OSDN Git Service

qmiga/qemu.git
2 years agoconfigure: remove duplicate help messages
Paolo Bonzini [Fri, 13 May 2022 09:37:57 +0000 (11:37 +0200)]
configure: remove duplicate help messages

These messages are already emitted by scripts/meson-parse-buildoptions.sh.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoconfigure: remove another dead variable
Paolo Bonzini [Fri, 13 May 2022 16:06:37 +0000 (18:06 +0200)]
configure: remove another dead variable

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agobuild: remove useless dependency
Paolo Bonzini [Fri, 13 May 2022 16:06:29 +0000 (18:06 +0200)]
build: remove useless dependency

qemu-plugins.symbols is now processed in Meson.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agointroduce -audio as a replacement for -soundhw
Paolo Bonzini [Wed, 27 Apr 2022 10:27:46 +0000 (12:27 +0200)]
introduce -audio as a replacement for -soundhw

-audio is used like "-audio pa,model=sb16".  It is almost as simple as
-soundhw, but it reuses the -audiodev parsing machinery and attaches an
audiodev to the newly-created device.  The main 'feature' is that
it knows about adding the codec device for model=intel-hda, and adding
the audiodev to the codec device.

In the future, it could be extended to support default models or
builtin devices, just like -nic, or even a default backend.  For now,
keep it simple.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agosoundhw: move help handling to vl.c
Paolo Bonzini [Wed, 27 Apr 2022 11:15:23 +0000 (13:15 +0200)]
soundhw: move help handling to vl.c

This will allow processing "-audio model=help" even if the backend
part of the option is missing.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agosoundhw: unify initialization for ISA and PCI soundhw
Paolo Bonzini [Wed, 27 Apr 2022 10:06:59 +0000 (12:06 +0200)]
soundhw: unify initialization for ISA and PCI soundhw

Use qdev_new instead of distinguishing isa_create_simple/pci_create_simple.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agosoundhw: extract soundhw help to a separate function
Paolo Bonzini [Wed, 27 Apr 2022 10:26:48 +0000 (12:26 +0200)]
soundhw: extract soundhw help to a separate function

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agosoundhw: remove ability to create multiple soundcards
Paolo Bonzini [Wed, 27 Apr 2022 10:04:47 +0000 (12:04 +0200)]
soundhw: remove ability to create multiple soundcards

The usefulness of enabling a dozen soundcards is dubious.  Simplify the
code by allowing a single instance of -soundhw, with no support for
parsing either comma-separated values or 'soundhw all'.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agorng: make opened property read-only
Paolo Bonzini [Mon, 9 May 2022 09:46:03 +0000 (11:46 +0200)]
rng: make opened property read-only

The ``opened=on`` option in the command line or QMP ``object-add`` either had
no effect (if ``opened`` was the last option) or caused errors.  The property
is therefore useless and was deprecated in 6.0; make it read-only now.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agocrypto: make loaded property read-only
Paolo Bonzini [Mon, 9 May 2022 09:42:23 +0000 (11:42 +0200)]
crypto: make loaded property read-only

The ``loaded=on`` option in the command line or QMP ``object-add`` either had
no effect (if ``loaded`` was the last option) or caused options to be
effectively ignored as if they were not given.  The property is therefore
useless and was deprecated in 6.0; make it read-only now.

The patch is best reviewed with "-b".

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Support Arch LBR in CPUID enumeration
Yang Weijiang [Tue, 15 Feb 2022 19:52:58 +0000 (14:52 -0500)]
target/i386: Support Arch LBR in CPUID enumeration

If CPUID.(EAX=07H, ECX=0):EDX[19] is set to 1, the processor
supports Architectural LBRs. In this case, CPUID leaf 01CH
indicates details of the Architectural LBRs capabilities.
XSAVE support for Architectural LBRs is enumerated in
CPUID.(EAX=0DH, ECX=0FH).

Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-9-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: introduce helper to access supported CPUID
Paolo Bonzini [Fri, 29 Apr 2022 18:52:52 +0000 (20:52 +0200)]
target/i386: introduce helper to access supported CPUID

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Enable Arch LBR migration states in vmstate
Yang Weijiang [Tue, 15 Feb 2022 19:52:57 +0000 (14:52 -0500)]
target/i386: Enable Arch LBR migration states in vmstate

The Arch LBR record MSRs and control MSRs will be migrated
to destination guest if the vcpus were running with Arch
LBR active.

Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-8-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Add MSR access interface for Arch LBR
Yang Weijiang [Tue, 15 Feb 2022 19:52:56 +0000 (14:52 -0500)]
target/i386: Add MSR access interface for Arch LBR

In the first generation of Arch LBR, the max support
Arch LBR depth is 32, both host and guest use the value
to set depth MSR. This can simplify the implementation
of patch given the side-effect of mismatch of host/guest
depth MSR: XRSTORS will reset all recording MSRs to 0s
if the saved depth mismatches MSR_ARCH_LBR_DEPTH.

In most of the cases Arch LBR is not in active status,
so check the control bit before save/restore the big
chunck of Arch LBR MSRs.

Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-7-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Add XSAVES support for Arch LBR
Yang Weijiang [Tue, 15 Feb 2022 19:52:55 +0000 (14:52 -0500)]
target/i386: Add XSAVES support for Arch LBR

Define Arch LBR bit in XSS and save/restore structure
for XSAVE area size calculation.

Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-6-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Enable support for XSAVES based features
Yang Weijiang [Tue, 15 Feb 2022 19:52:54 +0000 (14:52 -0500)]
target/i386: Enable support for XSAVES based features

There're some new features, including Arch LBR, depending
on XSAVES/XRSTORS support, the new instructions will
save/restore data based on feature bits enabled in XCR0 | XSS.
This patch adds the basic support for related CPUID enumeration
and meanwhile changes the name from FEAT_XSAVE_COMP_{LO|HI} to
FEAT_XSAVE_XCR0_{LO|HI} to differentiate clearly the feature
bits in XCR0 and those in XSS.

Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-5-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Add kvm_get_one_msr helper
Yang Weijiang [Tue, 15 Feb 2022 19:52:53 +0000 (14:52 -0500)]
target/i386: Add kvm_get_one_msr helper

When try to get one msr from KVM, I found there's no such kind of
existing interface while kvm_put_one_msr() is there. So here comes
the patch. It'll remove redundant preparation code before finally
call KVM_GET_MSRS IOCTL.

No functional change intended.

Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-4-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: Add lbr-fmt vPMU option to support guest LBR
Yang Weijiang [Tue, 15 Feb 2022 19:52:52 +0000 (14:52 -0500)]
target/i386: Add lbr-fmt vPMU option to support guest LBR

The Last Branch Recording (LBR) is a performance monitor unit (PMU)
feature on Intel processors which records a running trace of the most
recent branches taken by the processor in the LBR stack. This option
indicates the LBR format to enable for guest perf.

The LBR feature is enabled if below conditions are met:
1) KVM is enabled and the PMU is enabled.
2) msr-based-feature IA32_PERF_CAPABILITIES is supporterd on KVM.
3) Supported returned value for lbr_fmt from above msr is non-zero.
4) Guest vcpu model does support FEAT_1_ECX.CPUID_EXT_PDCM.
5) User-provided lbr-fmt value doesn't violate its bitmask (0x3f).
6) Target guest LBR format matches that of host.

Co-developed-by: Like Xu <like.xu@linux.intel.com>
Signed-off-by: Like Xu <like.xu@linux.intel.com>
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-3-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoqdev-properties: Add a new macro with bitmask check for uint64_t property
Yang Weijiang [Tue, 15 Feb 2022 19:52:51 +0000 (14:52 -0500)]
qdev-properties: Add a new macro with bitmask check for uint64_t property

The DEFINE_PROP_UINT64_CHECKMASK maro applies certain mask check agaist
user-supplied property value, reject the value if it violates the bitmask.

Co-developed-by: Like Xu <like.xu@linux.intel.com>
Signed-off-by: Like Xu <like.xu@linux.intel.com>
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-2-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoi386/cpu: Remove the deprecated cpu model 'Icelake-Client'
Robert Hoo [Mon, 14 Mar 2022 08:50:59 +0000 (16:50 +0800)]
i386/cpu: Remove the deprecated cpu model 'Icelake-Client'

Icelake, is the codename for Intel 3rd generation Xeon Scalable server
processors. There isn't ever client variants. This "Icelake-Client" CPU
model was added wrongly and imaginarily.

It has been deprecated since v5.2, now it's time to remove it completely
from code.

Signed-off-by: Robert Hoo <robert.hu@linux.intel.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1647247859-4947-1-git-send-email-robert.hu@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoqga-vss: Use the proper operator to free memory
Konstantin Kostiuk [Thu, 12 May 2022 15:49:09 +0000 (18:49 +0300)]
qga-vss: Use the proper operator to free memory

volume_name_wchar is allocated by 'void* operator new [](long long unsigned int)

Signed-off-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220512154909.331481-1-kkostiuk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoqga-vss: Add auto generated headers to dependencies
Konstantin Kostiuk [Thu, 12 May 2022 15:49:06 +0000 (18:49 +0300)]
qga-vss: Add auto generated headers to dependencies

Signed-off-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Message-Id: <20220512154906.331399-1-kkostiuk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoWHPX: fixed TPR/CR8 translation issues affecting VM debugging
Ivan Shcherbakov [Sat, 14 May 2022 05:50:21 +0000 (07:50 +0200)]
WHPX: fixed TPR/CR8 translation issues affecting VM debugging

This patch fixes the following error that would occur when trying to resume
a WHPX-accelerated VM from a breakpoint:

    qemu: WHPX: Failed to set interrupt state registers, hr=c0350005

The error arises from an incorrect CR8 value being passed to
WHvSetVirtualProcessorRegisters() that doesn't match the
value set via WHvSetVirtualProcessorInterruptControllerState2().

Signed-off-by: Ivan Shcherbakov <ivan@sysprogs.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoMerge tag 'linux-headers-v5.18-rc6' of https://gitlab.com/alex.williamson/qemu into...
Richard Henderson [Fri, 13 May 2022 16:45:17 +0000 (09:45 -0700)]
Merge tag 'linux-headers-v5.18-rc6' of https://gitlab.com/alex.williamson/qemu into staging

 * Linux header update to v5.18-rc6 and vfio file massaging (Alex Williamson)

# -----BEGIN PGP SIGNATURE-----
#
# iQJPBAABCAA5FiEEQvbATlQL0amee4qQI5ubbjuwiyIFAmJ+ajgbHGFsZXgud2ls
# bGlhbXNvbkByZWRoYXQuY29tAAoJECObm247sIsiqVUQAJV8iZ7UrgX9LpCo/kBg
# VuPdzv8NVoaxe33FkkMlJYIcm1wCIC+lc/nKczkJg5rdfD4/E0oNyv8DER+2q2jC
# Q94nq3CoSwnbNeRaAc6dcA/Hg9VL5VAwE8S/raDt+DbCBndR0o7zejm2CHZIYPxF
# GZXgexNThYcSjNsfkP2Sju/SwjWSFRHXFeTg5P6BOm2qWVGMNmvlnarPTffWOC0Y
# 6sT2VUykzahVSro5WxyuLdH114KI8QRZzSybMkLMpt3uaxZsFoAhO4FYZX4owiUn
# rSoHS0am86KCRG3msD8UjKKy/R4dP0hXW41xAr4HwRZWouNtwGJVtd1mA6kF5euL
# 23dQwSQeHvgFx0gY0FiFih/shZ24qvTDZDz8hNqgi3q42Tv8upghAyWL7v/aPpsy
# 94RH3eNSBkNEvWdK/e40LpuN14voRMLov4XcVirP2kbBFWoU4SSIZmASSLG/OeQn
# tQLatE53SMiXBabMRKeVCiyMxnoUtOd8fe97XFEONFIEU0NOABt5Zoldtbq4kwoQ
# tcUydKB66df3uWccb6Tpl3V7h6b8sBL8trfBUD1qANjcDbEzQVUVBTmg+DRwzs8d
# HQ4f6YQaJoyZlVDgC8GVA268ubAam8FEz1piro4gz8MaZJCeCdx5Teec0yCwjBL+
# IlNS08zjV5fHFZuMPd/qswjk
# =gegh
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 13 May 2022 07:24:56 AM PDT
# gpg:                using RSA key 42F6C04E540BD1A99E7B8A90239B9B6E3BB08B22
# gpg:                issuer "alex.williamson@redhat.com"
# gpg: Good signature from "Alex Williamson <alex.williamson@redhat.com>" [undefined]
# gpg:                 aka "Alex Williamson <alex@shazbot.org>" [undefined]
# gpg:                 aka "Alex Williamson <alwillia@redhat.com>" [undefined]
# gpg:                 aka "Alex Williamson <alex.l.williamson@gmail.com>" [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: 42F6 C04E 540B D1A9 9E7B  8A90 239B 9B6E 3BB0 8B22

* tag 'linux-headers-v5.18-rc6' of https://gitlab.com/alex.williamson/qemu:
  linux-headers: Update to v5.18-rc6

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agolinux-headers: Update to v5.18-rc6
Alex Williamson [Fri, 13 May 2022 14:20:08 +0000 (08:20 -0600)]
linux-headers: Update to v5.18-rc6

Update to c5eb0a61238d ("Linux 5.18-rc6").  Mechanical search and
replace of vfio defines with white space massaging.

Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2 years agoMerge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging
Richard Henderson [Thu, 12 May 2022 17:52:15 +0000 (10:52 -0700)]
Merge tag 'for-upstream' of https://gitlab.com/bonzini/qemu into staging

* small cleanups for pc-bios/optionrom Makefiles
* checkpatch: fix g_malloc check
* fix mremap() and RDMA detection
* confine igd-passthrough-isa-bridge to Xen-enabled builds
* cover PCI in arm-virt machine qtests
* add -M boot and -M mem compound properties
* bump SLIRP submodule
* support CFI with system libslirp (>= 4.7)
* clean up CoQueue wakeup functions
* fix vhost-vsock regression
* fix --disable-vnc compilation
* other minor bugfixes

# -----BEGIN PGP SIGNATURE-----
#
# iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmJ8/KMUHHBib256aW5p
# QHJlZGhhdC5jb20ACgkQv/vSX3jHroNTTAf9Et1C8iZn+OlZi99wMEeMy8a4mIE5
# CpkBpFphhkBvt3AH7XNsCyL4Gea4QgsI7nOIEVUwvW7gPf85PiBUX8mjrIVg3x1k
# bmMEwMKSTYPmDieAnYBP9zCqZQXNYP8L8WxVs2jFY2GXZ2ZogODYFbvCY4yEEB72
# UR6uIvQRdpiB6BEj8UZ+5i+sDtb0zxqrjzUz8T/PJC9/2JSNgi+sAWWQoQT3PPU7
# R7z2nmEa1VeVLPP6mUHvJKhBltVXF+LyIjQHvo+Tp9tSqp9JwXfFBNQ5W/MFes2D
# skF47N7PdgKRH9Dp4r0j+MqBwoAq86+ao+MKsbQ1Gb91HhoCWt/MrVrVyg==
# =1E6P
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 12 May 2022 05:25:07 AM PDT
# gpg:                using RSA key F13338574B662389866C7682BFFBD25F78C7AE83
# gpg:                issuer "pbonzini@redhat.com"
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [undefined]
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>" [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: 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' of https://gitlab.com/bonzini/qemu: (27 commits)
  vmxcap: add tertiary execution controls
  vl: make machine type deprecation a warning
  meson: link libpng independent of vnc
  vhost-backend: do not depend on CONFIG_VHOST_VSOCK
  coroutine-lock: qemu_co_queue_restart_all is a coroutine-only qemu_co_enter_all
  coroutine-lock: introduce qemu_co_queue_enter_all
  coroutine-lock: qemu_co_queue_next is a coroutine-only qemu_co_enter_next
  net: slirp: allow CFI with libslirp >= 4.7
  net: slirp: add support for CFI-friendly timer API
  net: slirp: switch to slirp_new
  net: slirp: introduce a wrapper struct for QemuTimer
  slirp: bump submodule past 4.7 release
  machine: move more memory validation to Machine object
  machine: make memory-backend a link property
  machine: add mem compound property
  machine: add boot compound property
  machine: use QAPI struct for boot configuration
  tests/qtest/libqos: Add generic pci host bridge in arm-virt machine
  tests/qtest/libqos: Skip hotplug tests if pci root bus is not hotpluggable
  tests/qtest/libqos/pci: Introduce pio_limit
  ...

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agoMerge tag 'for-upstream' of git://repo.or.cz/qemu/kevin into staging
Richard Henderson [Thu, 12 May 2022 15:37:28 +0000 (08:37 -0700)]
Merge tag 'for-upstream' of git://repo.or.cz/qemu/kevin into staging

Block layer patches

- coroutine: Fix crashes due to too large pool batch size
- fdc: Prevent end-of-track overrun
- nbd: MULTI_CONN for shared writable exports
- iotests test runner improvements

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAmJ9KCkRHGt3b2xmQHJl
# ZGhhdC5jb20ACgkQfwmycsiPL9ZtSRAAmYDFBPqxfutpFXM7kIKwL6COXJC12MOx
# Tmu8cDiGB/jNChdi3kl6I5h5njzo3U0ZlL/Ign6EzHoeoXLAPSeUWmuRsARwsZ+A
# rL61gf6yrMjAo45FZuIS0GlMDk8BauRwPl9qPWeqQcrtOMYpxwZfyFGmcMpQgAOI
# MSC1I8p3FA7oJhGpKIHDPOjaZA97Lm2rLnDIwZ4f0YgssbybFBcFCXOQbhpsVhLy
# Tjp/L+qRUtna9xBsPHQvHZW0kITQbCQPdX+oVqqUmwzSvuHqfXKe1YppyPjBt/S0
# H7nxtx4HOgP0lP5Kea+wbIRAk9Da5uaOW8hlMWRLShEKv1iTUenQSKteBB6CD03t
# GD9ze1kGoR9b6szw795BXxZxcWii0cn359lIVHeKR/U8zDuz5w3zhyl0klK8xeJy
# nj+JErLwQ7BD8kNR+7WAfXTF3tk2dQao1AvsBjn087KjMiJ/Mg8HY4K2zrjBUrHL
# DLTyAIjzct3BWJDZ02fb5jb8pHmIP3JO6m9Zvjm7ibP65BqJOwIXUTFpbgnrOg45
# oFLDV4JgC4Hh4GEtdm+UhQE51A0VVW5pDaqWTdWkCcuk3QgxUdM3Wm3SW6pw1Gvb
# T0X0j5RgF/k3YrW576R/VIy6z4YPbzAtiG4O/zSlsujHoDcVNWnxApgSB/unaDh8
# LNkFPGEMeSs=
# =JmTm
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 12 May 2022 08:30:49 AM PDT
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]

* tag 'for-upstream' of git://repo.or.cz/qemu/kevin:
  qemu-iotests: inline common.config into common.rc
  nbd/server: Allow MULTI_CONN for shared writable exports
  qemu-nbd: Pass max connections to blockdev layer
  tests/qtest/fdc-test: Add a regression test for CVE-2021-3507
  hw/block/fdc: Prevent end-of-track overrun (CVE-2021-3507)
  .gitlab-ci.d: export meson testlog.txt as an artifact
  tests/qemu-iotests: print intent to run a test in TAP mode
  iotests/testrunner: Flush after run_test()
  coroutine: Revert to constant batch size
  coroutine: Rename qemu_coroutine_inc/dec_pool_size()

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agoqemu-iotests: inline common.config into common.rc
Paolo Bonzini [Thu, 5 May 2022 09:47:23 +0000 (11:47 +0200)]
qemu-iotests: inline common.config into common.rc

common.rc has some complicated logic to find the common.config that
dates back to xfstests and is completely unnecessary now.  Just include
the contents of the file.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220505094723.732116-1-pbonzini@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agovmxcap: add tertiary execution controls
Paolo Bonzini [Wed, 11 May 2022 16:39:12 +0000 (18:39 +0200)]
vmxcap: add tertiary execution controls

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agovl: make machine type deprecation a warning
Paolo Bonzini [Wed, 11 May 2022 17:50:43 +0000 (13:50 -0400)]
vl: make machine type deprecation a warning

error_report should generally be followed by a failure; if we can proceed
anyway, that is just a warning and should be communicated properly to
the user with warn_report.

Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20220511175043.27327-1-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agonbd/server: Allow MULTI_CONN for shared writable exports
Eric Blake [Thu, 12 May 2022 00:49:24 +0000 (19:49 -0500)]
nbd/server: Allow MULTI_CONN for shared writable exports

According to the NBD spec, a server that advertises
NBD_FLAG_CAN_MULTI_CONN promises that multiple client connections will
not see any cache inconsistencies: when properly separated by a single
flush, actions performed by one client will be visible to another
client, regardless of which client did the flush.

We always satisfy these conditions in qemu - even when we support
multiple clients, ALL clients go through a single point of reference
into the block layer, with no local caching.  The effect of one client
is instantly visible to the next client.  Even if our backend were a
network device, we argue that any multi-path caching effects that
would cause inconsistencies in back-to-back actions not seeing the
effect of previous actions would be a bug in that backend, and not the
fault of caching in qemu.  As such, it is safe to unconditionally
advertise CAN_MULTI_CONN for any qemu NBD server situation that
supports parallel clients.

Note, however, that we don't want to advertise CAN_MULTI_CONN when we
know that a second client cannot connect (for historical reasons,
qemu-nbd defaults to a single connection while nbd-server-add and QMP
commands default to unlimited connections; but we already have
existing means to let either style of NBD server creation alter those
defaults).  This is visible by no longer advertising MULTI_CONN for
'qemu-nbd -r' without -e, as in the iotest nbd-qemu-allocation.

The harder part of this patch is setting up an iotest to demonstrate
behavior of multiple NBD clients to a single server.  It might be
possible with parallel qemu-io processes, but I found it easier to do
in python with the help of libnbd, and help from Nir and Vladimir in
writing the test.

Signed-off-by: Eric Blake <eblake@redhat.com>
Suggested-by: Nir Soffer <nsoffer@redhat.com>
Suggested-by: Vladimir Sementsov-Ogievskiy <v.sementsov-og@mail.ru>
Message-Id: <20220512004924.417153-3-eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agoqemu-nbd: Pass max connections to blockdev layer
Eric Blake [Thu, 12 May 2022 00:49:23 +0000 (19:49 -0500)]
qemu-nbd: Pass max connections to blockdev layer

The next patch wants to adjust whether the NBD server code advertises
MULTI_CONN based on whether it is known if the server limits to
exactly one client.  For a server started by QMP, this information is
obtained through nbd_server_start (which can support more than one
export); but for qemu-nbd (which supports exactly one export), it is
controlled only by the command-line option -e/--shared.  Since we
already have a hook function used by qemu-nbd, it's easiest to just
alter its signature to fit our needs.

Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <20220512004924.417153-2-eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agotests/qtest/fdc-test: Add a regression test for CVE-2021-3507
Philippe Mathieu-Daudé [Thu, 18 Nov 2021 11:57:33 +0000 (12:57 +0100)]
tests/qtest/fdc-test: Add a regression test for CVE-2021-3507

Add the reproducer from https://gitlab.com/qemu-project/qemu/-/issues/339

Without the previous commit, when running 'make check-qtest-i386'
with QEMU configured with '--enable-sanitizers' we get:

  ==4028352==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x619000062a00 at pc 0x5626d03c491a bp 0x7ffdb4199410 sp 0x7ffdb4198bc0
  READ of size 786432 at 0x619000062a00 thread T0
      #0 0x5626d03c4919 in __asan_memcpy (qemu-system-i386+0x1e65919)
      #1 0x5626d1c023cc in flatview_write_continue softmmu/physmem.c:2787:13
      #2 0x5626d1bf0c0f in flatview_write softmmu/physmem.c:2822:14
      #3 0x5626d1bf0798 in address_space_write softmmu/physmem.c:2914:18
      #4 0x5626d1bf0f37 in address_space_rw softmmu/physmem.c:2924:16
      #5 0x5626d1bf14c8 in cpu_physical_memory_rw softmmu/physmem.c:2933:5
      #6 0x5626d0bd5649 in cpu_physical_memory_write include/exec/cpu-common.h:82:5
      #7 0x5626d0bd0a07 in i8257_dma_write_memory hw/dma/i8257.c:452:9
      #8 0x5626d09f825d in fdctrl_transfer_handler hw/block/fdc.c:1616:13
      #9 0x5626d0a048b4 in fdctrl_start_transfer hw/block/fdc.c:1539:13
      #10 0x5626d09f4c3e in fdctrl_write_data hw/block/fdc.c:2266:13
      #11 0x5626d09f22f7 in fdctrl_write hw/block/fdc.c:829:9
      #12 0x5626d1c20bc5 in portio_write softmmu/ioport.c:207:17

  0x619000062a00 is located 0 bytes to the right of 512-byte region [0x619000062800,0x619000062a00)
  allocated by thread T0 here:
      #0 0x5626d03c66ec in posix_memalign (qemu-system-i386+0x1e676ec)
      #1 0x5626d2b988d4 in qemu_try_memalign util/oslib-posix.c:210:11
      #2 0x5626d2b98b0c in qemu_memalign util/oslib-posix.c:226:27
      #3 0x5626d09fbaf0 in fdctrl_realize_common hw/block/fdc.c:2341:20
      #4 0x5626d0a150ed in isabus_fdc_realize hw/block/fdc-isa.c:113:5
      #5 0x5626d2367935 in device_set_realized hw/core/qdev.c:531:13

  SUMMARY: AddressSanitizer: heap-buffer-overflow (qemu-system-i386+0x1e65919) in __asan_memcpy
  Shadow bytes around the buggy address:
    0x0c32800044f0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
    0x0c3280004500: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    0x0c3280004510: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    0x0c3280004520: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
    0x0c3280004530: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
  =>0x0c3280004540:[fa]fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
    0x0c3280004550: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
    0x0c3280004560: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
    0x0c3280004570: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
    0x0c3280004580: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
    0x0c3280004590: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
  Shadow byte legend (one shadow byte represents 8 application bytes):
    Addressable:           00
    Heap left redzone:       fa
    Freed heap region:       fd
  ==4028352==ABORTING

[ kwolf: Added snapshot=on to prevent write file lock failure ]

Reported-by: Alexander Bulekov <alxndr@bu.edu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alexander Bulekov <alxndr@bu.edu>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agohw/block/fdc: Prevent end-of-track overrun (CVE-2021-3507)
Philippe Mathieu-Daudé [Thu, 18 Nov 2021 11:57:32 +0000 (12:57 +0100)]
hw/block/fdc: Prevent end-of-track overrun (CVE-2021-3507)

Per the 82078 datasheet, if the end-of-track (EOT byte in
the FIFO) is more than the number of sectors per side, the
command is terminated unsuccessfully:

* 5.2.5 DATA TRANSFER TERMINATION

  The 82078 supports terminal count explicitly through
  the TC pin and implicitly through the underrun/over-
  run and end-of-track (EOT) functions. For full sector
  transfers, the EOT parameter can define the last
  sector to be transferred in a single or multisector
  transfer. If the last sector to be transferred is a par-
  tial sector, the host can stop transferring the data in
  mid-sector, and the 82078 will continue to complete
  the sector as if a hardware TC was received. The
  only difference between these implicit functions and
  TC is that they return "abnormal termination" result
  status. Such status indications can be ignored if they
  were expected.

* 6.1.3 READ TRACK

  This command terminates when the EOT specified
  number of sectors have been read. If the 82078
  does not find an I D Address Mark on the diskette
  after the second· occurrence of a pulse on the
  INDX# pin, then it sets the IC code in Status Regis-
  ter 0 to "01" (Abnormal termination), sets the MA bit
  in Status Register 1 to "1", and terminates the com-
  mand.

* 6.1.6 VERIFY

  Refer to Table 6-6 and Table 6-7 for information
  concerning the values of MT and EC versus SC and
  EOT value.

* Table 6·6. Result Phase Table

* Table 6-7. Verify Command Result Phase Table

Fix by aborting the transfer when EOT > # Sectors Per Side.

Cc: qemu-stable@nongnu.org
Cc: Hervé Poussineau <hpoussin@reactos.org>
Fixes: baca51faff0 ("floppy driver: disk geometry auto detect")
Reported-by: Alexander Bulekov <alxndr@bu.edu>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/339
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20211118115733.4038610-2-philmd@redhat.com>
Reviewed-by: Hanna Reitz <hreitz@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agomeson: link libpng independent of vnc
Kshitij Suri [Tue, 10 May 2022 16:19:32 +0000 (16:19 +0000)]
meson: link libpng independent of vnc

Currently png support is dependent on vnc for linking object file to
libpng. This commit makes the parameter independent of vnc as it breaks
system emulator with --disable-vnc unless --disable-png is added.

Fixes: 9a0a119a38 ("Added parameter to take screenshot with screendump as PNG", 2022-04-27)
Signed-off-by: Kshitij Suri <kshitij.suri@nutanix.com>
Message-Id: <20220510161932.228481-1-kshitij.suri@nutanix.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agovhost-backend: do not depend on CONFIG_VHOST_VSOCK
Paolo Bonzini [Wed, 11 May 2022 07:40:35 +0000 (09:40 +0200)]
vhost-backend: do not depend on CONFIG_VHOST_VSOCK

The vsock callbacks .vhost_vsock_set_guest_cid and
.vhost_vsock_set_running are the only ones to be conditional
on #ifdef CONFIG_VHOST_VSOCK.  This is different from any other
device-dependent callbacks like .vhost_scsi_set_endpoint, and it
also broke when CONFIG_VHOST_VSOCK was changed to a per-target
symbol.

It would be possible to also use the CONFIG_DEVICES include, but
really there is no reason for most virtio files to be per-target
so just remove the #ifdef to fix the issue.

Reported-by: Dov Murik <dovmurik@linux.ibm.com>
Fixes: 9972ae314f ("build: move vhost-vsock configuration to Kconfig")
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agocoroutine-lock: qemu_co_queue_restart_all is a coroutine-only qemu_co_enter_all
Paolo Bonzini [Wed, 27 Apr 2022 13:08:30 +0000 (15:08 +0200)]
coroutine-lock: qemu_co_queue_restart_all is a coroutine-only qemu_co_enter_all

qemu_co_queue_restart_all is basically the same as qemu_co_enter_all
but without a QemuLockable argument.  That's perfectly fine, but only as
long as the function is marked coroutine_fn.  If used outside coroutine
context, qemu_co_queue_wait will attempt to take the lock and that
is just broken: if you are calling qemu_co_queue_restart_all outside
coroutine context, the lock is going to be a QemuMutex which cannot be
taken twice by the same thread.

The patch adds the marker to qemu_co_queue_restart_all and to its sole
non-coroutine_fn caller; it then reimplements the function in terms of
qemu_co_enter_all_impl, to remove duplicated code and to clarify that the
latter also works in coroutine context.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20220427130830.150180-4-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agocoroutine-lock: introduce qemu_co_queue_enter_all
Paolo Bonzini [Wed, 27 Apr 2022 13:08:29 +0000 (15:08 +0200)]
coroutine-lock: introduce qemu_co_queue_enter_all

Because qemu_co_queue_restart_all does not release the lock, it should
be used only in coroutine context.  Introduce a new function that,
like qemu_co_enter_next, does release the lock, and use it whenever
qemu_co_queue_restart_all was used outside coroutine context.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20220427130830.150180-3-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agocoroutine-lock: qemu_co_queue_next is a coroutine-only qemu_co_enter_next
Paolo Bonzini [Wed, 27 Apr 2022 13:08:28 +0000 (15:08 +0200)]
coroutine-lock: qemu_co_queue_next is a coroutine-only qemu_co_enter_next

qemu_co_queue_next is basically the same as qemu_co_enter_next but
without a QemuLockable argument.  That's perfectly fine, but only
as long as the function is marked coroutine_fn.  If used outside
coroutine context, qemu_co_queue_wait will attempt to take the lock
and that is just broken: if you are calling qemu_co_queue_next outside
coroutine context, the lock is going to be a QemuMutex which cannot be
taken twice by the same thread.

The patch adds the marker and reimplements qemu_co_queue_next in terms of
qemu_co_enter_next_impl, to remove duplicated code and to clarify that the
latter also works in coroutine context.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20220427130830.150180-2-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agonet: slirp: allow CFI with libslirp >= 4.7
Paolo Bonzini [Mon, 11 Apr 2022 07:41:27 +0000 (09:41 +0200)]
net: slirp: allow CFI with libslirp >= 4.7

slirp 4.7 introduces a new CFI-friendly timer callback that does
not pass function pointers within libslirp as callbacks for timers.
Check the version number and, if it is new enough, allow using CFI
even with a system libslirp.

Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Marc-André Lureau <malureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agonet: slirp: add support for CFI-friendly timer API
Paolo Bonzini [Mon, 11 Apr 2022 07:39:16 +0000 (09:39 +0200)]
net: slirp: add support for CFI-friendly timer API

libslirp 4.7 introduces a CFI-friendly version of the .timer_new callback.
The new callback replaces the function pointer with an enum; invoking the
callback is done with a new function slirp_handle_timer.

Support the new API so that CFI can be made compatible with using a system
libslirp.

Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Marc-André Lureau <malureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agonet: slirp: switch to slirp_new
Paolo Bonzini [Mon, 11 Apr 2022 08:16:36 +0000 (10:16 +0200)]
net: slirp: switch to slirp_new

Replace slirp_init with slirp_new, so that a more recent cfg.version
can be specified.  The function appeared in version 4.1.0.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agonet: slirp: introduce a wrapper struct for QemuTimer
Paolo Bonzini [Mon, 11 Apr 2022 07:26:06 +0000 (09:26 +0200)]
net: slirp: introduce a wrapper struct for QemuTimer

This struct will be extended in the next few patches to support the
new slirp_handle_timer() call.  For that we need to store an additional
"int" for each SLIRP timer, in addition to the cb_opaque.

Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Reviewed-by: Marc-André Lureau <malureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoslirp: bump submodule past 4.7 release
Paolo Bonzini [Wed, 4 May 2022 20:15:14 +0000 (22:15 +0200)]
slirp: bump submodule past 4.7 release

Version 4.7 of slirp provides a new timer API that works better with CFI,
together with several other improvements:

* Allow disabling the internal DHCP server !22
* Support Unix sockets in hostfwd !103
* IPv6 DNS proxying support !110
* bootp: add support for UEFI HTTP boot !111

and bugfixes.

The submodule update also includes 2 commits to fix warnings in the
Win32 build.

Reviewed-by: Marc-André Lureau <malureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agomachine: move more memory validation to Machine object
Paolo Bonzini [Thu, 14 Apr 2022 16:53:00 +0000 (12:53 -0400)]
machine: move more memory validation to Machine object

This allows setting memory properties without going through
vl.c, and have them validated just the same.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220414165300.555321-6-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agomachine: make memory-backend a link property
Paolo Bonzini [Thu, 14 Apr 2022 16:52:59 +0000 (12:52 -0400)]
machine: make memory-backend a link property

Handle HostMemoryBackend creation and setting of ms->ram entirely in
machine_run_board_init.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220414165300.555321-5-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agomachine: add mem compound property
Paolo Bonzini [Thu, 14 Apr 2022 16:52:58 +0000 (12:52 -0400)]
machine: add mem compound property

Make -m syntactic sugar for a compound property "-machine
mem.{size,max-size,slots}".  The new property does not have
the magic conversion to megabytes of unsuffixed arguments,
and also does not understand that "0" means the default size
(you have to leave it out to get the default).  This means
that we need to convert the QemuOpts by hand to a QDict.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220414165300.555321-4-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agomachine: add boot compound property
Paolo Bonzini [Thu, 14 Apr 2022 16:52:57 +0000 (12:52 -0400)]
machine: add boot compound property

Make -boot syntactic sugar for a compound property "-machine boot.{order,menu,...}".
machine_boot_parse is replaced by the setter for the property.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220414165300.555321-3-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agomachine: use QAPI struct for boot configuration
Paolo Bonzini [Thu, 14 Apr 2022 16:52:56 +0000 (12:52 -0400)]
machine: use QAPI struct for boot configuration

As part of converting -boot to a property with a QAPI type, define
the struct and use it throughout QEMU to access boot configuration.
machine_boot_parse takes care of doing the QemuOpts->QAPI conversion by
hand, for now.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220414165300.555321-2-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years ago.gitlab-ci.d: export meson testlog.txt as an artifact
Daniel P. Berrangé [Mon, 9 May 2022 12:41:34 +0000 (13:41 +0100)]
.gitlab-ci.d: export meson testlog.txt as an artifact

When running 'make check' we only get a summary of progress on the
console. Fortunately meson/ninja have saved the raw test output to a
logfile. Exposing this log will make it easier to debug failures that
happen in CI.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220509124134.867431-3-berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agotests/qemu-iotests: print intent to run a test in TAP mode
Daniel P. Berrangé [Mon, 9 May 2022 12:41:33 +0000 (13:41 +0100)]
tests/qemu-iotests: print intent to run a test in TAP mode

When running I/O tests using TAP output mode, we get a single TAP test
with a sub-test reported for each I/O test that is run. The output looks
something like this:

 1..123
 ok qcow2 011
 ok qcow2 012
 ok qcow2 013
 ok qcow2 217
 ...

If everything runs or fails normally this is fine, but periodically we
have been seeing the test harness abort early before all 123 tests have
been run, just leaving a fairly useless message like

  TAP parsing error: Too few tests run (expected 123, got 107)

we have no idea which tests were running at the time the test harness
abruptly exited. This change causes us to print a message about our
intent to run each test, so we have a record of what is active at the
time the harness exits abnormally.

 1..123
 # running qcow2 011
 ok qcow2 011
 # running qcow2 012
 ok qcow2 012
 # running qcow2 013
 ok qcow2 013
 # running qcow2 217
 ok qcow2 217
 ...

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220509124134.867431-2-berrange@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agoiotests/testrunner: Flush after run_test()
Hanna Reitz [Fri, 6 May 2022 13:42:15 +0000 (15:42 +0200)]
iotests/testrunner: Flush after run_test()

When stdout is not a terminal, the buffer may not be flushed at each end
of line, so we should flush after each test is done.  This is especially
apparent when run by check-block, in two ways:

First, when running make check-block -jX with X > 1, progress indication
was missing, even though testrunner.py does theoretically print each
test's status once it has been run, even in multi-processing mode.
Flushing after each test restores this progress indication.

Second, sometimes make check-block failed altogether, with an error
message that "too few tests [were] run".  I presume that's because one
worker process in the job pool did not get to flush its stdout before
the main process exited, and so meson did not get to see that worker's
test results.  In any case, by flushing at the end of run_test(), the
problem has disappeared for me.

Signed-off-by: Hanna Reitz <hreitz@redhat.com>
Message-Id: <20220506134215.10086-1-hreitz@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agocoroutine: Revert to constant batch size
Kevin Wolf [Tue, 10 May 2022 15:10:20 +0000 (17:10 +0200)]
coroutine: Revert to constant batch size

Commit 4c41c69e changed the way the coroutine pool is sized because for
virtio-blk devices with a large queue size and heavy I/O, it was just
too small and caused coroutines to be deleted and reallocated soon
afterwards. The change made the size dynamic based on the number of
queues and the queue size of virtio-blk devices.

There are two important numbers here: Slightly simplified, when a
coroutine terminates, it is generally stored in the global release pool
up to a certain pool size, and if the pool is full, it is freed.
Conversely, when allocating a new coroutine, the coroutines in the
release pool are reused if the pool already has reached a certain
minimum size (the batch size), otherwise we allocate new coroutines.

The problem after commit 4c41c69e is that it not only increases the
maximum pool size (which is the intended effect), but also the batch
size for reusing coroutines (which is a bug). It means that in cases
with many devices and/or a large queue size (which defaults to the
number of vcpus for virtio-blk-pci), many thousand coroutines could be
sitting in the release pool without being reused.

This is not only a waste of memory and allocations, but it actually
makes the QEMU process likely to hit the vm.max_map_count limit on Linux
because each coroutine requires two mappings (its stack and the guard
page for the stack), causing it to abort() in qemu_alloc_stack() because
when the limit is hit, mprotect() starts to fail with ENOMEM.

In order to fix the problem, change the batch size back to 64 to avoid
uselessly accumulating coroutines in the release pool, but keep the
dynamic maximum pool size so that coroutines aren't freed too early
in heavy I/O scenarios.

Note that this fix doesn't strictly make it impossible to hit the limit,
but this would only happen if most of the coroutines are actually in use
at the same time, not just sitting in a pool. This is the same behaviour
as we already had before commit 4c41c69e. Fully preventing this would
require allowing qemu_coroutine_create() to return an error, but it
doesn't seem to be a scenario that people hit in practice.

Cc: qemu-stable@nongnu.org
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2079938
Fixes: 4c41c69e05fe28c0f95f8abd2ebf407e95a4f04b
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20220510151020.105528-3-kwolf@redhat.com>
Tested-by: Hiroki Narukawa <hnarukaw@yahoo-corp.jp>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agocoroutine: Rename qemu_coroutine_inc/dec_pool_size()
Kevin Wolf [Tue, 10 May 2022 15:10:19 +0000 (17:10 +0200)]
coroutine: Rename qemu_coroutine_inc/dec_pool_size()

It's true that these functions currently affect the batch size in which
coroutines are reused (i.e. moved from the global release pool to the
allocation pool of a specific thread), but this is a bug and will be
fixed in a separate patch.

In fact, the comment in the header file already just promises that it
influences the pool size, so reflect this in the name of the functions.
As a nice side effect, the shorter function name makes some line
wrapping unnecessary.

Cc: qemu-stable@nongnu.org
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20220510151020.105528-2-kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2 years agotests/qtest/libqos: Add generic pci host bridge in arm-virt machine
Eric Auger [Wed, 4 May 2022 15:20:25 +0000 (17:20 +0200)]
tests/qtest/libqos: Add generic pci host bridge in arm-virt machine

Up to now the virt-machine node contains a virtio-mmio node.
However no driver produces any PCI interface node. Hence, PCI
tests cannot be run with aarch64 binary.

Add a GPEX driver node that produces a pci interface node. This latter
then can be consumed by all the pci tests. One of the first motivation
was to be able to run the virtio-iommu-pci tests.

We still face an issue with pci hotplug tests as hotplug cannot happen
on the pcie root bus and require a generic root port. This will be
addressed later on.

We force cpu=max along with aarch64/virt machine as some PCI tests
require high MMIO regions to be available.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Message-Id: <20220504152025.1785704-4-eric.auger@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotests/qtest/libqos: Skip hotplug tests if pci root bus is not hotpluggable
Eric Auger [Wed, 4 May 2022 15:20:24 +0000 (17:20 +0200)]
tests/qtest/libqos: Skip hotplug tests if pci root bus is not hotpluggable

ARM does not not support hotplug on pcie.0. Add a flag on the bus
which tells if devices can be hotplugged and skip hotplug tests
if the bus cannot be hotplugged. This is a temporary solution to
enable the other pci tests on aarch64.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20220504152025.1785704-3-eric.auger@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotests/qtest/libqos/pci: Introduce pio_limit
Eric Auger [Wed, 4 May 2022 15:20:23 +0000 (17:20 +0200)]
tests/qtest/libqos/pci: Introduce pio_limit

At the moment the IO space limit is hardcoded to
QPCI_PIO_LIMIT = 0x10000. When accesses are performed to a bar,
the base address of this latter is compared against the limit
to decide whether we perform an IO or a memory access.

On ARM, we cannot keep this PIO limit as the arm-virt machine
uses [0x3eff0000, 0x3f000000 ] for the IO space map and we
are mandated to allocate at 0x0.

Add a new flag in QPCIBar indicating whether it is an IO bar
or a memory bar. This flag is set on QPCIBar allocation and
provisionned based on the BAR configuration. Then the new flag
is used in access functions and in iomap() function.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20220504152025.1785704-2-eric.auger@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agohw/xen/xen_pt: Resolve igd_passthrough_isa_bridge_create() indirection
Bernhard Beschow [Sat, 26 Mar 2022 16:58:24 +0000 (17:58 +0100)]
hw/xen/xen_pt: Resolve igd_passthrough_isa_bridge_create() indirection

Now that igd_passthrough_isa_bridge_create() is implemented within the
xen context it may use Xen* data types directly and become
xen_igd_passthrough_isa_bridge_create(). This resolves an indirection.

Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Message-Id: <20220326165825.30794-3-shentey@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agohw/xen/xen_pt: Confine igd-passthrough-isa-bridge to XEN
Bernhard Beschow [Sat, 26 Mar 2022 16:58:23 +0000 (17:58 +0100)]
hw/xen/xen_pt: Confine igd-passthrough-isa-bridge to XEN

igd-passthrough-isa-bridge is only requested in xen_pt but was
implemented in pc_piix.c. This caused xen_pt to dependend on i386/pc
which is hereby resolved.

Signed-off-by: Bernhard Beschow <shentey@gmail.com>
Acked-by: Anthony PERARD <anthony.perard@citrix.com>
Message-Id: <20220326165825.30794-2-shentey@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agomeson: Make mremap() detecting works correctly
Juan Quintela [Mon, 2 May 2022 13:11:19 +0000 (15:11 +0200)]
meson: Make mremap() detecting works correctly

Without this (at least in Fedora 35) it don't detect mremap()
correctly.

Signed-off-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20220502131119.2345-1-quintela@redhat.com>
[Also switch the LEGACY_RDMA_REG_MR test to cc.links, otherwise
 Debian fails to build. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agocheckpatch: fix g_malloc check
Paolo Bonzini [Wed, 6 Apr 2022 12:27:52 +0000 (14:27 +0200)]
checkpatch: fix g_malloc check

Use the string equality operator "eq", and ensure that $1 is defined by
using "(try|)" instead of "(try)?".  The alternative "((?:try)?)" is
longer and less readable.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agotarget/i386: do not consult nonexistent host leaves
Paolo Bonzini [Fri, 29 Apr 2022 19:16:28 +0000 (21:16 +0200)]
target/i386: do not consult nonexistent host leaves

When cache_info_passthrough is requested, QEMU passes the host values
of the cache information CPUID leaves down to the guest.  However,
it blindly assumes that the CPUID leaf exists on the host, and this
cannot be guaranteed: for example, KVM has recently started to
synthesize AMD leaves up to 0x80000021 in order to provide accurate
CPU bug information to guests.

Querying a nonexistent host leaf fills the output arguments of
host_cpuid with data that (albeit deterministic) is nonsensical
as cache information, namely the data in the highest Intel CPUID
leaf.  If said highest leaf is not ECX-dependent, this can even
cause an infinite loop when kvm_arch_init_vcpu prepares the input
to KVM_SET_CPUID2.  The infinite loop is only terminated by an
abort() when the array gets full.

Reported-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agopc-bios/optionrom: compile with -Wno-array-bounds
Paolo Bonzini [Wed, 13 Apr 2022 12:22:12 +0000 (14:22 +0200)]
pc-bios/optionrom: compile with -Wno-array-bounds

Avoids the following bogus warning:

pvh_main.c: In function ‘pvh_load_kernel’:
pvh_main.c:101:42: warning: array subscript 0 is outside array bounds of ‘uint16_t[0]’ {aka ‘short unsigned int[]’} [-Warray-bounds]
  101 |         uint32_t ebda_paddr = ((uint32_t)*((uint16_t *)EBDA_BASE_ADDR)) << 4;
      |                                          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Cc: qemu-stable@nongnu.org
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agopc-bios/optionrom: detect -fno-pie
Paolo Bonzini [Tue, 29 Mar 2022 11:01:58 +0000 (13:01 +0200)]
pc-bios/optionrom: detect -fno-pie

Do not rely on the detection that was done in the configure script,
since in the future we may want to cross-compile this file.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2 years agoMerge tag 'pull-misc-2022-05-11' of git://repo.or.cz/qemu/armbru into staging
Richard Henderson [Wed, 11 May 2022 16:00:26 +0000 (09:00 -0700)]
Merge tag 'pull-misc-2022-05-11' of git://repo.or.cz/qemu/armbru into staging

Miscellaneous patches patches for 2022-05-11

# -----BEGIN PGP SIGNATURE-----
#
# iQJGBAABCAAwFiEENUvIs9frKmtoZ05fOHC0AOuRhlMFAmJ7zwISHGFybWJydUBy
# ZWRoYXQuY29tAAoJEDhwtADrkYZThuAQAJdSuj5fpY8EXxhuS3Rc8uHPrz6lP+nZ
# kwxKPOldwFdmkXRJ8qrjcc/BXxiJU3pxmSRvFZ8miCFMrb4Vd16sUzD6PeKb1jr8
# JsrvXcsaWn4f/p0v0WraamwSQeZUMjqsZPgZut93qfJoKmgTaxoZnR+ZDHFKoQJS
# qBrHL/5+RPxSugLa6IEpSQwy80jd0tMBaG/e8V+JxzgFM5jzOExwXtfUujzS92Lr
# NgapnbEZrpqErBC1xhpetQ8Q5I4r0kkLj4Exm/ClNtIM2GByJxI8x2DE+NJZNDnm
# g/tvVKUhEl6cOywQRajAJ/LrhUpVSkz6wsczv35rhRS+1FoCb+PRKr42SxZGI2rB
# tZLYt4ouoSGk2pYiudoIBKsIR1Svu7Cmg4YzOL9yvqF0BS3cRDvPgm3QFvoeErjL
# EML7b41zLdIkbvujsJ7HJqVL44QmMSu13PcLUtDvLh+ivpL9wIUQn3ji+rfsgqh+
# RYw4niJ9JO3N3/VwEhlymc9kRSTgZ6rdIWPrtQ5ACwTADAv30++opxAlksE6mo0m
# TYrqyTG2FHGOKm+5Q4Lyx1heHJDUAE3dlRIhGt8KqD6UKlpSfIVIUU2ztjZK4JQ5
# n85LOLZkE9ejbvbpnLX8hgKfouVKKYwFagc/ZA649cIXvC8YDxdOwvhjEVCxa+V5
# dQbpQsekXf9G
# =jOTx
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 11 May 2022 07:58:10 AM PDT
# gpg:                using RSA key 354BC8B3D7EB2A6B68674E5F3870B400EB918653
# gpg:                issuer "armbru@redhat.com"
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" [undefined]
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.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: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* tag 'pull-misc-2022-05-11' of git://repo.or.cz/qemu/armbru:
  Clean up decorations and whitespace around header guards
  Normalize header guard symbol definition
  Clean up ill-advised or unusual header guards
  Clean up header guards that don't match their file name

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agoClean up decorations and whitespace around header guards
Markus Armbruster [Fri, 6 May 2022 13:49:11 +0000 (15:49 +0200)]
Clean up decorations and whitespace around header guards

Cleaned up with scripts/clean-header-guards.pl.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20220506134911.2856099-5-armbru@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
2 years agoNormalize header guard symbol definition
Markus Armbruster [Fri, 6 May 2022 13:49:10 +0000 (15:49 +0200)]
Normalize header guard symbol definition

We commonly define the header guard symbol without an explicit value.
Normalize the exceptions.

Done with scripts/clean-header-guards.pl.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20220506134911.2856099-4-armbru@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
2 years agoClean up ill-advised or unusual header guards
Markus Armbruster [Fri, 6 May 2022 13:49:09 +0000 (15:49 +0200)]
Clean up ill-advised or unusual header guards

Leading underscores are ill-advised because such identifiers are
reserved.  Trailing underscores are merely ugly.  Strip both.

Our header guards commonly end in _H.  Normalize the exceptions.

Macros should be ALL_CAPS.  Normalize the exception.

Done with scripts/clean-header-guards.pl.

include/hw/xen/interface/ and tools/virtiofsd/ left alone, because
these were imported from Xen and libfuse respectively.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20220506134911.2856099-3-armbru@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
2 years agoClean up header guards that don't match their file name
Markus Armbruster [Fri, 6 May 2022 13:49:08 +0000 (15:49 +0200)]
Clean up header guards that don't match their file name

Header guard symbols should match their file name to make guard
collisions less likely.

Cleaned up with scripts/clean-header-guards.pl, followed by some
renaming of new guard symbols picked by the script to better ones.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20220506134911.2856099-2-armbru@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
[Change to generated file ebpf/rss.bpf.skeleton.h backed out]

2 years agoMerge tag 'block-pull-request' of https://gitlab.com/stefanha/qemu into staging
Richard Henderson [Mon, 9 May 2022 18:07:04 +0000 (11:07 -0700)]
Merge tag 'block-pull-request' of https://gitlab.com/stefanha/qemu into staging

Pull request

- Add new thread-pool-min/thread-pool-max parameters to control the thread pool
  used for async I/O.

- Fix virtio-scsi IOThread 100% CPU consumption QEMU 7.0 regression.

# -----BEGIN PGP SIGNATURE-----
#
# iQEzBAABCAAdFiEEhpWov9P5fNqsNXdanKSrs4Grc8gFAmJ5DqgACgkQnKSrs4Gr
# c8iAqAf/WEJzEso0Hu3UUYJi2lAXpLxWPjoNBlPdQlKIJ/I0zQIF0P7GeCifF+0l
# iMjgBv0ofyAuV47gaTJlVrAR75+hJ/IXNDhnu3UuvNWfVOqvksgw6kuHkMo9A2hC
# 4tIHEU9J8jbQSSdQTaZR8Zj4FX1/zcxMBAXT3YO3De6zo78RatBTuNP4dsZzt8bI
# Qs1a4A0p2ScNXK8EcF4QwAWfoxu9OPPzN52DBCNxcIcnn0SUab4NbDxzpRV4ZhDP
# 08WoafI5O+2Kb36QysJN01LqajHrClG/fozrPzBLq5aZUK3xewJGB1hEdGTLkkmz
# NJNBg5Ldszwj4PDZ1dFU3/03aigb3g==
# =t5eR
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 09 May 2022 05:52:56 AM PDT
# gpg:                using RSA key 8695A8BFD3F97CDAAC35775A9CA4ABB381AB73C8
# gpg: Good signature from "Stefan Hajnoczi <stefanha@redhat.com>" [full]
# gpg:                 aka "Stefan Hajnoczi <stefanha@gmail.com>" [full]

* tag 'block-pull-request' of https://gitlab.com/stefanha/qemu:
  virtio-scsi: move request-related items from .h to .c
  virtio-scsi: clean up virtio_scsi_handle_cmd_vq()
  virtio-scsi: clean up virtio_scsi_handle_ctrl_vq()
  virtio-scsi: clean up virtio_scsi_handle_event_vq()
  virtio-scsi: don't waste CPU polling the event virtqueue
  virtio-scsi: fix ctrl and event handler functions in dataplane mode
  util/event-loop-base: Introduce options to set the thread pool size
  util/main-loop: Introduce the main loop into QOM
  Introduce event-loop-base abstract class

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agoMerge tag 'pull-target-arm-20220509' of https://git.linaro.org/people/pmaydell/qemu...
Richard Henderson [Mon, 9 May 2022 16:33:53 +0000 (09:33 -0700)]
Merge tag 'pull-target-arm-20220509' of https://git.linaro.org/people/pmaydell/qemu-arm into staging

target-arm queue:
 * MAINTAINERS/.mailmap: update email for Leif Lindholm
 * hw/arm: add version information to sbsa-ref machine DT
 * Enable new features for -cpu max:
   FEAT_Debugv8p2, FEAT_Debugv8p4, FEAT_RAS (minimal version only),
   FEAT_IESB, FEAT_CSV2, FEAT_CSV2_2, FEAT_CSV3, FEAT_DGH
 * Emulate Cortex-A76
 * Emulate Neoverse-N1
 * Fix the virt board default NUMA topology

# -----BEGIN PGP SIGNATURE-----
#
# iQJNBAABCAA3FiEE4aXFk81BneKOgxXPPCUl7RQ2DN4FAmJ5AbsZHHBldGVyLm1h
# eWRlbGxAbGluYXJvLm9yZwAKCRA8JSXtFDYM3vyFEACZZ6tRVJYB6YpIzI7rho9x
# hVQIMTc4D5lmVetJnbLdLazifIy60oIOtSKV3Y3oj5DLMcsf6NITrPaFPWNRX3Nm
# mcbTCT5FGj8i7b1CkpEylLwvRQbIaoz2GnJPckdYelxxAq1uJNog3fmoG8nVtJ1F
# HfXVCVkZGQyiyr6Y2/zn3vpdp9n6/4RymN8ugizkcgIRII87DKV+DNDalw613JG4
# 5xxBOGkYzo5DZM8TgL8Ylmb5Jy9XY0EN1xpkyHFOg6gi0B3UZTxHq5SvK6NFoZLJ
# ogyhmMh6IjEfhUIDCtWG9VCoPyWpOXAFoh7D7akFVB4g2SIvBvcuGzFxCAsh5q3K
# s+9CgNX1SZpJQkT1jLjQlNzoUhh8lNc7QvhPWVrbAj3scc+1xVnS5MJsokEV21Cx
# /bp3mFwCL+Q4gjsMKx1nKSvxLv8xlxRtIilmlfj+wvpkenIfIwHYjbvItJTlAy1L
# +arx8fqImNQorxO6oMjOuAlSbNnDKup5qvwGghyu/qz/YEnGQVzN6gI324Km081L
# 1u31H/B3C2rj3qMsYMp5yOqgprXi1D5c6wfYIpLD/C4UfHgIlRiprawZPDM7fAhX
# vxhUhhj3e9OgkbC9yqd6SUR2Uk3YaQlp319LyoZa3VKSvjBTciFsMXXnIV1UitYp
# BGtz8+FypPVkYH7zQB9c7Q==
# =ey1m
# -----END PGP SIGNATURE-----
# gpg: Signature made Mon 09 May 2022 04:57:47 AM PDT
# gpg:                using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg:                issuer "peter.maydell@linaro.org"
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>" [full]
# gpg:                 aka "Peter Maydell <pmaydell@gmail.com>" [full]
# gpg:                 aka "Peter Maydell <pmaydell@chiark.greenend.org.uk>" [full]

* tag 'pull-target-arm-20220509' of https://git.linaro.org/people/pmaydell/qemu-arm: (32 commits)
  hw/acpi/aml-build: Use existing CPU topology to build PPTT table
  hw/arm/virt: Fix CPU's default NUMA node ID
  qtest/numa-test: Correct CPU and NUMA association in aarch64_numa_cpu()
  hw/arm/virt: Consider SMP configuration in CPU topology
  qtest/numa-test: Specify CPU topology in aarch64_numa_cpu()
  qapi/machine.json: Add cluster-id
  hw/arm: add versioning to sbsa-ref machine DT
  target/arm: Define neoverse-n1
  target/arm: Define cortex-a76
  target/arm: Enable FEAT_DGH for -cpu max
  target/arm: Enable FEAT_CSV3 for -cpu max
  target/arm: Enable FEAT_CSV2_2 for -cpu max
  target/arm: Enable FEAT_CSV2 for -cpu max
  target/arm: Enable FEAT_IESB for -cpu max
  target/arm: Enable FEAT_RAS for -cpu max
  target/arm: Implement ESB instruction
  target/arm: Implement virtual SError exceptions
  target/arm: Enable SCR and HCR bits for RAS
  target/arm: Add minimal RAS registers
  target/arm: Enable FEAT_Debugv8p4 for -cpu max
  ...

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agohw/acpi/aml-build: Use existing CPU topology to build PPTT table
Gavin Shan [Tue, 3 May 2022 14:03:04 +0000 (22:03 +0800)]
hw/acpi/aml-build: Use existing CPU topology to build PPTT table

When the PPTT table is built, the CPU topology is re-calculated, but
it's unecessary because the CPU topology has been populated in
virt_possible_cpu_arch_ids() on arm/virt machine.

This reworks build_pptt() to avoid by reusing the existing IDs in
ms->possible_cpus. Currently, the only user of build_pptt() is
arm/virt machine.

Signed-off-by: Gavin Shan <gshan@redhat.com>
Tested-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Yanan Wang <wangyanan55@huawei.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Message-id: 20220503140304.855514-7-gshan@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agohw/arm/virt: Fix CPU's default NUMA node ID
Gavin Shan [Tue, 3 May 2022 14:03:03 +0000 (22:03 +0800)]
hw/arm/virt: Fix CPU's default NUMA node ID

When CPU-to-NUMA association isn't explicitly provided by users,
the default one is given by mc->get_default_cpu_node_id(). However,
the CPU topology isn't fully considered in the default association
and this causes CPU topology broken warnings on booting Linux guest.

For example, the following warning messages are observed when the
Linux guest is booted with the following command lines.

  /home/gavin/sandbox/qemu.main/build/qemu-system-aarch64 \
  -accel kvm -machine virt,gic-version=host               \
  -cpu host                                               \
  -smp 6,sockets=2,cores=3,threads=1                      \
  -m 1024M,slots=16,maxmem=64G                            \
  -object memory-backend-ram,id=mem0,size=128M            \
  -object memory-backend-ram,id=mem1,size=128M            \
  -object memory-backend-ram,id=mem2,size=128M            \
  -object memory-backend-ram,id=mem3,size=128M            \
  -object memory-backend-ram,id=mem4,size=128M            \
  -object memory-backend-ram,id=mem4,size=384M            \
  -numa node,nodeid=0,memdev=mem0                         \
  -numa node,nodeid=1,memdev=mem1                         \
  -numa node,nodeid=2,memdev=mem2                         \
  -numa node,nodeid=3,memdev=mem3                         \
  -numa node,nodeid=4,memdev=mem4                         \
  -numa node,nodeid=5,memdev=mem5
         :
  alternatives: patching kernel code
  BUG: arch topology borken
  the CLS domain not a subset of the MC domain
  <the above error log repeats>
  BUG: arch topology borken
  the DIE domain not a subset of the NODE domain

With current implementation of mc->get_default_cpu_node_id(),
CPU#0 to CPU#5 are associated with NODE#0 to NODE#5 separately.
That's incorrect because CPU#0/1/2 should be associated with same
NUMA node because they're seated in same socket.

This fixes the issue by considering the socket ID when the default
CPU-to-NUMA association is provided in virt_possible_cpu_arch_ids().
With this applied, no more CPU topology broken warnings are seen
from the Linux guest. The 6 CPUs are associated with NODE#0/1, but
there are no CPUs associated with NODE#2/3/4/5.

Signed-off-by: Gavin Shan <gshan@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Yanan Wang <wangyanan55@huawei.com>
Message-id: 20220503140304.855514-6-gshan@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoqtest/numa-test: Correct CPU and NUMA association in aarch64_numa_cpu()
Gavin Shan [Tue, 3 May 2022 14:03:02 +0000 (22:03 +0800)]
qtest/numa-test: Correct CPU and NUMA association in aarch64_numa_cpu()

In aarch64_numa_cpu(), the CPU and NUMA association is something
like below. Two threads in the same core/cluster/socket are
associated with two individual NUMA nodes, which is unreal as
Igor Mammedov mentioned. We don't expect the association to break
NUMA-to-socket boundary, which matches with the real world.

    NUMA-node  socket  cluster   core   thread
    ------------------------------------------
        0       0        0        0      0
        1       0        0        0      1

This corrects the topology for CPUs and their association with
NUMA nodes. After this patch is applied, the CPU and NUMA
association becomes something like below, which looks real.
Besides, socket/cluster/core/thread IDs are all checked when
the NUMA node IDs are verified. It helps to check if the CPU
topology is properly populated or not.

    NUMA-node  socket  cluster   core   thread
    ------------------------------------------
       0        1        0        0       0
       1        0        0        0       0

Suggested-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Gavin Shan <gshan@redhat.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-id: 20220503140304.855514-5-gshan@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agohw/arm/virt: Consider SMP configuration in CPU topology
Gavin Shan [Tue, 3 May 2022 14:03:01 +0000 (22:03 +0800)]
hw/arm/virt: Consider SMP configuration in CPU topology

Currently, the SMP configuration isn't considered when the CPU
topology is populated. In this case, it's impossible to provide
the default CPU-to-NUMA mapping or association based on the socket
ID of the given CPU.

This takes account of SMP configuration when the CPU topology
is populated. The die ID for the given CPU isn't assigned since
it's not supported on arm/virt machine. Besides, the used SMP
configuration in qtest/numa-test/aarch64_numa_cpu() is corrcted
to avoid testing failure

Signed-off-by: Gavin Shan <gshan@redhat.com>
Reviewed-by: Yanan Wang <wangyanan55@huawei.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-id: 20220503140304.855514-4-gshan@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoqtest/numa-test: Specify CPU topology in aarch64_numa_cpu()
Gavin Shan [Tue, 3 May 2022 14:03:00 +0000 (22:03 +0800)]
qtest/numa-test: Specify CPU topology in aarch64_numa_cpu()

The CPU topology isn't enabled on arm/virt machine yet, but we're
going to do it in next patch. After the CPU topology is enabled by
next patch, "thread-id=1" becomes invalid because the CPU core is
preferred on arm/virt machine. It means these two CPUs have 0/1
as their core IDs, but their thread IDs are all 0. It will trigger
test failure as the following message indicates:

  [14/21 qemu:qtest+qtest-aarch64 / qtest-aarch64/numa-test  ERROR
  1.48s   killed by signal 6 SIGABRT
  >>> G_TEST_DBUS_DAEMON=/home/gavin/sandbox/qemu.main/tests/dbus-vmstate-daemon.sh \
      QTEST_QEMU_STORAGE_DAEMON_BINARY=./storage-daemon/qemu-storage-daemon         \
      QTEST_QEMU_BINARY=./qemu-system-aarch64                                       \
      QTEST_QEMU_IMG=./qemu-img MALLOC_PERTURB_=83                                  \
      /home/gavin/sandbox/qemu.main/build/tests/qtest/numa-test --tap -k
  ――――――――――――――――――――――――――――――――――――――――――――――
  stderr:
  qemu-system-aarch64: -numa cpu,node-id=0,thread-id=1: no match found

This fixes the issue by providing comprehensive SMP configurations
in aarch64_numa_cpu(). The SMP configurations aren't used before
the CPU topology is enabled in next patch.

Signed-off-by: Gavin Shan <gshan@redhat.com>
Reviewed-by: Yanan Wang <wangyanan55@huawei.com>
Message-id: 20220503140304.855514-3-gshan@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoqapi/machine.json: Add cluster-id
Gavin Shan [Tue, 3 May 2022 14:02:59 +0000 (22:02 +0800)]
qapi/machine.json: Add cluster-id

This adds cluster-id in CPU instance properties, which will be used
by arm/virt machine. Besides, the cluster-id is also verified or
dumped in various spots:

  * hw/core/machine.c::machine_set_cpu_numa_node() to associate
    CPU with its NUMA node.

  * hw/core/machine.c::machine_numa_finish_cpu_init() to record
    CPU slots with no NUMA mapping set.

  * hw/core/machine-hmp-cmds.c::hmp_hotpluggable_cpus() to dump
    cluster-id.

Signed-off-by: Gavin Shan <gshan@redhat.com>
Reviewed-by: Yanan Wang <wangyanan55@huawei.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
Message-id: 20220503140304.855514-2-gshan@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agohw/arm: add versioning to sbsa-ref machine DT
Leif Lindholm [Thu, 5 May 2022 11:39:47 +0000 (12:39 +0100)]
hw/arm: add versioning to sbsa-ref machine DT

The sbsa-ref machine is continuously evolving. Some of the changes we
want to make in the near future, to align with real components (e.g.
the GIC-700), will break compatibility for existing firmware.

Introduce two new properties to the DT generated on machine generation:
- machine-version-major
  To be incremented when a platform change makes the machine
  incompatible with existing firmware.
- machine-version-minor
  To be incremented when functionality is added to the machine
  without causing incompatibility with existing firmware.
  to be reset to 0 when machine-version-major is incremented.

This versioning scheme is *neither*:
- A QEMU versioned machine type; a given version of QEMU will emulate
  a given version of the platform.
- A reflection of level of SBSA (now SystemReady SR) support provided.

The version will increment on guest-visible functional changes only,
akin to a revision ID register found on a physical platform.

These properties are both introduced with the value 0.
(Hence, a machine where the DT is lacking these nodes is equivalent
to version 0.0.)

Signed-off-by: Leif Lindholm <quic_llindhol@quicinc.com>
Message-id: 20220505113947.75714-1-quic_llindhol@quicinc.com
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Radoslaw Biernacki <rad@semihalf.com>
Cc: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Define neoverse-n1
Richard Henderson [Fri, 6 May 2022 18:02:42 +0000 (13:02 -0500)]
target/arm: Define neoverse-n1

Enable the n1 for virt and sbsa board use.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-25-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Define cortex-a76
Richard Henderson [Fri, 6 May 2022 18:02:41 +0000 (13:02 -0500)]
target/arm: Define cortex-a76

Enable the a76 for virt and sbsa board use.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-24-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_DGH for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:40 +0000 (13:02 -0500)]
target/arm: Enable FEAT_DGH for -cpu max

This extension concerns not merging memory access, which TCG does
not implement.  Thus we can trivially enable this feature.
Add a comment to handle_hint for the DGH instruction, but no code.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-23-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_CSV3 for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:39 +0000 (13:02 -0500)]
target/arm: Enable FEAT_CSV3 for -cpu max

This extension concerns cache speculation, which TCG does
not implement.  Thus we can trivially enable this feature.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-22-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_CSV2_2 for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:38 +0000 (13:02 -0500)]
target/arm: Enable FEAT_CSV2_2 for -cpu max

There is no branch prediction in TCG, therefore there is no
need to actually include the context number into the predictor.
Therefore all we need to do is add the state for SCXTNUM_ELx.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-21-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_CSV2 for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:37 +0000 (13:02 -0500)]
target/arm: Enable FEAT_CSV2 for -cpu max

This extension concerns branch speculation, which TCG does
not implement.  Thus we can trivially enable this feature.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-20-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_IESB for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:36 +0000 (13:02 -0500)]
target/arm: Enable FEAT_IESB for -cpu max

This feature is AArch64 only, and applies to physical SErrors,
which QEMU does not implement, thus the feature is a nop.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-19-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_RAS for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:35 +0000 (13:02 -0500)]
target/arm: Enable FEAT_RAS for -cpu max

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-18-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Implement ESB instruction
Richard Henderson [Fri, 6 May 2022 18:02:34 +0000 (13:02 -0500)]
target/arm: Implement ESB instruction

Check for and defer any pending virtual SError.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-17-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Implement virtual SError exceptions
Richard Henderson [Fri, 6 May 2022 18:02:33 +0000 (13:02 -0500)]
target/arm: Implement virtual SError exceptions

Virtual SError exceptions are raised by setting HCR_EL2.VSE,
and are routed to EL1 just like other virtual exceptions.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-16-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable SCR and HCR bits for RAS
Richard Henderson [Fri, 6 May 2022 18:02:32 +0000 (13:02 -0500)]
target/arm: Enable SCR and HCR bits for RAS

Enable writes to the TERR and TEA bits when RAS is enabled.
These bits are otherwise RES0.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-15-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Add minimal RAS registers
Richard Henderson [Fri, 6 May 2022 18:02:31 +0000 (13:02 -0500)]
target/arm: Add minimal RAS registers

Add only the system registers required to implement zero error
records.  This means that all values for ERRSELR are out of range,
which means that it and all of the indexed error record registers
need not be implemented.

Add the EL2 registers required for injecting virtual SError.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-14-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_Debugv8p4 for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:30 +0000 (13:02 -0500)]
target/arm: Enable FEAT_Debugv8p4 for -cpu max

This extension concerns changes to the External Debug interface,
with Secure and Non-secure access to the debug registers, and all
of it is outside the scope of QEMU.  Indicating support for this
is mandatory with FEAT_SEL2, which we do implement.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-13-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Enable FEAT_Debugv8p2 for -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:29 +0000 (13:02 -0500)]
target/arm: Enable FEAT_Debugv8p2 for -cpu max

The only portion of FEAT_Debugv8p2 that is relevant to QEMU
is CONTEXTIDR_EL2, which is also conditionally implemented
with FEAT_VHE.  The rest of the debug extension concerns the
External debug interface, which is outside the scope of QEMU.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-12-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Use field names for manipulating EL2 and EL3 modes
Richard Henderson [Fri, 6 May 2022 18:02:28 +0000 (13:02 -0500)]
target/arm: Use field names for manipulating EL2 and EL3 modes

Use FIELD_DP{32,64} to manipulate id_pfr1 and id_aa64pfr0
during arm_cpu_realizefn.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-11-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Annotate arm_max_initfn with FEAT identifiers
Richard Henderson [Fri, 6 May 2022 18:02:27 +0000 (13:02 -0500)]
target/arm: Annotate arm_max_initfn with FEAT identifiers

Update the legacy feature names to the current names.
Provide feature names for id changes that were not marked.
Sort the field updates into increasing bitfield order.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-10-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Split out aa32_max_features
Richard Henderson [Fri, 6 May 2022 18:02:26 +0000 (13:02 -0500)]
target/arm: Split out aa32_max_features

Share the code to set AArch32 max features so that we no
longer have code drift between qemu{-system,}-{arm,aarch64}.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-9-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Set ID_DFR0.PerfMon for qemu-system-arm -cpu max
Richard Henderson [Fri, 6 May 2022 18:02:25 +0000 (13:02 -0500)]
target/arm: Set ID_DFR0.PerfMon for qemu-system-arm -cpu max

We set this for qemu-system-aarch64, but failed to do so
for the strictly 32-bit emulation.

Fixes: 3bec78447a9 ("target/arm: Provide ARMv8.4-PMU in '-cpu max'")
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-8-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Update qemu-system-arm -cpu max to cortex-a57
Richard Henderson [Fri, 6 May 2022 18:02:24 +0000 (13:02 -0500)]
target/arm: Update qemu-system-arm -cpu max to cortex-a57

Instead of starting with cortex-a15 and adding v8 features to
a v7 cpu, begin with a v8 cpu stripped of its aarch64 features.
This fixes the long-standing to-do where we only enabled v8
features for user-only.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-7-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Move cortex impdef sysregs to cpu_tcg.c
Richard Henderson [Fri, 6 May 2022 18:02:23 +0000 (13:02 -0500)]
target/arm: Move cortex impdef sysregs to cpu_tcg.c

Previously we were defining some of these in user-only mode,
but none of them are accessible from user-only, therefore
define them only in system mode.

This will shortly be used from cpu_tcg.c also.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-6-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Adjust definition of CONTEXTIDR_EL2
Richard Henderson [Fri, 6 May 2022 18:02:22 +0000 (13:02 -0500)]
target/arm: Adjust definition of CONTEXTIDR_EL2

This register is present for either VHE or Debugv8p2.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-5-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotarget/arm: Merge zcr reginfo
Richard Henderson [Fri, 6 May 2022 18:02:21 +0000 (13:02 -0500)]
target/arm: Merge zcr reginfo

Drop zcr_no_el2_reginfo and merge the 3 registers into one array,
now that ZCR_EL2 can be squashed to RES0 and ZCR_EL3 dropped
while registering.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20220506180242.216785-4-richard.henderson@linaro.org
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>