicates that 28 bits are copied, so the statement about
use of 26 bits is _most likely_ incorrect and the corresponding code
remains the same as for standard mips instruction set.

Signed-off-by: Marcin Nowakowski <marcin.nowakowski@fungible.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230216051717.3911212-2-marcin.nowakowski@fungible.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agotarget/mips: Replace [g_]assert(0) -> g_assert_not_reached()
Philippe Mathieu-Daudé [Tue, 21 Feb 2023 22:47:55 +0000 (23:47 +0100)]
target/mips: Replace [g_]assert(0) -> g_assert_not_reached()

In order to avoid warnings such commit c0a6665c3c ("target/i386:
Remove compilation errors when -Werror=maybe-uninitialized"),
replace all assert(0) and g_assert(0) by g_assert_not_reached().

Remove any code following g_assert_not_reached().

See previous commit for rationale.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230221232520.14480-4-philmd@linaro.org>

17 months agodocs/system: Remove "mips" board from target-mips.rst
Jiaxun Yang [Thu, 2 Feb 2023 13:21:36 +0000 (13:21 +0000)]
docs/system: Remove "mips" board from target-mips.rst

This board had been removed long ago in commit f169413c27
("hw/mips: Remove the 'r4k' machine")

Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230202132138.30945-2-jiaxun.yang@flygoat.com>
[PMD: Mention commit f169413c27]
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
17 months agodocs: Update Xen-on-KVM documentation for PV disk support
David Woodhouse [Tue, 7 Mar 2023 10:54:31 +0000 (10:54 +0000)]
docs: Update Xen-on-KVM documentation for PV disk support

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agoMAINTAINERS: Add entry for Xen on KVM emulation
David Woodhouse [Tue, 7 Mar 2023 09:59:59 +0000 (09:59 +0000)]
MAINTAINERS: Add entry for Xen on KVM emulation

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agoi386/xen: Initialize Xen backends from pc_basic_device_init() for emulation
David Woodhouse [Wed, 15 Feb 2023 15:10:00 +0000 (16:10 +0100)]
i386/xen: Initialize Xen backends from pc_basic_device_init() for emulation

Now that all the work is done to enable the PV backends to work without
actual Xen, instantiate the bus from pc_basic_device_init() for emulated
mode.

This allows us finally to launch an emulated Xen guest with PV disk.

   qemu-system-x86_64 -serial mon:stdio -M q35 -cpu host -display none \
     -m 1G -smp 2 -accel kvm,xen-version=0x4000a,kernel-irqchip=split \
     -kernel bzImage -append "console=ttyS0 root=/dev/xvda1" \
     -drive file=/var/lib/libvirt/images/fedora28.qcow2,if=none,id=disk \
     -device xen-disk,drive=disk,vdev=xvda

If we use -M pc instead of q35, we can even add an IDE disk and boot a
guest image normally through grub. But q35 gives us AHCI and that isn't
unplugged by the Xen magic, so the guests ends up seeing "both" disks.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Implement soft reset for emulated gnttab
David Woodhouse [Tue, 10 Jan 2023 01:09:04 +0000 (01:09 +0000)]
hw/xen: Implement soft reset for emulated gnttab

This is only part of it; we will also need to get the PV back end drivers
to tear down their own mappings (or do it for them, but they kind of need
to stop using the pointers too).

Some more work on the actual PV back ends and xen-bus code is going to be
needed to really make soft reset and migration fully functional, and this
part is the basis for that.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Map guest XENSTORE_PFN grant in emulated Xenstore
David Woodhouse [Sat, 7 Jan 2023 13:54:07 +0000 (13:54 +0000)]
hw/xen: Map guest XENSTORE_PFN grant in emulated Xenstore

We don't actually access the guest's page through the grant, because
this isn't real Xen, and we can just use the page we gave it in the
first place. Map the grant anyway, mostly for cosmetic purposes so it
*looks* like it's in use in the guest-visible grant table.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add emulated implementation of XenStore operations
David Woodhouse [Thu, 19 Jan 2023 00:04:31 +0000 (00:04 +0000)]
hw/xen: Add emulated implementation of XenStore operations

Now that we have an internal implementation of XenStore, we can populate
the xenstore_backend_ops to allow PV backends to talk to it.

Watches can't be processed with immediate callbacks because that would
call back into XenBus code recursively. Defer them to a QEMUBH to be run
as appropriate from the main loop. We use a QEMUBH per XS handle, and it
walks all the watches (there shouldn't be many per handle) to fire any
which have pending events. We *could* have done it differently but this
allows us to use the same struct watch_event as we have for the guest
side, and keeps things relatively simple.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add emulated implementation of grant table operations
David Woodhouse [Fri, 6 Jan 2023 09:59:28 +0000 (09:59 +0000)]
hw/xen: Add emulated implementation of grant table operations

This is limited to mapping a single grant at a time, because under Xen the
pages are mapped *contiguously* into qemu's address space, and that's very
hard to do when those pages actually come from anonymous mappings in qemu
in the first place.

Eventually perhaps we can look at using shared mappings of actual objects
for system RAM, and then we can make new mappings of the same backing
store (be it deleted files, shmem, whatever). But for now let's stick to
a page at a time.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Hook up emulated implementation for event channel operations
David Woodhouse [Sun, 1 Jan 2023 23:49:25 +0000 (23:49 +0000)]
hw/xen: Hook up emulated implementation for event channel operations

We provided the backend-facing evtchn functions very early on as part of
the core Xen platform support, since things like timers and xenstore need
to use them.

By what may or may not be an astonishing coincidence, those functions
just *happen* all to have exactly the right function prototypes to slot
into the evtchn_backend_ops table and be called by the PV backends.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Only advertise ring-page-order for xen-block if gnttab supports it
David Woodhouse [Mon, 30 Jan 2023 16:27:07 +0000 (17:27 +0100)]
hw/xen: Only advertise ring-page-order for xen-block if gnttab supports it

Whem emulating Xen, multi-page grants are distinctly non-trivial and we
have elected not to support them for the time being. Don't advertise
them to the guest.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Avoid crash when backend watch fires too early
Paul Durrant [Mon, 30 Jan 2023 14:35:28 +0000 (14:35 +0000)]
hw/xen: Avoid crash when backend watch fires too early

The xen-block code ends up calling aio_poll() through blkconf_geometry(),
which means we see watch events during the indirect call to
xendev_class->realize() in xen_device_realize(). Unfortunately this call
is made before populating the initial frontend and backend device nodes
in xenstore and hence xen_block_frontend_changed() (which is called from
a watch event) fails to read the frontend's 'state' node, and hence
believes the device is being torn down. This in-turn sets the backend
state to XenbusStateClosed and causes the device to be deleted before it
is fully set up, leading to the crash.
By simply moving the call to xendev_class->realize() after the initial
xenstore nodes are populated, this sorry state of affairs is avoided.

Reported-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Build PV backend drivers for CONFIG_XEN_BUS
David Woodhouse [Mon, 2 Jan 2023 01:26:04 +0000 (01:26 +0000)]
hw/xen: Build PV backend drivers for CONFIG_XEN_BUS

Now that we have the redirectable Xen backend operations we can build the
PV backends even without the Xen libraries.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Rename xen_common.h to xen_native.h
David Woodhouse [Mon, 2 Jan 2023 00:39:13 +0000 (00:39 +0000)]
hw/xen: Rename xen_common.h to xen_native.h

This header is now only for native Xen code, not PV backends that may be
used in Xen emulation. Since the toolstack libraries may depend on the
specific version of Xen headers that they pull in (and will set the
__XEN_TOOLS__ macro to enable internal definitions that they depend on),
the rule is that xen_native.h (and thus the toolstack library headers)
must be included *before* any of the headers in include/hw/xen/interface.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Use XEN_PAGE_SIZE in PV backend drivers
David Woodhouse [Sat, 7 Jan 2023 16:47:43 +0000 (16:47 +0000)]
hw/xen: Use XEN_PAGE_SIZE in PV backend drivers

XC_PAGE_SIZE comes from the actual Xen libraries, while XEN_PAGE_SIZE is
provided by QEMU itself in xen_backend_ops.h. For backends which may be
built for emulation mode, use the latter.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Move xenstore_store_pv_console_info to xen_console.c
David Woodhouse [Sat, 7 Jan 2023 16:17:51 +0000 (16:17 +0000)]
hw/xen: Move xenstore_store_pv_console_info to xen_console.c

There's no need for this to be in the Xen accel code, and as we want to
use the Xen console support with KVM-emulated Xen we'll want to have a
platform-agnostic version of it. Make it use GString to build up the
path while we're at it.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add xenstore operations to allow redirection to internal emulation
Paul Durrant [Mon, 2 Jan 2023 11:05:16 +0000 (11:05 +0000)]
hw/xen: Add xenstore operations to allow redirection to internal emulation

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add foreignmem operations to allow redirection to internal emulation
David Woodhouse [Mon, 2 Jan 2023 01:13:46 +0000 (01:13 +0000)]
hw/xen: Add foreignmem operations to allow redirection to internal emulation

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Pass grant ref to gnttab unmap operation
David Woodhouse [Tue, 10 Jan 2023 00:03:49 +0000 (00:03 +0000)]
hw/xen: Pass grant ref to gnttab unmap operation

The previous commit introduced redirectable gnttab operations fairly
much like-for-like, with the exception of the extra arguments to the
->open() call which were always NULL/0 anyway.

This *changes* the arguments to the ->unmap() operation to include the
original ref# that was mapped. Under real Xen it isn't necessary; all we
need to do from QEMU is munmap(), then the kernel will release the grant,
and Xen does the tracking/refcounting for the guest.

When we have emulated grant tables though, we need to do all that for
ourselves. So let's have the back ends keep track of what they mapped
and pass it in to the ->unmap() method for us.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add gnttab operations to allow redirection to internal emulation
David Woodhouse [Sun, 1 Jan 2023 21:31:37 +0000 (21:31 +0000)]
hw/xen: Add gnttab operations to allow redirection to internal emulation

Move the existing code using libxengnttab to xen-operations.c and allow
the operations to be redirected so that we can add emulation of grant
table mapping for backend drivers.

In emulation, mapping more than one grant ref to be virtually contiguous
would be fairly difficult. The best way to do it might be to make the
ram_block mappings actually backed by a file (shmem or a deleted file,
perhaps) so that we can have multiple *shared* mappings of it. But that
would be fairly intrusive.

Making the backend drivers cope with page *lists* instead of expecting
the mapping to be contiguous is also non-trivial, since some structures
would actually *cross* page boundaries (e.g. the 32-bit blkif responses
which are 12 bytes).

So for now, we'll support only single-page mappings in emulation. Add a
XEN_GNTTAB_OP_FEATURE_MAP_MULTIPLE flag to indicate that the native Xen
implementation *does* support multi-page maps, and a helper function to
query it.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add evtchn operations to allow redirection to internal emulation
David Woodhouse [Sun, 1 Jan 2023 17:54:41 +0000 (17:54 +0000)]
hw/xen: Add evtchn operations to allow redirection to internal emulation

The existing implementation calling into the real libxenevtchn moves to
a new file hw/xen/xen-operations.c, and is called via a function table
which in a subsequent commit will also be able to invoke the emulated
event channel support.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Create initial XenStore nodes
Paul Durrant [Tue, 24 Jan 2023 09:34:06 +0000 (09:34 +0000)]
hw/xen: Create initial XenStore nodes

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Implement core serialize/deserialize methods for xenstore_impl
David Woodhouse [Tue, 31 Jan 2023 15:00:54 +0000 (15:00 +0000)]
hw/xen: Implement core serialize/deserialize methods for xenstore_impl

This implements the basic migration support in the back end, with unit
tests that give additional confidence in the node-counting already in
the tree.

However, the existing PV back ends like xen-disk don't support migration
yet. They will reset the ring and fail to continue where they left off.
We will fix that in future, but not in time for the 8.0 release.

Since there's also an open question of whether we want to serialize the
full XenStore or only the guest-owned nodes in /local/domain/${domid},
for now just mark the XenStore device as unmigratable.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Implement XenStore permissions
Paul Durrant [Mon, 23 Jan 2023 16:21:16 +0000 (16:21 +0000)]
hw/xen: Implement XenStore permissions

Store perms as a GList of strings, check permissions.

Signed-off-by: Paul Durrant <pdurrant@amazon.com>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Watches on XenStore transactions
David Woodhouse [Sun, 22 Jan 2023 22:59:49 +0000 (22:59 +0000)]
hw/xen: Watches on XenStore transactions

Firing watches on the nodes that still exist is relatively easy; just
walk the tree and look at the nodes with refcount of one.

Firing watches on *deleted* nodes is more fun. We add 'modified_in_tx'
and 'deleted_in_tx' flags to each node. Nodes with those flags cannot
be shared, as they will always be unique to the transaction in which
they were created.

When xs_node_walk would need to *create* a node as scaffolding and it
encounters a deleted_in_tx node, it can resurrect it simply by clearing
its deleted_in_tx flag. If that node originally had any *data*, they're
gone, and the modified_in_tx flag will have been set when it was first
deleted.

We then attempt to send appropriate watches when the transaction is
committed, properly delete the deleted_in_tx nodes, and remove the
modified_in_tx flag from the others.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Implement XenStore transactions
David Woodhouse [Sun, 22 Jan 2023 22:05:37 +0000 (22:05 +0000)]
hw/xen: Implement XenStore transactions

Given that the whole thing supported copy on write from the beginning,
transactions end up being fairly simple. On starting a transaction, just
take a ref of the existing root; swap it back in on a successful commit.

The main tree has a transaction ID too, and we keep a record of the last
transaction ID given out. if the main tree is ever modified when it isn't
the latest, it gets a new transaction ID.

A commit can only succeed if the main tree hasn't moved on since it was
forked. Strictly speaking, the XenStore protocol allows a transaction to
succeed as long as nothing *it* read or wrote has changed in the interim,
but no implementations do that; *any* change is sufficient to abort a
transaction.

This does not yet fire watches on the changed nodes on a commit. That bit
is more fun and will come in a follow-on commit.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Implement XenStore watches
David Woodhouse [Sun, 22 Jan 2023 18:38:23 +0000 (18:38 +0000)]
hw/xen: Implement XenStore watches

Starts out fairly simple: a hash table of watches based on the path.

Except there can be multiple watches on the same path, so the watch ends
up being a simple linked list, and the head of that list is in the hash
table. Which makes removal a bit of a PITA but it's not so bad; we just
special-case "I had to remove the head of the list and now I have to
replace it in / remove it from the hash table". And if we don't remove
the head, it's a simple linked-list operation.

We do need to fire watches on *deleted* nodes, so instead of just a simple
xs_node_unref() on the topmost victim, we need to recurse down and fire
watches on them all.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add basic XenStore tree walk and write/read/directory support
David Woodhouse [Fri, 20 Jan 2023 01:36:38 +0000 (01:36 +0000)]
hw/xen: Add basic XenStore tree walk and write/read/directory support

This is a fairly simple implementation of a copy-on-write tree.

The node walk function starts off at the root, with 'inplace == true'.
If it ever encounters a node with a refcount greater than one (including
the root node), then that node is shared with other trees, and cannot
be modified in place, so the inplace flag is cleared and we copy on
write from there on down.

Xenstore write has 'mkdir -p' semantics and will create the intermediate
nodes if they don't already exist, so in that case we flip the inplace
flag back to true as we populate the newly-created nodes.

We put a copy of the absolute path into the buffer in the struct walk_op,
with *two* NUL terminators at the end. As xs_node_walk() goes down the
tree, it replaces the next '/' separator with a NUL so that it can use
the 'child name' in place. The next recursion down then puts the '/'
back and repeats the exercise for the next path element... if it doesn't
hit that *second* NUL termination which indicates the true end of the
path.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agohw/xen: Add xenstore wire implementation and implementation stubs
David Woodhouse [Wed, 18 Jan 2023 18:55:47 +0000 (18:55 +0000)]
hw/xen: Add xenstore wire implementation and implementation stubs

This implements the basic wire protocol for the XenStore commands, punting
all the actual implementation to xs_impl_* functions which all just return
errors for now.

Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
17 months agoMerge tag 'for-upstream-mb' of https://gitlab.com/bonzini/qemu into staging
Peter Maydell [Tue, 7 Mar 2023 17:02:06 +0000 (17:02 +0000)]
Merge tag 'for-upstream-mb' of https://gitlab.com/bonzini/qemu into staging

* Fix missing memory barriers
* Fix comments about memory ordering

# -----BEGIN PGP SIGNATURE-----
#
# iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmQHIqoUHHBib256aW5p
# QHJlZGhhdC5jb20ACgkQv/vSX3jHroPYBwgArUaS0KGrBM1XmRUUpXnJokmA37n8
# ft477na+XW+p9VYi27B0R01P8j+AkCrAO0Ir1MLG7axjn5KiRMnbf2uBgqasEREv
# repJEXsqISoxA6vvAvnehKHAI9zu8b7frRc/30b6EOrrZpn0JKePSNRTyBu2seGO
# NFDXPVA2Wom+xXaNSEGt0dmoJ6AzEVIZKhUIwyvUWOC7MXuuIkRWn9/nySUdvEt0
# RIFPPk7JCjnEc32vb4Xnq/Ncsy20tMIM1hlDxMOVNq3brjeSCzS0PPPSjE/X5OtW
# Yn5YS0nCyD7wjP2dkXI4I1lUPxUUx6LvMz1aGbJCfyjSX41mNES/agoGgA==
# =KEUo
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 07 Mar 2023 11:40:26 GMT
# gpg:                using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg:                issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>" [full]
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4  E2F7 7E15 100C CD36 69B1
#      Subkey fingerprint: F133 3857 4B66 2389 866C  7682 BFFB D25F 78C7 AE83

* tag 'for-upstream-mb' of https://gitlab.com/bonzini/qemu:
  async: clarify usage of barriers in the polling case
  async: update documentation of the memory barriers
  physmem: add missing memory barrier
  qemu-coroutine-lock: add smp_mb__after_rmw()
  aio-wait: switch to smp_mb__after_rmw()
  edu: add smp_mb__after_rmw()
  qemu-thread-win32: cleanup, fix, document QemuEvent
  qemu-thread-posix: cleanup, fix, document QemuEvent
  qatomic: add smp_mb__before/after_rmw()

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agohw/arm/aspeed: Modified BMC FRU byte data in yosemitev2
Karthikeyan Pasupathi [Tue, 7 Mar 2023 15:53:18 +0000 (16:53 +0100)]
hw/arm/aspeed: Modified BMC FRU byte data in yosemitev2

Modified BMC FRU data in yosemite v2 platform.

Tested: Tested and Verified in yosemitev2 platform.

Fixes: 34f73a81e6 ("hw/arm/aspeed: Adding new machine Yosemitev2 in QEMU")
Signed-off-by: Karthikeyan Pasupathi <pkarthikeyan1509@gmail.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Message-Id: <20230307104833.3587947-1-pkarthikeyan1509@gmail.com>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
17 months agohw/arm/aspeed: Added TMP421 type sensor's support in tiogapass
Karthikeyan Pasupathi [Tue, 7 Mar 2023 15:53:18 +0000 (16:53 +0100)]
hw/arm/aspeed: Added TMP421 type sensor's support in tiogapass

Added TMP421 type sensor support in tiogapass platform.

Tested: Tested and verified in tiogapass platform.

Signed-off-by: Karthikeyan Pasupathi <pkarthikeyan1509@gmail.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Message-Id: <20230307103334.3586755-1-pkarthikeyan1509@gmail.com>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
17 months agohw/arm/aspeed: Added TMP421 type sensor's support in yosemitev2
Karthikeyan Pasupathi [Tue, 7 Mar 2023 15:53:18 +0000 (16:53 +0100)]
hw/arm/aspeed: Added TMP421 type sensor's support in yosemitev2

Added TMP421 type support in yosemite v2 platform.

Tested: Tested and verified in yosemite V2 platform.

Signed-off-by: Karthikeyan Pasupathi <pkarthikeyan1509@gmail.com>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Message-Id: <20230307095239.3583613-1-pkarthikeyan1509@gmail.com>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
17 months agopflash: Fix blk_pread_nonzeroes()
Kevin Wolf [Tue, 7 Mar 2023 15:53:18 +0000 (16:53 +0100)]
pflash: Fix blk_pread_nonzeroes()

Commit a4b15a8b introduced a new function blk_pread_nonzeroes(). Instead
of reading directly from the root node of the BlockBackend, it reads
from its 'file' child node. This can happen to mostly work for raw
images (as long as the 'raw' format driver is in use, but not actually
doing anything), but it breaks everything else.

Fix it to read from the root node instead.

Fixes: a4b15a8b9ef25b44fa92a4825312622600c1f37c
Reported-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20230307140230.59158-1-kwolf@redhat.com>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
17 months agom25p80: Improve error when the backend file size does not match the device
Cédric Le Goater [Tue, 7 Mar 2023 15:53:18 +0000 (16:53 +0100)]
m25p80: Improve error when the backend file size does not match the device

Currently, when a block backend is attached to a m25p80 device and the
associated file size does not match the flash model, QEMU complains
with the error message "failed to read the initial flash content".
This is confusing for the user.

Instead, use helper blk_check_size_and_read_all() introduced by commit
06f1521795 ("pflash: Require backend size to match device, improve
errors").

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Peter Delevoryas <peter@pjd.dev>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20221115151000.2080833-1-clg@kaod.org>
Signed-off-by: Cédric Le Goater <clg@kaod.org>
17 months agogitlab-ci.d/crossbuilds: Drop the i386 system emulation job
Thomas Huth [Mon, 6 Mar 2023 08:46:56 +0000 (09:46 +0100)]
gitlab-ci.d/crossbuilds: Drop the i386 system emulation job

Hardly anybody still uses 32-bit x86 environments for running QEMU with
full system emulation, so let's stop wasting our scarce CI minutes with
this job.

(There are still the 32-bit MinGW and TCI jobs around for having
some compile test coverage on 32-bit)

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
Message-Id: <20230306084658.29709-4-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agodocs/about/deprecated: Deprecate 32-bit x86 hosts for system emulation
Thomas Huth [Mon, 6 Mar 2023 08:46:54 +0000 (09:46 +0100)]
docs/about/deprecated: Deprecate 32-bit x86 hosts for system emulation

Hardly anybody still uses 32-bit x86 hosts today, so we should start
deprecating them to stop wasting our time and CI minutes here.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Wilfred Mallawa <wilfred.mallawa@wdc.com>
Message-Id: <20230306084658.29709-2-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agoinclude/hw/i386: Clean up includes in x86.h
Thomas Huth [Wed, 8 Feb 2023 14:34:48 +0000 (15:34 +0100)]
include/hw/i386: Clean up includes in x86.h

nmi.h and notify.h are not needed here, drop them to speed up
the compiling a little bit.

Message-Id: <20230210111438.1114600-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agotest: Check vnc enable before compiling vnc test
Juan Quintela [Wed, 1 Mar 2023 10:44:50 +0000 (11:44 +0100)]
test: Check vnc enable before compiling vnc test

Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20230301104450.1017-1-quintela@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agoHexagon (meson.build): define min bison version
Matheus Tavares Bernardino [Tue, 7 Feb 2023 14:52:31 +0000 (11:52 -0300)]
Hexagon (meson.build): define min bison version

Hexagon's idef-parser machinery uses some bison features that are not
available at older versions. The most preeminent example (as it can
be used as a sentinel) is "%define parse.error verbose". This was
introduced in version 3.0 of the tool, which is able to compile
qemu-hexagon just fine. However, compilation fails with the previous
minor bison release, v2.7. So let's assert the minimum version at
meson.build to give a more comprehensive error message for those trying
to compile QEMU.

[1]: https://www.gnu.org/software/bison/manual/html_node/_0025define-Summary.html#index-_0025define-parse_002eerror

Signed-off-by: Matheus Tavares Bernardino <quic_mathbern@quicinc.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alessandro Di Federico <ale@rev.ng>
Reviewed-by: Taylor Simpson <tsimpson@quicinc.com>
Message-Id: <a6763f9f7b89ea310ab86f9a2b311a05254a1acd.1675779233.git.quic_mathbern@quicinc.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agodocs/about/build-platforms: Refine the distro support policy
Thomas Huth [Thu, 23 Feb 2023 19:32:57 +0000 (20:32 +0100)]
docs/about/build-platforms: Refine the distro support policy

For long-term distributions that release a new version only very
seldom, we limit the support to five years after the initial release.
Otherwise, we might need to support distros like openSUSE 15 for
up to 7 or even more years in total due to our "two more years
after the next major release" rule, which is just way too much to
handle in a project like QEMU that only has limited human resources.

Message-Id: <20230223193257.1068205-1-thuth@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
17 months agoMerge tag 'pull-riscv-to-apply-20230306' of https://gitlab.com/palmer-dabbelt/qemu...
Peter Maydell [Tue, 7 Mar 2023 12:53:00 +0000 (12:53 +0000)]
Merge tag 'pull-riscv-to-apply-20230306' of https://gitlab.com/palmer-dabbelt/qemu into staging

Sixth RISC-V PR for 8.0

* Support for the Zicbiom, ZCicboz, and Zicbop extensions.
* OpenSBI has been updated to version 1.2, see
  <https://github.com/riscv-software-src/opensbi/releases/tag/v1.2> for
  the release notes.
* Support for setting the virtual address width (ie, sv39/sv48/sv57) on
  the command line.
* Support for ACPI on RISC-V.

# -----BEGIN PGP SIGNATURE-----
#
# iQJHBAABCAAxFiEEKzw3R0RoQ7JKlDp6LhMZ81+7GIkFAmQGYGgTHHBhbG1lckBk
# YWJiZWx0LmNvbQAKCRAuExnzX7sYidmyEAC6FEMbbFM5D++qR6w6xM6hXgzcrev6
# s1kyRRNVa45uSA78ti/Zi0hsDLNf7ZsNPndF0OIkkO5iAE0OVm3LU7tV1TqKcT82
# Dd9VXxe93zEmfnuJazHrMa54SXPhhnNdWHtKlZ6vBfZpbxgx0FFs50xkCsrM5LQZ
# hYHxQUqPWQTvF2MdDHrxCuLcdKl+Wg3ysCcgRh2d049KUBrIu6vNaHC2+AGRjCbj
# BkrGCkB82fTmVJjzAcVWQxLoAV12pCbJS4og1GtP8hA7WevtB39tbPin9siBKRZp
# QBeiIsg0nebkpmZGrb+xWVwlIBNe9yYwJa0KmveQk8v7L5RIzjM1mtDL91VrVljC
# KC2tfT570m0Iq2NoFMb3wd/kESHFzVDM/g+XYqRd4KSoiCNP/RbqYNQBwbMc31Tr
# E27xfA1D8w2vem0Rk20x3KgPf1Z5OmGXjq6YObTpnAzG8cZlA37qKBP+ortt5aHX
# GZSg3CAwknHHVajd4aaegkPsHxm1tRvoTfh38MwkPSNxaA9GD0nz0k9xaYDmeZ2L
# olfanNsaQEwcVUId31+7sAENg1TZU0fnj879/nxkMUCazVTdL8/mz+IoTTx0QCST
# 3+9ATWcyJUlmjbDKIs7kr1L+wJdvvHEJggPAbbPI8ekpXaLZvUYOT6ObzYKNAmwY
# wELQBn8QKXcLVA==
# =5gAt
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 06 Mar 2023 21:51:36 GMT
# gpg:                using RSA key 2B3C3747446843B24A943A7A2E1319F35FBB1889
# gpg:                issuer "palmer@dabbelt.com"
# gpg: Good signature from "Palmer Dabbelt <palmer@dabbelt.com>" [unknown]
# gpg:                 aka "Palmer Dabbelt <palmerdabbelt@google.com>" [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: 00CE 76D1 8349 60DF CE88  6DF8 EF4C A150 2CCB AB41
#      Subkey fingerprint: 2B3C 3747 4468 43B2 4A94  3A7A 2E13 19F3 5FBB 1889

* tag 'pull-riscv-to-apply-20230306' of https://gitlab.com/palmer-dabbelt/qemu: (22 commits)
  MAINTAINERS: Add entry for RISC-V ACPI
  hw/riscv/virt.c: Initialize the ACPI tables
  hw/riscv/virt: virt-acpi-build.c: Add RHCT Table
  hw/riscv/virt: virt-acpi-build.c: Add RINTC in MADT
  hw/riscv/virt: Enable basic ACPI infrastructure
  hw/riscv/virt: Add memmap pointer to RiscVVirtState
  hw/riscv/virt: Add a switch to disable ACPI
  hw/riscv/virt: Add OEM_ID and OEM_TABLE_ID fields
  riscv: Correctly set the device-tree entry 'mmu-type'
  riscv: Introduce satp mode hw capabilities
  riscv: Allow user to set the satp mode
  riscv: Change type of valid_vm_1_10_[32|64] to bool
  riscv: Pass Object to register_cpu_props instead of DeviceState
  roms/opensbi: Upgrade from v1.1 to v1.2
  gitlab/opensbi: Move to docker:stable
  hw: intc: Use cpu_by_arch_id to fetch CPU state
  target/riscv: cpu: Implement get_arch_id callback
  disas/riscv Fix ctzw disassemble
  hw/riscv/virt.c: add cbo[mz]-block-size fdt properties
  target/riscv: add Zicbop cbo.prefetch{i, r, m} placeholder
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agoasync: clarify usage of barriers in the polling case
Paolo Bonzini [Mon, 6 Mar 2023 09:43:52 +0000 (10:43 +0100)]
async: clarify usage of barriers in the polling case

Explain that aio_context_notifier_poll() relies on
aio_notify_accept() to catch all the memory writes that were
done before ctx->notified was set to true.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoasync: update documentation of the memory barriers
Paolo Bonzini [Mon, 6 Mar 2023 09:15:06 +0000 (10:15 +0100)]
async: update documentation of the memory barriers

Ever since commit 8c6b0356b539 ("util/async: make bh_aio_poll() O(1)",
2020-02-22), synchronization between qemu_bh_schedule() and aio_bh_poll()
is happening when the bottom half is enqueued in the bh_list; not
when the flags are set.  Update the documentation to match.

Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agophysmem: add missing memory barrier
Paolo Bonzini [Fri, 3 Mar 2023 13:36:32 +0000 (14:36 +0100)]
physmem: add missing memory barrier

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoqemu-coroutine-lock: add smp_mb__after_rmw()
Paolo Bonzini [Fri, 3 Mar 2023 09:52:59 +0000 (10:52 +0100)]
qemu-coroutine-lock: add smp_mb__after_rmw()

mutex->from_push and mutex->handoff in qemu-coroutine-lock implement
the familiar pattern:

   write a                                  write b
   smp_mb()                                 smp_mb()
   read b                                   read a

The memory barrier is required by the C memory model even after a
SEQ_CST read-modify-write operation such as QSLIST_INSERT_HEAD_ATOMIC.
Add it and avoid the unclear qatomic_mb_read() operation.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoaio-wait: switch to smp_mb__after_rmw()
Paolo Bonzini [Fri, 3 Mar 2023 10:03:52 +0000 (11:03 +0100)]
aio-wait: switch to smp_mb__after_rmw()

The barrier comes after an atomic increment, so it is enough to use
smp_mb__after_rmw(); this avoids a double barrier on x86 systems.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoedu: add smp_mb__after_rmw()
Paolo Bonzini [Thu, 2 Mar 2023 10:16:13 +0000 (11:16 +0100)]
edu: add smp_mb__after_rmw()

Ensure ordering between clearing the COMPUTING flag and checking
IRQFACT, and between setting the IRQFACT flag and checking
COMPUTING.  This ensures that no wakeups are lost.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoqemu-thread-win32: cleanup, fix, document QemuEvent
Paolo Bonzini [Thu, 2 Mar 2023 10:22:50 +0000 (11:22 +0100)]
qemu-thread-win32: cleanup, fix, document QemuEvent

QemuEvent is currently broken on ARM due to missing memory barriers
after qatomic_*().  Apart from adding the memory barrier, a closer look
reveals some unpaired memory barriers that are not really needed and
complicated the functions unnecessarily.  Also, it is relying on
a memory barrier in ResetEvent(); the barrier _ought_ to be there
but there is really no documentation about it, so make it explicit.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoqemu-thread-posix: cleanup, fix, document QemuEvent
Paolo Bonzini [Thu, 2 Mar 2023 10:19:52 +0000 (11:19 +0100)]
qemu-thread-posix: cleanup, fix, document QemuEvent

QemuEvent is currently broken on ARM due to missing memory barriers
after qatomic_*().  Apart from adding the memory barrier, a closer look
reveals some unpaired memory barriers too.  Document more clearly what
is going on.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoqatomic: add smp_mb__before/after_rmw()
Paolo Bonzini [Thu, 2 Mar 2023 10:10:56 +0000 (11:10 +0100)]
qatomic: add smp_mb__before/after_rmw()

On ARM, seqcst loads and stores (which QEMU does not use) are compiled
respectively as LDAR and STLR instructions.  Even though LDAR is
also used for load-acquire operations, it also waits for all STLRs to
leave the store buffer.  Thus, LDAR and STLR alone are load-acquire
and store-release operations, but LDAR also provides store-against-load
ordering as long as the previous store is a STLR.

Compare this to ARMv7, where store-release is DMB+STR and load-acquire
is LDR+DMB, but an additional DMB is needed between store-seqcst and
load-seqcst (e.g. DMB+STR+DMB+LDR+DMB); or with x86, where MOV provides
load-acquire and store-release semantics and the two can be reordered.

Likewise, on ARM sequentially consistent read-modify-write operations only
need to use LDAXR and STLXR respectively for the load and the store, while
on x86 they need to use the stronger LOCK prefix.

In a strange twist of events, however, the _stronger_ semantics
of the ARM instructions can end up causing bugs on ARM, not on x86.
The problems occur when seqcst atomics are mixed with relaxed atomics.

QEMU's atomics try to bridge the Linux API (that most of the developers
are familiar with) and the C11 API, and the two have a substantial
difference:

- in Linux, strongly-ordered atomics such as atomic_add_return() affect
  the global ordering of _all_ memory operations, including for example
  READ_ONCE()/WRITE_ONCE()

- in C11, sequentially consistent atomics (except for seq-cst fences)
  only affect the ordering of sequentially consistent operations.
  In particular, since relaxed loads are done with LDR on ARM, they are
  not ordered against seqcst stores (which are done with STLR).

QEMU implements high-level synchronization primitives with the idea that
the primitives contain the necessary memory barriers, and the callers can
use relaxed atomics (qatomic_read/qatomic_set) or even regular accesses.
This is very much incompatible with the C11 view that seqcst accesses
are only ordered against other seqcst accesses, and requires using seqcst
fences as in the following example:

   qatomic_set(&y, 1);            qatomic_set(&x, 1);
   smp_mb();                      smp_mb();
   ... qatomic_read(&x) ...       ... qatomic_read(&y) ...

When a qatomic_*() read-modify write operation is used instead of one
or both stores, developers that are more familiar with the Linux API may
be tempted to omit the smp_mb(), which will work on x86 but not on ARM.

This nasty difference between Linux and C11 read-modify-write operations
has already caused issues in util/async.c and more are being found.
Provide something similar to Linux smp_mb__before/after_atomic(); this
has the double function of documenting clearly why there is a memory
barrier, and avoiding a double barrier on x86 and s390x systems.

The new macro can already be put to use in qatomic_mb_set().

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
17 months agoMerge tag 'pull-target-arm-20230306' of https://git.linaro.org/people/pmaydell/qemu...
Peter Maydell [Tue, 7 Mar 2023 09:58:43 +0000 (09:58 +0000)]
Merge tag 'pull-target-arm-20230306' of https://git.linaro.org/people/pmaydell/qemu-arm into staging

target-arm queue:
 * allwinner-h3: Fix I2C controller model for Sun6i SoCs
 * allwinner-h3: Add missing i2c controllers
 * Expose M-profile system registers to gdbstub
 * Expose pauth information to gdbstub
 * Support direct boot for Linux/arm64 EFI zboot images
 * Fix incorrect stage 2 MMU setup validation

# -----BEGIN PGP SIGNATURE-----
#
# iQJNBAABCAA3FiEE4aXFk81BneKOgxXPPCUl7RQ2DN4FAmQGB+wZHHBldGVyLm1h
# eWRlbGxAbGluYXJvLm9yZwAKCRA8JSXtFDYM3gdQEACVfgbs77mxbOb6u8yWHKGZ
# tVnQr9KZMv2lmwt5H3ROJPXznchrIIAwdMeRgKnbI+lC5jTq9L+Q8RJch3t/EbAd
# f0VMyiPe3DzCbCrAR9cW6EWzbYnEVo3Ioj4k7qjxK6u1BIKhXz99DLYd1KRdTxnx
# BAYmcl857Uir1q2FrBVMZ/ItCLbk4ejn+YaDIawNue2/s1oGa+we473x9rosCFvp
# L9bzT3R46e0o+Mfkn1OYRmgCmURTalWPpWAxyOUFR9YbrzXleLgAKEB3o3PPcvls
# u26uxztyRMqje1q06VjUzwaLw7zN9XPhmir+NXX7KXp2/x9PZjApOpPtt0kl+6qe
# FbByKfl24O9w/OKewsJw+udCBYdYrRPm6tWv2D71iAwjBUzBJgNGe5VPRdPFtPDn
# uSRO65o34w1nPzRpAheUciZueiabYrVmIgVltFxj0JlrKGfgiYHPLVyU0Uu0K/A7
# F2kUEQIzIcWdo+c8SlvlWOEA2ojVd/KoLVLgndqr40Tk5pbc65TRS08kkVVl4cMT
# jUGscl7Dyxe+yo8+nHdycAJpnKYDllJOh2JbGv3r2FqCy5FMuIqW4hHeuUxwpE+O
# nxm7lzjnaVHSAFHdzhk9x4E4uH/GTcdWzX1EsmpgGqe5oejLJOrCINb+Dj44+Y8h
# 8aGRvE7kxMs11upxc7BcAw==
# =KIMt
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 06 Mar 2023 15:34:04 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-20230306' of https://git.linaro.org/people/pmaydell/qemu-arm: (21 commits)
  hw: arm: allwinner-h3: Fix and complete H3 i2c devices
  hw: allwinner-i2c: Fix TWI_CNTR_INT_FLAG on SUN6i SoCs
  hw: arm: Support direct boot for Linux/arm64 EFI zboot images
  target/arm: Rewrite check_s2_mmu_setup
  target/arm: Diagnose incorrect usage of arm_is_secure subroutines
  target/arm: Stub arm_hcr_el2_eff for m-profile
  target/arm: Handle m-profile in arm_is_secure
  target/arm: Implement gdbstub m-profile systemreg and secext
  target/arm: Export arm_v7m_get_sp_ptr
  target/arm: Export arm_v7m_mrs_control
  target/arm: Implement gdbstub pauth extension
  target/arm: Create pauth_ptr_mask
  target/arm: Simplify iteration over bit widths
  target/arm: Add name argument to output_vector_union_type
  target/arm: Fix svep width in arm_gen_dynamic_svereg_xml
  target/arm: Hoist pred_width in arm_gen_dynamic_svereg_xml
  target/arm: Simplify register counting in arm_gen_dynamic_svereg_xml
  target/arm: Split out output_vector_union_type
  target/arm: Move arm_gen_dynamic_svereg_xml to gdbstub64.c
  target/arm: Unexport arm_gen_dynamic_sysreg_xml
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agoMerge tag 'nvme-next-pull-request' of https://gitlab.com/birkelund/qemu into staging
Peter Maydell [Tue, 7 Mar 2023 09:58:25 +0000 (09:58 +0000)]
Merge tag 'nvme-next-pull-request' of https://gitlab.com/birkelund/qemu into staging

hw/nvme updates

* basic support for directives
* simple support for endurance groups
* emulation of flexible data placement (tp4146)

# -----BEGIN PGP SIGNATURE-----
#
# iQEzBAABCgAdFiEEUigzqnXi3OaiR2bATeGvMW1PDekFAmQF+doACgkQTeGvMW1P
# DenIEgf+MLsRQ3kKUmsgVNnPuR69M0COfyaz0AnfX6YEIL9ukFJQPsmASfPmHof5
# tCYIFyKEpZt/givmzSI1jdpm0uX2MRwLGLYRdNhEPVjo+TfGda15x7DgpBEduqjq
# mChUS2wrmgP9TZne+kTAU28pUpU7hcfrt1RkDOO86W8oJmpBeIyGe6vikVhQppKW
# fAIKvhNfN3p5Kxq1fhE6I5YzKd2vvKtBvPpZp2uFe6LHXEcVV/FPcTx3Ph+um/o6
# ScmmxowT4Wqk4EgXh1ohephlxB89aWgwLNHLHcfte6UCU9x4eSmTC2T3pf7piBaE
# pGLpzPoYk6BAurwrMuxCxYgStl6SzQ==
# =CNSk
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 06 Mar 2023 14:34:02 GMT
# gpg:                using RSA key 522833AA75E2DCE6A24766C04DE1AF316D4F0DE9
# gpg: Good signature from "Klaus Jensen <its@irrelevant.dk>" [full]
# gpg:                 aka "Klaus Jensen <k.jensen@samsung.com>" [full]
# Primary key fingerprint: DDCA 4D9C 9EF9 31CC 3468  4272 63D5 6FC5 E55D A838
#      Subkey fingerprint: 5228 33AA 75E2 DCE6 A247  66C0 4DE1 AF31 6D4F 0DE9

* tag 'nvme-next-pull-request' of https://gitlab.com/birkelund/qemu:
  hw/nvme: flexible data placement emulation
  hw/nvme: basic directives support
  hw/nvme: add basic endurance group support
  hw/nvme: store a pointer to the NvmeSubsystem in the NvmeNamespace
  hw/nvme: move adjustment of data_units{read,written}

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agoMerge tag 'pull-xen-20230306' of https://xenbits.xen.org/git-http/people/aperard...
Peter Maydell [Tue, 7 Mar 2023 09:58:13 +0000 (09:58 +0000)]
Merge tag 'pull-xen-20230306' of https://xenbits.xen.org/git-http/people/aperard/qemu-dm into staging

Xen queue:

- fix for graphic passthrough with 'xenfv' machine
- fix uninitialized variable

# -----BEGIN PGP SIGNATURE-----
#
# iQEzBAABCgAdFiEE+AwAYwjiLP2KkueYDPVXL9f7Va8FAmQF8fgACgkQDPVXL9f7
# Va/nSQf/XVfmhe2W1ailKJxuvGeMLRW/tmY/dsNAZNXXBMjRYEaF4Eps51pjYdb7
# 6UUY/atT1fm9v/AYhxc+k8weIE/mxCDbaRStQUzHlrWPof1NsmEeYZ3NVdVq5w7s
# FmDCR+yiP2tcrBPhPD0aFBB7Lsayfy0P5qLFMMeeerlkZmk1O3fB04EKtus3YD1r
# hVSH+H8i5b8vg0d/5fGGrRzKalh5E2xGGUfz4ukp3+AYWNCl2m65K0JsX42+G79b
# Cg+OpeNp9CEXZSUvkfVoRxH9OJp6GpGZIHA9U3nvH31KR4OnDeCSZuCiPvoUuvZT
# Q0fd8eA4DRTEtt9gJ+ecQEpON5dcSA==
# =kvNV
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 06 Mar 2023 14:00:24 GMT
# gpg:                using RSA key F80C006308E22CFD8A92E7980CF5572FD7FB55AF
# gpg: Good signature from "Anthony PERARD <anthony.perard@gmail.com>" [marginal]
# gpg:                 aka "Anthony PERARD <anthony.perard@citrix.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg:          It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 5379 2F71 024C 600F 778A  7161 D8D5 7199 DF83 42C8
#      Subkey fingerprint: F80C 0063 08E2 2CFD 8A92  E798 0CF5 572F D7FB 55AF

* tag 'pull-xen-20230306' of https://xenbits.xen.org/git-http/people/aperard/qemu-dm:
  hw/xen/xen_pt: fix uninitialized variable
  xen/pt: reserve PCI slot 2 for Intel igd-passthru

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agoMAINTAINERS: Add entry for RISC-V ACPI
Sunil V L [Thu, 2 Mar 2023 09:12:12 +0000 (14:42 +0530)]
MAINTAINERS: Add entry for RISC-V ACPI

RISC-V ACPI related functionality for virt machine is added in
virt-acpi-build.c. Add the maintainer entry after moving the
ARM ACPI entry under the main ACPI entry.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-9-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw/riscv/virt.c: Initialize the ACPI tables
Sunil V L [Thu, 2 Mar 2023 09:12:11 +0000 (14:42 +0530)]
hw/riscv/virt.c: Initialize the ACPI tables

Initialize the ACPI tables if the acpi option is not
disabled.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-8-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw/riscv/virt: virt-acpi-build.c: Add RHCT Table
Sunil V L [Thu, 2 Mar 2023 09:12:10 +0000 (14:42 +0530)]
hw/riscv/virt: virt-acpi-build.c: Add RHCT Table

RISC-V ACPI platforms need to provide RISC-V Hart Capabilities
Table (RHCT). Add this to the ACPI tables.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-7-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw/riscv/virt: virt-acpi-build.c: Add RINTC in MADT
Sunil V L [Thu, 2 Mar 2023 09:12:09 +0000 (14:42 +0530)]
hw/riscv/virt: virt-acpi-build.c: Add RINTC in MADT

Add Multiple APIC Description Table (MADT) with the
RINTC structure for each cpu.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-6-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw/riscv/virt: Enable basic ACPI infrastructure
Sunil V L [Thu, 2 Mar 2023 09:12:08 +0000 (14:42 +0530)]
hw/riscv/virt: Enable basic ACPI infrastructure

Add basic ACPI infrastructure for RISC-V with below tables.
        1) DSDT with below basic objects
                - CPUs
                - fw_cfg
        2) FADT revision 6 with HW_REDUCED flag
        3) XSDT
        4) RSDP

Add this functionality in a new file virt-acpi-build.c and enable
building this infrastructure.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-5-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw/riscv/virt: Add memmap pointer to RiscVVirtState
Sunil V L [Thu, 2 Mar 2023 09:12:07 +0000 (14:42 +0530)]
hw/riscv/virt: Add memmap pointer to RiscVVirtState

memmap needs to be exported outside of virt.c so that
modules like acpi can use it. Hence, add a pointer field
in RiscVVirtState structure and initialize it with the
memorymap.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-4-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw/riscv/virt: Add a switch to disable ACPI
Sunil V L [Thu, 2 Mar 2023 09:12:06 +0000 (14:42 +0530)]
hw/riscv/virt: Add a switch to disable ACPI

ACPI will be enabled by default. Add a switch to turn off
for testing and debug purposes.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-3-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw/riscv/virt: Add OEM_ID and OEM_TABLE_ID fields
Sunil V L [Thu, 2 Mar 2023 09:12:05 +0000 (14:42 +0530)]
hw/riscv/virt: Add OEM_ID and OEM_TABLE_ID fields

ACPI needs OEM_ID and OEM_TABLE_ID for the machine. Add these fields
in the RISCVVirtState structure and initialize with default values.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Message-ID: <20230302091212.999767-2-sunilvl@ventanamicro.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agoriscv: Correctly set the device-tree entry 'mmu-type'
Alexandre Ghiti [Fri, 3 Mar 2023 13:12:52 +0000 (14:12 +0100)]
riscv: Correctly set the device-tree entry 'mmu-type'

The 'mmu-type' should reflect what the hardware is capable of so use the
new satp_mode field in RISCVCPUConfig to do that.

Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-ID: <20230303131252.892893-6-alexghiti@rivosinc.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agoriscv: Introduce satp mode hw capabilities
Alexandre Ghiti [Fri, 3 Mar 2023 13:12:51 +0000 (14:12 +0100)]
riscv: Introduce satp mode hw capabilities

Currently, the max satp mode is set with the only constraint that it must be
implemented in QEMU, i.e. set in valid_vm_1_10_[32|64].

But we actually need to add another level of constraint: what the hw is
actually capable of, because currently, a linux booting on a sifive-u54
boots in sv57 mode which is incompatible with the cpu's sv39 max
capability.

So add a new bitmap to RISCVSATPMap which contains this capability and
initialize it in every XXX_cpu_init.

Finally:
- valid_vm_1_10_[32|64] constrains which satp mode the CPU can use
- the CPU hw capabilities constrains what the user may select
- the user's selection then constrains what's available to the guest
  OS.

Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-ID: <20230303131252.892893-5-alexghiti@rivosinc.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agoriscv: Allow user to set the satp mode
Alexandre Ghiti [Fri, 3 Mar 2023 13:12:50 +0000 (14:12 +0100)]
riscv: Allow user to set the satp mode

RISC-V specifies multiple sizes for addressable memory and Linux probes for
the machine's support at startup via the satp CSR register (done in
csr.c:validate_vm).

As per the specification, sv64 must support sv57, which in turn must
support sv48...etc. So we can restrict machine support by simply setting the
"highest" supported mode and the bare mode is always supported.

You can set the satp mode using the new properties "sv32", "sv39", "sv48",
"sv57" and "sv64" as follows:
-cpu rv64,sv57=on  # Linux will boot using sv57 scheme
-cpu rv64,sv39=on  # Linux will boot using sv39 scheme
-cpu rv64,sv57=off # Linux will boot using sv48 scheme
-cpu rv64          # Linux will boot using sv57 scheme by default

We take the highest level set by the user:
-cpu rv64,sv48=on,sv57=on # Linux will boot using sv57 scheme

We make sure that invalid configurations are rejected:
-cpu rv64,sv39=off,sv48=on # sv39 must be supported if higher modes are
                           # enabled

We accept "redundant" configurations:
-cpu rv64,sv48=on,sv57=off # Linux will boot using sv48 scheme

And contradictory configurations:
-cpu rv64,sv48=on,sv48=off # Linux will boot using sv39 scheme

Co-Developed-by: Ludovic Henry <ludovic@rivosinc.com>
Signed-off-by: Ludovic Henry <ludovic@rivosinc.com>
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Acked-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-ID: <20230303131252.892893-4-alexghiti@rivosinc.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agoriscv: Change type of valid_vm_1_10_[32|64] to bool
Alexandre Ghiti [Fri, 3 Mar 2023 13:12:49 +0000 (14:12 +0100)]
riscv: Change type of valid_vm_1_10_[32|64] to bool

This array is actually used as a boolean so swap its current char type
to a boolean and at the same time, change the type of validate_vm to
bool since it returns valid_vm_1_10_[32|64].

Suggested-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Message-ID: <20230303131252.892893-3-alexghiti@rivosinc.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agoriscv: Pass Object to register_cpu_props instead of DeviceState
Alexandre Ghiti [Fri, 3 Mar 2023 13:12:48 +0000 (14:12 +0100)]
riscv: Pass Object to register_cpu_props instead of DeviceState

One can extract the DeviceState pointer from the Object pointer, so pass
the Object for future commits to access other fields of Object.

No functional changes intended.

Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Frank Chang <frank.chang@sifive.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Bin Meng <bmeng@tinylab.org>
Message-ID: <20230303131252.892893-2-alexghiti@rivosinc.com>
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
17 months agohw: arm: allwinner-h3: Fix and complete H3 i2c devices
qianfan Zhao [Mon, 20 Feb 2023 08:12:52 +0000 (16:12 +0800)]
hw: arm: allwinner-h3: Fix and complete H3 i2c devices

Allwinner h3 has 4 twi(i2c) devices named twi0, twi1, twi2 and r_twi.
The registers are compatible with TYPE_AW_I2C_SUN6I, write 1 to clear
control register's INT_FLAG bit.

Signed-off-by: qianfan Zhao <qianfanguijin@163.com>
Reviewed-by: Strahinja Jankovic <strahinja.p.jankovic@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
17 months agohw/nvme: flexible data placement emulation
Jesper Devantier [Mon, 20 Feb 2023 11:59:26 +0000 (12:59 +0100)]
hw/nvme: flexible data placement emulation

Add emulation of TP4146 ("Flexible Data Placement").

Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Jesper Devantier <j.devantier@samsung.com>
Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
17 months agohw/nvme: basic directives support
Gollu Appalanaidu [Mon, 20 Feb 2023 11:59:25 +0000 (12:59 +0100)]
hw/nvme: basic directives support

Add support for the Directive Send and Recv commands and the Identify
directive.

Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Gollu Appalanaidu <anaidu.gollu@samsung.com>
Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
17 months agohw/nvme: add basic endurance group support
Klaus Jensen [Mon, 20 Feb 2023 11:59:24 +0000 (12:59 +0100)]
hw/nvme: add basic endurance group support

Add the mandatory Endurance Group identify data structures and log
pages.

For now, all namespaces in a subsystem belongs to a single Endurance
Group.

Reviewed-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Klaus Jensen <k.jensen@samsung.com>