#include "cpu.h"
#include "hw/arm/arm.h"
-/* Check that cpu.h's idea of coprocessor fields matches KVM's */
-#if (CP_REG_SIZE_SHIFT != KVM_REG_SIZE_SHIFT) || \
- (CP_REG_SIZE_MASK != KVM_REG_SIZE_MASK) || \
- (CP_REG_SIZE_U32 != KVM_REG_SIZE_U32) || \
- (CP_REG_SIZE_U64 != KVM_REG_SIZE_U64) || \
- (CP_REG_ARM != KVM_REG_ARM)
-#error mismatch between cpu.h and KVM header definitions
-#endif
-
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
KVM_CAP_LAST_INFO
};
-int kvm_arch_init(KVMState *s)
+bool kvm_arm_create_scratch_host_vcpu(const uint32_t *cpus_to_try,
+ int *fdarray,
+ struct kvm_vcpu_init *init)
{
- /* For ARM interrupt delivery is always asynchronous,
- * whether we are using an in-kernel VGIC or not.
- */
- kvm_async_interrupts_allowed = true;
- return 0;
-}
+ int ret, kvmfd = -1, vmfd = -1, cpufd = -1;
-unsigned long kvm_arch_vcpu_id(CPUState *cpu)
-{
- return cpu->cpu_index;
-}
+ kvmfd = qemu_open("/dev/kvm", O_RDWR);
+ if (kvmfd < 0) {
+ goto err;
+ }
+ vmfd = ioctl(kvmfd, KVM_CREATE_VM, 0);
+ if (vmfd < 0) {
+ goto err;
+ }
+ cpufd = ioctl(vmfd, KVM_CREATE_VCPU, 0);
+ if (cpufd < 0) {
+ goto err;
+ }
-static bool reg_syncs_via_tuple_list(uint64_t regidx)
-{
- /* Return true if the regidx is a register we should synchronize
- * via the cpreg_tuples array (ie is not a core reg we sync by
- * hand in kvm_arch_get/put_registers())
- */
- switch (regidx & KVM_REG_ARM_COPROC_MASK) {
- case KVM_REG_ARM_CORE:
- case KVM_REG_ARM_VFP:
- return false;
- default:
- return true;
+ ret = ioctl(vmfd, KVM_ARM_PREFERRED_TARGET, init);
+ if (ret >= 0) {
+ ret = ioctl(cpufd, KVM_ARM_VCPU_INIT, init);
+ if (ret < 0) {
+ goto err;
+ }
+ } else {
+ /* Old kernel which doesn't know about the
+ * PREFERRED_TARGET ioctl: we know it will only support
+ * creating one kind of guest CPU which is its preferred
+ * CPU type.
+ */
+ while (*cpus_to_try != QEMU_KVM_ARM_TARGET_NONE) {
+ init->target = *cpus_to_try++;
+ memset(init->features, 0, sizeof(init->features));
+ ret = ioctl(cpufd, KVM_ARM_VCPU_INIT, init);
+ if (ret >= 0) {
+ break;
+ }
+ }
+ if (ret < 0) {
+ goto err;
+ }
+ }
+
+ fdarray[0] = kvmfd;
+ fdarray[1] = vmfd;
+ fdarray[2] = cpufd;
+
+ return true;
+
+err:
+ if (cpufd >= 0) {
+ close(cpufd);
}
+ if (vmfd >= 0) {
+ close(vmfd);
+ }
+ if (kvmfd >= 0) {
+ close(kvmfd);
+ }
+
+ return false;
}
-static int compare_u64(const void *a, const void *b)
+void kvm_arm_destroy_scratch_host_vcpu(int *fdarray)
{
- return *(uint64_t *)a - *(uint64_t *)b;
+ int i;
+
+ for (i = 2; i >= 0; i--) {
+ close(fdarray[i]);
+ }
}
-int kvm_arch_init_vcpu(CPUState *cs)
+static void kvm_arm_host_cpu_class_init(ObjectClass *oc, void *data)
{
- struct kvm_vcpu_init init;
- int i, ret, arraylen;
- uint64_t v;
- struct kvm_one_reg r;
- struct kvm_reg_list rl;
- struct kvm_reg_list *rlp;
- ARMCPU *cpu = ARM_CPU(cs);
-
- init.target = KVM_ARM_TARGET_CORTEX_A15;
- memset(init.features, 0, sizeof(init.features));
- ret = kvm_vcpu_ioctl(cs, KVM_ARM_VCPU_INIT, &init);
- if (ret) {
- return ret;
- }
- /* Query the kernel to make sure it supports 32 VFP
- * registers: QEMU's "cortex-a15" CPU is always a
- * VFP-D32 core. The simplest way to do this is just
- * to attempt to read register d31.
- */
- r.id = KVM_REG_ARM | KVM_REG_SIZE_U64 | KVM_REG_ARM_VFP | 31;
- r.addr = (uintptr_t)(&v);
- ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
- if (ret == -ENOENT) {
- return -EINVAL;
- }
+ ARMHostCPUClass *ahcc = ARM_HOST_CPU_CLASS(oc);
- /* Populate the cpreg list based on the kernel's idea
- * of what registers exist (and throw away the TCG-created list).
+ /* All we really need to set up for the 'host' CPU
+ * is the feature bits -- we rely on the fact that the
+ * various ID register values in ARMCPU are only used for
+ * TCG CPUs.
*/
- rl.n = 0;
- ret = kvm_vcpu_ioctl(cs, KVM_GET_REG_LIST, &rl);
- if (ret != -E2BIG) {
- return ret;
- }
- rlp = g_malloc(sizeof(struct kvm_reg_list) + rl.n * sizeof(uint64_t));
- rlp->n = rl.n;
- ret = kvm_vcpu_ioctl(cs, KVM_GET_REG_LIST, rlp);
- if (ret) {
- goto out;
+ if (!kvm_arm_get_host_cpu_features(ahcc)) {
+ fprintf(stderr, "Failed to retrieve host CPU features!\n");
+ abort();
}
- /* Sort the list we get back from the kernel, since cpreg_tuples
- * must be in strictly ascending order.
- */
- qsort(&rlp->reg, rlp->n, sizeof(rlp->reg[0]), compare_u64);
-
- for (i = 0, arraylen = 0; i < rlp->n; i++) {
- if (!reg_syncs_via_tuple_list(rlp->reg[i])) {
- continue;
- }
- switch (rlp->reg[i] & KVM_REG_SIZE_MASK) {
- case KVM_REG_SIZE_U32:
- case KVM_REG_SIZE_U64:
- break;
- default:
- fprintf(stderr, "Can't handle size of register in kernel list\n");
- ret = -EINVAL;
- goto out;
- }
+}
- arraylen++;
- }
+static void kvm_arm_host_cpu_initfn(Object *obj)
+{
+ ARMHostCPUClass *ahcc = ARM_HOST_CPU_GET_CLASS(obj);
+ ARMCPU *cpu = ARM_CPU(obj);
+ CPUARMState *env = &cpu->env;
- cpu->cpreg_indexes = g_renew(uint64_t, cpu->cpreg_indexes, arraylen);
- cpu->cpreg_values = g_renew(uint64_t, cpu->cpreg_values, arraylen);
- cpu->cpreg_vmstate_indexes = g_renew(uint64_t, cpu->cpreg_vmstate_indexes,
- arraylen);
- cpu->cpreg_vmstate_values = g_renew(uint64_t, cpu->cpreg_vmstate_values,
- arraylen);
- cpu->cpreg_array_len = arraylen;
- cpu->cpreg_vmstate_array_len = arraylen;
-
- for (i = 0, arraylen = 0; i < rlp->n; i++) {
- uint64_t regidx = rlp->reg[i];
- if (!reg_syncs_via_tuple_list(regidx)) {
- continue;
- }
- cpu->cpreg_indexes[arraylen] = regidx;
- arraylen++;
- }
- assert(cpu->cpreg_array_len == arraylen);
+ cpu->kvm_target = ahcc->target;
+ cpu->dtb_compatible = ahcc->dtb_compatible;
+ env->features = ahcc->features;
+}
- if (!write_kvmstate_to_list(cpu)) {
- /* Shouldn't happen unless kernel is inconsistent about
- * what registers exist.
- */
- fprintf(stderr, "Initial read of kernel register state failed\n");
- ret = -EINVAL;
- goto out;
- }
+static const TypeInfo host_arm_cpu_type_info = {
+ .name = TYPE_ARM_HOST_CPU,
+#ifdef TARGET_AARCH64
+ .parent = TYPE_AARCH64_CPU,
+#else
+ .parent = TYPE_ARM_CPU,
+#endif
+ .instance_init = kvm_arm_host_cpu_initfn,
+ .class_init = kvm_arm_host_cpu_class_init,
+ .class_size = sizeof(ARMHostCPUClass),
+};
- /* Save a copy of the initial register values so that we can
- * feed it back to the kernel on VCPU reset.
+int kvm_arch_init(KVMState *s)
+{
+ /* For ARM interrupt delivery is always asynchronous,
+ * whether we are using an in-kernel VGIC or not.
*/
- cpu->cpreg_reset_values = g_memdup(cpu->cpreg_values,
- cpu->cpreg_array_len *
- sizeof(cpu->cpreg_values[0]));
+ kvm_async_interrupts_allowed = true;
+
+ type_register_static(&host_arm_cpu_type_info);
+
+ return 0;
+}
-out:
- g_free(rlp);
- return ret;
+unsigned long kvm_arch_vcpu_id(CPUState *cpu)
+{
+ return cpu->cpu_index;
}
/* We track all the KVM devices which need their memory addresses
*/
typedef struct KVMDevice {
struct kvm_arm_device_addr kda;
+ struct kvm_device_attr kdattr;
MemoryRegion *mr;
QSLIST_ENTRY(KVMDevice) entries;
+ int dev_fd;
} KVMDevice;
static QSLIST_HEAD(kvm_devices_head, KVMDevice) kvm_devices_head;
.region_del = kvm_arm_devlistener_del,
};
+static void kvm_arm_set_device_addr(KVMDevice *kd)
+{
+ struct kvm_device_attr *attr = &kd->kdattr;
+ int ret;
+
+ /* If the device control API is available and we have a device fd on the
+ * KVMDevice struct, let's use the newer API
+ */
+ if (kd->dev_fd >= 0) {
+ uint64_t addr = kd->kda.addr;
+ attr->addr = (uintptr_t)&addr;
+ ret = kvm_device_ioctl(kd->dev_fd, KVM_SET_DEVICE_ATTR, attr);
+ } else {
+ ret = kvm_vm_ioctl(kvm_state, KVM_ARM_SET_DEVICE_ADDR, &kd->kda);
+ }
+
+ if (ret < 0) {
+ fprintf(stderr, "Failed to set device address: %s\n",
+ strerror(-ret));
+ abort();
+ }
+}
+
static void kvm_arm_machine_init_done(Notifier *notifier, void *data)
{
KVMDevice *kd, *tkd;
memory_listener_unregister(&devlistener);
QSLIST_FOREACH_SAFE(kd, &kvm_devices_head, entries, tkd) {
if (kd->kda.addr != -1) {
- if (kvm_vm_ioctl(kvm_state, KVM_ARM_SET_DEVICE_ADDR,
- &kd->kda) < 0) {
- fprintf(stderr, "KVM_ARM_SET_DEVICE_ADDRESS failed: %s\n",
- strerror(errno));
- abort();
- }
+ kvm_arm_set_device_addr(kd);
}
+ memory_region_unref(kd->mr);
g_free(kd);
}
}
.notify = kvm_arm_machine_init_done,
};
-void kvm_arm_register_device(MemoryRegion *mr, uint64_t devid)
+void kvm_arm_register_device(MemoryRegion *mr, uint64_t devid, uint64_t group,
+ uint64_t attr, int dev_fd)
{
KVMDevice *kd;
kd->mr = mr;
kd->kda.id = devid;
kd->kda.addr = -1;
+ kd->kdattr.flags = 0;
+ kd->kdattr.group = group;
+ kd->kdattr.attr = attr;
+ kd->dev_fd = dev_fd;
QSLIST_INSERT_HEAD(&kvm_devices_head, kd, entries);
+ memory_region_ref(kd->mr);
}
bool write_kvmstate_to_list(ARMCPU *cpu)
return ok;
}
-typedef struct Reg {
- uint64_t id;
- int offset;
-} Reg;
-
-#define COREREG(KERNELNAME, QEMUFIELD) \
- { \
- KVM_REG_ARM | KVM_REG_SIZE_U32 | \
- KVM_REG_ARM_CORE | KVM_REG_ARM_CORE_REG(KERNELNAME), \
- offsetof(CPUARMState, QEMUFIELD) \
- }
-
-#define VFPSYSREG(R) \
- { \
- KVM_REG_ARM | KVM_REG_SIZE_U32 | KVM_REG_ARM_VFP | \
- KVM_REG_ARM_VFP_##R, \
- offsetof(CPUARMState, vfp.xregs[ARM_VFP_##R]) \
- }
-
-static const Reg regs[] = {
- /* R0_usr .. R14_usr */
- COREREG(usr_regs.uregs[0], regs[0]),
- COREREG(usr_regs.uregs[1], regs[1]),
- COREREG(usr_regs.uregs[2], regs[2]),
- COREREG(usr_regs.uregs[3], regs[3]),
- COREREG(usr_regs.uregs[4], regs[4]),
- COREREG(usr_regs.uregs[5], regs[5]),
- COREREG(usr_regs.uregs[6], regs[6]),
- COREREG(usr_regs.uregs[7], regs[7]),
- COREREG(usr_regs.uregs[8], usr_regs[0]),
- COREREG(usr_regs.uregs[9], usr_regs[1]),
- COREREG(usr_regs.uregs[10], usr_regs[2]),
- COREREG(usr_regs.uregs[11], usr_regs[3]),
- COREREG(usr_regs.uregs[12], usr_regs[4]),
- COREREG(usr_regs.uregs[13], banked_r13[0]),
- COREREG(usr_regs.uregs[14], banked_r14[0]),
- /* R13, R14, SPSR for SVC, ABT, UND, IRQ banks */
- COREREG(svc_regs[0], banked_r13[1]),
- COREREG(svc_regs[1], banked_r14[1]),
- COREREG(svc_regs[2], banked_spsr[1]),
- COREREG(abt_regs[0], banked_r13[2]),
- COREREG(abt_regs[1], banked_r14[2]),
- COREREG(abt_regs[2], banked_spsr[2]),
- COREREG(und_regs[0], banked_r13[3]),
- COREREG(und_regs[1], banked_r14[3]),
- COREREG(und_regs[2], banked_spsr[3]),
- COREREG(irq_regs[0], banked_r13[4]),
- COREREG(irq_regs[1], banked_r14[4]),
- COREREG(irq_regs[2], banked_spsr[4]),
- /* R8_fiq .. R14_fiq and SPSR_fiq */
- COREREG(fiq_regs[0], fiq_regs[0]),
- COREREG(fiq_regs[1], fiq_regs[1]),
- COREREG(fiq_regs[2], fiq_regs[2]),
- COREREG(fiq_regs[3], fiq_regs[3]),
- COREREG(fiq_regs[4], fiq_regs[4]),
- COREREG(fiq_regs[5], banked_r13[5]),
- COREREG(fiq_regs[6], banked_r14[5]),
- COREREG(fiq_regs[7], banked_spsr[5]),
- /* R15 */
- COREREG(usr_regs.uregs[15], regs[15]),
- /* VFP system registers */
- VFPSYSREG(FPSID),
- VFPSYSREG(MVFR1),
- VFPSYSREG(MVFR0),
- VFPSYSREG(FPEXC),
- VFPSYSREG(FPINST),
- VFPSYSREG(FPINST2),
-};
-
-int kvm_arch_put_registers(CPUState *cs, int level)
-{
- ARMCPU *cpu = ARM_CPU(cs);
- CPUARMState *env = &cpu->env;
- struct kvm_one_reg r;
- int mode, bn;
- int ret, i;
- uint32_t cpsr, fpscr;
-
- /* Make sure the banked regs are properly set */
- mode = env->uncached_cpsr & CPSR_M;
- bn = bank_number(mode);
- if (mode == ARM_CPU_MODE_FIQ) {
- memcpy(env->fiq_regs, env->regs + 8, 5 * sizeof(uint32_t));
- } else {
- memcpy(env->usr_regs, env->regs + 8, 5 * sizeof(uint32_t));
- }
- env->banked_r13[bn] = env->regs[13];
- env->banked_r14[bn] = env->regs[14];
- env->banked_spsr[bn] = env->spsr;
-
- /* Now we can safely copy stuff down to the kernel */
- for (i = 0; i < ARRAY_SIZE(regs); i++) {
- r.id = regs[i].id;
- r.addr = (uintptr_t)(env) + regs[i].offset;
- ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
- }
-
- /* Special cases which aren't a single CPUARMState field */
- cpsr = cpsr_read(env);
- r.id = KVM_REG_ARM | KVM_REG_SIZE_U32 |
- KVM_REG_ARM_CORE | KVM_REG_ARM_CORE_REG(usr_regs.ARM_cpsr);
- r.addr = (uintptr_t)(&cpsr);
- ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
-
- /* VFP registers */
- r.id = KVM_REG_ARM | KVM_REG_SIZE_U64 | KVM_REG_ARM_VFP;
- for (i = 0; i < 32; i++) {
- r.addr = (uintptr_t)(&env->vfp.regs[i]);
- ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
- r.id++;
- }
-
- r.id = KVM_REG_ARM | KVM_REG_SIZE_U32 | KVM_REG_ARM_VFP |
- KVM_REG_ARM_VFP_FPSCR;
- fpscr = vfp_get_fpscr(env);
- r.addr = (uintptr_t)&fpscr;
- ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
-
- /* Note that we do not call write_cpustate_to_list()
- * here, so we are only writing the tuple list back to
- * KVM. This is safe because nothing can change the
- * CPUARMState cp15 fields (in particular gdb accesses cannot)
- * and so there are no changes to sync. In fact syncing would
- * be wrong at this point: for a constant register where TCG and
- * KVM disagree about its value, the preceding write_list_to_cpustate()
- * would not have had any effect on the CPUARMState value (since the
- * register is read-only), and a write_cpustate_to_list() here would
- * then try to write the TCG value back into KVM -- this would either
- * fail or incorrectly change the value the guest sees.
- *
- * If we ever want to allow the user to modify cp15 registers via
- * the gdb stub, we would need to be more clever here (for instance
- * tracking the set of registers kvm_arch_get_registers() successfully
- * managed to update the CPUARMState with, and only allowing those
- * to be written back up into the kernel).
- */
- if (!write_list_to_kvmstate(cpu)) {
- return EINVAL;
- }
-
- return ret;
-}
-
-int kvm_arch_get_registers(CPUState *cs)
-{
- ARMCPU *cpu = ARM_CPU(cs);
- CPUARMState *env = &cpu->env;
- struct kvm_one_reg r;
- int mode, bn;
- int ret, i;
- uint32_t cpsr, fpscr;
-
- for (i = 0; i < ARRAY_SIZE(regs); i++) {
- r.id = regs[i].id;
- r.addr = (uintptr_t)(env) + regs[i].offset;
- ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
- }
-
- /* Special cases which aren't a single CPUARMState field */
- r.id = KVM_REG_ARM | KVM_REG_SIZE_U32 |
- KVM_REG_ARM_CORE | KVM_REG_ARM_CORE_REG(usr_regs.ARM_cpsr);
- r.addr = (uintptr_t)(&cpsr);
- ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
- cpsr_write(env, cpsr, 0xffffffff);
-
- /* Make sure the current mode regs are properly set */
- mode = env->uncached_cpsr & CPSR_M;
- bn = bank_number(mode);
- if (mode == ARM_CPU_MODE_FIQ) {
- memcpy(env->regs + 8, env->fiq_regs, 5 * sizeof(uint32_t));
- } else {
- memcpy(env->regs + 8, env->usr_regs, 5 * sizeof(uint32_t));
- }
- env->regs[13] = env->banked_r13[bn];
- env->regs[14] = env->banked_r14[bn];
- env->spsr = env->banked_spsr[bn];
-
- /* VFP registers */
- r.id = KVM_REG_ARM | KVM_REG_SIZE_U64 | KVM_REG_ARM_VFP;
- for (i = 0; i < 32; i++) {
- r.addr = (uintptr_t)(&env->vfp.regs[i]);
- ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
- r.id++;
- }
-
- r.id = KVM_REG_ARM | KVM_REG_SIZE_U32 | KVM_REG_ARM_VFP |
- KVM_REG_ARM_VFP_FPSCR;
- r.addr = (uintptr_t)&fpscr;
- ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
- if (ret) {
- return ret;
- }
- vfp_set_fpscr(env, fpscr);
-
- if (!write_kvmstate_to_list(cpu)) {
- return EINVAL;
- }
- /* Note that it's OK to have registers which aren't in CPUState,
- * so we can ignore a failure return here.
- */
- write_list_to_cpustate(cpu);
-
- return 0;
-}
-
void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
{
}
return 0;
}
-void kvm_arch_reset_vcpu(CPUState *cs)
-{
- /* Feed the kernel back its initial register state */
- ARMCPU *cpu = ARM_CPU(cs);
-
- memmove(cpu->cpreg_values, cpu->cpreg_reset_values,
- cpu->cpreg_array_len * sizeof(cpu->cpreg_values[0]));
-
- if (!write_list_to_kvmstate(cpu)) {
- abort();
- }
-}
-
bool kvm_arch_stop_on_emulation_error(CPUState *cs)
{
return true;
void kvm_arch_init_irq_routing(KVMState *s)
{
}
+
+int kvm_arch_irqchip_create(KVMState *s)
+{
+ int ret;
+
+ /* If we can create the VGIC using the newer device control API, we
+ * let the device do this when it initializes itself, otherwise we
+ * fall back to the old API */
+
+ ret = kvm_create_device(s, KVM_DEV_TYPE_ARM_VGIC_V2, true);
+ if (ret == 0) {
+ return 1;
+ }
+
+ return 0;
+}
projects / qemu.git / blobdiff