kvm/i386: refactor kvm_arch_init and split it into smaller functions

kvm_arch_init() enables a lot of vm capabilities. Refactor them into separate
smaller functions. Energy MSR related operations also moved to its own
function. There should be no functional impact.

Signed-off-by: Ani Sinha <anisinha@redhat.com>
Link: https://lore.kernel.org/r/20240903124143.39345-2-anisinha@redhat.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This commit is contained in:
Ani Sinha 2024-09-03 18:11:42 +05:30 committed by Paolo Bonzini
parent 804dfbe3ef
commit 0cc42e63bb

View File

@ -3005,10 +3005,185 @@ int kvm_arch_get_default_type(MachineState *ms)
return 0;
}
static int kvm_vm_enable_exception_payload(KVMState *s)
{
int ret = 0;
has_exception_payload = kvm_check_extension(s, KVM_CAP_EXCEPTION_PAYLOAD);
if (has_exception_payload) {
ret = kvm_vm_enable_cap(s, KVM_CAP_EXCEPTION_PAYLOAD, 0, true);
if (ret < 0) {
error_report("kvm: Failed to enable exception payload cap: %s",
strerror(-ret));
}
}
return ret;
}
static int kvm_vm_enable_triple_fault_event(KVMState *s)
{
int ret = 0;
has_triple_fault_event = \
kvm_check_extension(s,
KVM_CAP_X86_TRIPLE_FAULT_EVENT);
if (has_triple_fault_event) {
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT, 0, true);
if (ret < 0) {
error_report("kvm: Failed to enable triple fault event cap: %s",
strerror(-ret));
}
}
return ret;
}
static int kvm_vm_set_identity_map_addr(KVMState *s, uint64_t *identity_base)
{
/*
* On older Intel CPUs, KVM uses vm86 mode to emulate 16-bit code directly.
* In order to use vm86 mode, an EPT identity map and a TSS are needed.
* Since these must be part of guest physical memory, we need to allocate
* them, both by setting their start addresses in the kernel and by
* creating a corresponding e820 entry. We need 4 pages before the BIOS,
* so this value allows up to 16M BIOSes.
*/
*identity_base = 0xfeffc000;
return kvm_vm_ioctl(s, KVM_SET_IDENTITY_MAP_ADDR, identity_base);
}
static int kvm_vm_set_nr_mmu_pages(KVMState *s)
{
uint64_t shadow_mem;
int ret = 0;
shadow_mem = object_property_get_int(OBJECT(s),
"kvm-shadow-mem",
&error_abort);
if (shadow_mem != -1) {
shadow_mem /= 4096;
ret = kvm_vm_ioctl(s, KVM_SET_NR_MMU_PAGES, shadow_mem);
}
return ret;
}
static int kvm_vm_set_tss_addr(KVMState *s, uint64_t identity_base)
{
/* Set TSS base one page after EPT identity map. */
return kvm_vm_ioctl(s, KVM_SET_TSS_ADDR, identity_base);
}
static int kvm_vm_enable_disable_exits(KVMState *s)
{
int disable_exits = kvm_check_extension(s, KVM_CAP_X86_DISABLE_EXITS);
/* Work around for kernel header with a typo. TODO: fix header and drop. */
#if defined(KVM_X86_DISABLE_EXITS_HTL) && !defined(KVM_X86_DISABLE_EXITS_HLT)
#define KVM_X86_DISABLE_EXITS_HLT KVM_X86_DISABLE_EXITS_HTL
#endif
if (disable_exits) {
disable_exits &= (KVM_X86_DISABLE_EXITS_MWAIT |
KVM_X86_DISABLE_EXITS_HLT |
KVM_X86_DISABLE_EXITS_PAUSE |
KVM_X86_DISABLE_EXITS_CSTATE);
}
return kvm_vm_enable_cap(s, KVM_CAP_X86_DISABLE_EXITS, 0,
disable_exits);
}
static int kvm_vm_enable_bus_lock_exit(KVMState *s)
{
int ret = 0;
ret = kvm_check_extension(s, KVM_CAP_X86_BUS_LOCK_EXIT);
if (!(ret & KVM_BUS_LOCK_DETECTION_EXIT)) {
error_report("kvm: bus lock detection unsupported");
return -ENOTSUP;
}
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_BUS_LOCK_EXIT, 0,
KVM_BUS_LOCK_DETECTION_EXIT);
if (ret < 0) {
error_report("kvm: Failed to enable bus lock detection cap: %s",
strerror(-ret));
}
return ret;
}
static int kvm_vm_enable_notify_vmexit(KVMState *s)
{
int ret = 0;
if (s->notify_vmexit != NOTIFY_VMEXIT_OPTION_DISABLE) {
uint64_t notify_window_flags =
((uint64_t)s->notify_window << 32) |
KVM_X86_NOTIFY_VMEXIT_ENABLED |
KVM_X86_NOTIFY_VMEXIT_USER;
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_NOTIFY_VMEXIT, 0,
notify_window_flags);
if (ret < 0) {
error_report("kvm: Failed to enable notify vmexit cap: %s",
strerror(-ret));
}
}
return ret;
}
static int kvm_vm_enable_userspace_msr(KVMState *s)
{
int ret = kvm_vm_enable_cap(s, KVM_CAP_X86_USER_SPACE_MSR, 0,
KVM_MSR_EXIT_REASON_FILTER);
if (ret < 0) {
error_report("Could not enable user space MSRs: %s",
strerror(-ret));
exit(1);
}
if (!kvm_filter_msr(s, MSR_CORE_THREAD_COUNT,
kvm_rdmsr_core_thread_count, NULL)) {
error_report("Could not install MSR_CORE_THREAD_COUNT handler!");
exit(1);
}
return 0;
}
static void kvm_vm_enable_energy_msrs(KVMState *s)
{
bool r;
if (s->msr_energy.enable == true) {
r = kvm_filter_msr(s, MSR_RAPL_POWER_UNIT,
kvm_rdmsr_rapl_power_unit, NULL);
if (!r) {
error_report("Could not install MSR_RAPL_POWER_UNIT \
handler");
exit(1);
}
r = kvm_filter_msr(s, MSR_PKG_POWER_LIMIT,
kvm_rdmsr_pkg_power_limit, NULL);
if (!r) {
error_report("Could not install MSR_PKG_POWER_LIMIT \
handler");
exit(1);
}
r = kvm_filter_msr(s, MSR_PKG_POWER_INFO,
kvm_rdmsr_pkg_power_info, NULL);
if (!r) {
error_report("Could not install MSR_PKG_POWER_INFO \
handler");
exit(1);
}
r = kvm_filter_msr(s, MSR_PKG_ENERGY_STATUS,
kvm_rdmsr_pkg_energy_status, NULL);
if (!r) {
error_report("Could not install MSR_PKG_ENERGY_STATUS \
handler");
exit(1);
}
}
return;
}
int kvm_arch_init(MachineState *ms, KVMState *s)
{
uint64_t identity_base = 0xfffbc000;
uint64_t shadow_mem;
int ret;
struct utsname utsname;
Error *local_err = NULL;
@ -3038,24 +3213,14 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
hv_vpindex_settable = kvm_check_extension(s, KVM_CAP_HYPERV_VP_INDEX);
has_exception_payload = kvm_check_extension(s, KVM_CAP_EXCEPTION_PAYLOAD);
if (has_exception_payload) {
ret = kvm_vm_enable_cap(s, KVM_CAP_EXCEPTION_PAYLOAD, 0, true);
if (ret < 0) {
error_report("kvm: Failed to enable exception payload cap: %s",
strerror(-ret));
return ret;
}
ret = kvm_vm_enable_exception_payload(s);
if (ret < 0) {
return ret;
}
has_triple_fault_event = kvm_check_extension(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT);
if (has_triple_fault_event) {
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_TRIPLE_FAULT_EVENT, 0, true);
if (ret < 0) {
error_report("kvm: Failed to enable triple fault event cap: %s",
strerror(-ret));
return ret;
}
ret = kvm_vm_enable_triple_fault_event(s);
if (ret < 0) {
return ret;
}
if (s->xen_version) {
@ -3086,22 +3251,12 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
uname(&utsname);
lm_capable_kernel = strcmp(utsname.machine, "x86_64") == 0;
/*
* On older Intel CPUs, KVM uses vm86 mode to emulate 16-bit code directly.
* In order to use vm86 mode, an EPT identity map and a TSS are needed.
* Since these must be part of guest physical memory, we need to allocate
* them, both by setting their start addresses in the kernel and by
* creating a corresponding e820 entry. We need 4 pages before the BIOS,
* so this value allows up to 16M BIOSes.
*/
identity_base = 0xfeffc000;
ret = kvm_vm_ioctl(s, KVM_SET_IDENTITY_MAP_ADDR, &identity_base);
ret = kvm_vm_set_identity_map_addr(s, &identity_base);
if (ret < 0) {
return ret;
}
/* Set TSS base one page after EPT identity map. */
ret = kvm_vm_ioctl(s, KVM_SET_TSS_ADDR, identity_base + 0x1000);
ret = kvm_vm_set_tss_addr(s, identity_base + 0x1000);
if (ret < 0) {
return ret;
}
@ -3109,13 +3264,9 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
/* Tell fw_cfg to notify the BIOS to reserve the range. */
e820_add_entry(identity_base, 0x4000, E820_RESERVED);
shadow_mem = object_property_get_int(OBJECT(s), "kvm-shadow-mem", &error_abort);
if (shadow_mem != -1) {
shadow_mem /= 4096;
ret = kvm_vm_ioctl(s, KVM_SET_NR_MMU_PAGES, shadow_mem);
if (ret < 0) {
return ret;
}
ret = kvm_vm_set_nr_mmu_pages(s);
if (ret < 0) {
return ret;
}
if (kvm_check_extension(s, KVM_CAP_X86_SMM) &&
@ -3126,20 +3277,7 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
}
if (enable_cpu_pm) {
int disable_exits = kvm_check_extension(s, KVM_CAP_X86_DISABLE_EXITS);
/* Work around for kernel header with a typo. TODO: fix header and drop. */
#if defined(KVM_X86_DISABLE_EXITS_HTL) && !defined(KVM_X86_DISABLE_EXITS_HLT)
#define KVM_X86_DISABLE_EXITS_HLT KVM_X86_DISABLE_EXITS_HTL
#endif
if (disable_exits) {
disable_exits &= (KVM_X86_DISABLE_EXITS_MWAIT |
KVM_X86_DISABLE_EXITS_HLT |
KVM_X86_DISABLE_EXITS_PAUSE |
KVM_X86_DISABLE_EXITS_CSTATE);
}
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_DISABLE_EXITS, 0,
disable_exits);
ret = kvm_vm_enable_disable_exits(s);
if (ret < 0) {
error_report("kvm: guest stopping CPU not supported: %s",
strerror(-ret));
@ -3150,16 +3288,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
X86MachineState *x86ms = X86_MACHINE(ms);
if (x86ms->bus_lock_ratelimit > 0) {
ret = kvm_check_extension(s, KVM_CAP_X86_BUS_LOCK_EXIT);
if (!(ret & KVM_BUS_LOCK_DETECTION_EXIT)) {
error_report("kvm: bus lock detection unsupported");
return -ENOTSUP;
}
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_BUS_LOCK_EXIT, 0,
KVM_BUS_LOCK_DETECTION_EXIT);
ret = kvm_vm_enable_bus_lock_exit(s);
if (ret < 0) {
error_report("kvm: Failed to enable bus lock detection cap: %s",
strerror(-ret));
return ret;
}
ratelimit_init(&bus_lock_ratelimit_ctrl);
@ -3168,80 +3298,25 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
}
}
if (s->notify_vmexit != NOTIFY_VMEXIT_OPTION_DISABLE &&
kvm_check_extension(s, KVM_CAP_X86_NOTIFY_VMEXIT)) {
uint64_t notify_window_flags =
((uint64_t)s->notify_window << 32) |
KVM_X86_NOTIFY_VMEXIT_ENABLED |
KVM_X86_NOTIFY_VMEXIT_USER;
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_NOTIFY_VMEXIT, 0,
notify_window_flags);
if (ret < 0) {
error_report("kvm: Failed to enable notify vmexit cap: %s",
strerror(-ret));
return ret;
}
}
if (kvm_vm_check_extension(s, KVM_CAP_X86_USER_SPACE_MSR)) {
bool r;
ret = kvm_vm_enable_cap(s, KVM_CAP_X86_USER_SPACE_MSR, 0,
KVM_MSR_EXIT_REASON_FILTER);
if (ret) {
error_report("Could not enable user space MSRs: %s",
strerror(-ret));
exit(1);
if (kvm_check_extension(s, KVM_CAP_X86_NOTIFY_VMEXIT)) {
ret = kvm_vm_enable_notify_vmexit(s);
if (ret < 0) {
return ret;
}
}
r = kvm_filter_msr(s, MSR_CORE_THREAD_COUNT,
kvm_rdmsr_core_thread_count, NULL);
if (!r) {
error_report("Could not install MSR_CORE_THREAD_COUNT handler: %s",
strerror(-ret));
exit(1);
if (kvm_vm_check_extension(s, KVM_CAP_X86_USER_SPACE_MSR)) {
ret = kvm_vm_enable_userspace_msr(s);
if (ret < 0) {
return ret;
}
if (s->msr_energy.enable == true) {
r = kvm_filter_msr(s, MSR_RAPL_POWER_UNIT,
kvm_rdmsr_rapl_power_unit, NULL);
if (!r) {
error_report("Could not install MSR_RAPL_POWER_UNIT \
handler: %s",
strerror(-ret));
kvm_vm_enable_energy_msrs(s);
if (kvm_msr_energy_thread_init(s, ms)) {
error_report("kvm : error RAPL feature requirement not met");
exit(1);
}
r = kvm_filter_msr(s, MSR_PKG_POWER_LIMIT,
kvm_rdmsr_pkg_power_limit, NULL);
if (!r) {
error_report("Could not install MSR_PKG_POWER_LIMIT \
handler: %s",
strerror(-ret));
exit(1);
}
r = kvm_filter_msr(s, MSR_PKG_POWER_INFO,
kvm_rdmsr_pkg_power_info, NULL);
if (!r) {
error_report("Could not install MSR_PKG_POWER_INFO \
handler: %s",
strerror(-ret));
exit(1);
}
r = kvm_filter_msr(s, MSR_PKG_ENERGY_STATUS,
kvm_rdmsr_pkg_energy_status, NULL);
if (!r) {
error_report("Could not install MSR_PKG_ENERGY_STATUS \
handler: %s",
strerror(-ret));
exit(1);
}
r = kvm_msr_energy_thread_init(s, ms);
if (r) {
error_report("kvm : error RAPL feature requirement not meet");
exit(1);
}
}
}