Rework the seemingly generic x86_cpuinit_ops.early_percpu_clock_init hook into a dedicated PV sched_clock hook, as the only reason the hook exists is to allow kvmclock to enable its PV clock on secondary CPUs before the kernel tries to reference sched_clock, e.g. when grabbing a timestamp for printk.
Rearranging the hook doesn't exactly reduce complexity; arguably it does the opposite. But as-is, it's practically impossible to understand *why* kvmclock needs to do early configuration. Signed-off-by: Sean Christopherson <[email protected]> --- arch/x86/include/asm/timer.h | 8 ++++++-- arch/x86/include/asm/x86_init.h | 2 -- arch/x86/kernel/kvmclock.c | 13 ++++++------- arch/x86/kernel/smpboot.c | 3 ++- arch/x86/kernel/tsc.c | 16 +++++++++++++++- arch/x86/kernel/x86_init.c | 1 - 6 files changed, 29 insertions(+), 14 deletions(-) diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h index ca5c95d48c03..ab1271bd9c3b 100644 --- a/arch/x86/include/asm/timer.h +++ b/arch/x86/include/asm/timer.h @@ -15,14 +15,18 @@ extern bool using_native_sched_clock(void); #ifdef CONFIG_PARAVIRT int __init __paravirt_set_sched_clock(u64 (*func)(void), bool stable, - void (*save)(void), void (*restore)(void)); + void (*save)(void), void (*restore)(void), + void (*start_secondary)); static __always_inline void paravirt_set_sched_clock(u64 (*func)(void), void (*save)(void), void (*restore)(void)) { - (void)__paravirt_set_sched_clock(func, true, save, restore); + (void)__paravirt_set_sched_clock(func, true, save, restore, NULL); } +void paravirt_sched_clock_start_secondary(void); +#else +static inline void paravirt_sched_clock_start_secondary(void) { } #endif /* diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h index 6c8a6ead84f6..d1b3f18ea41f 100644 --- a/arch/x86/include/asm/x86_init.h +++ b/arch/x86/include/asm/x86_init.h @@ -187,13 +187,11 @@ struct x86_init_ops { /** * struct x86_cpuinit_ops - platform specific cpu hotplug setups * @setup_percpu_clockev: set up the per cpu clock event device - * @early_percpu_clock_init: early init of the per cpu clock event device * @fixup_cpu_id: fixup function for cpuinfo_x86::topo.pkg_id * @parallel_bringup: Parallel bringup control */ struct x86_cpuinit_ops { void (*setup_percpu_clockev)(void); - void (*early_percpu_clock_init)(void); void (*fixup_cpu_id)(struct cpuinfo_x86 *c, int node); bool parallel_bringup; }; diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index 0578bc448b1b..62c8ea2e6769 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -127,12 +127,13 @@ static void kvm_save_sched_clock_state(void) kvmclock_disable(); } -#ifdef CONFIG_SMP -static void kvm_setup_secondary_clock(void) +static void kvm_setup_secondary_sched_clock(void) { + if (WARN_ON_ONCE(!IS_ENABLED(CONFIG_SMP))) + return; + kvm_register_clock("secondary cpu, sched_clock setup"); } -#endif static void kvm_restore_sched_clock_state(void) { @@ -352,7 +353,8 @@ static __init void kvm_sched_clock_init(bool stable) { if (__paravirt_set_sched_clock(kvm_sched_clock_read, stable, kvm_save_sched_clock_state, - kvm_restore_sched_clock_state)) + kvm_restore_sched_clock_state, + kvm_setup_secondary_sched_clock)) return; kvm_sched_clock_offset = kvm_clock_read(); @@ -437,9 +439,6 @@ void __init kvmclock_init(void) x86_platform.get_wallclock = kvm_get_wallclock; x86_platform.set_wallclock = kvm_set_wallclock; -#ifdef CONFIG_SMP - x86_cpuinit.early_percpu_clock_init = kvm_setup_secondary_clock; -#endif kvm_get_preset_lpj(); clocksource_register_hz(&kvm_clock, NSEC_PER_SEC); diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 294a8ea60298..318ae70e5e7b 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -78,6 +78,7 @@ #include <asm/io_apic.h> #include <asm/fpu/api.h> #include <asm/setup.h> +#include <asm/timer.h> #include <asm/uv/uv.h> #include <asm/microcode.h> #include <asm/i8259.h> @@ -275,7 +276,7 @@ static void notrace __noendbr start_secondary(void *unused) cpu_init(); fpu__init_cpu(); rcutree_report_cpu_starting(raw_smp_processor_id()); - x86_cpuinit.early_percpu_clock_init(); + paravirt_sched_clock_start_secondary(); ap_starting(); diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 7a261214fa3e..f78e86494dec 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -280,8 +280,19 @@ bool using_native_sched_clock(void) return static_call_query(pv_sched_clock) == native_sched_clock; } +#ifdef CONFIG_SMP +static void (*pv_sched_clock_start_secondary)(void) __ro_after_init; + +void paravirt_sched_clock_start_secondary(void) +{ + if (pv_sched_clock_start_secondary) + pv_sched_clock_start_secondary(); +} +#endif + int __init __paravirt_set_sched_clock(u64 (*func)(void), bool stable, - void (*save)(void), void (*restore)(void)) + void (*save)(void), void (*restore)(void), + void (*start_secondary)) { /* * Don't replace TSC with a PV clock when running as a CoCo guest and @@ -298,6 +309,9 @@ int __init __paravirt_set_sched_clock(u64 (*func)(void), bool stable, static_call_update(pv_sched_clock, func); x86_platform.save_sched_clock_state = save; x86_platform.restore_sched_clock_state = restore; +#ifdef CONFIG_SMP + pv_sched_clock_start_secondary = start_secondary; +#endif return 0; } #else diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index ebefb77c37bb..cbb5ee613ed5 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -128,7 +128,6 @@ struct x86_init_ops x86_init __initdata = { }; struct x86_cpuinit_ops x86_cpuinit = { - .early_percpu_clock_init = x86_init_noop, .setup_percpu_clockev = setup_secondary_APIC_clock, .parallel_bringup = true, }; -- 2.54.0.563.g4f69b47b94-goog
