virt.c 44.9 KB
Newer Older
P
Peter Maydell 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
/*
 * 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.
 */

#include "hw/sysbus.h"
#include "hw/arm/arm.h"
#include "hw/arm/primecell.h"
34
#include "hw/arm/virt.h"
P
Peter Maydell 已提交
35 36
#include "hw/devices.h"
#include "net/net.h"
37
#include "sysemu/block-backend.h"
P
Peter Maydell 已提交
38 39 40 41
#include "sysemu/device_tree.h"
#include "sysemu/sysemu.h"
#include "sysemu/kvm.h"
#include "hw/boards.h"
42
#include "hw/loader.h"
P
Peter Maydell 已提交
43 44 45
#include "exec/address-spaces.h"
#include "qemu/bitops.h"
#include "qemu/error-report.h"
46
#include "hw/pci-host/gpex.h"
47
#include "hw/arm/virt-acpi-build.h"
48 49
#include "hw/arm/sysbus-fdt.h"
#include "hw/platform-bus.h"
50
#include "hw/arm/fdt.h"
51
#include "hw/intc/arm_gic_common.h"
52
#include "kvm_arm.h"
53
#include "hw/smbios/smbios.h"
54
#include "qapi/visitor.h"
P
Peter Maydell 已提交
55 56

/* Number of external interrupt lines to configure the GIC with */
57
#define NUM_IRQS 256
P
Peter Maydell 已提交
58

59 60 61 62
#define PLATFORM_BUS_NUM_IRQS 64

static ARMPlatformBusSystemParams platform_bus_params;

P
Peter Maydell 已提交
63 64 65 66 67 68 69 70 71
typedef struct VirtBoardInfo {
    struct arm_boot_info bootinfo;
    const char *cpu_model;
    const MemMapEntry *memmap;
    const int *irqmap;
    int smp_cpus;
    void *fdt;
    int fdt_size;
    uint32_t clock_phandle;
72
    uint32_t gic_phandle;
73
    uint32_t v2m_phandle;
P
Peter Maydell 已提交
74 75
} VirtBoardInfo;

76 77 78 79 80 81 82
typedef struct {
    MachineClass parent;
    VirtBoardInfo *daughterboard;
} VirtMachineClass;

typedef struct {
    MachineState parent;
83
    bool secure;
84
    bool highmem;
85
    int32_t gic_version;
86 87
} VirtMachineState;

88
#define TYPE_VIRT_MACHINE   MACHINE_TYPE_NAME("virt")
89 90 91 92 93 94 95
#define VIRT_MACHINE(obj) \
    OBJECT_CHECK(VirtMachineState, (obj), TYPE_VIRT_MACHINE)
#define VIRT_MACHINE_GET_CLASS(obj) \
    OBJECT_GET_CLASS(VirtMachineClass, obj, TYPE_VIRT_MACHINE)
#define VIRT_MACHINE_CLASS(klass) \
    OBJECT_CLASS_CHECK(VirtMachineClass, klass, TYPE_VIRT_MACHINE)

P
Peter Maydell 已提交
96 97 98 99 100 101 102 103 104
/* 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.
P
Peter Maydell 已提交
105 106
 * Note that devices should generally be placed at multiples of 0x10000,
 * to accommodate guests using 64K pages.
P
Peter Maydell 已提交
107 108 109
 */
static const MemMapEntry a15memmap[] = {
    /* Space up to 0x8000000 is reserved for a boot ROM */
110 111
    [VIRT_FLASH] =              {          0, 0x08000000 },
    [VIRT_CPUPERIPHS] =         { 0x08000000, 0x00020000 },
P
Peter Maydell 已提交
112
    /* GIC distributor and CPU interfaces sit inside the CPU peripheral space */
113 114 115
    [VIRT_GIC_DIST] =           { 0x08000000, 0x00010000 },
    [VIRT_GIC_CPU] =            { 0x08010000, 0x00010000 },
    [VIRT_GIC_V2M] =            { 0x08020000, 0x00001000 },
116 117 118 119
    /* The space in between here is reserved for GICv3 CPU/vCPU/HYP */
    [VIRT_GIC_ITS] =            { 0x08080000, 0x00020000 },
    /* This redistributor space allows up to 2*64kB*123 CPUs */
    [VIRT_GIC_REDIST] =         { 0x080A0000, 0x00F60000 },
120 121
    [VIRT_UART] =               { 0x09000000, 0x00001000 },
    [VIRT_RTC] =                { 0x09010000, 0x00001000 },
M
Marc Marí 已提交
122
    [VIRT_FW_CFG] =             { 0x09020000, 0x00000018 },
S
Shannon Zhao 已提交
123
    [VIRT_GPIO] =               { 0x09030000, 0x00001000 },
124
    [VIRT_MMIO] =               { 0x0a000000, 0x00000200 },
P
Peter Maydell 已提交
125
    /* ...repeating for a total of NUM_VIRTIO_TRANSPORTS, each of that size */
126 127 128 129 130
    [VIRT_PLATFORM_BUS] =       { 0x0c000000, 0x02000000 },
    [VIRT_PCIE_MMIO] =          { 0x10000000, 0x2eff0000 },
    [VIRT_PCIE_PIO] =           { 0x3eff0000, 0x00010000 },
    [VIRT_PCIE_ECAM] =          { 0x3f000000, 0x01000000 },
    [VIRT_MEM] =                { 0x40000000, 30ULL * 1024 * 1024 * 1024 },
131 132
    /* Second PCIe window, 512GB wide at the 512GB boundary */
    [VIRT_PCIE_MMIO_HIGH] =   { 0x8000000000ULL, 0x8000000000ULL },
P
Peter Maydell 已提交
133 134 135 136
};

static const int a15irqmap[] = {
    [VIRT_UART] = 1,
P
Peter Maydell 已提交
137
    [VIRT_RTC] = 2,
138
    [VIRT_PCIE] = 3, /* ... to 6 */
S
Shannon Zhao 已提交
139
    [VIRT_GPIO] = 7,
P
Peter Maydell 已提交
140
    [VIRT_MMIO] = 16, /* ...to 16 + NUM_VIRTIO_TRANSPORTS - 1 */
141
    [VIRT_GIC_V2M] = 48, /* ...to 48 + NUM_GICV2M_SPIS - 1 */
142
    [VIRT_PLATFORM_BUS] = 112, /* ...to 112 + PLATFORM_BUS_NUM_IRQS -1 */
P
Peter Maydell 已提交
143 144 145 146 147 148 149 150
};

static VirtBoardInfo machines[] = {
    {
        .cpu_model = "cortex-a15",
        .memmap = a15memmap,
        .irqmap = a15irqmap,
    },
151 152 153 154 155
    {
        .cpu_model = "cortex-a53",
        .memmap = a15memmap,
        .irqmap = a15irqmap,
    },
156 157 158 159 160
    {
        .cpu_model = "cortex-a57",
        .memmap = a15memmap,
        .irqmap = a15irqmap,
    },
P
Peter Maydell 已提交
161 162 163 164 165
    {
        .cpu_model = "host",
        .memmap = a15memmap,
        .irqmap = a15irqmap,
    },
P
Peter Maydell 已提交
166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191
};

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 */
192 193 194
    qemu_fdt_setprop_string(fdt, "/", "compatible", "linux,dummy-virt");
    qemu_fdt_setprop_cell(fdt, "/", "#address-cells", 0x2);
    qemu_fdt_setprop_cell(fdt, "/", "#size-cells", 0x2);
P
Peter Maydell 已提交
195 196 197 198 199

    /*
     * /chosen and /memory nodes must exist for load_dtb
     * to fill in necessary properties later
     */
200 201 202
    qemu_fdt_add_subnode(fdt, "/chosen");
    qemu_fdt_add_subnode(fdt, "/memory");
    qemu_fdt_setprop_string(fdt, "/memory", "device_type", "memory");
P
Peter Maydell 已提交
203 204 205 206 207 208

    /* 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.
     */
209 210 211 212 213 214
    vbi->clock_phandle = qemu_fdt_alloc_phandle(fdt);
    qemu_fdt_add_subnode(fdt, "/apb-pclk");
    qemu_fdt_setprop_string(fdt, "/apb-pclk", "compatible", "fixed-clock");
    qemu_fdt_setprop_cell(fdt, "/apb-pclk", "#clock-cells", 0x0);
    qemu_fdt_setprop_cell(fdt, "/apb-pclk", "clock-frequency", 24000000);
    qemu_fdt_setprop_string(fdt, "/apb-pclk", "clock-output-names",
P
Peter Maydell 已提交
215
                                "clk24mhz");
216
    qemu_fdt_setprop_cell(fdt, "/apb-pclk", "phandle", vbi->clock_phandle);
P
Peter Maydell 已提交
217

218 219 220 221
}

static void fdt_add_psci_node(const VirtBoardInfo *vbi)
{
222 223 224 225
    uint32_t cpu_suspend_fn;
    uint32_t cpu_off_fn;
    uint32_t cpu_on_fn;
    uint32_t migrate_fn;
226 227 228
    void *fdt = vbi->fdt;
    ARMCPU *armcpu = ARM_CPU(qemu_get_cpu(0));

229 230 231 232
    qemu_fdt_add_subnode(fdt, "/psci");
    if (armcpu->psci_version == 2) {
        const char comp[] = "arm,psci-0.2\0arm,psci";
        qemu_fdt_setprop(fdt, "/psci", "compatible", comp, sizeof(comp));
233

234 235 236 237 238 239 240 241 242
        cpu_off_fn = QEMU_PSCI_0_2_FN_CPU_OFF;
        if (arm_feature(&armcpu->env, ARM_FEATURE_AARCH64)) {
            cpu_suspend_fn = QEMU_PSCI_0_2_FN64_CPU_SUSPEND;
            cpu_on_fn = QEMU_PSCI_0_2_FN64_CPU_ON;
            migrate_fn = QEMU_PSCI_0_2_FN64_MIGRATE;
        } else {
            cpu_suspend_fn = QEMU_PSCI_0_2_FN_CPU_SUSPEND;
            cpu_on_fn = QEMU_PSCI_0_2_FN_CPU_ON;
            migrate_fn = QEMU_PSCI_0_2_FN_MIGRATE;
243
        }
244 245
    } else {
        qemu_fdt_setprop_string(fdt, "/psci", "compatible", "arm,psci");
246

247 248 249 250
        cpu_suspend_fn = QEMU_PSCI_0_1_FN_CPU_SUSPEND;
        cpu_off_fn = QEMU_PSCI_0_1_FN_CPU_OFF;
        cpu_on_fn = QEMU_PSCI_0_1_FN_CPU_ON;
        migrate_fn = QEMU_PSCI_0_1_FN_MIGRATE;
P
Peter Maydell 已提交
251
    }
252 253 254 255 256 257 258 259 260 261 262 263

    /* We adopt the PSCI spec's nomenclature, and use 'conduit' to refer
     * to the instruction that should be used to invoke PSCI functions.
     * However, the device tree binding uses 'method' instead, so that is
     * what we should use here.
     */
    qemu_fdt_setprop_string(fdt, "/psci", "method", "hvc");

    qemu_fdt_setprop_cell(fdt, "/psci", "cpu_suspend", cpu_suspend_fn);
    qemu_fdt_setprop_cell(fdt, "/psci", "cpu_off", cpu_off_fn);
    qemu_fdt_setprop_cell(fdt, "/psci", "cpu_on", cpu_on_fn);
    qemu_fdt_setprop_cell(fdt, "/psci", "migrate", migrate_fn);
P
Peter Maydell 已提交
264 265
}

266
static void fdt_add_timer_nodes(const VirtBoardInfo *vbi, int gictype)
P
Peter Maydell 已提交
267 268 269 270 271
{
    /* 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.
     */
272
    ARMCPU *armcpu;
P
Peter Maydell 已提交
273 274
    uint32_t irqflags = GIC_FDT_IRQ_FLAGS_EDGE_LO_HI;

275 276 277 278 279
    if (gictype == 2) {
        irqflags = deposit32(irqflags, GIC_FDT_IRQ_PPI_CPU_START,
                             GIC_FDT_IRQ_PPI_CPU_WIDTH,
                             (1 << vbi->smp_cpus) - 1);
    }
P
Peter Maydell 已提交
280

281
    qemu_fdt_add_subnode(vbi->fdt, "/timer");
282 283 284 285 286 287 288 289 290 291

    armcpu = ARM_CPU(qemu_get_cpu(0));
    if (arm_feature(&armcpu->env, ARM_FEATURE_V8)) {
        const char compat[] = "arm,armv8-timer\0arm,armv7-timer";
        qemu_fdt_setprop(vbi->fdt, "/timer", "compatible",
                         compat, sizeof(compat));
    } else {
        qemu_fdt_setprop_string(vbi->fdt, "/timer", "compatible",
                                "arm,armv7-timer");
    }
292
    qemu_fdt_setprop_cells(vbi->fdt, "/timer", "interrupts",
293 294 295 296
                       GIC_FDT_IRQ_TYPE_PPI, ARCH_TIMER_S_EL1_IRQ, irqflags,
                       GIC_FDT_IRQ_TYPE_PPI, ARCH_TIMER_NS_EL1_IRQ, irqflags,
                       GIC_FDT_IRQ_TYPE_PPI, ARCH_TIMER_VIRT_IRQ, irqflags,
                       GIC_FDT_IRQ_TYPE_PPI, ARCH_TIMER_NS_EL2_IRQ, irqflags);
P
Peter Maydell 已提交
297 298 299 300 301
}

static void fdt_add_cpu_nodes(const VirtBoardInfo *vbi)
{
    int cpu;
302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324
    int addr_cells = 1;

    /*
     * From Documentation/devicetree/bindings/arm/cpus.txt
     *  On ARM v8 64-bit systems value should be set to 2,
     *  that corresponds to the MPIDR_EL1 register size.
     *  If MPIDR_EL1[63:32] value is equal to 0 on all CPUs
     *  in the system, #address-cells can be set to 1, since
     *  MPIDR_EL1[63:32] bits are not used for CPUs
     *  identification.
     *
     *  Here we actually don't know whether our system is 32- or 64-bit one.
     *  The simplest way to go is to examine affinity IDs of all our CPUs. If
     *  at least one of them has Aff3 populated, we set #address-cells to 2.
     */
    for (cpu = 0; cpu < vbi->smp_cpus; cpu++) {
        ARMCPU *armcpu = ARM_CPU(qemu_get_cpu(cpu));

        if (armcpu->mp_affinity & ARM_AFF3_MASK) {
            addr_cells = 2;
            break;
        }
    }
P
Peter Maydell 已提交
325

326
    qemu_fdt_add_subnode(vbi->fdt, "/cpus");
327
    qemu_fdt_setprop_cell(vbi->fdt, "/cpus", "#address-cells", addr_cells);
328
    qemu_fdt_setprop_cell(vbi->fdt, "/cpus", "#size-cells", 0x0);
P
Peter Maydell 已提交
329 330 331 332 333

    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));

334 335 336
        qemu_fdt_add_subnode(vbi->fdt, nodename);
        qemu_fdt_setprop_string(vbi->fdt, nodename, "device_type", "cpu");
        qemu_fdt_setprop_string(vbi->fdt, nodename, "compatible",
P
Peter Maydell 已提交
337 338 339
                                    armcpu->dtb_compatible);

        if (vbi->smp_cpus > 1) {
340
            qemu_fdt_setprop_string(vbi->fdt, nodename,
P
Peter Maydell 已提交
341 342 343
                                        "enable-method", "psci");
        }

344 345 346 347 348 349 350 351
        if (addr_cells == 2) {
            qemu_fdt_setprop_u64(vbi->fdt, nodename, "reg",
                                 armcpu->mp_affinity);
        } else {
            qemu_fdt_setprop_cell(vbi->fdt, nodename, "reg",
                                  armcpu->mp_affinity);
        }

P
Peter Maydell 已提交
352 353 354 355
        g_free(nodename);
    }
}

356
static void fdt_add_v2m_gic_node(VirtBoardInfo *vbi)
P
Peter Maydell 已提交
357
{
358 359 360 361 362 363 364 365 366 367
    vbi->v2m_phandle = qemu_fdt_alloc_phandle(vbi->fdt);
    qemu_fdt_add_subnode(vbi->fdt, "/intc/v2m");
    qemu_fdt_setprop_string(vbi->fdt, "/intc/v2m", "compatible",
                            "arm,gic-v2m-frame");
    qemu_fdt_setprop(vbi->fdt, "/intc/v2m", "msi-controller", NULL, 0);
    qemu_fdt_setprop_sized_cells(vbi->fdt, "/intc/v2m", "reg",
                                 2, vbi->memmap[VIRT_GIC_V2M].base,
                                 2, vbi->memmap[VIRT_GIC_V2M].size);
    qemu_fdt_setprop_cell(vbi->fdt, "/intc/v2m", "phandle", vbi->v2m_phandle);
}
P
Peter Maydell 已提交
368

369
static void fdt_add_gic_node(VirtBoardInfo *vbi, int type)
370
{
371 372
    vbi->gic_phandle = qemu_fdt_alloc_phandle(vbi->fdt);
    qemu_fdt_setprop_cell(vbi->fdt, "/", "interrupt-parent", vbi->gic_phandle);
P
Peter Maydell 已提交
373

374 375 376
    qemu_fdt_add_subnode(vbi->fdt, "/intc");
    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "#interrupt-cells", 3);
    qemu_fdt_setprop(vbi->fdt, "/intc", "interrupt-controller", NULL, 0);
377 378 379
    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "#address-cells", 0x2);
    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "#size-cells", 0x2);
    qemu_fdt_setprop(vbi->fdt, "/intc", "ranges", NULL, 0);
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398
    if (type == 3) {
        qemu_fdt_setprop_string(vbi->fdt, "/intc", "compatible",
                                "arm,gic-v3");
        qemu_fdt_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_REDIST].base,
                                     2, vbi->memmap[VIRT_GIC_REDIST].size);
    } else {
        /* 'cortex-a15-gic' means 'GIC v2' */
        qemu_fdt_setprop_string(vbi->fdt, "/intc", "compatible",
                                "arm,cortex-a15-gic");
        qemu_fdt_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);
    }

399
    qemu_fdt_setprop_cell(vbi->fdt, "/intc", "phandle", vbi->gic_phandle);
P
Peter Maydell 已提交
400 401
}

402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420
static void create_v2m(VirtBoardInfo *vbi, qemu_irq *pic)
{
    int i;
    int irq = vbi->irqmap[VIRT_GIC_V2M];
    DeviceState *dev;

    dev = qdev_create(NULL, "arm-gicv2m");
    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, vbi->memmap[VIRT_GIC_V2M].base);
    qdev_prop_set_uint32(dev, "base-spi", irq);
    qdev_prop_set_uint32(dev, "num-spi", NUM_GICV2M_SPIS);
    qdev_init_nofail(dev);

    for (i = 0; i < NUM_GICV2M_SPIS; i++) {
        sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]);
    }

    fdt_add_v2m_gic_node(vbi);
}

421
static void create_gic(VirtBoardInfo *vbi, qemu_irq *pic, int type, bool secure)
422
{
423
    /* We create a standalone GIC */
424 425
    DeviceState *gicdev;
    SysBusDevice *gicbusdev;
426
    const char *gictype;
427 428
    int i;

429
    gictype = (type == 3) ? gicv3_class_name() : gic_class_name();
430 431

    gicdev = qdev_create(NULL, gictype);
432
    qdev_prop_set_uint32(gicdev, "revision", type);
433 434 435 436 437
    qdev_prop_set_uint32(gicdev, "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(gicdev, "num-irq", NUM_IRQS + 32);
438 439 440
    if (!kvm_irqchip_in_kernel()) {
        qdev_prop_set_bit(gicdev, "has-security-extensions", secure);
    }
441 442 443
    qdev_init_nofail(gicdev);
    gicbusdev = SYS_BUS_DEVICE(gicdev);
    sysbus_mmio_map(gicbusdev, 0, vbi->memmap[VIRT_GIC_DIST].base);
444 445 446 447 448
    if (type == 3) {
        sysbus_mmio_map(gicbusdev, 1, vbi->memmap[VIRT_GIC_REDIST].base);
    } else {
        sysbus_mmio_map(gicbusdev, 1, vbi->memmap[VIRT_GIC_CPU].base);
    }
449 450 451 452 453 454 455

    /* Wire the outputs from each CPU's generic timer to the
     * appropriate GIC PPI inputs, and the GIC's IRQ output to
     * the CPU's IRQ input.
     */
    for (i = 0; i < smp_cpus; i++) {
        DeviceState *cpudev = DEVICE(qemu_get_cpu(i));
456
        int ppibase = NUM_IRQS + i * GIC_INTERNAL + GIC_NR_SGIS;
457 458 459
        int irq;
        /* Mapping from the output timer irq lines from the CPU to the
         * GIC PPI inputs we use for the virt board.
460
         */
461 462 463 464 465 466 467 468 469 470 471 472
        const int timer_irq[] = {
            [GTIMER_PHYS] = ARCH_TIMER_NS_EL1_IRQ,
            [GTIMER_VIRT] = ARCH_TIMER_VIRT_IRQ,
            [GTIMER_HYP]  = ARCH_TIMER_NS_EL2_IRQ,
            [GTIMER_SEC]  = ARCH_TIMER_S_EL1_IRQ,
        };

        for (irq = 0; irq < ARRAY_SIZE(timer_irq); irq++) {
            qdev_connect_gpio_out(cpudev, irq,
                                  qdev_get_gpio_in(gicdev,
                                                   ppibase + timer_irq[irq]));
        }
473 474

        sysbus_connect_irq(gicbusdev, i, qdev_get_gpio_in(cpudev, ARM_CPU_IRQ));
475 476
        sysbus_connect_irq(gicbusdev, i + smp_cpus,
                           qdev_get_gpio_in(cpudev, ARM_CPU_FIQ));
477 478 479 480 481 482
    }

    for (i = 0; i < NUM_IRQS; i++) {
        pic[i] = qdev_get_gpio_in(gicdev, i);
    }

483
    fdt_add_gic_node(vbi, type);
484

485 486 487
    if (type == 2) {
        create_v2m(vbi, pic);
    }
488 489
}

P
Peter Maydell 已提交
490 491 492 493 494 495 496 497 498 499 500 501
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);
502
    qemu_fdt_add_subnode(vbi->fdt, nodename);
P
Peter Maydell 已提交
503
    /* Note that we can't use setprop_string because of the embedded NUL */
504
    qemu_fdt_setprop(vbi->fdt, nodename, "compatible",
P
Peter Maydell 已提交
505
                         compat, sizeof(compat));
506
    qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
P
Peter Maydell 已提交
507
                                     2, base, 2, size);
508
    qemu_fdt_setprop_cells(vbi->fdt, nodename, "interrupts",
P
Peter Maydell 已提交
509
                               GIC_FDT_IRQ_TYPE_SPI, irq,
510
                               GIC_FDT_IRQ_FLAGS_LEVEL_HI);
511
    qemu_fdt_setprop_cells(vbi->fdt, nodename, "clocks",
P
Peter Maydell 已提交
512
                               vbi->clock_phandle, vbi->clock_phandle);
513
    qemu_fdt_setprop(vbi->fdt, nodename, "clock-names",
P
Peter Maydell 已提交
514
                         clocknames, sizeof(clocknames));
515

516
    qemu_fdt_setprop_string(vbi->fdt, "/chosen", "stdout-path", nodename);
P
Peter Maydell 已提交
517 518 519
    g_free(nodename);
}

P
Peter Maydell 已提交
520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536
static void create_rtc(const VirtBoardInfo *vbi, qemu_irq *pic)
{
    char *nodename;
    hwaddr base = vbi->memmap[VIRT_RTC].base;
    hwaddr size = vbi->memmap[VIRT_RTC].size;
    int irq = vbi->irqmap[VIRT_RTC];
    const char compat[] = "arm,pl031\0arm,primecell";

    sysbus_create_simple("pl031", base, pic[irq]);

    nodename = g_strdup_printf("/pl031@%" PRIx64, base);
    qemu_fdt_add_subnode(vbi->fdt, nodename);
    qemu_fdt_setprop(vbi->fdt, nodename, "compatible", compat, sizeof(compat));
    qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
                                 2, base, 2, size);
    qemu_fdt_setprop_cells(vbi->fdt, nodename, "interrupts",
                           GIC_FDT_IRQ_TYPE_SPI, irq,
537
                           GIC_FDT_IRQ_FLAGS_LEVEL_HI);
P
Peter Maydell 已提交
538 539 540 541 542
    qemu_fdt_setprop_cell(vbi->fdt, nodename, "clocks", vbi->clock_phandle);
    qemu_fdt_setprop_string(vbi->fdt, nodename, "clock-names", "apb_pclk");
    g_free(nodename);
}

S
Shannon Zhao 已提交
543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568
static void create_gpio(const VirtBoardInfo *vbi, qemu_irq *pic)
{
    char *nodename;
    hwaddr base = vbi->memmap[VIRT_GPIO].base;
    hwaddr size = vbi->memmap[VIRT_GPIO].size;
    int irq = vbi->irqmap[VIRT_GPIO];
    const char compat[] = "arm,pl061\0arm,primecell";

    sysbus_create_simple("pl061", base, pic[irq]);

    nodename = g_strdup_printf("/pl061@%" PRIx64, base);
    qemu_fdt_add_subnode(vbi->fdt, nodename);
    qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
                                 2, base, 2, size);
    qemu_fdt_setprop(vbi->fdt, nodename, "compatible", compat, sizeof(compat));
    qemu_fdt_setprop_cell(vbi->fdt, nodename, "#gpio-cells", 2);
    qemu_fdt_setprop(vbi->fdt, nodename, "gpio-controller", NULL, 0);
    qemu_fdt_setprop_cells(vbi->fdt, nodename, "interrupts",
                           GIC_FDT_IRQ_TYPE_SPI, irq,
                           GIC_FDT_IRQ_FLAGS_LEVEL_HI);
    qemu_fdt_setprop_cell(vbi->fdt, nodename, "clocks", vbi->clock_phandle);
    qemu_fdt_setprop_string(vbi->fdt, nodename, "clock-names", "apb_pclk");

    g_free(nodename);
}

P
Peter Maydell 已提交
569 570 571 572 573
static void create_virtio_devices(const VirtBoardInfo *vbi, qemu_irq *pic)
{
    int i;
    hwaddr size = vbi->memmap[VIRT_MMIO].size;

574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599
    /* We create the transports in forwards order. Since qbus_realize()
     * prepends (not appends) new child buses, the incrementing loop below will
     * create a list of virtio-mmio buses with decreasing base addresses.
     *
     * When a -device option is processed from the command line,
     * qbus_find_recursive() picks the next free virtio-mmio bus in forwards
     * order. The upshot is that -device options in increasing command line
     * order are mapped to virtio-mmio buses with decreasing base addresses.
     *
     * When this code was originally written, that arrangement ensured that the
     * guest Linux kernel would give the lowest "name" (/dev/vda, eth0, etc) to
     * the first -device on the command line. (The end-to-end order is a
     * function of this loop, qbus_realize(), qbus_find_recursive(), and the
     * guest kernel's name-to-address assignment strategy.)
     *
     * Meanwhile, the kernel's traversal seems to have been reversed; see eg.
     * the message, if not necessarily the code, of commit 70161ff336.
     * Therefore the loop now establishes the inverse of the original intent.
     *
     * Unfortunately, we can't counteract the kernel change by reversing the
     * loop; it would break existing command lines.
     *
     * In any case, the kernel makes no guarantee about the stability of
     * enumeration order of virtio devices (as demonstrated by it changing
     * between kernel versions). For reliable and stable identification
     * of disks users must use UUIDs or similar mechanisms.
P
Peter Maydell 已提交
600 601 602 603 604 605 606 607
     */
    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]);
    }

608 609 610 611 612 613 614
    /* We add dtb nodes in reverse order so that they appear in the finished
     * device tree lowest address first.
     *
     * Note that this mapping is independent of the loop above. The previous
     * loop influences virtio device to virtio transport assignment, whereas
     * this loop controls how virtio transports are laid out in the dtb.
     */
P
Peter Maydell 已提交
615 616 617 618 619 620
    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);
621 622 623 624 625 626 627 628
        qemu_fdt_add_subnode(vbi->fdt, nodename);
        qemu_fdt_setprop_string(vbi->fdt, nodename,
                                "compatible", "virtio,mmio");
        qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
                                     2, base, 2, size);
        qemu_fdt_setprop_cells(vbi->fdt, nodename, "interrupts",
                               GIC_FDT_IRQ_TYPE_SPI, irq,
                               GIC_FDT_IRQ_FLAGS_EDGE_LO_HI);
P
Peter Maydell 已提交
629 630 631 632
        g_free(nodename);
    }
}

633 634 635 636 637 638 639 640 641 642
static void create_one_flash(const char *name, hwaddr flashbase,
                             hwaddr flashsize)
{
    /* Create and map a single flash device. We use the same
     * parameters as the flash devices on the Versatile Express board.
     */
    DriveInfo *dinfo = drive_get_next(IF_PFLASH);
    DeviceState *dev = qdev_create(NULL, "cfi.pflash01");
    const uint64_t sectorlength = 256 * 1024;

643 644 645
    if (dinfo) {
        qdev_prop_set_drive(dev, "drive", blk_by_legacy_dinfo(dinfo),
                            &error_abort);
646 647 648 649 650 651
    }

    qdev_prop_set_uint32(dev, "num-blocks", flashsize / sectorlength);
    qdev_prop_set_uint64(dev, "sector-length", sectorlength);
    qdev_prop_set_uint8(dev, "width", 4);
    qdev_prop_set_uint8(dev, "device-width", 2);
652
    qdev_prop_set_bit(dev, "big-endian", false);
653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
    qdev_prop_set_uint16(dev, "id0", 0x89);
    qdev_prop_set_uint16(dev, "id1", 0x18);
    qdev_prop_set_uint16(dev, "id2", 0x00);
    qdev_prop_set_uint16(dev, "id3", 0x00);
    qdev_prop_set_string(dev, "name", name);
    qdev_init_nofail(dev);

    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, flashbase);
}

static void create_flash(const VirtBoardInfo *vbi)
{
    /* Create two flash devices to fill the VIRT_FLASH space in the memmap.
     * Any file passed via -bios goes in the first of these.
     */
    hwaddr flashsize = vbi->memmap[VIRT_FLASH].size / 2;
    hwaddr flashbase = vbi->memmap[VIRT_FLASH].base;
    char *nodename;

    if (bios_name) {
G
Gonglei 已提交
673
        char *fn;
674
        int image_size;
675 676 677 678 679 680 681 682

        if (drive_get(IF_PFLASH, 0, 0)) {
            error_report("The contents of the first flash device may be "
                         "specified with -bios or with -drive if=pflash... "
                         "but you cannot use both options at once");
            exit(1);
        }
        fn = qemu_find_file(QEMU_FILE_TYPE_BIOS, bios_name);
683 684 685 686 687 688 689
        if (!fn) {
            error_report("Could not find ROM image '%s'", bios_name);
            exit(1);
        }
        image_size = load_image_targphys(fn, flashbase, flashsize);
        g_free(fn);
        if (image_size < 0) {
690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707
            error_report("Could not load ROM image '%s'", bios_name);
            exit(1);
        }
    }

    create_one_flash("virt.flash0", flashbase, flashsize);
    create_one_flash("virt.flash1", flashbase + flashsize, flashsize);

    nodename = g_strdup_printf("/flash@%" PRIx64, flashbase);
    qemu_fdt_add_subnode(vbi->fdt, nodename);
    qemu_fdt_setprop_string(vbi->fdt, nodename, "compatible", "cfi-flash");
    qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
                                 2, flashbase, 2, flashsize,
                                 2, flashbase + flashsize, 2, flashsize);
    qemu_fdt_setprop_cell(vbi->fdt, nodename, "bank-width", 4);
    g_free(nodename);
}

M
Marc Marí 已提交
708
static void create_fw_cfg(const VirtBoardInfo *vbi, AddressSpace *as)
L
Laszlo Ersek 已提交
709 710 711 712 713
{
    hwaddr base = vbi->memmap[VIRT_FW_CFG].base;
    hwaddr size = vbi->memmap[VIRT_FW_CFG].size;
    char *nodename;

M
Marc Marí 已提交
714
    fw_cfg_init_mem_wide(base + 8, base, 8, base + 16, as);
L
Laszlo Ersek 已提交
715 716 717 718 719 720 721 722 723 724

    nodename = g_strdup_printf("/fw-cfg@%" PRIx64, base);
    qemu_fdt_add_subnode(vbi->fdt, nodename);
    qemu_fdt_setprop_string(vbi->fdt, nodename,
                            "compatible", "qemu,fw-cfg-mmio");
    qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
                                 2, base, 2, size);
    g_free(nodename);
}

725 726 727 728
static void create_pcie_irq_map(const VirtBoardInfo *vbi, uint32_t gic_phandle,
                                int first_irq, const char *nodename)
{
    int devfn, pin;
729
    uint32_t full_irq_map[4 * 4 * 10] = { 0 };
730 731 732 733 734 735 736 737 738 739 740 741
    uint32_t *irq_map = full_irq_map;

    for (devfn = 0; devfn <= 0x18; devfn += 0x8) {
        for (pin = 0; pin < 4; pin++) {
            int irq_type = GIC_FDT_IRQ_TYPE_SPI;
            int irq_nr = first_irq + ((pin + PCI_SLOT(devfn)) % PCI_NUM_PINS);
            int irq_level = GIC_FDT_IRQ_FLAGS_LEVEL_HI;
            int i;

            uint32_t map[] = {
                devfn << 8, 0, 0,                           /* devfn */
                pin + 1,                                    /* PCI pin */
742
                gic_phandle, 0, 0, irq_type, irq_nr, irq_level }; /* GIC irq */
743 744

            /* Convert map to big endian */
745
            for (i = 0; i < 10; i++) {
746 747
                irq_map[i] = cpu_to_be32(map[i]);
            }
748
            irq_map += 10;
749 750 751 752 753 754 755 756 757 758 759
        }
    }

    qemu_fdt_setprop(vbi->fdt, nodename, "interrupt-map",
                     full_irq_map, sizeof(full_irq_map));

    qemu_fdt_setprop_cells(vbi->fdt, nodename, "interrupt-map-mask",
                           0x1800, 0, 0, /* devfn (PCI_SLOT(3)) */
                           0x7           /* PCI irq */);
}

760 761
static void create_pcie(const VirtBoardInfo *vbi, qemu_irq *pic,
                        bool use_highmem)
762
{
763 764
    hwaddr base_mmio = vbi->memmap[VIRT_PCIE_MMIO].base;
    hwaddr size_mmio = vbi->memmap[VIRT_PCIE_MMIO].size;
765 766
    hwaddr base_mmio_high = vbi->memmap[VIRT_PCIE_MMIO_HIGH].base;
    hwaddr size_mmio_high = vbi->memmap[VIRT_PCIE_MMIO_HIGH].size;
767 768 769 770 771 772
    hwaddr base_pio = vbi->memmap[VIRT_PCIE_PIO].base;
    hwaddr size_pio = vbi->memmap[VIRT_PCIE_PIO].size;
    hwaddr base_ecam = vbi->memmap[VIRT_PCIE_ECAM].base;
    hwaddr size_ecam = vbi->memmap[VIRT_PCIE_ECAM].size;
    hwaddr base = base_mmio;
    int nr_pcie_buses = size_ecam / PCIE_MMCFG_SIZE_MIN;
773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802
    int irq = vbi->irqmap[VIRT_PCIE];
    MemoryRegion *mmio_alias;
    MemoryRegion *mmio_reg;
    MemoryRegion *ecam_alias;
    MemoryRegion *ecam_reg;
    DeviceState *dev;
    char *nodename;
    int i;

    dev = qdev_create(NULL, TYPE_GPEX_HOST);
    qdev_init_nofail(dev);

    /* Map only the first size_ecam bytes of ECAM space */
    ecam_alias = g_new0(MemoryRegion, 1);
    ecam_reg = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0);
    memory_region_init_alias(ecam_alias, OBJECT(dev), "pcie-ecam",
                             ecam_reg, 0, size_ecam);
    memory_region_add_subregion(get_system_memory(), base_ecam, ecam_alias);

    /* Map the MMIO window into system address space so as to expose
     * the section of PCI MMIO space which starts at the same base address
     * (ie 1:1 mapping for that part of PCI MMIO space visible through
     * the window).
     */
    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);

803 804 805 806 807 808 809 810 811 812
    if (use_highmem) {
        /* Map high MMIO space */
        MemoryRegion *high_mmio_alias = g_new0(MemoryRegion, 1);

        memory_region_init_alias(high_mmio_alias, OBJECT(dev), "pcie-mmio-high",
                                 mmio_reg, base_mmio_high, size_mmio_high);
        memory_region_add_subregion(get_system_memory(), base_mmio_high,
                                    high_mmio_alias);
    }

813
    /* Map IO port space */
814
    sysbus_mmio_map(SYS_BUS_DEVICE(dev), 2, base_pio);
815 816 817 818 819 820 821 822 823 824 825 826 827 828 829

    for (i = 0; i < GPEX_NUM_IRQS; i++) {
        sysbus_connect_irq(SYS_BUS_DEVICE(dev), i, pic[irq + i]);
    }

    nodename = g_strdup_printf("/pcie@%" PRIx64, base);
    qemu_fdt_add_subnode(vbi->fdt, nodename);
    qemu_fdt_setprop_string(vbi->fdt, nodename,
                            "compatible", "pci-host-ecam-generic");
    qemu_fdt_setprop_string(vbi->fdt, nodename, "device_type", "pci");
    qemu_fdt_setprop_cell(vbi->fdt, nodename, "#address-cells", 3);
    qemu_fdt_setprop_cell(vbi->fdt, nodename, "#size-cells", 2);
    qemu_fdt_setprop_cells(vbi->fdt, nodename, "bus-range", 0,
                           nr_pcie_buses - 1);

830 831 832 833
    if (vbi->v2m_phandle) {
        qemu_fdt_setprop_cells(vbi->fdt, nodename, "msi-parent",
                               vbi->v2m_phandle);
    }
834

835 836
    qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "reg",
                                 2, base_ecam, 2, size_ecam);
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853

    if (use_highmem) {
        qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "ranges",
                                     1, FDT_PCI_RANGE_IOPORT, 2, 0,
                                     2, base_pio, 2, size_pio,
                                     1, FDT_PCI_RANGE_MMIO, 2, base_mmio,
                                     2, base_mmio, 2, size_mmio,
                                     1, FDT_PCI_RANGE_MMIO_64BIT,
                                     2, base_mmio_high,
                                     2, base_mmio_high, 2, size_mmio_high);
    } else {
        qemu_fdt_setprop_sized_cells(vbi->fdt, nodename, "ranges",
                                     1, FDT_PCI_RANGE_IOPORT, 2, 0,
                                     2, base_pio, 2, size_pio,
                                     1, FDT_PCI_RANGE_MMIO, 2, base_mmio,
                                     2, base_mmio, 2, size_mmio);
    }
854 855

    qemu_fdt_setprop_cell(vbi->fdt, nodename, "#interrupt-cells", 1);
856
    create_pcie_irq_map(vbi, vbi->gic_phandle, irq, nodename);
857 858 859 860

    g_free(nodename);
}

861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901
static void create_platform_bus(VirtBoardInfo *vbi, qemu_irq *pic)
{
    DeviceState *dev;
    SysBusDevice *s;
    int i;
    ARMPlatformBusFDTParams *fdt_params = g_new(ARMPlatformBusFDTParams, 1);
    MemoryRegion *sysmem = get_system_memory();

    platform_bus_params.platform_bus_base = vbi->memmap[VIRT_PLATFORM_BUS].base;
    platform_bus_params.platform_bus_size = vbi->memmap[VIRT_PLATFORM_BUS].size;
    platform_bus_params.platform_bus_first_irq = vbi->irqmap[VIRT_PLATFORM_BUS];
    platform_bus_params.platform_bus_num_irqs = PLATFORM_BUS_NUM_IRQS;

    fdt_params->system_params = &platform_bus_params;
    fdt_params->binfo = &vbi->bootinfo;
    fdt_params->intc = "/intc";
    /*
     * register a machine init done notifier that creates the device tree
     * nodes of the platform bus and its children dynamic sysbus devices
     */
    arm_register_platform_bus_fdt_creator(fdt_params);

    dev = qdev_create(NULL, TYPE_PLATFORM_BUS_DEVICE);
    dev->id = TYPE_PLATFORM_BUS_DEVICE;
    qdev_prop_set_uint32(dev, "num_irqs",
        platform_bus_params.platform_bus_num_irqs);
    qdev_prop_set_uint32(dev, "mmio_size",
        platform_bus_params.platform_bus_size);
    qdev_init_nofail(dev);
    s = SYS_BUS_DEVICE(dev);

    for (i = 0; i < platform_bus_params.platform_bus_num_irqs; i++) {
        int irqn = platform_bus_params.platform_bus_first_irq + i;
        sysbus_connect_irq(s, i, pic[irqn]);
    }

    memory_region_add_subregion(sysmem,
                                platform_bus_params.platform_bus_base,
                                sysbus_mmio_get_region(s, 0));
}

P
Peter Maydell 已提交
902 903 904 905 906 907 908 909
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;
}

910 911 912 913 914
static void virt_build_smbios(VirtGuestInfo *guest_info)
{
    FWCfgState *fw_cfg = guest_info->fw_cfg;
    uint8_t *smbios_tables, *smbios_anchor;
    size_t smbios_tables_len, smbios_anchor_len;
915
    const char *product = "QEMU Virtual Machine";
916 917 918 919 920

    if (!fw_cfg) {
        return;
    }

921 922 923 924 925
    if (kvm_enabled()) {
        product = "KVM Virtual Machine";
    }

    smbios_set_defaults("QEMU", product,
926 927 928 929 930 931 932 933 934 935 936 937 938
                        "1.0", false, true, SMBIOS_ENTRY_POINT_30);

    smbios_get_tables(NULL, 0, &smbios_tables, &smbios_tables_len,
                      &smbios_anchor, &smbios_anchor_len);

    if (smbios_anchor) {
        fw_cfg_add_file(fw_cfg, "etc/smbios/smbios-tables",
                        smbios_tables, smbios_tables_len);
        fw_cfg_add_file(fw_cfg, "etc/smbios/smbios-anchor",
                        smbios_anchor, smbios_anchor_len);
    }
}

939 940 941 942 943 944
static
void virt_guest_info_machine_done(Notifier *notifier, void *data)
{
    VirtGuestInfoState *guest_info_state = container_of(notifier,
                                              VirtGuestInfoState, machine_done);
    virt_acpi_setup(&guest_info_state->info);
945
    virt_build_smbios(&guest_info_state->info);
946 947
}

948
static void machvirt_init(MachineState *machine)
P
Peter Maydell 已提交
949
{
950
    VirtMachineState *vms = VIRT_MACHINE(machine);
P
Peter Maydell 已提交
951 952
    qemu_irq pic[NUM_IRQS];
    MemoryRegion *sysmem = get_system_memory();
953
    int gic_version = vms->gic_version;
954
    int n, max_cpus;
P
Peter Maydell 已提交
955
    MemoryRegion *ram = g_new(MemoryRegion, 1);
956
    const char *cpu_model = machine->cpu_model;
P
Peter Maydell 已提交
957
    VirtBoardInfo *vbi;
958 959
    VirtGuestInfoState *guest_info_state = g_malloc0(sizeof *guest_info_state);
    VirtGuestInfo *guest_info = &guest_info_state->info;
960
    char **cpustr;
P
Peter Maydell 已提交
961 962 963 964 965

    if (!cpu_model) {
        cpu_model = "cortex-a15";
    }

966 967 968 969 970 971
    /* We can probe only here because during property set
     * KVM is not available yet
     */
    if (!gic_version) {
        gic_version = kvm_arm_vgic_probe();
        if (!gic_version) {
A
Andrew Jones 已提交
972 973
            error_report("Unable to determine GIC version supported by host");
            error_printf("KVM acceleration is probably not supported\n");
974 975 976 977
            exit(1);
        }
    }

978 979 980 981
    /* Separate the actual CPU model name from any appended features */
    cpustr = g_strsplit(cpu_model, ",", 2);

    vbi = find_machine_info(cpustr[0]);
P
Peter Maydell 已提交
982 983

    if (!vbi) {
984
        error_report("mach-virt: CPU %s not supported", cpustr[0]);
P
Peter Maydell 已提交
985 986 987
        exit(1);
    }

988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003
    /* The maximum number of CPUs depends on the GIC version, or on how
     * many redistributors we can fit into the memory map.
     */
    if (gic_version == 3) {
        max_cpus = vbi->memmap[VIRT_GIC_REDIST].size / 0x20000;
    } else {
        max_cpus = GIC_NCPU;
    }

    if (smp_cpus > max_cpus) {
        error_report("Number of SMP CPUs requested (%d) exceeds max CPUs "
                     "supported by machine 'mach-virt' (%d)",
                     smp_cpus, max_cpus);
        exit(1);
    }

P
Peter Maydell 已提交
1004 1005
    vbi->smp_cpus = smp_cpus;

1006
    if (machine->ram_size > vbi->memmap[VIRT_MEM].size) {
P
Peter Maydell 已提交
1007 1008 1009 1010 1011 1012 1013
        error_report("mach-virt: cannot model more than 30GB RAM");
        exit(1);
    }

    create_fdt(vbi);

    for (n = 0; n < smp_cpus; n++) {
1014 1015
        ObjectClass *oc = cpu_class_by_name(TYPE_ARM_CPU, cpustr[0]);
        CPUClass *cc = CPU_CLASS(oc);
P
Peter Maydell 已提交
1016
        Object *cpuobj;
1017
        Error *err = NULL;
1018
        char *cpuopts = g_strdup(cpustr[1]);
P
Peter Maydell 已提交
1019 1020

        if (!oc) {
A
Andrew Jones 已提交
1021
            error_report("Unable to find CPU definition");
P
Peter Maydell 已提交
1022 1023 1024 1025
            exit(1);
        }
        cpuobj = object_new(object_class_get_name(oc));

1026
        /* Handle any CPU options specified by the user */
1027 1028
        cc->parse_features(CPU(cpuobj), cpuopts, &err);
        g_free(cpuopts);
1029
        if (err) {
1030
            error_report_err(err);
1031 1032 1033
            exit(1);
        }

1034 1035 1036 1037
        if (!vms->secure) {
            object_property_set_bool(cpuobj, false, "has_el3", NULL);
        }

1038 1039 1040
        object_property_set_int(cpuobj, QEMU_PSCI_CONDUIT_HVC, "psci-conduit",
                                NULL);

P
Peter Maydell 已提交
1041 1042 1043 1044
        /* Secondary CPUs start in PSCI powered-down state */
        if (n > 0) {
            object_property_set_bool(cpuobj, true, "start-powered-off", NULL);
        }
P
Peter Maydell 已提交
1045 1046 1047 1048 1049 1050

        if (object_property_find(cpuobj, "reset-cbar", NULL)) {
            object_property_set_int(cpuobj, vbi->memmap[VIRT_CPUPERIPHS].base,
                                    "reset-cbar", &error_abort);
        }

P
Peter Maydell 已提交
1051 1052
        object_property_set_bool(cpuobj, true, "realized", NULL);
    }
1053
    g_strfreev(cpustr);
1054
    fdt_add_timer_nodes(vbi, gic_version);
P
Peter Maydell 已提交
1055
    fdt_add_cpu_nodes(vbi);
1056
    fdt_add_psci_node(vbi);
P
Peter Maydell 已提交
1057

1058 1059
    memory_region_allocate_system_memory(ram, NULL, "mach-virt.ram",
                                         machine->ram_size);
P
Peter Maydell 已提交
1060 1061
    memory_region_add_subregion(sysmem, vbi->memmap[VIRT_MEM].base, ram);

1062 1063
    create_flash(vbi);

1064
    create_gic(vbi, pic, gic_version, vms->secure);
P
Peter Maydell 已提交
1065 1066 1067

    create_uart(vbi, pic);

P
Peter Maydell 已提交
1068 1069
    create_rtc(vbi, pic);

1070
    create_pcie(vbi, pic, vms->highmem);
1071

S
Shannon Zhao 已提交
1072 1073
    create_gpio(vbi, pic);

P
Peter Maydell 已提交
1074 1075 1076 1077 1078 1079
    /* 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);

M
Marc Marí 已提交
1080
    create_fw_cfg(vbi, &address_space_memory);
1081 1082 1083 1084 1085 1086
    rom_set_fw(fw_cfg_find());

    guest_info->smp_cpus = smp_cpus;
    guest_info->fw_cfg = fw_cfg_find();
    guest_info->memmap = vbi->memmap;
    guest_info->irqmap = vbi->irqmap;
1087
    guest_info->use_highmem = vms->highmem;
1088
    guest_info->gic_version = gic_version;
1089 1090
    guest_info_state->machine_done.notify = virt_guest_info_machine_done;
    qemu_add_machine_init_done_notifier(&guest_info_state->machine_done);
L
Laszlo Ersek 已提交
1091

1092 1093 1094 1095
    vbi->bootinfo.ram_size = machine->ram_size;
    vbi->bootinfo.kernel_filename = machine->kernel_filename;
    vbi->bootinfo.kernel_cmdline = machine->kernel_cmdline;
    vbi->bootinfo.initrd_filename = machine->initrd_filename;
P
Peter Maydell 已提交
1096 1097 1098 1099
    vbi->bootinfo.nb_cpus = smp_cpus;
    vbi->bootinfo.board_id = -1;
    vbi->bootinfo.loader_start = vbi->memmap[VIRT_MEM].base;
    vbi->bootinfo.get_dtb = machvirt_dtb;
1100
    vbi->bootinfo.firmware_loaded = bios_name || drive_get(IF_PFLASH, 0, 0);
P
Peter Maydell 已提交
1101
    arm_load_kernel(ARM_CPU(first_cpu), &vbi->bootinfo);
1102 1103 1104 1105 1106 1107 1108 1109

    /*
     * arm_load_kernel machine init done notifier registration must
     * happen before the platform_bus_create call. In this latter,
     * another notifier is registered which adds platform bus nodes.
     * Notifiers are executed in registration reverse order.
     */
    create_platform_bus(vbi, pic);
P
Peter Maydell 已提交
1110 1111
}

1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125
static bool virt_get_secure(Object *obj, Error **errp)
{
    VirtMachineState *vms = VIRT_MACHINE(obj);

    return vms->secure;
}

static void virt_set_secure(Object *obj, bool value, Error **errp)
{
    VirtMachineState *vms = VIRT_MACHINE(obj);

    vms->secure = value;
}

1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139
static bool virt_get_highmem(Object *obj, Error **errp)
{
    VirtMachineState *vms = VIRT_MACHINE(obj);

    return vms->highmem;
}

static void virt_set_highmem(Object *obj, bool value, Error **errp)
{
    VirtMachineState *vms = VIRT_MACHINE(obj);

    vms->highmem = value;
}

1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
static char *virt_get_gic_version(Object *obj, Error **errp)
{
    VirtMachineState *vms = VIRT_MACHINE(obj);
    const char *val = vms->gic_version == 3 ? "3" : "2";

    return g_strdup(val);
}

static void virt_set_gic_version(Object *obj, const char *value, Error **errp)
{
    VirtMachineState *vms = VIRT_MACHINE(obj);

    if (!strcmp(value, "3")) {
        vms->gic_version = 3;
    } else if (!strcmp(value, "2")) {
        vms->gic_version = 2;
    } else if (!strcmp(value, "host")) {
        vms->gic_version = 0; /* Will probe later */
    } else {
A
Andrew Jones 已提交
1159 1160
        error_report("Invalid gic-version option value");
        error_printf("Allowed gic-version values are: 3, 2, host\n");
1161 1162 1163 1164
        exit(1);
    }
}

1165 1166 1167 1168
static void virt_instance_init(Object *obj)
{
    VirtMachineState *vms = VIRT_MACHINE(obj);

1169 1170 1171 1172 1173
    /* EL3 is disabled by default on virt: this makes us consistent
     * between KVM and TCG for this board, and it also allows us to
     * boot UEFI blobs which assume no TrustZone support.
     */
    vms->secure = false;
1174 1175 1176 1177 1178 1179
    object_property_add_bool(obj, "secure", virt_get_secure,
                             virt_set_secure, NULL);
    object_property_set_description(obj, "secure",
                                    "Set on/off to enable/disable the ARM "
                                    "Security Extensions (TrustZone)",
                                    NULL);
1180 1181 1182 1183 1184 1185 1186 1187 1188

    /* High memory is enabled by default */
    vms->highmem = true;
    object_property_add_bool(obj, "highmem", virt_get_highmem,
                             virt_set_highmem, NULL);
    object_property_set_description(obj, "highmem",
                                    "Set on/off to enable/disable using "
                                    "physical address space above 32 bits",
                                    NULL);
1189 1190 1191 1192 1193 1194 1195
    /* Default GIC type is v2 */
    vms->gic_version = 2;
    object_property_add_str(obj, "gic-version", virt_get_gic_version,
                        virt_set_gic_version, NULL);
    object_property_set_description(obj, "gic-version",
                                    "Set GIC version. "
                                    "Valid values are 2, 3 and host", NULL);
1196 1197
}

1198 1199 1200 1201 1202 1203
static void virt_class_init(ObjectClass *oc, void *data)
{
    MachineClass *mc = MACHINE_CLASS(oc);

    mc->desc = "ARM Virtual Machine",
    mc->init = machvirt_init;
1204 1205 1206
    /* Start max_cpus at the maximum QEMU supports. We'll further restrict
     * it later in machvirt_init, where we have more information about the
     * configuration of the particular instance.
1207
     */
1208
    mc->max_cpus = MAX_CPUMASK_BITS;
1209
    mc->has_dynamic_sysbus = true;
1210 1211
    mc->block_default_type = IF_VIRTIO;
    mc->no_cdrom = 1;
1212
    mc->pci_allow_0_address = true;
1213 1214 1215 1216 1217 1218
}

static const TypeInfo machvirt_info = {
    .name = TYPE_VIRT_MACHINE,
    .parent = TYPE_MACHINE,
    .instance_size = sizeof(VirtMachineState),
1219
    .instance_init = virt_instance_init,
1220 1221
    .class_size = sizeof(VirtMachineClass),
    .class_init = virt_class_init,
P
Peter Maydell 已提交
1222 1223 1224 1225
};

static void machvirt_machine_init(void)
{
1226
    type_register_static(&machvirt_info);
P
Peter Maydell 已提交
1227 1228 1229
}

machine_init(machvirt_machine_init);