^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * Xen time implementation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * This is implemented in terms of a clocksource driver which uses
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * the hypervisor clock as a nanosecond timebase, and a clockevent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * driver which uses the hypervisor's timer mechanism.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/interrupt.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/clocksource.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/clockchips.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/gfp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/pvclock_gtod.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/timekeeper_internal.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <asm/pvclock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <asm/xen/hypervisor.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <asm/xen/hypercall.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <xen/events.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <xen/features.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <xen/interface/xen.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <xen/interface/vcpu.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include "xen-ops.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) /* Minimum amount of time until next clock event fires */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #define TIMER_SLOP 100000
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) static u64 xen_sched_clock_offset __read_mostly;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) /* Get the TSC speed from Xen */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) static unsigned long xen_tsc_khz(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) struct pvclock_vcpu_time_info *info =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) &HYPERVISOR_shared_info->vcpu_info[0].time;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) setup_force_cpu_cap(X86_FEATURE_TSC_KNOWN_FREQ);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) return pvclock_tsc_khz(info);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) static u64 xen_clocksource_read(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) struct pvclock_vcpu_time_info *src;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) u64 ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) preempt_disable_notrace();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) src = &__this_cpu_read(xen_vcpu)->time;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) ret = pvclock_clocksource_read(src);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) preempt_enable_notrace();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) static u64 xen_clocksource_get_cycles(struct clocksource *cs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) return xen_clocksource_read();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) static u64 xen_sched_clock(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) return xen_clocksource_read() - xen_sched_clock_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) static void xen_read_wallclock(struct timespec64 *ts)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) struct shared_info *s = HYPERVISOR_shared_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) struct pvclock_wall_clock *wall_clock = &(s->wc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) struct pvclock_vcpu_time_info *vcpu_time;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) vcpu_time = &get_cpu_var(xen_vcpu)->time;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) pvclock_read_wallclock(wall_clock, vcpu_time, ts);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) put_cpu_var(xen_vcpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) static void xen_get_wallclock(struct timespec64 *now)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) xen_read_wallclock(now);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) static int xen_set_wallclock(const struct timespec64 *now)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) return -ENODEV;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) static int xen_pvclock_gtod_notify(struct notifier_block *nb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) unsigned long was_set, void *priv)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) /* Protected by the calling core code serialization */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) static struct timespec64 next_sync;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) struct xen_platform_op op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) struct timespec64 now;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) struct timekeeper *tk = priv;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) static bool settime64_supported = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) now.tv_sec = tk->xtime_sec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) now.tv_nsec = (long)(tk->tkr_mono.xtime_nsec >> tk->tkr_mono.shift);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) * We only take the expensive HV call when the clock was set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) * or when the 11 minutes RTC synchronization time elapsed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) if (!was_set && timespec64_compare(&now, &next_sync) < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) return NOTIFY_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) again:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) if (settime64_supported) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) op.cmd = XENPF_settime64;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) op.u.settime64.mbz = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) op.u.settime64.secs = now.tv_sec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) op.u.settime64.nsecs = now.tv_nsec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) op.u.settime64.system_time = xen_clocksource_read();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) op.cmd = XENPF_settime32;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) op.u.settime32.secs = now.tv_sec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) op.u.settime32.nsecs = now.tv_nsec;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) op.u.settime32.system_time = xen_clocksource_read();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) ret = HYPERVISOR_platform_op(&op);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) if (ret == -ENOSYS && settime64_supported) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) settime64_supported = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) goto again;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) if (ret < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) return NOTIFY_BAD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) * Move the next drift compensation time 11 minutes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) * ahead. That's emulating the sync_cmos_clock() update for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) * the hardware RTC.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) next_sync = now;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) next_sync.tv_sec += 11 * 60;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) return NOTIFY_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) static struct notifier_block xen_pvclock_gtod_notifier = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) .notifier_call = xen_pvclock_gtod_notify,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) static int xen_cs_enable(struct clocksource *cs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) vclocks_set_used(VDSO_CLOCKMODE_PVCLOCK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) static struct clocksource xen_clocksource __read_mostly = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) .name = "xen",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) .rating = 400,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) .read = xen_clocksource_get_cycles,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) .mask = CLOCKSOURCE_MASK(64),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) .flags = CLOCK_SOURCE_IS_CONTINUOUS,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) .enable = xen_cs_enable,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) Xen clockevent implementation
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) Xen has two clockevent implementations:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) The old timer_op one works with all released versions of Xen prior
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) to version 3.0.4. This version of the hypervisor provides a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) single-shot timer with nanosecond resolution. However, sharing the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) same event channel is a 100Hz tick which is delivered while the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) vcpu is running. We don't care about or use this tick, but it will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) cause the core time code to think the timer fired too soon, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) will end up resetting it each time. It could be filtered, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) doing so has complications when the ktime clocksource is not yet
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) the xen clocksource (ie, at boot time).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) The new vcpu_op-based timer interface allows the tick timer period
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) to be changed or turned off. The tick timer is not useful as a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) periodic timer because events are only delivered to running vcpus.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) The one-shot timer can report when a timeout is in the past, so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) set_next_event is capable of returning -ETIME when appropriate.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) This interface is used when available.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) Get a hypervisor absolute time. In theory we could maintain an
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) offset between the kernel's time and the hypervisor's time, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) apply that to a kernel's absolute timeout. Unfortunately the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) hypervisor and kernel times can drift even if the kernel is using
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) the Xen clocksource, because ntp can warp the kernel's clocksource.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) static s64 get_abs_timeout(unsigned long delta)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) return xen_clocksource_read() + delta;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) static int xen_timerop_shutdown(struct clock_event_device *evt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) /* cancel timeout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) HYPERVISOR_set_timer_op(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) static int xen_timerop_set_next_event(unsigned long delta,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) struct clock_event_device *evt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) WARN_ON(!clockevent_state_oneshot(evt));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) if (HYPERVISOR_set_timer_op(get_abs_timeout(delta)) < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) /* We may have missed the deadline, but there's no real way of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) knowing for sure. If the event was in the past, then we'll
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) get an immediate interrupt. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) static struct clock_event_device xen_timerop_clockevent __ro_after_init = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) .name = "xen",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) .features = CLOCK_EVT_FEAT_ONESHOT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) .max_delta_ns = 0xffffffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) .max_delta_ticks = 0xffffffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) .min_delta_ns = TIMER_SLOP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) .min_delta_ticks = TIMER_SLOP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) .mult = 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) .shift = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) .rating = 500,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) .set_state_shutdown = xen_timerop_shutdown,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) .set_next_event = xen_timerop_set_next_event,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) static int xen_vcpuop_shutdown(struct clock_event_device *evt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) if (HYPERVISOR_vcpu_op(VCPUOP_stop_singleshot_timer, xen_vcpu_nr(cpu),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) NULL) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, xen_vcpu_nr(cpu),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) NULL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) static int xen_vcpuop_set_oneshot(struct clock_event_device *evt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, xen_vcpu_nr(cpu),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) NULL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) static int xen_vcpuop_set_next_event(unsigned long delta,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) struct clock_event_device *evt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) struct vcpu_set_singleshot_timer single;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) WARN_ON(!clockevent_state_oneshot(evt));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) single.timeout_abs_ns = get_abs_timeout(delta);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) /* Get an event anyway, even if the timeout is already expired */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) single.flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) ret = HYPERVISOR_vcpu_op(VCPUOP_set_singleshot_timer, xen_vcpu_nr(cpu),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) &single);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) BUG_ON(ret != 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) static struct clock_event_device xen_vcpuop_clockevent __ro_after_init = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) .name = "xen",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) .features = CLOCK_EVT_FEAT_ONESHOT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) .max_delta_ns = 0xffffffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) .max_delta_ticks = 0xffffffff,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) .min_delta_ns = TIMER_SLOP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) .min_delta_ticks = TIMER_SLOP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) .mult = 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) .shift = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) .rating = 500,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) .set_state_shutdown = xen_vcpuop_shutdown,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) .set_state_oneshot = xen_vcpuop_set_oneshot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) .set_next_event = xen_vcpuop_set_next_event,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) static const struct clock_event_device *xen_clockevent =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) &xen_timerop_clockevent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) struct xen_clock_event_device {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) struct clock_event_device evt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) char name[16];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) static DEFINE_PER_CPU(struct xen_clock_event_device, xen_clock_events) = { .evt.irq = -1 };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) static irqreturn_t xen_timer_interrupt(int irq, void *dev_id)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) struct clock_event_device *evt = this_cpu_ptr(&xen_clock_events.evt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) irqreturn_t ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) ret = IRQ_NONE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) if (evt->event_handler) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) evt->event_handler(evt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) ret = IRQ_HANDLED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) void xen_teardown_timer(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) struct clock_event_device *evt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) evt = &per_cpu(xen_clock_events, cpu).evt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) if (evt->irq >= 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) unbind_from_irqhandler(evt->irq, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) evt->irq = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) void xen_setup_timer(int cpu)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) struct xen_clock_event_device *xevt = &per_cpu(xen_clock_events, cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) struct clock_event_device *evt = &xevt->evt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) int irq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) WARN(evt->irq >= 0, "IRQ%d for CPU%d is already allocated\n", evt->irq, cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) if (evt->irq >= 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) xen_teardown_timer(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) printk(KERN_INFO "installing Xen timer for CPU %d\n", cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) snprintf(xevt->name, sizeof(xevt->name), "timer%d", cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) irq = bind_virq_to_irqhandler(VIRQ_TIMER, cpu, xen_timer_interrupt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) IRQF_PERCPU|IRQF_NOBALANCING|IRQF_TIMER|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) IRQF_FORCE_RESUME|IRQF_EARLY_RESUME,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) xevt->name, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) (void)xen_set_irq_priority(irq, XEN_IRQ_PRIORITY_MAX);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) memcpy(evt, xen_clockevent, sizeof(*evt));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) evt->cpumask = cpumask_of(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) evt->irq = irq;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) void xen_setup_cpu_clockevents(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) clockevents_register_device(this_cpu_ptr(&xen_clock_events.evt));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) void xen_timer_resume(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) int cpu;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) if (xen_clockevent != &xen_vcpuop_clockevent)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) for_each_online_cpu(cpu) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) xen_vcpu_nr(cpu), NULL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) static const struct pv_time_ops xen_time_ops __initconst = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) .sched_clock = xen_sched_clock,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) .steal_clock = xen_steal_clock,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) static struct pvclock_vsyscall_time_info *xen_clock __read_mostly;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) static u64 xen_clock_value_saved;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) void xen_save_time_memory_area(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) struct vcpu_register_time_memory_area t;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) xen_clock_value_saved = xen_clocksource_read() - xen_sched_clock_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) if (!xen_clock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) t.addr.v = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) ret = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_time_memory_area, 0, &t);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) if (ret != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) pr_notice("Cannot save secondary vcpu_time_info (err %d)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) clear_page(xen_clock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) void xen_restore_time_memory_area(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) struct vcpu_register_time_memory_area t;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) if (!xen_clock)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) t.addr.v = &xen_clock->pvti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) ret = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_time_memory_area, 0, &t);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) * We don't disable VDSO_CLOCKMODE_PVCLOCK entirely if it fails to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) * register the secondary time info with Xen or if we migrated to a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) * host without the necessary flags. On both of these cases what
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) * happens is either process seeing a zeroed out pvti or seeing no
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) * PVCLOCK_TSC_STABLE_BIT bit set. Userspace checks the latter and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) * if 0, it discards the data in pvti and fallbacks to a system
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) * call for a reliable timestamp.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) if (ret != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) pr_notice("Cannot restore secondary vcpu_time_info (err %d)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) /* Need pvclock_resume() before using xen_clocksource_read(). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) pvclock_resume();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) xen_sched_clock_offset = xen_clocksource_read() - xen_clock_value_saved;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) static void xen_setup_vsyscall_time_info(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) struct vcpu_register_time_memory_area t;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) struct pvclock_vsyscall_time_info *ti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) ti = (struct pvclock_vsyscall_time_info *)get_zeroed_page(GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) if (!ti)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) t.addr.v = &ti->pvti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) ret = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_time_memory_area, 0, &t);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) if (ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) pr_notice("xen: VDSO_CLOCKMODE_PVCLOCK not supported (err %d)\n", ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) free_page((unsigned long)ti);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) * If primary time info had this bit set, secondary should too since
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) * it's the same data on both just different memory regions. But we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) * still check it in case hypervisor is buggy.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) if (!(ti->pvti.flags & PVCLOCK_TSC_STABLE_BIT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) t.addr.v = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) ret = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_time_memory_area,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) 0, &t);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) if (!ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) free_page((unsigned long)ti);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) pr_notice("xen: VDSO_CLOCKMODE_PVCLOCK not supported (tsc unstable)\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) xen_clock = ti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) pvclock_set_pvti_cpu0_va(xen_clock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) xen_clocksource.vdso_clock_mode = VDSO_CLOCKMODE_PVCLOCK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) static void __init xen_time_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) struct pvclock_vcpu_time_info *pvti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) struct timespec64 tp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) /* As Dom0 is never moved, no penalty on using TSC there */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) if (xen_initial_domain())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) xen_clocksource.rating = 275;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) clocksource_register_hz(&xen_clocksource, NSEC_PER_SEC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, xen_vcpu_nr(cpu),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) NULL) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) /* Successfully turned off 100Hz tick, so we have the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) vcpuop-based timer interface */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) printk(KERN_DEBUG "Xen: using vcpuop timer interface\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) xen_clockevent = &xen_vcpuop_clockevent;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) /* Set initial system time with full resolution */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) xen_read_wallclock(&tp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) do_settimeofday64(&tp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) setup_force_cpu_cap(X86_FEATURE_TSC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) * We check ahead on the primary time info if this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) * bit is supported hence speeding up Xen clocksource.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) pvti = &__this_cpu_read(xen_vcpu)->time;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) if (pvti->flags & PVCLOCK_TSC_STABLE_BIT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) pvclock_set_flags(PVCLOCK_TSC_STABLE_BIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) xen_setup_vsyscall_time_info();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) xen_setup_runstate_info(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) xen_setup_timer(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) xen_setup_cpu_clockevents();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) xen_time_setup_guest();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) if (xen_initial_domain())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) pvclock_gtod_register_notifier(&xen_pvclock_gtod_notifier);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) void __init xen_init_time_ops(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) xen_sched_clock_offset = xen_clocksource_read();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) pv_ops.time = xen_time_ops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) x86_init.timers.timer_init = xen_time_init;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) x86_init.timers.setup_percpu_clockev = x86_init_noop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) x86_cpuinit.setup_percpu_clockev = x86_init_noop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) x86_platform.calibrate_tsc = xen_tsc_khz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) x86_platform.get_wallclock = xen_get_wallclock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) /* Dom0 uses the native method to set the hardware RTC. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) if (!xen_initial_domain())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) x86_platform.set_wallclock = xen_set_wallclock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) #ifdef CONFIG_XEN_PVHVM
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) static void xen_hvm_setup_cpu_clockevents(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) int cpu = smp_processor_id();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) xen_setup_runstate_info(cpu);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) * xen_setup_timer(cpu) - snprintf is bad in atomic context. Hence
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) * doing it xen_hvm_cpu_notify (which gets called by smp_init during
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) * early bootup and also during CPU hotplug events).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) xen_setup_cpu_clockevents();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) void __init xen_hvm_init_time_ops(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) * vector callback is needed otherwise we cannot receive interrupts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) * on cpu > 0 and at this point we don't know how many cpus are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) * available.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) if (!xen_have_vector_callback)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) if (!xen_feature(XENFEAT_hvm_safe_pvclock)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) pr_info("Xen doesn't support pvclock on HVM, disable pv timer");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) xen_sched_clock_offset = xen_clocksource_read();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) pv_ops.time = xen_time_ops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) x86_init.timers.setup_percpu_clockev = xen_time_init;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) x86_cpuinit.setup_percpu_clockev = xen_hvm_setup_cpu_clockevents;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) x86_platform.calibrate_tsc = xen_tsc_khz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) x86_platform.get_wallclock = xen_get_wallclock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) x86_platform.set_wallclock = xen_set_wallclock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) /* Kernel parameter to specify Xen timer slop */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) static int __init parse_xen_timer_slop(char *ptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) unsigned long slop = memparse(ptr, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) xen_timerop_clockevent.min_delta_ns = slop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) xen_timerop_clockevent.min_delta_ticks = slop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) xen_vcpuop_clockevent.min_delta_ns = slop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) xen_vcpuop_clockevent.min_delta_ticks = slop;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) early_param("xen_timer_slop", parse_xen_timer_slop);
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags