Re: [Qemu-devel] [PATCH v9 07/11] hw/arm: Add 'virt' platform

[Christoffer Dall <christoffer.dall@linaro.org>]

On Fri, Nov 22, 2013 at 05:17:14PM +0000, Peter Maydell wrote:
> Add 'virt' platform support corresponding to arch/arm/mach-virt
> in the Linux kernel tree. This has no platform-specific code but
> can use any device whose kernel driver is is able to work purely
> from a device tree node. We use this to instantiate a minimal
> set of devices: a GIC and some virtio-mmio transports.
> 
> Signed-off-by: John Rigby <john.rigby@linaro.org>
> [PMM:
>  Significantly overhauled:
>  * renamed user-facing machine to just "virt"
>  * removed the A9 support (it can't work since the A9 has no
>    generic timers)
>  * added virtio-mmio transports instead of random set of 'soc' devices
>    (though we retain a pl011 UART)
>  * instead of updating io_base as we step through adding devices,
>    define a memory map with an array (similar to vexpress)
>  * similarly, define irqmap with an array
>  * folded in some minor fixes from John's aarch64-support patch
>  * rather than explicitly doing endian-swapping on FDT cells,
>    use fdt APIs that let us just pass in host-endian values
>    and let the fdt layer take care of the swapping
>  * miscellaneous minor code cleanups and style fixes
> ]
> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
> ---
>  hw/arm/Makefile.objs |    2 +-
>  hw/arm/virt.c        |  444 ++++++++++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 445 insertions(+), 1 deletion(-)
>  create mode 100644 hw/arm/virt.c
> 
> diff --git a/hw/arm/Makefile.objs b/hw/arm/Makefile.objs
> index 3671b42..78b5614 100644
> --- a/hw/arm/Makefile.objs
> +++ b/hw/arm/Makefile.objs
> @@ -1,7 +1,7 @@
>  obj-y += boot.o collie.o exynos4_boards.o gumstix.o highbank.o
>  obj-y += integratorcp.o kzm.o mainstone.o musicpal.o nseries.o
>  obj-y += omap_sx1.o palm.o realview.o spitz.o stellaris.o
> -obj-y += tosa.o versatilepb.o vexpress.o xilinx_zynq.o z2.o
> +obj-y += tosa.o versatilepb.o vexpress.o virt.o xilinx_zynq.o z2.o
>  
>  obj-y += armv7m.o exynos4210.o pxa2xx.o pxa2xx_gpio.o pxa2xx_pic.o
>  obj-y += omap1.o omap2.o strongarm.o
> diff --git a/hw/arm/virt.c b/hw/arm/virt.c
> new file mode 100644
> index 0000000..1e42cc2
> --- /dev/null
> +++ b/hw/arm/virt.c
> @@ -0,0 +1,444 @@
> +/*
> + * ARM mach-virt emulation
> + *
> + * Copyright (c) 2013 Linaro Limited
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms and conditions of the GNU General Public License,
> + * version 2 or later, as published by the Free Software Foundation.
> + *
> + * This program is distributed in the hope it will be useful, but WITHOUT
> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
> + * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
> + * more details.
> + *
> + * You should have received a copy of the GNU General Public License along with
> + * this program.  If not, see <http://www.gnu.org/licenses/>.
> + *
> + * Emulate a virtual board which works by passing Linux all the information
> + * it needs about what devices are present via the device tree.
> + * There are some restrictions about what we can do here:
> + *  + we can only present devices whose Linux drivers will work based
> + *    purely on the device tree with no platform data at all
> + *  + we want to present a very stripped-down minimalist platform,
> + *    both because this reduces the security attack surface from the guest
> + *    and also because it reduces our exposure to being broken when
> + *    the kernel updates its device tree bindings and requires further
> + *    information in a device binding that we aren't providing.
> + * This is essentially the same approach kvmtool uses.

nit: questionable value of this last line in the comment.

> + */
> +
> +#include "hw/sysbus.h"
> +#include "hw/arm/arm.h"
> +#include "hw/arm/primecell.h"
> +#include "hw/devices.h"
> +#include "net/net.h"
> +#include "sysemu/device_tree.h"
> +#include "sysemu/sysemu.h"
> +#include "sysemu/kvm.h"
> +#include "hw/boards.h"
> +#include "exec/address-spaces.h"
> +#include "qemu/bitops.h"
> +#include "qemu/error-report.h"
> +
> +#define NUM_VIRTIO_TRANSPORTS 32
> +
> +/* Number of external interrupt lines to configure the GIC with */
> +#define NUM_IRQS 128
> +
> +#define GIC_FDT_IRQ_TYPE_SPI 0
> +#define GIC_FDT_IRQ_TYPE_PPI 1
> +
> +#define GIC_FDT_IRQ_FLAGS_EDGE_LO_HI 1
> +#define GIC_FDT_IRQ_FLAGS_EDGE_HI_LO 2
> +#define GIC_FDT_IRQ_FLAGS_LEVEL_HI 4
> +#define GIC_FDT_IRQ_FLAGS_LEVEL_LO 8
> +
> +#define GIC_FDT_IRQ_PPI_CPU_START 8
> +#define GIC_FDT_IRQ_PPI_CPU_WIDTH 8
> +
> +enum {
> +    VIRT_FLASH,
> +    VIRT_MEM,
> +    VIRT_CPUPERIPHS,
> +    VIRT_GIC_DIST,
> +    VIRT_GIC_CPU,
> +    VIRT_UART,
> +    VIRT_MMIO,
> +};
> +
> +typedef struct MemMapEntry {
> +    hwaddr base;
> +    hwaddr size;
> +} MemMapEntry;
> +
> +typedef struct VirtBoardInfo {
> +    struct arm_boot_info bootinfo;
> +    const char *cpu_model;
> +    const char *qdevname;
> +    const char *gic_compatible;
> +    const MemMapEntry *memmap;
> +    const int *irqmap;
> +    int smp_cpus;
> +    void *fdt;
> +    int fdt_size;
> +    uint32_t clock_phandle;
> +} VirtBoardInfo;
> +
> +/* Addresses and sizes of our components.
> + * 0..128MB is space for a flash device so we can run bootrom code such as UEFI.
> + * 128MB..256MB is used for miscellaneous device I/O.
> + * 256MB..1GB is reserved for possible future PCI support (ie where the
> + * PCI memory window will go if we add a PCI host controller).
> + * 1GB and up is RAM (which may happily spill over into the
> + * high memory region beyond 4GB).
> + * This represents a compromise between how much RAM can be given to
> + * a 32 bit VM and leaving space for expansion and in particular for PCI.
> + */
> +static const MemMapEntry a15memmap[] = {
> +    /* Space up to 0x8000000 is reserved for a boot ROM */
> +    [VIRT_FLASH] = { 0, 0x8000000 },
> +    [VIRT_CPUPERIPHS] = { 0x8000000, 0x8000 },
> +    /* GIC distributor and CPU interfaces sit inside the CPU peripheral space */
> +    [VIRT_GIC_DIST] = { 0x8001000, 0x1000 },
> +    [VIRT_GIC_CPU] = { 0x8002000, 0x1000 },
> +    [VIRT_UART] = { 0x9000000, 0x1000 },
> +    [VIRT_MMIO] = { 0xa000000, 0x200 },
> +    /* ...repeating for a total of NUM_VIRTIO_TRANSPORTS, each of that size */
> +    /* 0x10000000 .. 0x40000000 reserved for PCI */
> +    [VIRT_MEM] = { 0x40000000, 30ULL * 1024 * 1024 * 1024 },
> +};
> +
> +static const int a15irqmap[] = {
> +    [VIRT_UART] = 1,
> +    [VIRT_MMIO] = 16, /* ...to 16 + NUM_VIRTIO_TRANSPORTS - 1 */
> +};
> +
> +static VirtBoardInfo machines[] = {
> +    {
> +        .cpu_model = "cortex-a15",
> +        .qdevname = "a15mpcore_priv",
> +        .gic_compatible = "arm,cortex-a15-gic",
> +        .memmap = a15memmap,
> +        .irqmap = a15irqmap,
> +    },
> +};
> +
> +static VirtBoardInfo *find_machine_info(const char *cpu)
> +{
> +    int i;
> +
> +    for (i = 0; i < ARRAY_SIZE(machines); i++) {
> +        if (strcmp(cpu, machines[i].cpu_model) == 0) {
> +            return &machines[i];
> +        }
> +    }
> +    return NULL;
> +}
> +
> +static void create_fdt(VirtBoardInfo *vbi)
> +{
> +    void *fdt = create_device_tree(&vbi->fdt_size);
> +
> +    if (!fdt) {
> +        error_report("create_device_tree() failed");
> +        exit(1);
> +    }
> +
> +    vbi->fdt = fdt;
> +
> +    /* Header */
> +    qemu_devtree_setprop_string(fdt, "/", "compatible", "linux,dummy-virt");
> +    qemu_devtree_setprop_cell(fdt, "/", "#address-cells", 0x2);
> +    qemu_devtree_setprop_cell(fdt, "/", "#size-cells", 0x2);
> +
> +    /*
> +     * /chosen and /memory nodes must exist for load_dtb
> +     * to fill in necessary properties later
> +     */
> +    qemu_devtree_add_subnode(fdt, "/chosen");
> +    qemu_devtree_add_subnode(fdt, "/memory");
> +    qemu_devtree_setprop_string(fdt, "/memory", "device_type", "memory");
> +
> +    /* Clock node, for the benefit of the UART. The kernel device tree
> +     * binding documentation claims the PL011 node clock properties are
> +     * optional but in practice if you omit them the kernel refuses to
> +     * probe for the device.
> +     */
> +    vbi->clock_phandle = qemu_devtree_alloc_phandle(fdt);
> +    qemu_devtree_add_subnode(fdt, "/apb-pclk");
> +    qemu_devtree_setprop_string(fdt, "/apb-pclk", "compatible", "fixed-clock");
> +    qemu_devtree_setprop_cell(fdt, "/apb-pclk", "#clock-cells", 0x0);
> +    qemu_devtree_setprop_cell(fdt, "/apb-pclk", "clock-frequency", 24000000);
> +    qemu_devtree_setprop_string(fdt, "/apb-pclk", "clock-output-names",
> +                                "clk24mhz");
> +    qemu_devtree_setprop_cell(fdt, "/apb-pclk", "phandle", vbi->clock_phandle);
> +
> +    /* No PSCI for TCG yet */
> +    if (kvm_enabled()) {
> +        qemu_devtree_add_subnode(fdt, "/psci");
> +        qemu_devtree_setprop_string(fdt, "/psci", "compatible", "arm,psci");
> +        qemu_devtree_setprop_string(fdt, "/psci", "method", "hvc");
> +        qemu_devtree_setprop_cell(fdt, "/psci", "cpu_suspend",
> +                                  PSCI_FN_CPU_SUSPEND);
> +        qemu_devtree_setprop_cell(fdt, "/psci", "cpu_off", PSCI_FN_CPU_OFF);
> +        qemu_devtree_setprop_cell(fdt, "/psci", "cpu_on", PSCI_FN_CPU_ON);
> +        qemu_devtree_setprop_cell(fdt, "/psci", "migrate", PSCI_FN_MIGRATE);
> +    }
> +}
> +
> +static void fdt_add_timer_nodes(const VirtBoardInfo *vbi)
> +{
> +    /* Note that on A15 h/w these interrupts are level-triggered,
> +     * but for the GIC implementation provided by both QEMU and KVM
> +     * they are edge-triggered.
> +     */
> +    uint32_t irqflags = GIC_FDT_IRQ_FLAGS_EDGE_LO_HI;
> +
> +    irqflags = deposit32(irqflags, GIC_FDT_IRQ_PPI_CPU_START,
> +                         GIC_FDT_IRQ_PPI_CPU_WIDTH, (1 << vbi->smp_cpus) - 1);
> +
> +    qemu_devtree_add_subnode(vbi->fdt, "/timer");
> +    qemu_devtree_setprop_string(vbi->fdt, "/timer",
> +                                "compatible", "arm,armv7-timer");
> +    qemu_devtree_setprop_cells(vbi->fdt, "/timer", "interrupts",
> +                               GIC_FDT_IRQ_TYPE_PPI, 13, irqflags,
> +                               GIC_FDT_IRQ_TYPE_PPI, 14, irqflags,
> +                               GIC_FDT_IRQ_TYPE_PPI, 11, irqflags,
> +                               GIC_FDT_IRQ_TYPE_PPI, 10, irqflags);
> +}
> +
> +static void fdt_add_cpu_nodes(const VirtBoardInfo *vbi)
> +{
> +    int cpu;
> +
> +    qemu_devtree_add_subnode(vbi->fdt, "/cpus");
> +    qemu_devtree_setprop_cell(vbi->fdt, "/cpus", "#address-cells", 0x1);
> +    qemu_devtree_setprop_cell(vbi->fdt, "/cpus", "#size-cells", 0x0);
> +
> +    for (cpu = vbi->smp_cpus - 1; cpu >= 0; cpu--) {
> +        char *nodename = g_strdup_printf("/cpus/cpu@%d", cpu);
> +        ARMCPU *armcpu = ARM_CPU(qemu_get_cpu(cpu));
> +
> +        qemu_devtree_add_subnode(vbi->fdt, nodename);
> +        qemu_devtree_setprop_string(vbi->fdt, nodename, "device_type", "cpu");
> +        qemu_devtree_setprop_string(vbi->fdt, nodename, "compatible",
> +                                    armcpu->dtb_compatible);
> +
> +        if (vbi->smp_cpus > 1) {
> +            qemu_devtree_setprop_string(vbi->fdt, nodename,
> +                                        "enable-method", "psci");
> +        }
> +
> +        qemu_devtree_setprop_cell(vbi->fdt, nodename, "reg", cpu);
> +        g_free(nodename);
> +    }
> +}
> +
> +static void fdt_add_gic_node(const VirtBoardInfo *vbi)
> +{
> +    uint32_t gic_phandle;
> +
> +    gic_phandle = qemu_devtree_alloc_phandle(vbi->fdt);
> +    qemu_devtree_setprop_cell(vbi->fdt, "/", "interrupt-parent", gic_phandle);
> +
> +    qemu_devtree_add_subnode(vbi->fdt, "/intc");
> +    qemu_devtree_setprop_string(vbi->fdt, "/intc", "compatible",
> +                                vbi->gic_compatible);
> +    qemu_devtree_setprop_cell(vbi->fdt, "/intc", "#interrupt-cells", 3);
> +    qemu_devtree_setprop(vbi->fdt, "/intc", "interrupt-controller", NULL, 0);
> +    qemu_devtree_setprop_sized_cells(vbi->fdt, "/intc", "reg",
> +                                     2, vbi->memmap[VIRT_GIC_DIST].base,
> +                                     2, vbi->memmap[VIRT_GIC_DIST].size,
> +                                     2, vbi->memmap[VIRT_GIC_CPU].base,
> +                                     2, vbi->memmap[VIRT_GIC_CPU].size);
> +    qemu_devtree_setprop_cell(vbi->fdt, "/intc", "phandle", gic_phandle);
> +}
> +
> +static void create_uart(const VirtBoardInfo *vbi, qemu_irq *pic)
> +{
> +    char *nodename;
> +    hwaddr base = vbi->memmap[VIRT_UART].base;
> +    hwaddr size = vbi->memmap[VIRT_UART].size;
> +    int irq = vbi->irqmap[VIRT_UART];
> +    const char compat[] = "arm,pl011\0arm,primecell";
> +    const char clocknames[] = "uartclk\0apb_pclk";
> +
> +    sysbus_create_simple("pl011", base, pic[irq]);
> +
> +    nodename = g_strdup_printf("/pl011@%" PRIx64, base);
> +    qemu_devtree_add_subnode(vbi->fdt, nodename);
> +    /* Note that we can't use setprop_string because of the embedded NUL */
> +    qemu_devtree_setprop(vbi->fdt, nodename, "compatible",
> +                         compat, sizeof(compat));
> +    qemu_devtree_setprop_sized_cells(vbi->fdt, nodename, "reg",
> +                                     2, base, 2, size);
> +    qemu_devtree_setprop_cells(vbi->fdt, nodename, "interrupts",
> +                               GIC_FDT_IRQ_TYPE_SPI, irq,
> +                               GIC_FDT_IRQ_FLAGS_EDGE_LO_HI);
> +    qemu_devtree_setprop_cells(vbi->fdt, nodename, "clocks",
> +                               vbi->clock_phandle, vbi->clock_phandle);
> +    qemu_devtree_setprop(vbi->fdt, nodename, "clock-names",
> +                         clocknames, sizeof(clocknames));
> +    g_free(nodename);
> +}
> +
> +static void create_virtio_devices(const VirtBoardInfo *vbi, qemu_irq *pic)
> +{
> +    int i;
> +    hwaddr size = vbi->memmap[VIRT_MMIO].size;
> +
> +    /* Note that we have to create the transports in forwards order
> +     * so that command line devices are inserted lowest address first,
> +     * and then add dtb nodes in reverse order so that they appear in
> +     * the finished device tree lowest address first.
> +     */
> +    for (i = 0; i < NUM_VIRTIO_TRANSPORTS; i++) {
> +        int irq = vbi->irqmap[VIRT_MMIO] + i;
> +        hwaddr base = vbi->memmap[VIRT_MMIO].base + i * size;
> +
> +        sysbus_create_simple("virtio-mmio", base, pic[irq]);
> +    }
> +
> +    for (i = NUM_VIRTIO_TRANSPORTS - 1; i >= 0; i--) {
> +        char *nodename;
> +        int irq = vbi->irqmap[VIRT_MMIO] + i;
> +        hwaddr base = vbi->memmap[VIRT_MMIO].base + i * size;
> +
> +        nodename = g_strdup_printf("/virtio_mmio@%" PRIx64, base);
> +        qemu_devtree_add_subnode(vbi->fdt, nodename);
> +        qemu_devtree_setprop_string(vbi->fdt, nodename,
> +                                    "compatible", "virtio,mmio");
> +        qemu_devtree_setprop_sized_cells(vbi->fdt, nodename, "reg",
> +                                         2, base, 2, size);
> +        qemu_devtree_setprop_cells(vbi->fdt, nodename, "interrupts",
> +                                   GIC_FDT_IRQ_TYPE_SPI, irq,
> +                                   GIC_FDT_IRQ_FLAGS_EDGE_LO_HI);
> +        g_free(nodename);
> +    }
> +}
> +
> +static void *machvirt_dtb(const struct arm_boot_info *binfo, int *fdt_size)
> +{
> +    const VirtBoardInfo *board = (const VirtBoardInfo *)binfo;
> +
> +    *fdt_size = board->fdt_size;
> +    return board->fdt;
> +}
> +
> +static void machvirt_init(QEMUMachineInitArgs *args)
> +{
> +    qemu_irq pic[NUM_IRQS];
> +    MemoryRegion *sysmem = get_system_memory();
> +    int n;
> +    MemoryRegion *ram = g_new(MemoryRegion, 1);
> +    DeviceState *dev;
> +    SysBusDevice *busdev;
> +    const char *cpu_model = args->cpu_model;
> +    VirtBoardInfo *vbi;
> +
> +    if (!cpu_model) {
> +        cpu_model = "cortex-a15";
> +    }
> +
> +    vbi = find_machine_info(cpu_model);
> +
> +    if (!vbi) {
> +        error_report("mach-virt: CPU %s not supported", cpu_model);
> +        exit(1);
> +    }
> +
> +    vbi->smp_cpus = smp_cpus;
> +
> +    /*
> +     * Only supported method of starting secondary CPUs is PSCI and
> +     * PSCI is not yet supported with TCG, so limit smp_cpus to 1
> +     * if we're not using KVM.
> +     */
> +    if (!kvm_enabled() && smp_cpus > 1) {
> +        error_report("mach-virt: must enable KVM to use multiple CPUs");
> +        exit(1);
> +    }
> +
> +    if (args->ram_size > vbi->memmap[VIRT_MEM].size) {
> +        error_report("mach-virt: cannot model more than 30GB RAM");
> +        exit(1);
> +    }
> +
> +    create_fdt(vbi);
> +    fdt_add_timer_nodes(vbi);
> +
> +    for (n = 0; n < smp_cpus; n++) {
> +        ObjectClass *oc = cpu_class_by_name(TYPE_ARM_CPU, cpu_model);
> +        Object *cpuobj;
> +
> +        if (!oc) {
> +            fprintf(stderr, "Unable to find CPU definition\n");
> +            exit(1);
> +        }
> +        cpuobj = object_new(object_class_get_name(oc));
> +
> +        /* Secondary CPUs start in PSCI powered-down state */
> +        if (n > 0) {
> +            object_property_set_bool(cpuobj, true, "start-powered-off", NULL);
> +        }
> +        object_property_set_bool(cpuobj, true, "realized", NULL);
> +    }
> +    fdt_add_cpu_nodes(vbi);
> +
> +    memory_region_init_ram(ram, NULL, "mach-virt.ram", args->ram_size);
> +    vmstate_register_ram_global(ram);
> +    memory_region_add_subregion(sysmem, vbi->memmap[VIRT_MEM].base, ram);
> +
> +    dev = qdev_create(NULL, vbi->qdevname);
> +    qdev_prop_set_uint32(dev, "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(dev, "num-irq", NUM_IRQS + 32);
> +    qdev_init_nofail(dev);
> +    busdev = SYS_BUS_DEVICE(dev);
> +    sysbus_mmio_map(busdev, 0, vbi->memmap[VIRT_CPUPERIPHS].base);
> +    fdt_add_gic_node(vbi);
> +    for (n = 0; n < smp_cpus; n++) {
> +        DeviceState *cpudev = DEVICE(qemu_get_cpu(n));
> +
> +        sysbus_connect_irq(busdev, n, qdev_get_gpio_in(cpudev, ARM_CPU_IRQ));
> +    }
> +
> +    for (n = 0; n < NUM_IRQS; n++) {
> +        pic[n] = qdev_get_gpio_in(dev, n);
> +    }
> +
> +    create_uart(vbi, pic);
> +
> +    /* Create mmio transports, so the user can create virtio backends
> +     * (which will be automatically plugged in to the transports). If
> +     * no backend is created the transport will just sit harmlessly idle.
> +     */
> +    create_virtio_devices(vbi, pic);
> +
> +    vbi->bootinfo.ram_size = args->ram_size;
> +    vbi->bootinfo.kernel_filename = args->kernel_filename;
> +    vbi->bootinfo.kernel_cmdline = args->kernel_cmdline;
> +    vbi->bootinfo.initrd_filename = args->initrd_filename;
> +    vbi->bootinfo.nb_cpus = smp_cpus;
> +    vbi->bootinfo.board_id = -1;

board_id = -1 ?

> +    vbi->bootinfo.loader_start = vbi->memmap[VIRT_MEM].base;
> +    vbi->bootinfo.get_dtb = machvirt_dtb;
> +    arm_load_kernel(ARM_CPU(first_cpu), &vbi->bootinfo);
> +}
> +
> +static QEMUMachine machvirt_a15_machine = {
> +    .name = "virt",
> +    .desc = "ARM Virtual Machine",
> +    .init = machvirt_init,
> +    .max_cpus = 4,
> +};
> +
> +static void machvirt_machine_init(void)
> +{
> +    qemu_register_machine(&machvirt_a15_machine);
> +}
> +
> +machine_init(machvirt_machine_init);
> -- 
> 1.7.9.5
> 
> _______________________________________________
> kvmarm mailing list
> kvmarm@lists.cs.columbia.edu
> https://lists.cs.columbia.edu/cucslists/listinfo/kvmarm