2013-03-05 04:34:41 +04:00
|
|
|
/*
|
|
|
|
* ARM implementation of KVM hooks
|
|
|
|
*
|
|
|
|
* Copyright Christoffer Dall 2009-2010
|
|
|
|
*
|
|
|
|
* This work is licensed under the terms of the GNU GPL, version 2 or later.
|
|
|
|
* See the COPYING file in the top-level directory.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
2015-12-07 19:23:44 +03:00
|
|
|
#include "qemu/osdep.h"
|
2013-03-05 04:34:41 +04:00
|
|
|
#include <sys/ioctl.h>
|
|
|
|
|
|
|
|
#include <linux/kvm.h>
|
|
|
|
|
|
|
|
#include "qemu-common.h"
|
|
|
|
#include "qemu/timer.h"
|
2015-12-17 16:37:15 +03:00
|
|
|
#include "qemu/error-report.h"
|
Include qemu/main-loop.h less
In my "build everything" tree, changing qemu/main-loop.h triggers a
recompile of some 5600 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h). It includes block/aio.h,
which in turn includes qemu/event_notifier.h, qemu/notify.h,
qemu/processor.h, qemu/qsp.h, qemu/queue.h, qemu/thread-posix.h,
qemu/thread.h, qemu/timer.h, and a few more.
Include qemu/main-loop.h only where it's needed. Touching it now
recompiles only some 1700 objects. For block/aio.h and
qemu/event_notifier.h, these numbers drop from 5600 to 2800. For the
others, they shrink only slightly.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190812052359.30071-21-armbru@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
2019-08-12 08:23:50 +03:00
|
|
|
#include "qemu/main-loop.h"
|
2013-03-05 04:34:41 +04:00
|
|
|
#include "sysemu/sysemu.h"
|
|
|
|
#include "sysemu/kvm.h"
|
2019-03-04 13:13:34 +03:00
|
|
|
#include "sysemu/kvm_int.h"
|
2013-03-05 04:34:42 +04:00
|
|
|
#include "kvm_arm.h"
|
2013-03-05 04:34:41 +04:00
|
|
|
#include "cpu.h"
|
2018-05-04 20:05:52 +03:00
|
|
|
#include "trace.h"
|
2014-12-11 15:07:53 +03:00
|
|
|
#include "internals.h"
|
2018-05-04 20:05:52 +03:00
|
|
|
#include "hw/pci/pci.h"
|
2015-04-08 14:30:58 +03:00
|
|
|
#include "exec/memattrs.h"
|
2016-10-17 21:22:16 +03:00
|
|
|
#include "exec/address-spaces.h"
|
2015-12-17 19:16:08 +03:00
|
|
|
#include "hw/boards.h"
|
2019-08-12 08:23:42 +03:00
|
|
|
#include "hw/irq.h"
|
2015-12-15 15:16:16 +03:00
|
|
|
#include "qemu/log.h"
|
2013-03-05 04:34:41 +04:00
|
|
|
|
|
|
|
const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
|
|
|
|
KVM_CAP_LAST_INFO
|
|
|
|
};
|
|
|
|
|
2015-04-01 19:57:30 +03:00
|
|
|
static bool cap_has_mp_state;
|
2018-10-24 09:50:16 +03:00
|
|
|
static bool cap_has_inject_serror_esr;
|
2015-04-01 19:57:30 +03:00
|
|
|
|
2018-03-09 20:09:44 +03:00
|
|
|
static ARMHostCPUFeatures arm_host_cpu_features;
|
|
|
|
|
2014-06-19 21:06:26 +04:00
|
|
|
int kvm_arm_vcpu_init(CPUState *cs)
|
|
|
|
{
|
|
|
|
ARMCPU *cpu = ARM_CPU(cs);
|
|
|
|
struct kvm_vcpu_init init;
|
|
|
|
|
|
|
|
init.target = cpu->kvm_target;
|
|
|
|
memcpy(init.features, cpu->kvm_init_features, sizeof(init.features));
|
|
|
|
|
|
|
|
return kvm_vcpu_ioctl(cs, KVM_ARM_VCPU_INIT, &init);
|
|
|
|
}
|
|
|
|
|
2018-10-24 09:50:16 +03:00
|
|
|
void kvm_arm_init_serror_injection(CPUState *cs)
|
|
|
|
{
|
|
|
|
cap_has_inject_serror_esr = kvm_check_extension(cs->kvm_state,
|
|
|
|
KVM_CAP_ARM_INJECT_SERROR_ESR);
|
|
|
|
}
|
|
|
|
|
2013-11-22 21:17:17 +04:00
|
|
|
bool kvm_arm_create_scratch_host_vcpu(const uint32_t *cpus_to_try,
|
|
|
|
int *fdarray,
|
|
|
|
struct kvm_vcpu_init *init)
|
|
|
|
{
|
|
|
|
int ret, kvmfd = -1, vmfd = -1, cpufd = -1;
|
|
|
|
|
|
|
|
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;
|
|
|
|
}
|
|
|
|
|
2016-03-30 19:27:24 +03:00
|
|
|
if (!init) {
|
|
|
|
/* Caller doesn't want the VCPU to be initialized, so skip it */
|
|
|
|
goto finish;
|
|
|
|
}
|
|
|
|
|
2013-11-22 21:17:17 +04:00
|
|
|
ret = ioctl(vmfd, KVM_ARM_PREFERRED_TARGET, init);
|
|
|
|
if (ret >= 0) {
|
|
|
|
ret = ioctl(cpufd, KVM_ARM_VCPU_INIT, init);
|
|
|
|
if (ret < 0) {
|
|
|
|
goto err;
|
|
|
|
}
|
2016-03-30 19:27:24 +03:00
|
|
|
} else if (cpus_to_try) {
|
2013-11-22 21:17:17 +04:00
|
|
|
/* 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;
|
|
|
|
}
|
2016-03-30 19:27:24 +03:00
|
|
|
} else {
|
|
|
|
/* Treat a NULL cpus_to_try argument the same as an empty
|
|
|
|
* list, which means we will fail the call since this must
|
|
|
|
* be an old kernel which doesn't support PREFERRED_TARGET.
|
|
|
|
*/
|
|
|
|
goto err;
|
2013-11-22 21:17:17 +04:00
|
|
|
}
|
|
|
|
|
2016-03-30 19:27:24 +03:00
|
|
|
finish:
|
2013-11-22 21:17:17 +04:00
|
|
|
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;
|
|
|
|
}
|
|
|
|
|
|
|
|
void kvm_arm_destroy_scratch_host_vcpu(int *fdarray)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 2; i >= 0; i--) {
|
|
|
|
close(fdarray[i]);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-03-09 20:09:44 +03:00
|
|
|
void kvm_arm_set_cpu_features_from_host(ARMCPU *cpu)
|
2013-11-22 21:17:17 +04:00
|
|
|
{
|
2018-03-09 20:09:44 +03:00
|
|
|
CPUARMState *env = &cpu->env;
|
2013-11-22 21:17:17 +04:00
|
|
|
|
2018-03-09 20:09:44 +03:00
|
|
|
if (!arm_host_cpu_features.dtb_compatible) {
|
|
|
|
if (!kvm_enabled() ||
|
|
|
|
!kvm_arm_get_host_cpu_features(&arm_host_cpu_features)) {
|
|
|
|
/* We can't report this error yet, so flag that we need to
|
|
|
|
* in arm_cpu_realizefn().
|
|
|
|
*/
|
|
|
|
cpu->kvm_target = QEMU_KVM_ARM_TARGET_NONE;
|
|
|
|
cpu->host_cpu_probe_failed = true;
|
|
|
|
return;
|
|
|
|
}
|
2013-11-22 21:17:17 +04:00
|
|
|
}
|
2018-03-09 20:09:44 +03:00
|
|
|
|
|
|
|
cpu->kvm_target = arm_host_cpu_features.target;
|
|
|
|
cpu->dtb_compatible = arm_host_cpu_features.dtb_compatible;
|
2018-11-19 18:29:07 +03:00
|
|
|
cpu->isar = arm_host_cpu_features.isar;
|
2018-03-09 20:09:44 +03:00
|
|
|
env->features = arm_host_cpu_features.features;
|
2013-11-22 21:17:17 +04:00
|
|
|
}
|
|
|
|
|
2019-08-02 15:25:27 +03:00
|
|
|
bool kvm_arm_pmu_supported(CPUState *cpu)
|
|
|
|
{
|
|
|
|
KVMState *s = KVM_STATE(current_machine->accelerator);
|
|
|
|
|
|
|
|
return kvm_check_extension(s, KVM_CAP_ARM_PMU_V3);
|
|
|
|
}
|
|
|
|
|
2019-03-04 13:13:34 +03:00
|
|
|
int kvm_arm_get_max_vm_ipa_size(MachineState *ms)
|
|
|
|
{
|
|
|
|
KVMState *s = KVM_STATE(ms->accelerator);
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
ret = kvm_check_extension(s, KVM_CAP_ARM_VM_IPA_SIZE);
|
|
|
|
return ret > 0 ? ret : 40;
|
|
|
|
}
|
|
|
|
|
2015-02-04 18:43:51 +03:00
|
|
|
int kvm_arch_init(MachineState *ms, KVMState *s)
|
2013-03-05 04:34:41 +04:00
|
|
|
{
|
|
|
|
/* For ARM interrupt delivery is always asynchronous,
|
|
|
|
* whether we are using an in-kernel VGIC or not.
|
|
|
|
*/
|
|
|
|
kvm_async_interrupts_allowed = true;
|
2013-11-22 21:17:17 +04:00
|
|
|
|
2017-07-11 13:21:26 +03:00
|
|
|
/*
|
|
|
|
* PSCI wakes up secondary cores, so we always need to
|
|
|
|
* have vCPUs waiting in kernel space
|
|
|
|
*/
|
|
|
|
kvm_halt_in_kernel_allowed = true;
|
|
|
|
|
2015-04-01 19:57:30 +03:00
|
|
|
cap_has_mp_state = kvm_check_extension(s, KVM_CAP_MP_STATE);
|
|
|
|
|
2013-03-05 04:34:41 +04:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned long kvm_arch_vcpu_id(CPUState *cpu)
|
|
|
|
{
|
|
|
|
return cpu->cpu_index;
|
|
|
|
}
|
|
|
|
|
2013-03-05 04:34:42 +04:00
|
|
|
/* We track all the KVM devices which need their memory addresses
|
|
|
|
* passing to the kernel in a list of these structures.
|
|
|
|
* When board init is complete we run through the list and
|
|
|
|
* tell the kernel the base addresses of the memory regions.
|
|
|
|
* We use a MemoryListener to track mapping and unmapping of
|
|
|
|
* the regions during board creation, so the board models don't
|
|
|
|
* need to do anything special for the KVM case.
|
2018-06-22 15:28:35 +03:00
|
|
|
*
|
|
|
|
* Sometimes the address must be OR'ed with some other fields
|
|
|
|
* (for example for KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION).
|
|
|
|
* @kda_addr_ormask aims at storing the value of those fields.
|
2013-03-05 04:34:42 +04:00
|
|
|
*/
|
|
|
|
typedef struct KVMDevice {
|
|
|
|
struct kvm_arm_device_addr kda;
|
2014-02-26 21:20:00 +04:00
|
|
|
struct kvm_device_attr kdattr;
|
2018-06-22 15:28:35 +03:00
|
|
|
uint64_t kda_addr_ormask;
|
2013-03-05 04:34:42 +04:00
|
|
|
MemoryRegion *mr;
|
|
|
|
QSLIST_ENTRY(KVMDevice) entries;
|
2014-02-26 21:20:00 +04:00
|
|
|
int dev_fd;
|
2013-03-05 04:34:42 +04:00
|
|
|
} KVMDevice;
|
|
|
|
|
2018-12-06 13:58:10 +03:00
|
|
|
static QSLIST_HEAD(, KVMDevice) kvm_devices_head;
|
2013-03-05 04:34:42 +04:00
|
|
|
|
|
|
|
static void kvm_arm_devlistener_add(MemoryListener *listener,
|
|
|
|
MemoryRegionSection *section)
|
|
|
|
{
|
|
|
|
KVMDevice *kd;
|
|
|
|
|
|
|
|
QSLIST_FOREACH(kd, &kvm_devices_head, entries) {
|
|
|
|
if (section->mr == kd->mr) {
|
|
|
|
kd->kda.addr = section->offset_within_address_space;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void kvm_arm_devlistener_del(MemoryListener *listener,
|
|
|
|
MemoryRegionSection *section)
|
|
|
|
{
|
|
|
|
KVMDevice *kd;
|
|
|
|
|
|
|
|
QSLIST_FOREACH(kd, &kvm_devices_head, entries) {
|
|
|
|
if (section->mr == kd->mr) {
|
|
|
|
kd->kda.addr = -1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static MemoryListener devlistener = {
|
|
|
|
.region_add = kvm_arm_devlistener_add,
|
|
|
|
.region_del = kvm_arm_devlistener_del,
|
|
|
|
};
|
|
|
|
|
2014-02-26 21:20:00 +04:00
|
|
|
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;
|
2018-06-22 15:28:35 +03:00
|
|
|
|
|
|
|
addr |= kd->kda_addr_ormask;
|
2014-02-26 21:20:00 +04:00
|
|
|
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();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-03-05 04:34:42 +04:00
|
|
|
static void kvm_arm_machine_init_done(Notifier *notifier, void *data)
|
|
|
|
{
|
|
|
|
KVMDevice *kd, *tkd;
|
|
|
|
|
|
|
|
QSLIST_FOREACH_SAFE(kd, &kvm_devices_head, entries, tkd) {
|
|
|
|
if (kd->kda.addr != -1) {
|
2014-02-26 21:20:00 +04:00
|
|
|
kvm_arm_set_device_addr(kd);
|
2013-03-05 04:34:42 +04:00
|
|
|
}
|
2013-05-06 12:46:11 +04:00
|
|
|
memory_region_unref(kd->mr);
|
2018-06-22 15:28:35 +03:00
|
|
|
QSLIST_REMOVE_HEAD(&kvm_devices_head, entries);
|
2013-03-05 04:34:42 +04:00
|
|
|
g_free(kd);
|
|
|
|
}
|
2018-01-22 09:02:42 +03:00
|
|
|
memory_listener_unregister(&devlistener);
|
2013-03-05 04:34:42 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
static Notifier notify = {
|
|
|
|
.notify = kvm_arm_machine_init_done,
|
|
|
|
};
|
|
|
|
|
2014-02-26 21:20:00 +04:00
|
|
|
void kvm_arm_register_device(MemoryRegion *mr, uint64_t devid, uint64_t group,
|
2018-06-22 15:28:35 +03:00
|
|
|
uint64_t attr, int dev_fd, uint64_t addr_ormask)
|
2013-03-05 04:34:42 +04:00
|
|
|
{
|
|
|
|
KVMDevice *kd;
|
|
|
|
|
|
|
|
if (!kvm_irqchip_in_kernel()) {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (QSLIST_EMPTY(&kvm_devices_head)) {
|
2016-10-17 21:22:16 +03:00
|
|
|
memory_listener_register(&devlistener, &address_space_memory);
|
2013-03-05 04:34:42 +04:00
|
|
|
qemu_add_machine_init_done_notifier(¬ify);
|
|
|
|
}
|
|
|
|
kd = g_new0(KVMDevice, 1);
|
|
|
|
kd->mr = mr;
|
|
|
|
kd->kda.id = devid;
|
|
|
|
kd->kda.addr = -1;
|
2014-02-26 21:20:00 +04:00
|
|
|
kd->kdattr.flags = 0;
|
|
|
|
kd->kdattr.group = group;
|
|
|
|
kd->kdattr.attr = attr;
|
|
|
|
kd->dev_fd = dev_fd;
|
2018-06-22 15:28:35 +03:00
|
|
|
kd->kda_addr_ormask = addr_ormask;
|
2013-03-05 04:34:42 +04:00
|
|
|
QSLIST_INSERT_HEAD(&kvm_devices_head, kd, entries);
|
2013-05-06 12:46:11 +04:00
|
|
|
memory_region_ref(kd->mr);
|
2013-03-05 04:34:42 +04:00
|
|
|
}
|
|
|
|
|
2014-12-11 15:07:53 +03:00
|
|
|
static int compare_u64(const void *a, const void *b)
|
|
|
|
{
|
|
|
|
if (*(uint64_t *)a > *(uint64_t *)b) {
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
if (*(uint64_t *)a < *(uint64_t *)b) {
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-10-08 16:55:02 +03:00
|
|
|
/* Initialize the ARMCPU cpreg list according to the kernel's
|
2014-12-11 15:07:53 +03:00
|
|
|
* definition of what CPU registers it knows about (and throw away
|
|
|
|
* the previous TCG-created cpreg list).
|
|
|
|
*/
|
|
|
|
int kvm_arm_init_cpreg_list(ARMCPU *cpu)
|
|
|
|
{
|
|
|
|
struct kvm_reg_list rl;
|
|
|
|
struct kvm_reg_list *rlp;
|
|
|
|
int i, ret, arraylen;
|
|
|
|
CPUState *cs = CPU(cpu);
|
|
|
|
|
|
|
|
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;
|
|
|
|
}
|
|
|
|
/* 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 (!kvm_arm_reg_syncs_via_cpreg_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++;
|
|
|
|
}
|
|
|
|
|
|
|
|
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 (!kvm_arm_reg_syncs_via_cpreg_list(regidx)) {
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
cpu->cpreg_indexes[arraylen] = regidx;
|
|
|
|
arraylen++;
|
|
|
|
}
|
|
|
|
assert(cpu->cpreg_array_len == arraylen);
|
|
|
|
|
|
|
|
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;
|
|
|
|
}
|
|
|
|
|
|
|
|
out:
|
|
|
|
g_free(rlp);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2013-06-25 21:16:07 +04:00
|
|
|
bool write_kvmstate_to_list(ARMCPU *cpu)
|
|
|
|
{
|
|
|
|
CPUState *cs = CPU(cpu);
|
|
|
|
int i;
|
|
|
|
bool ok = true;
|
|
|
|
|
|
|
|
for (i = 0; i < cpu->cpreg_array_len; i++) {
|
|
|
|
struct kvm_one_reg r;
|
|
|
|
uint64_t regidx = cpu->cpreg_indexes[i];
|
|
|
|
uint32_t v32;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
r.id = regidx;
|
|
|
|
|
|
|
|
switch (regidx & KVM_REG_SIZE_MASK) {
|
|
|
|
case KVM_REG_SIZE_U32:
|
|
|
|
r.addr = (uintptr_t)&v32;
|
|
|
|
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
|
|
|
|
if (!ret) {
|
|
|
|
cpu->cpreg_values[i] = v32;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case KVM_REG_SIZE_U64:
|
|
|
|
r.addr = (uintptr_t)(cpu->cpreg_values + i);
|
|
|
|
ret = kvm_vcpu_ioctl(cs, KVM_GET_ONE_REG, &r);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
abort();
|
|
|
|
}
|
|
|
|
if (ret) {
|
|
|
|
ok = false;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return ok;
|
|
|
|
}
|
|
|
|
|
2015-07-21 13:18:45 +03:00
|
|
|
bool write_list_to_kvmstate(ARMCPU *cpu, int level)
|
2013-06-25 21:16:07 +04:00
|
|
|
{
|
|
|
|
CPUState *cs = CPU(cpu);
|
|
|
|
int i;
|
|
|
|
bool ok = true;
|
|
|
|
|
|
|
|
for (i = 0; i < cpu->cpreg_array_len; i++) {
|
|
|
|
struct kvm_one_reg r;
|
|
|
|
uint64_t regidx = cpu->cpreg_indexes[i];
|
|
|
|
uint32_t v32;
|
|
|
|
int ret;
|
|
|
|
|
2015-07-21 13:18:45 +03:00
|
|
|
if (kvm_arm_cpreg_level(regidx) > level) {
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2013-06-25 21:16:07 +04:00
|
|
|
r.id = regidx;
|
|
|
|
switch (regidx & KVM_REG_SIZE_MASK) {
|
|
|
|
case KVM_REG_SIZE_U32:
|
|
|
|
v32 = cpu->cpreg_values[i];
|
|
|
|
r.addr = (uintptr_t)&v32;
|
|
|
|
break;
|
|
|
|
case KVM_REG_SIZE_U64:
|
|
|
|
r.addr = (uintptr_t)(cpu->cpreg_values + i);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
abort();
|
|
|
|
}
|
|
|
|
ret = kvm_vcpu_ioctl(cs, KVM_SET_ONE_REG, &r);
|
|
|
|
if (ret) {
|
|
|
|
/* We might fail for "unknown register" and also for
|
|
|
|
* "you tried to set a register which is constant with
|
|
|
|
* a different value from what it actually contains".
|
|
|
|
*/
|
|
|
|
ok = false;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return ok;
|
|
|
|
}
|
|
|
|
|
2014-12-11 15:07:53 +03:00
|
|
|
void kvm_arm_reset_vcpu(ARMCPU *cpu)
|
|
|
|
{
|
2014-12-11 15:07:53 +03:00
|
|
|
int ret;
|
|
|
|
|
2014-12-11 15:07:53 +03:00
|
|
|
/* Re-init VCPU so that all registers are set to
|
|
|
|
* their respective reset values.
|
|
|
|
*/
|
2014-12-11 15:07:53 +03:00
|
|
|
ret = kvm_arm_vcpu_init(CPU(cpu));
|
|
|
|
if (ret < 0) {
|
|
|
|
fprintf(stderr, "kvm_arm_vcpu_init failed: %s\n", strerror(-ret));
|
|
|
|
abort();
|
|
|
|
}
|
|
|
|
if (!write_kvmstate_to_list(cpu)) {
|
|
|
|
fprintf(stderr, "write_kvmstate_to_list failed\n");
|
|
|
|
abort();
|
|
|
|
}
|
2019-05-07 14:55:02 +03:00
|
|
|
/*
|
|
|
|
* Sync the reset values also into the CPUState. This is necessary
|
|
|
|
* because the next thing we do will be a kvm_arch_put_registers()
|
|
|
|
* which will update the list values from the CPUState before copying
|
|
|
|
* the list values back to KVM. It's OK to ignore failure returns here
|
|
|
|
* for the same reason we do so in kvm_arch_get_registers().
|
|
|
|
*/
|
|
|
|
write_list_to_cpustate(cpu);
|
2014-12-11 15:07:53 +03:00
|
|
|
}
|
|
|
|
|
2015-04-01 19:57:30 +03:00
|
|
|
/*
|
|
|
|
* Update KVM's MP_STATE based on what QEMU thinks it is
|
|
|
|
*/
|
|
|
|
int kvm_arm_sync_mpstate_to_kvm(ARMCPU *cpu)
|
|
|
|
{
|
|
|
|
if (cap_has_mp_state) {
|
|
|
|
struct kvm_mp_state mp_state = {
|
2017-02-23 21:29:23 +03:00
|
|
|
.mp_state = (cpu->power_state == PSCI_OFF) ?
|
|
|
|
KVM_MP_STATE_STOPPED : KVM_MP_STATE_RUNNABLE
|
2015-04-01 19:57:30 +03:00
|
|
|
};
|
|
|
|
int ret = kvm_vcpu_ioctl(CPU(cpu), KVM_SET_MP_STATE, &mp_state);
|
|
|
|
if (ret) {
|
|
|
|
fprintf(stderr, "%s: failed to set MP_STATE %d/%s\n",
|
|
|
|
__func__, ret, strerror(-ret));
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Sync the KVM MP_STATE into QEMU
|
|
|
|
*/
|
|
|
|
int kvm_arm_sync_mpstate_to_qemu(ARMCPU *cpu)
|
|
|
|
{
|
|
|
|
if (cap_has_mp_state) {
|
|
|
|
struct kvm_mp_state mp_state;
|
|
|
|
int ret = kvm_vcpu_ioctl(CPU(cpu), KVM_GET_MP_STATE, &mp_state);
|
|
|
|
if (ret) {
|
|
|
|
fprintf(stderr, "%s: failed to get MP_STATE %d/%s\n",
|
|
|
|
__func__, ret, strerror(-ret));
|
|
|
|
abort();
|
|
|
|
}
|
2017-02-23 21:29:23 +03:00
|
|
|
cpu->power_state = (mp_state.mp_state == KVM_MP_STATE_STOPPED) ?
|
|
|
|
PSCI_OFF : PSCI_ON;
|
2015-04-01 19:57:30 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2018-10-24 09:50:16 +03:00
|
|
|
int kvm_put_vcpu_events(ARMCPU *cpu)
|
|
|
|
{
|
|
|
|
CPUARMState *env = &cpu->env;
|
|
|
|
struct kvm_vcpu_events events;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (!kvm_has_vcpu_events()) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
memset(&events, 0, sizeof(events));
|
|
|
|
events.exception.serror_pending = env->serror.pending;
|
|
|
|
|
|
|
|
/* Inject SError to guest with specified syndrome if host kernel
|
|
|
|
* supports it, otherwise inject SError without syndrome.
|
|
|
|
*/
|
|
|
|
if (cap_has_inject_serror_esr) {
|
|
|
|
events.exception.serror_has_esr = env->serror.has_esr;
|
|
|
|
events.exception.serror_esr = env->serror.esr;
|
|
|
|
}
|
|
|
|
|
|
|
|
ret = kvm_vcpu_ioctl(CPU(cpu), KVM_SET_VCPU_EVENTS, &events);
|
|
|
|
if (ret) {
|
|
|
|
error_report("failed to put vcpu events");
|
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int kvm_get_vcpu_events(ARMCPU *cpu)
|
|
|
|
{
|
|
|
|
CPUARMState *env = &cpu->env;
|
|
|
|
struct kvm_vcpu_events events;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
if (!kvm_has_vcpu_events()) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
memset(&events, 0, sizeof(events));
|
|
|
|
ret = kvm_vcpu_ioctl(CPU(cpu), KVM_GET_VCPU_EVENTS, &events);
|
|
|
|
if (ret) {
|
|
|
|
error_report("failed to get vcpu events");
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
env->serror.pending = events.exception.serror_pending;
|
|
|
|
env->serror.has_esr = events.exception.serror_has_esr;
|
|
|
|
env->serror.esr = events.exception.serror_esr;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-03-05 04:34:41 +04:00
|
|
|
void kvm_arch_pre_run(CPUState *cs, struct kvm_run *run)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2015-04-08 14:30:58 +03:00
|
|
|
MemTxAttrs kvm_arch_post_run(CPUState *cs, struct kvm_run *run)
|
2013-03-05 04:34:41 +04:00
|
|
|
{
|
2017-07-11 13:21:26 +03:00
|
|
|
ARMCPU *cpu;
|
|
|
|
uint32_t switched_level;
|
|
|
|
|
|
|
|
if (kvm_irqchip_in_kernel()) {
|
|
|
|
/*
|
|
|
|
* We only need to sync timer states with user-space interrupt
|
|
|
|
* controllers, so return early and save cycles if we don't.
|
|
|
|
*/
|
|
|
|
return MEMTXATTRS_UNSPECIFIED;
|
|
|
|
}
|
|
|
|
|
|
|
|
cpu = ARM_CPU(cs);
|
|
|
|
|
|
|
|
/* Synchronize our shadowed in-kernel device irq lines with the kvm ones */
|
|
|
|
if (run->s.regs.device_irq_level != cpu->device_irq_level) {
|
|
|
|
switched_level = cpu->device_irq_level ^ run->s.regs.device_irq_level;
|
|
|
|
|
|
|
|
qemu_mutex_lock_iothread();
|
|
|
|
|
|
|
|
if (switched_level & KVM_ARM_DEV_EL1_VTIMER) {
|
|
|
|
qemu_set_irq(cpu->gt_timer_outputs[GTIMER_VIRT],
|
|
|
|
!!(run->s.regs.device_irq_level &
|
|
|
|
KVM_ARM_DEV_EL1_VTIMER));
|
|
|
|
switched_level &= ~KVM_ARM_DEV_EL1_VTIMER;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (switched_level & KVM_ARM_DEV_EL1_PTIMER) {
|
|
|
|
qemu_set_irq(cpu->gt_timer_outputs[GTIMER_PHYS],
|
|
|
|
!!(run->s.regs.device_irq_level &
|
|
|
|
KVM_ARM_DEV_EL1_PTIMER));
|
|
|
|
switched_level &= ~KVM_ARM_DEV_EL1_PTIMER;
|
|
|
|
}
|
|
|
|
|
2017-09-04 17:21:54 +03:00
|
|
|
if (switched_level & KVM_ARM_DEV_PMU) {
|
|
|
|
qemu_set_irq(cpu->pmu_interrupt,
|
|
|
|
!!(run->s.regs.device_irq_level & KVM_ARM_DEV_PMU));
|
|
|
|
switched_level &= ~KVM_ARM_DEV_PMU;
|
|
|
|
}
|
2017-07-11 13:21:26 +03:00
|
|
|
|
|
|
|
if (switched_level) {
|
|
|
|
qemu_log_mask(LOG_UNIMP, "%s: unhandled in-kernel device IRQ %x\n",
|
|
|
|
__func__, switched_level);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We also mark unknown levels as processed to not waste cycles */
|
|
|
|
cpu->device_irq_level = run->s.regs.device_irq_level;
|
|
|
|
qemu_mutex_unlock_iothread();
|
|
|
|
}
|
|
|
|
|
2015-04-08 14:30:58 +03:00
|
|
|
return MEMTXATTRS_UNSPECIFIED;
|
2013-03-05 04:34:41 +04:00
|
|
|
}
|
|
|
|
|
2015-12-17 16:37:15 +03:00
|
|
|
|
2013-03-05 04:34:41 +04:00
|
|
|
int kvm_arch_handle_exit(CPUState *cs, struct kvm_run *run)
|
|
|
|
{
|
2015-12-17 16:37:15 +03:00
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
switch (run->exit_reason) {
|
|
|
|
case KVM_EXIT_DEBUG:
|
|
|
|
if (kvm_arm_handle_debug(cs, &run->debug.arch)) {
|
|
|
|
ret = EXCP_DEBUG;
|
|
|
|
} /* otherwise return to guest */
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
qemu_log_mask(LOG_UNIMP, "%s: un-handled exit reason %d\n",
|
|
|
|
__func__, run->exit_reason);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return ret;
|
2013-03-05 04:34:41 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
bool kvm_arch_stop_on_emulation_error(CPUState *cs)
|
|
|
|
{
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
int kvm_arch_process_async_events(CPUState *cs)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-12-17 16:37:15 +03:00
|
|
|
/* The #ifdef protections are until 32bit headers are imported and can
|
|
|
|
* be removed once both 32 and 64 bit reach feature parity.
|
|
|
|
*/
|
2013-03-05 04:34:41 +04:00
|
|
|
void kvm_arch_update_guest_debug(CPUState *cs, struct kvm_guest_debug *dbg)
|
|
|
|
{
|
2015-12-17 16:37:15 +03:00
|
|
|
#ifdef KVM_GUESTDBG_USE_SW_BP
|
|
|
|
if (kvm_sw_breakpoints_active(cs)) {
|
|
|
|
dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP;
|
|
|
|
}
|
|
|
|
#endif
|
2015-12-17 16:37:15 +03:00
|
|
|
#ifdef KVM_GUESTDBG_USE_HW
|
|
|
|
if (kvm_arm_hw_debug_active(cs)) {
|
|
|
|
dbg->control |= KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW;
|
|
|
|
kvm_arm_copy_hw_debug_data(&dbg->arch);
|
|
|
|
}
|
|
|
|
#endif
|
2013-03-05 04:34:41 +04:00
|
|
|
}
|
2013-06-12 11:26:52 +04:00
|
|
|
|
|
|
|
void kvm_arch_init_irq_routing(KVMState *s)
|
|
|
|
{
|
|
|
|
}
|
2014-02-26 21:20:00 +04:00
|
|
|
|
2015-12-17 19:16:08 +03:00
|
|
|
int kvm_arch_irqchip_create(MachineState *ms, KVMState *s)
|
2014-02-26 21:20:00 +04:00
|
|
|
{
|
2015-12-17 19:16:08 +03:00
|
|
|
if (machine_kernel_irqchip_split(ms)) {
|
|
|
|
perror("-machine kernel_irqchip=split is not supported on ARM.");
|
|
|
|
exit(1);
|
|
|
|
}
|
|
|
|
|
2014-02-26 21:20:00 +04:00
|
|
|
/* 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 */
|
2015-09-24 03:29:37 +03:00
|
|
|
return kvm_check_extension(s, KVM_CAP_DEVICE_CTRL);
|
|
|
|
}
|
2014-02-26 21:20:00 +04:00
|
|
|
|
2015-09-24 03:29:37 +03:00
|
|
|
int kvm_arm_vgic_probe(void)
|
|
|
|
{
|
|
|
|
if (kvm_create_device(kvm_state,
|
|
|
|
KVM_DEV_TYPE_ARM_VGIC_V3, true) == 0) {
|
|
|
|
return 3;
|
|
|
|
} else if (kvm_create_device(kvm_state,
|
|
|
|
KVM_DEV_TYPE_ARM_VGIC_V2, true) == 0) {
|
|
|
|
return 2;
|
|
|
|
} else {
|
|
|
|
return 0;
|
2014-02-26 21:20:00 +04:00
|
|
|
}
|
|
|
|
}
|
2015-01-09 11:04:40 +03:00
|
|
|
|
|
|
|
int kvm_arch_fixup_msi_route(struct kvm_irq_routing_entry *route,
|
2015-10-15 16:44:52 +03:00
|
|
|
uint64_t address, uint32_t data, PCIDevice *dev)
|
2015-01-09 11:04:40 +03:00
|
|
|
{
|
2018-05-04 20:05:52 +03:00
|
|
|
AddressSpace *as = pci_device_iommu_address_space(dev);
|
|
|
|
hwaddr xlat, len, doorbell_gpa;
|
|
|
|
MemoryRegionSection mrs;
|
|
|
|
MemoryRegion *mr;
|
|
|
|
int ret = 1;
|
|
|
|
|
|
|
|
if (as == &address_space_memory) {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* MSI doorbell address is translated by an IOMMU */
|
|
|
|
|
|
|
|
rcu_read_lock();
|
2018-05-31 16:50:52 +03:00
|
|
|
mr = address_space_translate(as, address, &xlat, &len, true,
|
|
|
|
MEMTXATTRS_UNSPECIFIED);
|
2018-05-04 20:05:52 +03:00
|
|
|
if (!mr) {
|
|
|
|
goto unlock;
|
|
|
|
}
|
|
|
|
mrs = memory_region_find(mr, xlat, 1);
|
|
|
|
if (!mrs.mr) {
|
|
|
|
goto unlock;
|
|
|
|
}
|
|
|
|
|
|
|
|
doorbell_gpa = mrs.offset_within_address_space;
|
|
|
|
memory_region_unref(mrs.mr);
|
|
|
|
|
|
|
|
route->u.msi.address_lo = doorbell_gpa;
|
|
|
|
route->u.msi.address_hi = doorbell_gpa >> 32;
|
|
|
|
|
|
|
|
trace_kvm_arm_fixup_msi_route(address, doorbell_gpa);
|
|
|
|
|
|
|
|
ret = 0;
|
|
|
|
|
|
|
|
unlock:
|
|
|
|
rcu_read_unlock();
|
|
|
|
return ret;
|
2015-01-09 11:04:40 +03:00
|
|
|
}
|
2015-06-02 16:56:23 +03:00
|
|
|
|
2016-07-14 08:56:31 +03:00
|
|
|
int kvm_arch_add_msi_route_post(struct kvm_irq_routing_entry *route,
|
|
|
|
int vector, PCIDevice *dev)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int kvm_arch_release_virq_post(int virq)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-06-02 16:56:23 +03:00
|
|
|
int kvm_arch_msi_data_to_gsi(uint32_t data)
|
|
|
|
{
|
|
|
|
return (data - 32) & 0xffff;
|
|
|
|
}
|