qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v7 15/15] hw/vmapple/vmapple: Add vmapple machine type


From: Akihiko Odaki
Subject: Re: [PATCH v7 15/15] hw/vmapple/vmapple: Add vmapple machine type
Date: Wed, 6 Nov 2024 18:16:08 +0900
User-agent: Mozilla Thunderbird

On 2024/11/06 18:10, Phil Dennis-Jordan wrote:


On Wed, 6 Nov 2024 at 08:42, Akihiko Odaki <akihiko.odaki@daynix.com <mailto:akihiko.odaki@daynix.com>> wrote:

    On 2024/11/06 0:30, Phil Dennis-Jordan wrote:
     > From: Alexander Graf <graf@amazon.com <mailto:graf@amazon.com>>
     >
     > Apple defines a new "vmapple" machine type as part of its proprietary
     > macOS Virtualization.Framework vmm. This machine type is similar
    to the
     > virt one, but with subtle differences in base devices, a few special
     > vmapple device additions and a vastly different boot chain.
     >
     > This patch reimplements this machine type in QEMU. To use it, you
     > have to have a readily installed version of macOS for VMApple,
     > run on macOS with -accel hvf, pass the Virtualization.Framework
     > boot rom (AVPBooter) in via -bios, pass the aux and root volume
    as pflash
     > and pass aux and root volume as virtio drives. In addition, you also
     > need to find the machine UUID and pass that as -M vmapple,uuid=
    parameter:
     >
     > $ qemu-system-aarch64 -accel hvf -M vmapple,uuid=0x1234 -m 4G \
     >      -bios /System/Library/Frameworks/Virtualization.framework/
    Versions/A/Resources/AVPBooter.vmapple2.bin
     >      -drive file=aux,if=pflash,format=raw \
     >      -drive file=root,if=pflash,format=raw \
     >      -drive file=aux,if=none,id=aux,format=raw \
     >      -device vmapple-virtio-aux,drive=aux \
     >      -drive file=root,if=none,id=root,format=raw \
     >      -device vmapple-virtio-root,drive=root
     >
     > With all these in place, you should be able to see macOS booting
     > successfully.
     >
     > Known issues:
     >   - Keyboard and mouse/tablet input is laggy. The reason for this is
     >     either that macOS's XHCI driver is broken when the device/
    platform
     >     does not support MSI/MSI-X, or there's some unfortunate interplay
     >     with Qemu's XHCI implementation in this scenario.
     >   - Currently only macOS 12 guests are supported. The boot
    process for
     >     13+ will need further investigation and adjustment.
     >
     > Signed-off-by: Alexander Graf <graf@amazon.com
    <mailto:graf@amazon.com>>
     > Co-authored-by: Phil Dennis-Jordan <phil@philjordan.eu
    <mailto:phil@philjordan.eu>>
     > Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu
    <mailto:phil@philjordan.eu>>
     > ---
     > v3:
     >   * Rebased on latest upstream, updated affinity and NIC creation
     > API usage
     >   * Included Apple-variant virtio-blk in build dependency
     >   * Updated API usage for setting 'redist-region-count' array-
    typed property on GIC.
     >   * Switched from virtio HID devices (for which macOS 12 does not
    contain
     >     drivers) to an XHCI USB controller and USB HID devices.
     >
     > v4:
     >   * Fixups for v4 changes to the other patches in the set.
     >   * Corrected the assert macro to use
     >   * Removed superfluous endian conversions corresponding to cfg's.
     >   * Init error handling improvement.
     >   * No need to select CPU type on TCG, as only HVF is supported.
     >   * Machine type version bumped to 9.2
     >   * #include order improved
     >
     > v5:
     >   * Fixed memory reservation for ecam alias region.
     >   * Better error handling setting properties on devices.
     >   * Simplified the machine ECID/UUID extraction script and
    actually created a
     >     file for it rather than quoting its code in documentation.
     >
     > v7:
     >   * Tiny error handling fix, un-inlined function.
     >
     >   MAINTAINERS                 |   1 +
     >   contrib/vmapple/uuid.sh     |   9 +
     >   docs/system/arm/vmapple.rst |  60 ++++
     >   docs/system/target-arm.rst  |   1 +
     >   hw/vmapple/Kconfig          |  20 ++
     >   hw/vmapple/meson.build      |   1 +
     >   hw/vmapple/vmapple.c        | 659 +++++++++++++++++++++++++++++
    +++++++
     >   7 files changed, 751 insertions(+)
     >   create mode 100755 contrib/vmapple/uuid.sh
     >   create mode 100644 docs/system/arm/vmapple.rst
     >   create mode 100644 hw/vmapple/vmapple.c
     >
     > diff --git a/MAINTAINERS b/MAINTAINERS
     > index 9d0d26cb65f..9591fd44a34 100644
     > --- a/MAINTAINERS
     > +++ b/MAINTAINERS
     > @@ -2767,6 +2767,7 @@ R: Phil Dennis-Jordan <phil@philjordan.eu
    <mailto:phil@philjordan.eu>>
     >   S: Maintained
     >   F: hw/vmapple/*
     >   F: include/hw/vmapple/*
     > +F: docs/system/arm/vmapple.rst
     >
     >   Subsystems
     >   ----------
     > diff --git a/contrib/vmapple/uuid.sh b/contrib/vmapple/uuid.sh
     > new file mode 100755
     > index 00000000000..956e8c3afed
     > --- /dev/null
     > +++ b/contrib/vmapple/uuid.sh
     > @@ -0,0 +1,9 @@
     > +#!/bin/sh
     > +# Used for converting a guest provisioned using
    Virtualization.framework
     > +# for use with the QEMU 'vmapple' aarch64 machine type.
     > +#
     > +# Extracts the Machine UUID from Virtualization.framework VM
    JSON file.
     > +# (as produced by 'macosvm', passed as command line argument)
     > +
     > +plutil -extract machineId raw "$1" | base64 -d | plutil -extract
    ECID raw -
     > +
     > diff --git a/docs/system/arm/vmapple.rst b/docs/system/arm/
    vmapple.rst
     > new file mode 100644
     > index 00000000000..6a634fa4572
     > --- /dev/null
     > +++ b/docs/system/arm/vmapple.rst
     > @@ -0,0 +1,60 @@
     > +VMApple machine emulation
     >
    
+========================================================================================
     > +
     > +VMApple is the device model that the macOS built-in hypervisor
    called "Virtualization.framework"
     > +exposes to Apple Silicon macOS guests. The "vmapple" machine
    model in QEMU implements the same
     > +device model, but does not use any code from
    Virtualization.Framework.
     > +
     > +Prerequisites
     > +-------------
     > +
     > +To run the vmapple machine model, you need to
     > +
     > + * Run on Apple Silicon
     > + * Run on macOS 12.0 or above
     > + * Have an already installed copy of a Virtualization.Framework
    macOS 12 virtual machine. I will
     > +   assume that you installed it using the macosvm CLI.
     > +
     > +First, we need to extract the UUID from the virtual machine that
    you installed. You can do this
     > +by running the shell script in contrib/vmapple/uuid.sh on the
    macosvm.json file.
     > +
     > +.. code-block:: bash
     > +  :caption: uuid.sh script to extract the UUID from a
    macosvm.json file
     > +
     > +  $ contrib/vmapple/uuid.sh "path/to/macosvm.json"
     > +
     > +Now we also need to trim the aux partition. It contains metadata
    that we can just discard:
     > +
     > +.. code-block:: bash
     > +  :caption: Command to trim the aux file
     > +
     > +  $ dd if="aux.img" of="aux.img.trimmed" bs=$(( 0x4000 )) skip=1
     > +
     > +How to run
     > +----------
     > +
     > +Then, we can launch QEMU with the Virtualization.Framework pre-
    boot environment and the readily
     > +installed target disk images. I recommend to port forward the
    VM's ssh and vnc ports to the host
     > +to get better interactive access into the target system:
     > +
     > +.. code-block:: bash
     > +  :caption: Example execution command line
     > +
     > +  $ UUID=$(uuid.sh macosvm.json)
     > +  $ AVPBOOTER=/System/Library/Frameworks/
    Virtualization.framework/Resources/AVPBooter.vmapple2.bin
     > +  $ AUX=aux.img.trimmed
     > +  $ DISK=disk.img
     > +  $ qemu-system-aarch64 \
     > +       -serial mon:stdio \
     > +       -m 4G \
     > +       -accel hvf \
     > +       -M vmapple,uuid=$UUID \
     > +       -bios $AVPBOOTER \
     > +        -drive file="$AUX",if=pflash,format=raw \
     > +        -drive file="$DISK",if=pflash,format=raw \
     > +       -drive file="$AUX",if=none,id=aux,format=raw \
     > +       -drive file="$DISK",if=none,id=root,format=raw \
     > +       -device vmapple-virtio-aux,drive=aux \
     > +       -device vmapple-virtio-root,drive=root \
     > +       -net
    user,ipv6=off,hostfwd=tcp::2222-:22,hostfwd=tcp::5901-:5900 \
     > +       -net nic,model=virtio-net-pci \
     > diff --git a/docs/system/target-arm.rst b/docs/system/target-arm.rst
     > index 9aaa9c414c9..3426f79100b 100644
     > --- a/docs/system/target-arm.rst
     > +++ b/docs/system/target-arm.rst
     > @@ -102,6 +102,7 @@ Board-specific documentation
     >      arm/stellaris
     >      arm/stm32
     >      arm/virt
     > +   arm/vmapple
     >      arm/xenpvh
     >      arm/xlnx-versal-virt
     >      arm/xlnx-zynq
     > diff --git a/hw/vmapple/Kconfig b/hw/vmapple/Kconfig
     > index bcd1be63e3c..6a4c4a7fa2e 100644
     > --- a/hw/vmapple/Kconfig
     > +++ b/hw/vmapple/Kconfig
     > @@ -10,3 +10,23 @@ config VMAPPLE_CFG
     >   config VMAPPLE_VIRTIO_BLK
     >       bool
     >
     > +config VMAPPLE
     > +    bool
     > +    depends on ARM
     > +    depends on HVF
     > +    default y if ARM
     > +    imply PCI_DEVICES
     > +    select ARM_GIC
     > +    select PLATFORM_BUS
     > +    select PCI_EXPRESS
     > +    select PCI_EXPRESS_GENERIC_BRIDGE
     > +    select PL011 # UART
     > +    select PL031 # RTC
     > +    select PL061 # GPIO
     > +    select GPIO_PWR
     > +    select PVPANIC_MMIO
     > +    select VMAPPLE_AES
     > +    select VMAPPLE_BDIF
     > +    select VMAPPLE_CFG
     > +    select MAC_PVG_MMIO
     > +    select VMAPPLE_VIRTIO_BLK
     > diff --git a/hw/vmapple/meson.build b/hw/vmapple/meson.build
     > index bf17cf906c9..e572f7d5602 100644
     > --- a/hw/vmapple/meson.build
     > +++ b/hw/vmapple/meson.build
> @@ -2,3 +2,4 @@ system_ss.add(when: 'CONFIG_VMAPPLE_AES', if_true: files('aes.c'))
     >   system_ss.add(when: 'CONFIG_VMAPPLE_BDIF', if_true:
    files('bdif.c'))
     >   system_ss.add(when: 'CONFIG_VMAPPLE_CFG',  if_true: files('cfg.c'))
     >   system_ss.add(when: 'CONFIG_VMAPPLE_VIRTIO_BLK',  if_true:
    files('virtio-blk.c'))
     > +specific_ss.add(when: 'CONFIG_VMAPPLE',     if_true:
    files('vmapple.c'))
     > diff --git a/hw/vmapple/vmapple.c b/hw/vmapple/vmapple.c
     > new file mode 100644
     > index 00000000000..aac3ec35856
     > --- /dev/null
     > +++ b/hw/vmapple/vmapple.c
     > @@ -0,0 +1,659 @@
     > +/*
     > + * VMApple machine emulation
     > + *
     > + * Copyright © 2023 Amazon.com, Inc. or its affiliates. All
    Rights Reserved.
     > + *
     > + * This work is licensed under the terms of the GNU GPL, version
    2 or later.
     > + * See the COPYING file in the top-level directory.
     > + *
     > + * SPDX-License-Identifier: GPL-2.0-or-later
     > + *
     > + * VMApple is the device model that the macOS built-in
    hypervisor called
     > + * "Virtualization.framework" exposes to Apple Silicon macOS
    guests. The
     > + * machine model in this file implements the same device model
    in QEMU, but
     > + * does not use any code from Virtualization.Framework.
     > + */
     > +
     > +#include "qemu/osdep.h"
     > +#include "qemu/bitops.h"
     > +#include "qemu/datadir.h"
     > +#include "qemu/error-report.h"
     > +#include "qemu/guest-random.h"
     > +#include "qemu/help-texts.h"
     > +#include "qemu/log.h"
     > +#include "qemu/module.h"
     > +#include "qemu/option.h"
     > +#include "qemu/units.h"
     > +#include "monitor/qdev.h"
     > +#include "hw/boards.h"
     > +#include "hw/irq.h"
     > +#include "hw/loader.h"
     > +#include "hw/qdev-properties.h"
     > +#include "hw/sysbus.h"
     > +#include "hw/usb.h"
     > +#include "hw/arm/boot.h"
     > +#include "hw/arm/primecell.h"
     > +#include "hw/char/pl011.h"
     > +#include "hw/intc/arm_gic.h"
     > +#include "hw/intc/arm_gicv3_common.h"
     > +#include "hw/misc/pvpanic.h"
     > +#include "hw/pci-host/gpex.h"
     > +#include "hw/usb/xhci.h"
     > +#include "hw/virtio/virtio-pci.h"
     > +#include "hw/vmapple/vmapple.h"
     > +#include "net/net.h"
     > +#include "qapi/error.h"
     > +#include "qapi/qmp/qlist.h"
     > +#include "qapi/visitor.h"
     > +#include "qapi/qapi-visit-common.h"
     > +#include "standard-headers/linux/input.h"
     > +#include "sysemu/hvf.h"
     > +#include "sysemu/kvm.h"
     > +#include "sysemu/reset.h"
     > +#include "sysemu/runstate.h"
     > +#include "sysemu/sysemu.h"
     > +#include "target/arm/internals.h"
     > +#include "target/arm/kvm_arm.h"
     > +
     > +struct VMAppleMachineClass {
     > +    MachineClass parent;
     > +};
     > +
     > +struct VMAppleMachineState {
     > +    MachineState parent;
     > +
     > +    Notifier machine_done;
     > +    struct arm_boot_info bootinfo;
     > +    MemMapEntry *memmap;
     > +    const int *irqmap;
     > +    DeviceState *gic;
     > +    DeviceState *cfg;
     > +    Notifier powerdown_notifier;
     > +    PCIBus *bus;
     > +    MemoryRegion fw_mr;
     > +    MemoryRegion ecam_alias;
     > +    uint64_t uuid;
     > +};
     > +
     > +#define DEFINE_VMAPPLE_MACHINE_LATEST(major, minor, latest) \
     > +    static void
    vmapple##major##_##minor##_class_init(ObjectClass *oc, \
     > +                                                    void *data) \
     > +    { \
     > +        MachineClass *mc = MACHINE_CLASS(oc); \
     > +        vmapple_machine_##major##_##minor##_options(mc); \
     > +        mc->desc = "QEMU " # major "." # minor " Apple Virtual
    Machine"; \
     > +        if (latest) { \
     > +            mc->alias = "vmapple"; \
     > +        } \
     > +    } \
     > +    static const TypeInfo machvmapple##major##_##minor##_info = { \
     > +        .name = MACHINE_TYPE_NAME("vmapple-" # major "." # minor), \
     > +        .parent = TYPE_VMAPPLE_MACHINE, \
     > +        .class_init = vmapple##major##_##minor##_class_init, \
     > +    }; \
     > +    static void machvmapple_machine_##major##_##minor##_init(void) \
     > +    { \
> + type_register_static(&machvmapple##major##_##minor##_info); \
     > +    } \
     > +    type_init(machvmapple_machine_##major##_##minor##_init);
     > +
     > +#define DEFINE_VMAPPLE_MACHINE_AS_LATEST(major, minor) \
     > +    DEFINE_VMAPPLE_MACHINE_LATEST(major, minor, true)
     > +#define DEFINE_VMAPPLE_MACHINE(major, minor) \
     > +    DEFINE_VMAPPLE_MACHINE_LATEST(major, minor, false)
     > +
     > +#define TYPE_VMAPPLE_MACHINE   MACHINE_TYPE_NAME("vmapple")
     > +OBJECT_DECLARE_TYPE(VMAppleMachineState, VMAppleMachineClass,
    VMAPPLE_MACHINE)
     > +
     > +/* Number of external interrupt lines to configure the GIC with */
     > +#define NUM_IRQS 256
     > +
     > +enum {
     > +    VMAPPLE_FIRMWARE,
     > +    VMAPPLE_CONFIG,
     > +    VMAPPLE_MEM,
     > +    VMAPPLE_GIC_DIST,
     > +    VMAPPLE_GIC_REDIST,
     > +    VMAPPLE_UART,
     > +    VMAPPLE_RTC,
     > +    VMAPPLE_PCIE,
     > +    VMAPPLE_PCIE_MMIO,
     > +    VMAPPLE_PCIE_ECAM,
     > +    VMAPPLE_GPIO,
     > +    VMAPPLE_PVPANIC,
     > +    VMAPPLE_APV_GFX,
     > +    VMAPPLE_APV_IOSFC,
     > +    VMAPPLE_AES_1,
     > +    VMAPPLE_AES_2,
     > +    VMAPPLE_BDOOR,
     > +    VMAPPLE_MEMMAP_LAST,
     > +};
     > +
     > +static MemMapEntry memmap[] = {
     > +    [VMAPPLE_FIRMWARE] =           { 0x00100000, 0x00100000 },
     > +    [VMAPPLE_CONFIG] =             { 0x00400000, 0x00010000 },
     > +
     > +    [VMAPPLE_GIC_DIST] =           { 0x10000000, 0x00010000 },
     > +    [VMAPPLE_GIC_REDIST] =         { 0x10010000, 0x00400000 },
     > +
     > +    [VMAPPLE_UART] =               { 0x20010000, 0x00010000 },
     > +    [VMAPPLE_RTC] =                { 0x20050000, 0x00001000 },
     > +    [VMAPPLE_GPIO] =               { 0x20060000, 0x00001000 },
     > +    [VMAPPLE_PVPANIC] =            { 0x20070000, 0x00000002 },
     > +    [VMAPPLE_BDOOR] =              { 0x30000000, 0x00200000 },
     > +    [VMAPPLE_APV_GFX] =            { 0x30200000, 0x00010000 },
     > +    [VMAPPLE_APV_IOSFC] =          { 0x30210000, 0x00010000 },
     > +    [VMAPPLE_AES_1] =              { 0x30220000, 0x00004000 },
     > +    [VMAPPLE_AES_2] =              { 0x30230000, 0x00004000 },
     > +    [VMAPPLE_PCIE_ECAM] =          { 0x40000000, 0x10000000 },
     > +    [VMAPPLE_PCIE_MMIO] =          { 0x50000000, 0x1fff0000 },
     > +
     > +    /* Actual RAM size depends on configuration */
     > +    [VMAPPLE_MEM] =                { 0x70000000ULL, GiB},
     > +};
     > +
     > +static const int irqmap[] = {
     > +    [VMAPPLE_UART] = 1,
     > +    [VMAPPLE_RTC] = 2,
     > +    [VMAPPLE_GPIO] = 0x5,
     > +    [VMAPPLE_APV_IOSFC] = 0x10,
     > +    [VMAPPLE_APV_GFX] = 0x11,
     > +    [VMAPPLE_AES_1] = 0x12,
     > +    [VMAPPLE_PCIE] = 0x20,
     > +};
     > +
     > +#define GPEX_NUM_IRQS 16
     > +
     > +static void create_bdif(VMAppleMachineState *vms, MemoryRegion *mem)
     > +{
     > +    DeviceState *bdif;
     > +    SysBusDevice *bdif_sb;
     > +    DriveInfo *di_aux = drive_get(IF_PFLASH, 0, 0);
     > +    DriveInfo *di_root = drive_get(IF_PFLASH, 0, 1);
     > +
     > +    if (!di_aux) {
     > +        error_report("No AUX device. Please specify one as
    pflash drive.");
     > +        exit(1);
     > +    }
     > +
     > +    if (!di_root) {
     > +        /* Fall back to the first IF_VIRTIO device as root device */
     > +        di_root = drive_get(IF_VIRTIO, 0, 0);
     > +    }
     > +
     > +    if (!di_root) {
     > +        error_report("No root device. Please specify one as
    virtio drive.");
     > +        exit(1);
     > +    }
     > +
     > +    /* PV backdoor device */
     > +    bdif = qdev_new(TYPE_VMAPPLE_BDIF);
     > +    bdif_sb = SYS_BUS_DEVICE(bdif);
     > +    sysbus_mmio_map(bdif_sb, 0, vms->memmap[VMAPPLE_BDOOR].base);
     > +
     > +    qdev_prop_set_drive(DEVICE(bdif), "aux",
    blk_by_legacy_dinfo(di_aux));
     > +    qdev_prop_set_drive(DEVICE(bdif), "root",
    blk_by_legacy_dinfo(di_root));
     > +
     > +    sysbus_realize_and_unref(bdif_sb, &error_fatal);
     > +}
     > +
     > +static void create_pvpanic(VMAppleMachineState *vms,
    MemoryRegion *mem)
     > +{
     > +    SysBusDevice *cfg;
     > +
     > +    vms->cfg = qdev_new(TYPE_PVPANIC_MMIO_DEVICE);
     > +    cfg = SYS_BUS_DEVICE(vms->cfg);
     > +    sysbus_mmio_map(cfg, 0, vms->memmap[VMAPPLE_PVPANIC].base);
     > +
     > +    sysbus_realize_and_unref(cfg, &error_fatal);
     > +}
     > +
     > +static void create_cfg(VMAppleMachineState *vms, MemoryRegion *mem)
     > +{
     > +    SysBusDevice *cfg;
     > +    MachineState *machine = MACHINE(vms);
     > +    uint32_t rnd = 1;
     > +
     > +    vms->cfg = qdev_new(TYPE_VMAPPLE_CFG);
     > +    cfg = SYS_BUS_DEVICE(vms->cfg);
     > +    sysbus_mmio_map(cfg, 0, vms->memmap[VMAPPLE_CONFIG].base);
     > +
     > +    qemu_guest_getrandom_nofail(&rnd, sizeof(rnd));
     > +
     > +    qdev_prop_set_uint32(vms->cfg, "nr-cpus", machine->smp.cpus);
     > +    qdev_prop_set_uint64(vms->cfg, "ecid", vms->uuid);
     > +    qdev_prop_set_uint64(vms->cfg, "ram-size", machine->ram_size);
     > +    qdev_prop_set_uint32(vms->cfg, "rnd", rnd);
     > +
     > +    sysbus_realize_and_unref(cfg, &error_fatal);
     > +}
     > +
     > +static void create_gfx(VMAppleMachineState *vms, MemoryRegion *mem)
     > +{
     > +    int irq_gfx = vms->irqmap[VMAPPLE_APV_GFX];
     > +    int irq_iosfc = vms->irqmap[VMAPPLE_APV_IOSFC];
     > +    SysBusDevice *gfx;
     > +
     > +    gfx = SYS_BUS_DEVICE(qdev_new("apple-gfx-mmio"));
     > +    sysbus_mmio_map(gfx, 0, vms->memmap[VMAPPLE_APV_GFX].base);
     > +    sysbus_mmio_map(gfx, 1, vms->memmap[VMAPPLE_APV_IOSFC].base);
     > +    sysbus_connect_irq(gfx, 0, qdev_get_gpio_in(vms->gic, irq_gfx));
     > +    sysbus_connect_irq(gfx, 1, qdev_get_gpio_in(vms->gic,
    irq_iosfc));
     > +    sysbus_realize_and_unref(gfx, &error_fatal);
     > +}
     > +
     > +static void create_aes(VMAppleMachineState *vms, MemoryRegion *mem)
     > +{
     > +    int irq = vms->irqmap[VMAPPLE_AES_1];
     > +    SysBusDevice *aes;
     > +
     > +    aes = SYS_BUS_DEVICE(qdev_new(TYPE_APPLE_AES));
     > +    sysbus_mmio_map(aes, 0, vms->memmap[VMAPPLE_AES_1].base);
     > +    sysbus_mmio_map(aes, 1, vms->memmap[VMAPPLE_AES_2].base);
     > +    sysbus_connect_irq(aes, 0, qdev_get_gpio_in(vms->gic, irq));
     > +    sysbus_realize_and_unref(aes, &error_fatal);
     > +}
     > +
     > +static int arm_gic_ppi_index(int cpu_nr, int ppi_index)
     > +{
     > +    return NUM_IRQS + cpu_nr * GIC_INTERNAL + ppi_index;
     > +}
     > +
     > +static void create_gic(VMAppleMachineState *vms, MemoryRegion *mem)
     > +{
     > +    MachineState *ms = MACHINE(vms);
     > +    /* We create a standalone GIC */
     > +    SysBusDevice *gicbusdev;
     > +    QList *redist_region_count;
     > +    int i;
     > +    unsigned int smp_cpus = ms->smp.cpus;
     > +
     > +    vms->gic = qdev_new(gicv3_class_name());
     > +    qdev_prop_set_uint32(vms->gic, "revision", 3);
     > +    qdev_prop_set_uint32(vms->gic, "num-cpu", smp_cpus);
     > +    /*
     > +     * Note that the num-irq property counts both internal and
    external
     > +     * interrupts; there are always 32 of the former (mandated
    by GIC spec).
     > +     */
     > +    qdev_prop_set_uint32(vms->gic, "num-irq", NUM_IRQS + 32);
     > +
     > +    uint32_t redist0_capacity =
     > +                vms->memmap[VMAPPLE_GIC_REDIST].size /
    GICV3_REDIST_SIZE;
     > +    uint32_t redist0_count = MIN(smp_cpus, redist0_capacity);
     > +
     > +    redist_region_count = qlist_new();
     > +    qlist_append_int(redist_region_count, redist0_count);
     > +    qdev_prop_set_array(vms->gic, "redist-region-count",
    redist_region_count);
     > +
     > +    gicbusdev = SYS_BUS_DEVICE(vms->gic);
     > +    sysbus_realize_and_unref(gicbusdev, &error_fatal);
     > +    sysbus_mmio_map(gicbusdev, 0, vms-
     >memmap[VMAPPLE_GIC_DIST].base);
     > +    sysbus_mmio_map(gicbusdev, 1, vms-
     >memmap[VMAPPLE_GIC_REDIST].base);
     > +
     > +    /*
     > +     * Wire the outputs from each CPU's generic timer and the GICv3
     > +     * maintenance interrupt signal to the appropriate GIC PPI
    inputs,
     > +     * and the GIC's IRQ/FIQ/VIRQ/VFIQ interrupt outputs to the
    CPU's inputs.
     > +     */
     > +    for (i = 0; i < smp_cpus; i++) {
     > +        DeviceState *cpudev = DEVICE(qemu_get_cpu(i));
     > +
     > +        /* Map the virt timer to PPI 27 */
     > +        qdev_connect_gpio_out(cpudev, GTIMER_VIRT,
     > +                              qdev_get_gpio_in(vms->gic,
> +  arm_gic_ppi_index(i, 27)));
     > +
     > +        /* Map the GIC IRQ and FIQ lines to CPU */
     > +        sysbus_connect_irq(gicbusdev, i,
    qdev_get_gpio_in(cpudev, ARM_CPU_IRQ));
     > +        sysbus_connect_irq(gicbusdev, i + smp_cpus,
     > +                           qdev_get_gpio_in(cpudev, ARM_CPU_FIQ));
     > +    }
     > +}
     > +
     > +static void create_uart(const VMAppleMachineState *vms, int uart,
     > +                        MemoryRegion *mem, Chardev *chr)
     > +{
     > +    hwaddr base = vms->memmap[uart].base;
     > +    int irq = vms->irqmap[uart];
     > +    DeviceState *dev = qdev_new(TYPE_PL011);
     > +    SysBusDevice *s = SYS_BUS_DEVICE(dev);
     > +
     > +    qdev_prop_set_chr(dev, "chardev", chr);
     > +    sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal);
     > +    memory_region_add_subregion(mem, base,
     > +                                sysbus_mmio_get_region(s, 0));
     > +    sysbus_connect_irq(s, 0, qdev_get_gpio_in(vms->gic, irq));
     > +}
     > +
     > +static void create_rtc(const VMAppleMachineState *vms)
     > +{
     > +    hwaddr base = vms->memmap[VMAPPLE_RTC].base;
     > +    int irq = vms->irqmap[VMAPPLE_RTC];
     > +
     > +    sysbus_create_simple("pl031", base, qdev_get_gpio_in(vms-
     >gic, irq));
     > +}
     > +
     > +static DeviceState *gpio_key_dev;
     > +static void vmapple_powerdown_req(Notifier *n, void *opaque)
     > +{
     > +    /* use gpio Pin 3 for power button event */
     > +    qemu_set_irq(qdev_get_gpio_in(gpio_key_dev, 0), 1);
     > +}
     > +
     > +static void create_gpio_devices(const VMAppleMachineState *vms,
    int gpio,
     > +                                MemoryRegion *mem)
     > +{
     > +    DeviceState *pl061_dev;
     > +    hwaddr base = vms->memmap[gpio].base;
     > +    int irq = vms->irqmap[gpio];
     > +    SysBusDevice *s;
     > +
     > +    pl061_dev = qdev_new("pl061");
     > +    /* Pull lines down to 0 if not driven by the PL061 */
     > +    qdev_prop_set_uint32(pl061_dev, "pullups", 0);
     > +    qdev_prop_set_uint32(pl061_dev, "pulldowns", 0xff);
     > +    s = SYS_BUS_DEVICE(pl061_dev);
     > +    sysbus_realize_and_unref(s, &error_fatal);
     > +    memory_region_add_subregion(mem, base,
    sysbus_mmio_get_region(s, 0));
     > +    sysbus_connect_irq(s, 0, qdev_get_gpio_in(vms->gic, irq));
     > +    gpio_key_dev = sysbus_create_simple("gpio-key", -1,
> + qdev_get_gpio_in(pl061_dev, 3));
     > +}
     > +
     > +static void vmapple_firmware_init(VMAppleMachineState *vms,
     > +                                  MemoryRegion *sysmem)
     > +{
     > +    hwaddr size = vms->memmap[VMAPPLE_FIRMWARE].size;
     > +    hwaddr base = vms->memmap[VMAPPLE_FIRMWARE].base;
     > +    const char *bios_name;
     > +    int image_size;
     > +    char *fname;
     > +
     > +    bios_name = MACHINE(vms)->firmware;
     > +    if (!bios_name) {
     > +        error_report("No firmware specified");
     > +        exit(1);
     > +    }
     > +
     > +    fname = qemu_find_file(QEMU_FILE_TYPE_BIOS, bios_name);
     > +    if (!fname) {
     > +        error_report("Could not find ROM image '%s'", bios_name);
     > +        exit(1);
     > +    }
     > +
     > +    memory_region_init_ram(&vms->fw_mr, NULL, "firmware", size,
    &error_fatal);
     > +    image_size = load_image_mr(fname, &vms->fw_mr);
     > +
     > +    g_free(fname);
     > +    if (image_size < 0) {
     > +        error_report("Could not load ROM image '%s'", bios_name);
     > +        exit(1);
     > +    }
     > +
     > +    memory_region_add_subregion(get_system_memory(), base, &vms-
     >fw_mr);
     > +}
     > +
     > +static void create_pcie(VMAppleMachineState *vms)
     > +{
     > +    hwaddr base_mmio = vms->memmap[VMAPPLE_PCIE_MMIO].base;
     > +    hwaddr size_mmio = vms->memmap[VMAPPLE_PCIE_MMIO].size;
     > +    hwaddr base_ecam = vms->memmap[VMAPPLE_PCIE_ECAM].base;
     > +    hwaddr size_ecam = vms->memmap[VMAPPLE_PCIE_ECAM].size;
     > +    int irq = vms->irqmap[VMAPPLE_PCIE];
     > +    MemoryRegion *mmio_alias;
     > +    MemoryRegion *mmio_reg;
     > +    MemoryRegion *ecam_reg;
     > +    DeviceState *dev;
     > +    int i;
     > +    PCIHostState *pci;
     > +    DeviceState *usb_controller;
     > +    USBBus *usb_bus;
     > +
     > +    dev = qdev_new(TYPE_GPEX_HOST);
     > +    qdev_prop_set_uint32(dev, "num-irqs", GPEX_NUM_IRQS);
     > +    sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal);
     > +
     > +    /* Map only the first size_ecam bytes of ECAM space */
     > +    ecam_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
     > +    memory_region_init_alias(&vms->ecam_alias, OBJECT(dev),
    "pcie-ecam",
     > +                             ecam_reg, 0, size_ecam);
     > +    memory_region_add_subregion(get_system_memory(), base_ecam,
     > +                                &vms->ecam_alias);
     > +
     > +    /*
     > +     * Map the MMIO window from [0x50000000-0x7fff0000] in PCI
    space into
     > +     * system address space at [0x50000000-0x7fff0000].
     > +     */
     > +    mmio_alias = g_new0(MemoryRegion, 1);
     > +    mmio_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 1);
     > +    memory_region_init_alias(mmio_alias, OBJECT(dev), "pcie-mmio",
     > +                             mmio_reg, base_mmio, size_mmio);
     > +    memory_region_add_subregion(get_system_memory(), base_mmio,
    mmio_alias);
     > +
     > +    for (i = 0; i < GPEX_NUM_IRQS; i++) {
     > +        sysbus_connect_irq(SYS_BUS_DEVICE(dev), i,
     > +                           qdev_get_gpio_in(vms->gic, irq + i));
     > +        gpex_set_irq_num(GPEX_HOST(dev), i, irq + i);
     > +    }
     > +
     > +    pci = PCI_HOST_BRIDGE(dev);
     > +    vms->bus = pci->bus;
     > +    g_assert(vms->bus);
     > +
     > +    while ((dev = qemu_create_nic_device("virtio-net-pci", true,
    NULL))) {
     > +        qdev_realize_and_unref(dev, BUS(vms->bus), &error_fatal);
     > +    }
     > +
     > +    usb_controller = qdev_new(TYPE_QEMU_XHCI);
     > +    qdev_realize_and_unref(usb_controller, BUS(pci->bus),
    &error_fatal);
     > +
     > +    usb_bus = USB_BUS(object_resolve_type_unambiguous(TYPE_USB_BUS,
> + &error_fatal));
     > +    usb_create_simple(usb_bus, "usb-kbd");
     > +    usb_create_simple(usb_bus, "usb-tablet");
     > +}
     > +
     > +static void vmapple_reset(void *opaque)
     > +{
     > +    VMAppleMachineState *vms = opaque;
     > +    hwaddr base = vms->memmap[VMAPPLE_FIRMWARE].base;
     > +
     > +    cpu_set_pc(first_cpu, base);
     > +}
     > +
     > +static void mach_vmapple_init(MachineState *machine)
     > +{
     > +    VMAppleMachineState *vms = VMAPPLE_MACHINE(machine);
     > +    MachineClass *mc = MACHINE_GET_CLASS(machine);
     > +    const CPUArchIdList *possible_cpus;
     > +    MemoryRegion *sysmem = get_system_memory();
     > +    int n;
     > +    unsigned int smp_cpus = machine->smp.cpus;
     > +    unsigned int max_cpus = machine->smp.max_cpus;
     > +
     > +    vms->memmap = memmap;
     > +    machine->usb = true;
     > +
     > +    possible_cpus = mc->possible_cpu_arch_ids(machine);
     > +    assert(possible_cpus->len == max_cpus);
     > +    for (n = 0; n < possible_cpus->len; n++) {
     > +        Object *cpu;
     > +        CPUState *cs;
     > +
     > +        if (n >= smp_cpus) {
     > +            break;
     > +        }
     > +
     > +        cpu = object_new(possible_cpus->cpus[n].type);
     > +        object_property_set_int(cpu, "mp-affinity",
     > +                                possible_cpus->cpus[n].arch_id,
    &error_fatal);
     > +
     > +        cs = CPU(cpu);
     > +        cs->cpu_index = n;
     > +
     > +        numa_cpu_pre_plug(&possible_cpus->cpus[cs->cpu_index],
    DEVICE(cpu),
     > +                          &error_fatal);
     > +
     > +        if (object_property_find(cpu, "has_el3")) {
     > +            object_property_set_bool(cpu, "has_el3", false,
    &error_fatal);
     > +        }
     > +        if (object_property_find(cpu, "has_el2")) {
     > +            object_property_set_bool(cpu, "has_el2", false,
    &error_fatal);
     > +        }
     > +        object_property_set_int(cpu, "psci-conduit",
    QEMU_PSCI_CONDUIT_HVC,
     > +                                &error_fatal);
     > +
     > +        /* Secondary CPUs start in PSCI powered-down state */
     > +        if (n > 0) {
     > +            object_property_set_bool(cpu, "start-powered-off", true,
     > +                                     &error_fatal);
     > +        }
     > +
     > +        object_property_set_link(cpu, "memory", OBJECT(sysmem),
    &error_abort);
     > +        qdev_realize(DEVICE(cpu), NULL, &error_fatal);
     > +        object_unref(cpu);
     > +    }
     > +
     > +    memory_region_add_subregion(sysmem, vms-
     >memmap[VMAPPLE_MEM].base,
     > +                                machine->ram);
     > +
     > +    create_gic(vms, sysmem);
     > +    create_bdif(vms, sysmem);
     > +    create_pvpanic(vms, sysmem);
     > +    create_aes(vms, sysmem);
     > +    create_gfx(vms, sysmem);
     > +    create_uart(vms, VMAPPLE_UART, sysmem, serial_hd(0));
     > +    create_rtc(vms);
     > +    create_pcie(vms);
     > +
     > +    create_gpio_devices(vms, VMAPPLE_GPIO, sysmem);
     > +
     > +    vmapple_firmware_init(vms, sysmem);
     > +    create_cfg(vms, sysmem);
     > +
     > +    /* connect powerdown request */
     > +    vms->powerdown_notifier.notify = vmapple_powerdown_req;
     > +    qemu_register_powerdown_notifier(&vms->powerdown_notifier);
     > +
     > +    vms->bootinfo.ram_size = machine->ram_size;
     > +    vms->bootinfo.board_id = -1;
     > +    vms->bootinfo.loader_start = vms->memmap[VMAPPLE_MEM].base;
     > +    vms->bootinfo.skip_dtb_autoload = true;
     > +    vms->bootinfo.firmware_loaded = true;
     > +    arm_load_kernel(ARM_CPU(first_cpu), machine, &vms->bootinfo);
     > +
     > +    qemu_register_reset(vmapple_reset, vms);
     > +}
     > +
     > +static CpuInstanceProperties
     > +vmapple_cpu_index_to_props(MachineState *ms, unsigned cpu_index)
     > +{
     > +    MachineClass *mc = MACHINE_GET_CLASS(ms);
     > +    const CPUArchIdList *possible_cpus = mc-
     >possible_cpu_arch_ids(ms);
     > +
     > +    assert(cpu_index < possible_cpus->len);
     > +    return possible_cpus->cpus[cpu_index].props;
     > +}
     > +
     > +
     > +static int64_t vmapple_get_default_cpu_node_id(const
    MachineState *ms, int idx)
     > +{
     > +    return idx % ms->numa_state->num_nodes;
     > +}
     > +
     > +static const CPUArchIdList
    *vmapple_possible_cpu_arch_ids(MachineState *ms)
     > +{
     > +    int n;
     > +    unsigned int max_cpus = ms->smp.max_cpus;
     > +
     > +    if (ms->possible_cpus) {
     > +        assert(ms->possible_cpus->len == max_cpus);
     > +        return ms->possible_cpus;
     > +    }
     > +
     > +    ms->possible_cpus = g_malloc0(sizeof(CPUArchIdList) +
     > +                                  sizeof(CPUArchId) * max_cpus);
     > +    ms->possible_cpus->len = max_cpus;
     > +    for (n = 0; n < ms->possible_cpus->len; n++) {
     > +        ms->possible_cpus->cpus[n].type = ms->cpu_type;
     > +        ms->possible_cpus->cpus[n].arch_id =
     > +            arm_build_mp_affinity(n, GICV3_TARGETLIST_BITS);
     > +        ms->possible_cpus->cpus[n].props.has_thread_id = true;
     > +        ms->possible_cpus->cpus[n].props.thread_id = n;
     > +    }
     > +    return ms->possible_cpus;
     > +}
     > +
     > +static void vmapple_get_uuid(Object *obj, Visitor *v, const char
    *name,
     > +                             void *opaque, Error **errp)
     > +{
     > +    VMAppleMachineState *vms = VMAPPLE_MACHINE(obj);
     > +
     > +    visit_type_uint64(v, name, &vms->uuid, errp);
     > +}
     > +
     > +static void vmapple_set_uuid(Object *obj, Visitor *v, const char
    *name,
     > +                             void *opaque, Error **errp)
     > +{
     > +    VMAppleMachineState *vms = VMAPPLE_MACHINE(obj);
     > +    Error *error = NULL;
     > +
     > +    visit_type_uint64(v, name, &vms->uuid, &error);
     > +    if (error) {
     > +        error_propagate(errp, error);
     > +        return;
     > +    }
     > +}
     > +
     > +static void vmapple_machine_class_init(ObjectClass *oc, void *data)
     > +{
     > +    MachineClass *mc = MACHINE_CLASS(oc);
     > +
     > +    mc->init = mach_vmapple_init;
     > +    mc->max_cpus = 32;
     > +    mc->block_default_type = IF_VIRTIO;
     > +    mc->no_cdrom = 1;
     > +    mc->pci_allow_0_address = true;
     > +    mc->minimum_page_bits = 12;
     > +    mc->possible_cpu_arch_ids = vmapple_possible_cpu_arch_ids;
     > +    mc->cpu_index_to_instance_props = vmapple_cpu_index_to_props;
     > +    mc->default_cpu_type = ARM_CPU_TYPE_NAME("host");
     > +    mc->get_default_cpu_node_id = vmapple_get_default_cpu_node_id;
     > +    mc->default_ram_id = "mach-vmapple.ram";
     > +
     > +    object_register_sugar_prop(TYPE_VIRTIO_PCI, "disable-legacy",
     > +                               "on", true);
     > +
     > +    object_class_property_add(oc, "uuid", "uint64",
    vmapple_get_uuid,
     > +                              vmapple_set_uuid, NULL, NULL);

    Use: object_class_property_add_uint64_ptr


Do I understand correctly that to use this, I need to move the uuid field out of VMAppleMachineState and turn it into a global/static variable? There are no active uses of object_class_property_add_uint64_ptr (or the uint32 version) in the Qemu code base to use as reference, but that's what it looks like based on the single use of object_class_property_add_uint8_ptr.

I'm not convinced this is an improvement to be honest. (And the lack of uptake of this API would suggest I'm not the only one who finds this an awkward API - on the other hand, there are lots of examples of object_class_property_add() being used for uint64-typed properties exactly as Alex has done here.)

I didn't mean to add a class property. object_property_add_uint64_ptr() will do what we want I think.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]