LoongArch: KVM: Add PV steal time support in guest side

Per-cpu struct kvm_steal_time is added here, its size is 64 bytes and
also defined as 64 bytes, so that the whole structure is in one physical
page.

When a VCPU is online, function pv_enable_steal_time() is called. This
function will pass guest physical address of struct kvm_steal_time and
tells hypervisor to enable steal time. When a vcpu is offline, physical
address is set as 0 and tells hypervisor to disable steal time.

Here is an output of vmstat on guest when there is workload on both host
and guest. It shows steal time stat information.

procs -----------memory---------- -----io---- -system-- ------cpu-----
 r  b   swpd   free  inact active   bi    bo   in   cs us sy id wa st
15  1      0 7583616 184112  72208    20    0  162   52 31  6 43  0 20
17  0      0 7583616 184704  72192    0     0 6318 6885  5 60  8  5 22
16  0      0 7583616 185392  72144    0     0 1766 1081  0 49  0  1 50
16  0      0 7583616 184816  72304    0     0 6300 6166  4 62 12  2 20
18  0      0 7583632 184480  72240    0     0 2814 1754  2 58  4  1 35

Signed-off-by: Bibo Mao <[email protected]>
Signed-off-by: Huacai Chen <[email protected]>

Author: bibo-mao

Documentation/admin-guide/kernel-parameters.txt

@@ -4073,9 +4073,9 @@
 			prediction) vulnerability. System may allow data
 			leaks with this option.
 
-	no-steal-acc	[X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,EARLY] Disable
-			paravirtualized steal time accounting. steal time is
-			computed, but won't influence scheduler behaviour
+	no-steal-acc	[X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY]
+			Disable paravirtualized steal time accounting. steal time
+			is computed, but won't influence scheduler behaviour
 
 	nosync		[HW,M68K] Disables sync negotiation for all devices.
 

arch/loongarch/Kconfig

@@ -649,6 +649,17 @@ config PARAVIRT
 	  over full virtualization.  However, when run without a hypervisor
 	  the kernel is theoretically slower and slightly larger.
 
+config PARAVIRT_TIME_ACCOUNTING
+	bool "Paravirtual steal time accounting"
+	depends on PARAVIRT
+	help
+	  Select this option to enable fine granularity task steal time
+	  accounting. Time spent executing other tasks in parallel with
+	  the current vCPU is discounted from the vCPU power. To account for
+	  that, there can be a small performance impact.
+
+	  If in doubt, say N here.
+
 endmenu
 
 config ARCH_SELECT_MEMORY_MODEL

arch/loongarch/include/asm/paravirt.h

@@ -18,6 +18,7 @@ static inline u64 paravirt_steal_clock(int cpu)
 }
 
 int __init pv_ipi_init(void);
+int __init pv_time_init(void);
 
 #else
 
@@ -26,5 +27,9 @@ static inline int pv_ipi_init(void)
 	return 0;
 }
 
+static inline int pv_time_init(void)
+{
+	return 0;
+}
 #endif // CONFIG_PARAVIRT
 #endif

arch/loongarch/kernel/paravirt.c

@@ -4,11 +4,14 @@
 #include <linux/interrupt.h>
 #include <linux/jump_label.h>
 #include <linux/kvm_para.h>
+#include <linux/reboot.h>
 #include <linux/static_call.h>
 #include <asm/paravirt.h>
 
+static int has_steal_clock;
 struct static_key paravirt_steal_enabled;
 struct static_key paravirt_steal_rq_enabled;
+static DEFINE_PER_CPU(struct kvm_steal_time, steal_time) __aligned(64);
 
 static u64 native_steal_clock(int cpu)
 {
@@ -17,6 +20,34 @@ static u64 native_steal_clock(int cpu)
 
 DEFINE_STATIC_CALL(pv_steal_clock, native_steal_clock);
 
+static bool steal_acc = true;
+
+static int __init parse_no_stealacc(char *arg)
+{
+	steal_acc = false;
+	return 0;
+}
+early_param("no-steal-acc", parse_no_stealacc);
+
+static u64 paravt_steal_clock(int cpu)
+{
+	int version;
+	u64 steal;
+	struct kvm_steal_time *src;
+
+	src = &per_cpu(steal_time, cpu);
+	do {
+
+		version = src->version;
+		virt_rmb(); /* Make sure that the version is read before the steal */
+		steal = src->steal;
+		virt_rmb(); /* Make sure that the steal is read before the next version */
+
+	} while ((version & 1) || (version != src->version));
+
+	return steal;
+}
+
 #ifdef CONFIG_SMP
 static void pv_send_ipi_single(int cpu, unsigned int action)
 {
@@ -149,3 +180,117 @@ int __init pv_ipi_init(void)
 
 	return 0;
 }
+
+static int pv_enable_steal_time(void)
+{
+	int cpu = smp_processor_id();
+	unsigned long addr;
+	struct kvm_steal_time *st;
+
+	if (!has_steal_clock)
+		return -EPERM;
+
+	st = &per_cpu(steal_time, cpu);
+	addr = per_cpu_ptr_to_phys(st);
+
+	/* The whole structure kvm_steal_time should be in one page */
+	if (PFN_DOWN(addr) != PFN_DOWN(addr + sizeof(*st))) {
+		pr_warn("Illegal PV steal time addr %lx\n", addr);
+		return -EFAULT;
+	}
+
+	addr |= KVM_STEAL_PHYS_VALID;
+	kvm_hypercall2(KVM_HCALL_FUNC_NOTIFY, KVM_FEATURE_STEAL_TIME, addr);
+
+	return 0;
+}
+
+static void pv_disable_steal_time(void)
+{
+	if (has_steal_clock)
+		kvm_hypercall2(KVM_HCALL_FUNC_NOTIFY, KVM_FEATURE_STEAL_TIME, 0);
+}
+
+#ifdef CONFIG_SMP
+static int pv_time_cpu_online(unsigned int cpu)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	pv_enable_steal_time();
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+static int pv_time_cpu_down_prepare(unsigned int cpu)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	pv_disable_steal_time();
+	local_irq_restore(flags);
+
+	return 0;
+}
+#endif
+
+static void pv_cpu_reboot(void *unused)
+{
+	pv_disable_steal_time();
+}
+
+static int pv_reboot_notify(struct notifier_block *nb, unsigned long code, void *unused)
+{
+	on_each_cpu(pv_cpu_reboot, NULL, 1);
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block pv_reboot_nb = {
+	.notifier_call  = pv_reboot_notify,
+};
+
+int __init pv_time_init(void)
+{
+	int r, feature;
+
+	if (!cpu_has_hypervisor)
+		return 0;
+	if (!kvm_para_available())
+		return 0;
+
+	feature = read_cpucfg(CPUCFG_KVM_FEATURE);
+	if (!(feature & KVM_FEATURE_STEAL_TIME))
+		return 0;
+
+	has_steal_clock = 1;
+	r = pv_enable_steal_time();
+	if (r < 0) {
+		has_steal_clock = 0;
+		return 0;
+	}
+	register_reboot_notifier(&pv_reboot_nb);
+
+#ifdef CONFIG_SMP
+	r = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
+				      "loongarch/pv_time:online",
+				      pv_time_cpu_online, pv_time_cpu_down_prepare);
+	if (r < 0) {
+		has_steal_clock = 0;
+		pr_err("Failed to install cpu hotplug callbacks\n");
+		return r;
+	}
+#endif
+
+	static_call_update(pv_steal_clock, paravt_steal_clock);
+
+	static_key_slow_inc(&paravirt_steal_enabled);
+#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING
+	if (steal_acc)
+		static_key_slow_inc(&paravirt_steal_rq_enabled);
+#endif
+
+	pr_info("Using paravirt steal-time\n");
+
+	return 0;
+}

arch/loongarch/kernel/time.c

@@ -15,6 +15,7 @@
 
 #include <asm/cpu-features.h>
 #include <asm/loongarch.h>
+#include <asm/paravirt.h>
 #include <asm/time.h>
 
 u64 cpu_clock_freq;
@@ -214,4 +215,5 @@ void __init time_init(void)
 
 	constant_clockevent_init();
 	constant_clocksource_init();
+	pv_time_init();
 }