linux/drivers/clocksource/hyperv_timer.c
Dexuan Cui 7f828d5fff clocksource: hyper-v: Use lapic timer in a TDX VM without paravisor
In a TDX VM without paravisor, currently the default timer is the Hyper-V
timer, which depends on the slow VM Reference Counter MSR: the Hyper-V TSC
page is not enabled in such a VM because the VM uses Invariant TSC as a
better clocksource and it's challenging to mark the Hyper-V TSC page shared
in very early boot.

Lower the rating of the Hyper-V timer so the local APIC timer becomes the
the default timer in such a VM, and print a warning in case Invariant TSC
is unavailable in such a VM. This change should cause no perceivable
performance difference.

Cc: stable@vger.kernel.org # 6.6+
Reviewed-by: Roman Kisel <romank@linux.microsoft.com>
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Michael Kelley <mhklinux@outlook.com>
Link: https://lore.kernel.org/r/20240621061614.8339-1-decui@microsoft.com
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Message-ID: <20240621061614.8339-1-decui@microsoft.com>
2024-06-24 07:06:29 +00:00

629 lines
17 KiB
C

// SPDX-License-Identifier: GPL-2.0
/*
* Clocksource driver for the synthetic counter and timers
* provided by the Hyper-V hypervisor to guest VMs, as described
* in the Hyper-V Top Level Functional Spec (TLFS). This driver
* is instruction set architecture independent.
*
* Copyright (C) 2019, Microsoft, Inc.
*
* Author: Michael Kelley <mikelley@microsoft.com>
*/
#include <linux/percpu.h>
#include <linux/cpumask.h>
#include <linux/clockchips.h>
#include <linux/clocksource.h>
#include <linux/sched_clock.h>
#include <linux/mm.h>
#include <linux/cpuhotplug.h>
#include <linux/interrupt.h>
#include <linux/irq.h>
#include <linux/acpi.h>
#include <linux/hyperv.h>
#include <clocksource/hyperv_timer.h>
#include <asm/hyperv-tlfs.h>
#include <asm/mshyperv.h>
static struct clock_event_device __percpu *hv_clock_event;
static u64 hv_sched_clock_offset __ro_after_init;
/*
* If false, we're using the old mechanism for stimer0 interrupts
* where it sends a VMbus message when it expires. The old
* mechanism is used when running on older versions of Hyper-V
* that don't support Direct Mode. While Hyper-V provides
* four stimer's per CPU, Linux uses only stimer0.
*
* Because Direct Mode does not require processing a VMbus
* message, stimer interrupts can be enabled earlier in the
* process of booting a CPU, and consistent with when timer
* interrupts are enabled for other clocksource drivers.
* However, for legacy versions of Hyper-V when Direct Mode
* is not enabled, setting up stimer interrupts must be
* delayed until VMbus is initialized and can process the
* interrupt message.
*/
static bool direct_mode_enabled;
static int stimer0_irq = -1;
static int stimer0_message_sint;
static __maybe_unused DEFINE_PER_CPU(long, stimer0_evt);
/*
* Common code for stimer0 interrupts coming via Direct Mode or
* as a VMbus message.
*/
void hv_stimer0_isr(void)
{
struct clock_event_device *ce;
ce = this_cpu_ptr(hv_clock_event);
ce->event_handler(ce);
}
EXPORT_SYMBOL_GPL(hv_stimer0_isr);
/*
* stimer0 interrupt handler for architectures that support
* per-cpu interrupts, which also implies Direct Mode.
*/
static irqreturn_t __maybe_unused hv_stimer0_percpu_isr(int irq, void *dev_id)
{
hv_stimer0_isr();
return IRQ_HANDLED;
}
static int hv_ce_set_next_event(unsigned long delta,
struct clock_event_device *evt)
{
u64 current_tick;
current_tick = hv_read_reference_counter();
current_tick += delta;
hv_set_msr(HV_MSR_STIMER0_COUNT, current_tick);
return 0;
}
static int hv_ce_shutdown(struct clock_event_device *evt)
{
hv_set_msr(HV_MSR_STIMER0_COUNT, 0);
hv_set_msr(HV_MSR_STIMER0_CONFIG, 0);
if (direct_mode_enabled && stimer0_irq >= 0)
disable_percpu_irq(stimer0_irq);
return 0;
}
static int hv_ce_set_oneshot(struct clock_event_device *evt)
{
union hv_stimer_config timer_cfg;
timer_cfg.as_uint64 = 0;
timer_cfg.enable = 1;
timer_cfg.auto_enable = 1;
if (direct_mode_enabled) {
/*
* When it expires, the timer will directly interrupt
* on the specified hardware vector/IRQ.
*/
timer_cfg.direct_mode = 1;
timer_cfg.apic_vector = HYPERV_STIMER0_VECTOR;
if (stimer0_irq >= 0)
enable_percpu_irq(stimer0_irq, IRQ_TYPE_NONE);
} else {
/*
* When it expires, the timer will generate a VMbus message,
* to be handled by the normal VMbus interrupt handler.
*/
timer_cfg.direct_mode = 0;
timer_cfg.sintx = stimer0_message_sint;
}
hv_set_msr(HV_MSR_STIMER0_CONFIG, timer_cfg.as_uint64);
return 0;
}
/*
* hv_stimer_init - Per-cpu initialization of the clockevent
*/
static int hv_stimer_init(unsigned int cpu)
{
struct clock_event_device *ce;
if (!hv_clock_event)
return 0;
ce = per_cpu_ptr(hv_clock_event, cpu);
ce->name = "Hyper-V clockevent";
ce->features = CLOCK_EVT_FEAT_ONESHOT;
ce->cpumask = cpumask_of(cpu);
/*
* Lower the rating of the Hyper-V timer in a TDX VM without paravisor,
* so the local APIC timer (lapic_clockevent) is the default timer in
* such a VM. The Hyper-V timer is not preferred in such a VM because
* it depends on the slow VM Reference Counter MSR (the Hyper-V TSC
* page is not enbled in such a VM because the VM uses Invariant TSC
* as a better clocksource and it's challenging to mark the Hyper-V
* TSC page shared in very early boot).
*/
if (!ms_hyperv.paravisor_present && hv_isolation_type_tdx())
ce->rating = 90;
else
ce->rating = 1000;
ce->set_state_shutdown = hv_ce_shutdown;
ce->set_state_oneshot = hv_ce_set_oneshot;
ce->set_next_event = hv_ce_set_next_event;
clockevents_config_and_register(ce,
HV_CLOCK_HZ,
HV_MIN_DELTA_TICKS,
HV_MAX_MAX_DELTA_TICKS);
return 0;
}
/*
* hv_stimer_cleanup - Per-cpu cleanup of the clockevent
*/
int hv_stimer_cleanup(unsigned int cpu)
{
struct clock_event_device *ce;
if (!hv_clock_event)
return 0;
/*
* In the legacy case where Direct Mode is not enabled
* (which can only be on x86/64), stimer cleanup happens
* relatively early in the CPU offlining process. We
* must unbind the stimer-based clockevent device so
* that the LAPIC timer can take over until clockevents
* are no longer needed in the offlining process. Note
* that clockevents_unbind_device() eventually calls
* hv_ce_shutdown().
*
* The unbind should not be done when Direct Mode is
* enabled because we may be on an architecture where
* there are no other clockevent devices to fallback to.
*/
ce = per_cpu_ptr(hv_clock_event, cpu);
if (direct_mode_enabled)
hv_ce_shutdown(ce);
else
clockevents_unbind_device(ce, cpu);
return 0;
}
EXPORT_SYMBOL_GPL(hv_stimer_cleanup);
/*
* These placeholders are overridden by arch specific code on
* architectures that need special setup of the stimer0 IRQ because
* they don't support per-cpu IRQs (such as x86/x64).
*/
void __weak hv_setup_stimer0_handler(void (*handler)(void))
{
};
void __weak hv_remove_stimer0_handler(void)
{
};
#ifdef CONFIG_ACPI
/* Called only on architectures with per-cpu IRQs (i.e., not x86/x64) */
static int hv_setup_stimer0_irq(void)
{
int ret;
ret = acpi_register_gsi(NULL, HYPERV_STIMER0_VECTOR,
ACPI_EDGE_SENSITIVE, ACPI_ACTIVE_HIGH);
if (ret < 0) {
pr_err("Can't register Hyper-V stimer0 GSI. Error %d", ret);
return ret;
}
stimer0_irq = ret;
ret = request_percpu_irq(stimer0_irq, hv_stimer0_percpu_isr,
"Hyper-V stimer0", &stimer0_evt);
if (ret) {
pr_err("Can't request Hyper-V stimer0 IRQ %d. Error %d",
stimer0_irq, ret);
acpi_unregister_gsi(stimer0_irq);
stimer0_irq = -1;
}
return ret;
}
static void hv_remove_stimer0_irq(void)
{
if (stimer0_irq == -1) {
hv_remove_stimer0_handler();
} else {
free_percpu_irq(stimer0_irq, &stimer0_evt);
acpi_unregister_gsi(stimer0_irq);
stimer0_irq = -1;
}
}
#else
static int hv_setup_stimer0_irq(void)
{
return 0;
}
static void hv_remove_stimer0_irq(void)
{
}
#endif
/* hv_stimer_alloc - Global initialization of the clockevent and stimer0 */
int hv_stimer_alloc(bool have_percpu_irqs)
{
int ret;
/*
* Synthetic timers are always available except on old versions of
* Hyper-V on x86. In that case, return as error as Linux will use a
* clockevent based on emulated LAPIC timer hardware.
*/
if (!(ms_hyperv.features & HV_MSR_SYNTIMER_AVAILABLE))
return -EINVAL;
hv_clock_event = alloc_percpu(struct clock_event_device);
if (!hv_clock_event)
return -ENOMEM;
direct_mode_enabled = ms_hyperv.misc_features &
HV_STIMER_DIRECT_MODE_AVAILABLE;
/*
* If Direct Mode isn't enabled, the remainder of the initialization
* is done later by hv_stimer_legacy_init()
*/
if (!direct_mode_enabled)
return 0;
if (have_percpu_irqs) {
ret = hv_setup_stimer0_irq();
if (ret)
goto free_clock_event;
} else {
hv_setup_stimer0_handler(hv_stimer0_isr);
}
/*
* Since we are in Direct Mode, stimer initialization
* can be done now with a CPUHP value in the same range
* as other clockevent devices.
*/
ret = cpuhp_setup_state(CPUHP_AP_HYPERV_TIMER_STARTING,
"clockevents/hyperv/stimer:starting",
hv_stimer_init, hv_stimer_cleanup);
if (ret < 0) {
hv_remove_stimer0_irq();
goto free_clock_event;
}
return ret;
free_clock_event:
free_percpu(hv_clock_event);
hv_clock_event = NULL;
return ret;
}
EXPORT_SYMBOL_GPL(hv_stimer_alloc);
/*
* hv_stimer_legacy_init -- Called from the VMbus driver to handle
* the case when Direct Mode is not enabled, and the stimer
* must be initialized late in the CPU onlining process.
*
*/
void hv_stimer_legacy_init(unsigned int cpu, int sint)
{
if (direct_mode_enabled)
return;
/*
* This function gets called by each vCPU, so setting the
* global stimer_message_sint value each time is conceptually
* not ideal, but the value passed in is always the same and
* it avoids introducing yet another interface into this
* clocksource driver just to set the sint in the legacy case.
*/
stimer0_message_sint = sint;
(void)hv_stimer_init(cpu);
}
EXPORT_SYMBOL_GPL(hv_stimer_legacy_init);
/*
* hv_stimer_legacy_cleanup -- Called from the VMbus driver to
* handle the case when Direct Mode is not enabled, and the
* stimer must be cleaned up early in the CPU offlining
* process.
*/
void hv_stimer_legacy_cleanup(unsigned int cpu)
{
if (direct_mode_enabled)
return;
(void)hv_stimer_cleanup(cpu);
}
EXPORT_SYMBOL_GPL(hv_stimer_legacy_cleanup);
/*
* Do a global cleanup of clockevents for the cases of kexec and
* vmbus exit
*/
void hv_stimer_global_cleanup(void)
{
int cpu;
/*
* hv_stime_legacy_cleanup() will stop the stimer if Direct
* Mode is not enabled, and fallback to the LAPIC timer.
*/
for_each_present_cpu(cpu) {
hv_stimer_legacy_cleanup(cpu);
}
if (!hv_clock_event)
return;
if (direct_mode_enabled) {
cpuhp_remove_state(CPUHP_AP_HYPERV_TIMER_STARTING);
hv_remove_stimer0_irq();
stimer0_irq = -1;
}
free_percpu(hv_clock_event);
hv_clock_event = NULL;
}
EXPORT_SYMBOL_GPL(hv_stimer_global_cleanup);
static __always_inline u64 read_hv_clock_msr(void)
{
/*
* Read the partition counter to get the current tick count. This count
* is set to 0 when the partition is created and is incremented in 100
* nanosecond units.
*
* Use hv_raw_get_msr() because this function is used from
* noinstr. Notable; while HV_MSR_TIME_REF_COUNT is a synthetic
* register it doesn't need the GHCB path.
*/
return hv_raw_get_msr(HV_MSR_TIME_REF_COUNT);
}
/*
* Code and definitions for the Hyper-V clocksources. Two
* clocksources are defined: one that reads the Hyper-V defined MSR, and
* the other that uses the TSC reference page feature as defined in the
* TLFS. The MSR version is for compatibility with old versions of
* Hyper-V and 32-bit x86. The TSC reference page version is preferred.
*/
static union {
struct ms_hyperv_tsc_page page;
u8 reserved[PAGE_SIZE];
} tsc_pg __bss_decrypted __aligned(PAGE_SIZE);
static struct ms_hyperv_tsc_page *tsc_page = &tsc_pg.page;
static unsigned long tsc_pfn;
unsigned long hv_get_tsc_pfn(void)
{
return tsc_pfn;
}
EXPORT_SYMBOL_GPL(hv_get_tsc_pfn);
struct ms_hyperv_tsc_page *hv_get_tsc_page(void)
{
return tsc_page;
}
EXPORT_SYMBOL_GPL(hv_get_tsc_page);
static __always_inline u64 read_hv_clock_tsc(void)
{
u64 cur_tsc, time;
/*
* The Hyper-V Top-Level Function Spec (TLFS), section Timers,
* subsection Refererence Counter, guarantees that the TSC and MSR
* times are in sync and monotonic. Therefore we can fall back
* to the MSR in case the TSC page indicates unavailability.
*/
if (!hv_read_tsc_page_tsc(tsc_page, &cur_tsc, &time))
time = read_hv_clock_msr();
return time;
}
static u64 notrace read_hv_clock_tsc_cs(struct clocksource *arg)
{
return read_hv_clock_tsc();
}
static u64 noinstr read_hv_sched_clock_tsc(void)
{
return (read_hv_clock_tsc() - hv_sched_clock_offset) *
(NSEC_PER_SEC / HV_CLOCK_HZ);
}
static void suspend_hv_clock_tsc(struct clocksource *arg)
{
union hv_reference_tsc_msr tsc_msr;
/* Disable the TSC page */
tsc_msr.as_uint64 = hv_get_msr(HV_MSR_REFERENCE_TSC);
tsc_msr.enable = 0;
hv_set_msr(HV_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
}
static void resume_hv_clock_tsc(struct clocksource *arg)
{
union hv_reference_tsc_msr tsc_msr;
/* Re-enable the TSC page */
tsc_msr.as_uint64 = hv_get_msr(HV_MSR_REFERENCE_TSC);
tsc_msr.enable = 1;
tsc_msr.pfn = tsc_pfn;
hv_set_msr(HV_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
}
#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
static int hv_cs_enable(struct clocksource *cs)
{
vclocks_set_used(VDSO_CLOCKMODE_HVCLOCK);
return 0;
}
#endif
static struct clocksource hyperv_cs_tsc = {
.name = "hyperv_clocksource_tsc_page",
.rating = 500,
.read = read_hv_clock_tsc_cs,
.mask = CLOCKSOURCE_MASK(64),
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
.suspend= suspend_hv_clock_tsc,
.resume = resume_hv_clock_tsc,
#ifdef HAVE_VDSO_CLOCKMODE_HVCLOCK
.enable = hv_cs_enable,
.vdso_clock_mode = VDSO_CLOCKMODE_HVCLOCK,
#else
.vdso_clock_mode = VDSO_CLOCKMODE_NONE,
#endif
};
static u64 notrace read_hv_clock_msr_cs(struct clocksource *arg)
{
return read_hv_clock_msr();
}
static struct clocksource hyperv_cs_msr = {
.name = "hyperv_clocksource_msr",
.rating = 495,
.read = read_hv_clock_msr_cs,
.mask = CLOCKSOURCE_MASK(64),
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
};
/*
* Reference to pv_ops must be inline so objtool
* detection of noinstr violations can work correctly.
*/
#ifdef CONFIG_GENERIC_SCHED_CLOCK
static __always_inline void hv_setup_sched_clock(void *sched_clock)
{
/*
* We're on an architecture with generic sched clock (not x86/x64).
* The Hyper-V sched clock read function returns nanoseconds, not
* the normal 100ns units of the Hyper-V synthetic clock.
*/
sched_clock_register(sched_clock, 64, NSEC_PER_SEC);
}
#elif defined CONFIG_PARAVIRT
static __always_inline void hv_setup_sched_clock(void *sched_clock)
{
/* We're on x86/x64 *and* using PV ops */
paravirt_set_sched_clock(sched_clock);
}
#else /* !CONFIG_GENERIC_SCHED_CLOCK && !CONFIG_PARAVIRT */
static __always_inline void hv_setup_sched_clock(void *sched_clock) {}
#endif /* CONFIG_GENERIC_SCHED_CLOCK */
static void __init hv_init_tsc_clocksource(void)
{
union hv_reference_tsc_msr tsc_msr;
/*
* If Hyper-V offers TSC_INVARIANT, then the virtualized TSC correctly
* handles frequency and offset changes due to live migration,
* pause/resume, and other VM management operations. So lower the
* Hyper-V Reference TSC rating, causing the generic TSC to be used.
* TSC_INVARIANT is not offered on ARM64, so the Hyper-V Reference
* TSC will be preferred over the virtualized ARM64 arch counter.
*/
if (ms_hyperv.features & HV_ACCESS_TSC_INVARIANT) {
hyperv_cs_tsc.rating = 250;
hyperv_cs_msr.rating = 245;
}
if (!(ms_hyperv.features & HV_MSR_REFERENCE_TSC_AVAILABLE))
return;
hv_read_reference_counter = read_hv_clock_tsc;
/*
* TSC page mapping works differently in root compared to guest.
* - In guest partition the guest PFN has to be passed to the
* hypervisor.
* - In root partition it's other way around: it has to map the PFN
* provided by the hypervisor.
* But it can't be mapped right here as it's too early and MMU isn't
* ready yet. So, we only set the enable bit here and will remap the
* page later in hv_remap_tsc_clocksource().
*
* It worth mentioning, that TSC clocksource read function
* (read_hv_clock_tsc) has a MSR-based fallback mechanism, used when
* TSC page is zeroed (which is the case until the PFN is remapped) and
* thus TSC clocksource will work even without the real TSC page
* mapped.
*/
tsc_msr.as_uint64 = hv_get_msr(HV_MSR_REFERENCE_TSC);
if (hv_root_partition)
tsc_pfn = tsc_msr.pfn;
else
tsc_pfn = HVPFN_DOWN(virt_to_phys(tsc_page));
tsc_msr.enable = 1;
tsc_msr.pfn = tsc_pfn;
hv_set_msr(HV_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
clocksource_register_hz(&hyperv_cs_tsc, NSEC_PER_SEC/100);
/*
* If TSC is invariant, then let it stay as the sched clock since it
* will be faster than reading the TSC page. But if not invariant, use
* the TSC page so that live migrations across hosts with different
* frequencies is handled correctly.
*/
if (!(ms_hyperv.features & HV_ACCESS_TSC_INVARIANT)) {
hv_sched_clock_offset = hv_read_reference_counter();
hv_setup_sched_clock(read_hv_sched_clock_tsc);
}
}
void __init hv_init_clocksource(void)
{
/*
* Try to set up the TSC page clocksource, then the MSR clocksource.
* At least one of these will always be available except on very old
* versions of Hyper-V on x86. In that case we won't have a Hyper-V
* clocksource, but Linux will still run with a clocksource based
* on the emulated PIT or LAPIC timer.
*
* Never use the MSR clocksource as sched clock. It's too slow.
* Better to use the native sched clock as the fallback.
*/
hv_init_tsc_clocksource();
if (ms_hyperv.features & HV_MSR_TIME_REF_COUNT_AVAILABLE)
clocksource_register_hz(&hyperv_cs_msr, NSEC_PER_SEC/100);
}
void __init hv_remap_tsc_clocksource(void)
{
if (!(ms_hyperv.features & HV_MSR_REFERENCE_TSC_AVAILABLE))
return;
if (!hv_root_partition) {
WARN(1, "%s: attempt to remap TSC page in guest partition\n",
__func__);
return;
}
tsc_page = memremap(tsc_pfn << HV_HYP_PAGE_SHIFT, sizeof(tsc_pg),
MEMREMAP_WB);
if (!tsc_page)
pr_err("Failed to remap Hyper-V TSC page.\n");
}