Merge branch 'x86-timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 timer updates from Thomas Gleixner:
 "These updates are related to TSC handling:

   - Support platforms which have synchronized TSCs but the boot CPU has
     a non zero TSC_ADJUST value, which is considered a firmware bug on
     normal systems.

     This applies to HPE/SGI UV platforms where the platform firmware
     uses TSC_ADJUST to ensure TSC synchronization across a huge number
     of sockets, but due to power on timings the boot CPU cannot be
     guaranteed to have a zero TSC_ADJUST register value.

   - Fix the ordering of udelay calibration and kvmclock_init()

   - Cleanup the udelay and calibration code"

* 'x86-timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/tsc: Mark cyc2ns_init() and detect_art() __init
  x86/platform/UV: Mark tsc_check_sync as an init function
  x86/tsc: Make CONFIG_X86_TSC=n build work again
  x86/platform/UV: Add check of TSC state set by UV BIOS
  x86/tsc: Provide a means to disable TSC ART
  x86/tsc: Drastically reduce the number of firmware bug warnings
  x86/tsc: Skip TSC test and error messages if already unstable
  x86/tsc: Add option that TSC on Socket 0 being non-zero is valid
  x86/timers: Move simple_udelay_calibration() past kvmclock_init()
  x86/timers: Make recalibrate_cpu_khz() void
  x86/timers: Move the simple udelay calibration to tsc.h

7 files changed, 148 insertions, 51 deletions

diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h
index 47457ab975fd..7365dd4acffb 100644
--- a/arch/x86/include/asm/timer.h
+++ b/arch/x86/include/asm/timer.h
@@ -9,7 +9,7 @@
 #define TICK_SIZE (tick_nsec / 1000)
 
 unsigned long long native_sched_clock(void);
-extern int recalibrate_cpu_khz(void);
+extern void recalibrate_cpu_khz(void);
 
 extern int no_timer_check;
 
diff --git a/arch/x86/include/asm/tsc.h b/arch/x86/include/asm/tsc.h
index 8da0efb13544..cf5d53c3f9ea 100644
--- a/arch/x86/include/asm/tsc.h
+++ b/arch/x86/include/asm/tsc.h
@@ -32,15 +32,22 @@ static inline cycles_t get_cycles(void)
 
 extern struct system_counterval_t convert_art_to_tsc(u64 art);
 
+extern void tsc_early_delay_calibrate(void);
 extern void tsc_init(void);
 extern void mark_tsc_unstable(char *reason);
 extern int unsynchronized_tsc(void);
 extern int check_tsc_unstable(void);
+extern void mark_tsc_async_resets(char *reason);
 extern unsigned long native_calibrate_cpu(void);
 extern unsigned long native_calibrate_tsc(void);
 extern unsigned long long native_sched_clock_from_tsc(u64 tsc);
 
 extern int tsc_clocksource_reliable;
+#ifdef CONFIG_X86_TSC
+extern bool tsc_async_resets;
+#else
+# define tsc_async_resets	false
+#endif
 
 /*
  * Boot-time check whether the TSCs are synchronized across
diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h
index 9cffb44a3cf5..036e26d63d9a 100644
--- a/arch/x86/include/asm/uv/uv_hub.h
+++ b/arch/x86/include/asm/uv/uv_hub.h
@@ -776,23 +776,36 @@ static inline int uv_num_possible_blades(void)
 extern void uv_nmi_setup(void);
 extern void uv_nmi_setup_hubless(void);
 
+/* BIOS/Kernel flags exchange MMR */
+#define UVH_BIOS_KERNEL_MMR		UVH_SCRATCH5
+#define UVH_BIOS_KERNEL_MMR_ALIAS	UVH_SCRATCH5_ALIAS
+#define UVH_BIOS_KERNEL_MMR_ALIAS_2	UVH_SCRATCH5_ALIAS_2
+
+/* TSC sync valid, set by BIOS */
+#define UVH_TSC_SYNC_MMR	UVH_BIOS_KERNEL_MMR
+#define UVH_TSC_SYNC_SHIFT	10
+#define UVH_TSC_SYNC_SHIFT_UV2K	16	/* UV2/3k have different bits */
+#define UVH_TSC_SYNC_MASK	3	/* 0011 */
+#define UVH_TSC_SYNC_VALID	3	/* 0011 */
+#define UVH_TSC_SYNC_INVALID	2	/* 0010 */
+
 /* BMC sets a bit this MMR non-zero before sending an NMI */
-#define UVH_NMI_MMR		UVH_SCRATCH5
-#define UVH_NMI_MMR_CLEAR	UVH_SCRATCH5_ALIAS
+#define UVH_NMI_MMR		UVH_BIOS_KERNEL_MMR
+#define UVH_NMI_MMR_CLEAR	UVH_BIOS_KERNEL_MMR_ALIAS
 #define UVH_NMI_MMR_SHIFT	63
-#define	UVH_NMI_MMR_TYPE	"SCRATCH5"
+#define UVH_NMI_MMR_TYPE	"SCRATCH5"
 
 /* Newer SMM NMI handler, not present in all systems */
 #define UVH_NMI_MMRX		UVH_EVENT_OCCURRED0
 #define UVH_NMI_MMRX_CLEAR	UVH_EVENT_OCCURRED0_ALIAS
 #define UVH_NMI_MMRX_SHIFT	UVH_EVENT_OCCURRED0_EXTIO_INT0_SHFT
-#define	UVH_NMI_MMRX_TYPE	"EXTIO_INT0"
+#define UVH_NMI_MMRX_TYPE	"EXTIO_INT0"
 
 /* Non-zero indicates newer SMM NMI handler present */
 #define UVH_NMI_MMRX_SUPPORTED	UVH_EXTIO_INT0_BROADCAST
 
 /* Indicates to BIOS that we want to use the newer SMM NMI handler */
-#define UVH_NMI_MMRX_REQ	UVH_SCRATCH5_ALIAS_2
+#define UVH_NMI_MMRX_REQ	UVH_BIOS_KERNEL_MMR_ALIAS_2
 #define UVH_NMI_MMRX_REQ_SHIFT	62
 
 struct uv_hub_nmi_s {
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index 0f7f925e8ae1..e1b8e8bf6b3c 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -154,6 +154,48 @@ static int __init early_get_pnodeid(void)
 	return pnode;
 }
 
+static void __init uv_tsc_check_sync(void)
+{
+	u64 mmr;
+	int sync_state;
+	int mmr_shift;
+	char *state;
+	bool valid;
+
+	/* Accommodate different UV arch BIOSes */
+	mmr = uv_early_read_mmr(UVH_TSC_SYNC_MMR);
+	mmr_shift =
+		is_uv1_hub() ? 0 :
+		is_uv2_hub() ? UVH_TSC_SYNC_SHIFT_UV2K : UVH_TSC_SYNC_SHIFT;
+	if (mmr_shift)
+		sync_state = (mmr >> mmr_shift) & UVH_TSC_SYNC_MASK;
+	else
+		sync_state = 0;
+
+	switch (sync_state) {
+	case UVH_TSC_SYNC_VALID:
+		state = "in sync";
+		valid = true;
+		break;
+
+	case UVH_TSC_SYNC_INVALID:
+		state = "unstable";
+		valid = false;
+		break;
+	default:
+		state = "unknown: assuming valid";
+		valid = true;
+		break;
+	}
+	pr_info("UV: TSC sync state from BIOS:0%d(%s)\n", sync_state, state);
+
+	/* Mark flag that says TSC != 0 is valid for socket 0 */
+	if (valid)
+		mark_tsc_async_resets("UV BIOS");
+	else
+		mark_tsc_unstable("UV BIOS");
+}
+
 /* [Copied from arch/x86/kernel/cpu/topology.c:detect_extended_topology()] */
 
 #define SMT_LEVEL			0	/* Leaf 0xb SMT level */
@@ -288,6 +330,7 @@ static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	}
 
 	pr_info("UV: OEM IDs %s/%s, System/HUB Types %d/%d, uv_apic %d\n", oem_id, oem_table_id, uv_system_type, uv_min_hub_revision_id, uv_apic);
+	uv_tsc_check_sync();
 
 	return uv_apic;
 
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 47073f6aeaf2..8af2e8d0c0a1 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -812,26 +812,6 @@ dump_kernel_offset(struct notifier_block *self, unsigned long v, void *p)
 	return 0;
 }
 
-static void __init simple_udelay_calibration(void)
-{
-	unsigned int tsc_khz, cpu_khz;
-	unsigned long lpj;
-
-	if (!boot_cpu_has(X86_FEATURE_TSC))
-		return;
-
-	cpu_khz = x86_platform.calibrate_cpu();
-	tsc_khz = x86_platform.calibrate_tsc();
-
-	tsc_khz = tsc_khz ? : cpu_khz;
-	if (!tsc_khz)
-		return;
-
-	lpj = tsc_khz * 1000;
-	do_div(lpj, HZ);
-	loops_per_jiffy = lpj;
-}
-
 /*
  * Determine if we were loaded by an EFI loader.  If so, then we have also been
  * passed the efi memmap, systab, etc., so we should use these data structures
@@ -1039,8 +1019,6 @@ void __init setup_arch(char **cmdline_p)
 	 */
 	init_hypervisor_platform();
 
-	simple_udelay_calibration();
-
 	x86_init.resources.probe_roms();
 
 	/* after parse_early_param, so could debug it */
@@ -1125,9 +1103,6 @@ void __init setup_arch(char **cmdline_p)
 	memblock_set_current_limit(ISA_END_ADDRESS);
 	e820__memblock_setup();
 
-	if (!early_xdbc_setup_hardware())
-		early_xdbc_register_console();
-
 	reserve_bios_regions();
 
 	if (efi_enabled(EFI_MEMMAP)) {
@@ -1233,6 +1208,10 @@ void __init setup_arch(char **cmdline_p)
 	kvmclock_init();
 #endif
 
+	tsc_early_delay_calibrate();
+	if (!early_xdbc_setup_hardware())
+		early_xdbc_register_console();
+
 	x86_init.paging.pagetable_init();
 
 	kasan_init();
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index ad2b925a808e..8ea117f8142e 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -112,7 +112,7 @@ static void cyc2ns_data_init(struct cyc2ns_data *data)
 	data->cyc2ns_offset = 0;
 }
 
-static void cyc2ns_init(int cpu)
+static void __init cyc2ns_init(int cpu)
 {
 	struct cyc2ns *c2n = &per_cpu(cyc2ns, cpu);
 
@@ -812,13 +812,13 @@ unsigned long native_calibrate_cpu(void)
 	return tsc_pit_min;
 }
 
-int recalibrate_cpu_khz(void)
+void recalibrate_cpu_khz(void)
 {
 #ifndef CONFIG_SMP
 	unsigned long cpu_khz_old = cpu_khz;
 
 	if (!boot_cpu_has(X86_FEATURE_TSC))
-		return -ENODEV;
+		return;
 
 	cpu_khz = x86_platform.calibrate_cpu();
 	tsc_khz = x86_platform.calibrate_tsc();
@@ -828,10 +828,6 @@ int recalibrate_cpu_khz(void)
 		cpu_khz = tsc_khz;
 	cpu_data(0).loops_per_jiffy = cpufreq_scale(cpu_data(0).loops_per_jiffy,
 						    cpu_khz_old, cpu_khz);
-
-	return 0;
-#else
-	return -ENODEV;
 #endif
 }
 
@@ -959,17 +955,21 @@ core_initcall(cpufreq_register_tsc_scaling);
 /*
  * If ART is present detect the numerator:denominator to convert to TSC
  */
-static void detect_art(void)
+static void __init detect_art(void)
 {
 	unsigned int unused[2];
 
 	if (boot_cpu_data.cpuid_level < ART_CPUID_LEAF)
 		return;
 
-	/* Don't enable ART in a VM, non-stop TSC and TSC_ADJUST required */
+	/*
+	 * Don't enable ART in a VM, non-stop TSC and TSC_ADJUST required,
+	 * and the TSC counter resets must not occur asynchronously.
+	 */
 	if (boot_cpu_has(X86_FEATURE_HYPERVISOR) ||
 	    !boot_cpu_has(X86_FEATURE_NONSTOP_TSC) ||
-	    !boot_cpu_has(X86_FEATURE_TSC_ADJUST))
+	    !boot_cpu_has(X86_FEATURE_TSC_ADJUST) ||
+	    tsc_async_resets)
 		return;
 
 	cpuid(ART_CPUID_LEAF, &art_to_tsc_denominator,
@@ -1263,6 +1263,25 @@ static int __init init_tsc_clocksource(void)
  */
 device_initcall(init_tsc_clocksource);
 
+void __init tsc_early_delay_calibrate(void)
+{
+	unsigned long lpj;
+
+	if (!boot_cpu_has(X86_FEATURE_TSC))
+		return;
+
+	cpu_khz = x86_platform.calibrate_cpu();
+	tsc_khz = x86_platform.calibrate_tsc();
+
+	tsc_khz = tsc_khz ? : cpu_khz;
+	if (!tsc_khz)
+		return;
+
+	lpj = tsc_khz * 1000;
+	do_div(lpj, HZ);
+	loops_per_jiffy = lpj;
+}
+
 void __init tsc_init(void)
 {
 	u64 lpj, cyc;
diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c
index e76a9881306b..ec534f978867 100644
--- a/arch/x86/kernel/tsc_sync.c
+++ b/arch/x86/kernel/tsc_sync.c
@@ -31,6 +31,20 @@ struct tsc_adjust {
 
 static DEFINE_PER_CPU(struct tsc_adjust, tsc_adjust);
 
+/*
+ * TSC's on different sockets may be reset asynchronously.
+ * This may cause the TSC ADJUST value on socket 0 to be NOT 0.
+ */
+bool __read_mostly tsc_async_resets;
+
+void mark_tsc_async_resets(char *reason)
+{
+	if (tsc_async_resets)
+		return;
+	tsc_async_resets = true;
+	pr_info("tsc: Marking TSC async resets true due to %s\n", reason);
+}
+
 void tsc_verify_tsc_adjust(bool resume)
 {
 	struct tsc_adjust *adj = this_cpu_ptr(&tsc_adjust);
@@ -39,6 +53,10 @@ void tsc_verify_tsc_adjust(bool resume)
 	if (!boot_cpu_has(X86_FEATURE_TSC_ADJUST))
 		return;
 
+	/* Skip unnecessary error messages if TSC already unstable */
+	if (check_tsc_unstable())
+		return;
+
 	/* Rate limit the MSR check */
 	if (!resume && time_before(jiffies, adj->nextcheck))
 		return;
@@ -72,12 +90,22 @@ static void tsc_sanitize_first_cpu(struct tsc_adjust *cur, s64 bootval,
 	 * non zero. We don't do that on non boot cpus because physical
 	 * hotplug should have set the ADJUST register to a value > 0 so
 	 * the TSC is in sync with the already running cpus.
+	 *
+	 * Also don't force the ADJUST value to zero if that is a valid value
+	 * for socket 0 as determined by the system arch.  This is required
+	 * when multiple sockets are reset asynchronously with each other
+	 * and socket 0 may not have an TSC ADJUST value of 0.
 	 */
 	if (bootcpu && bootval != 0) {
-		pr_warn(FW_BUG "TSC ADJUST: CPU%u: %lld force to 0\n", cpu,
-			bootval);
-		wrmsrl(MSR_IA32_TSC_ADJUST, 0);
-		bootval = 0;
+		if (likely(!tsc_async_resets)) {
+			pr_warn(FW_BUG "TSC ADJUST: CPU%u: %lld force to 0\n",
+				cpu, bootval);
+			wrmsrl(MSR_IA32_TSC_ADJUST, 0);
+			bootval = 0;
+		} else {
+			pr_info("TSC ADJUST: CPU%u: %lld NOT forced to 0\n",
+				cpu, bootval);
+		}
 	}
 	cur->adjusted = bootval;
 }
@@ -91,6 +119,10 @@ bool __init tsc_store_and_check_tsc_adjust(bool bootcpu)
 	if (!boot_cpu_has(X86_FEATURE_TSC_ADJUST))
 		return false;
 
+	/* Skip unnecessary error messages if TSC already unstable */
+	if (check_tsc_unstable())
+		return false;
+
 	rdmsrl(MSR_IA32_TSC_ADJUST, bootval);
 	cur->bootval = bootval;
 	cur->nextcheck = jiffies + HZ;
@@ -119,6 +151,13 @@ bool tsc_store_and_check_tsc_adjust(bool bootcpu)
 	cur->warned = false;
 
 	/*
+	 * If a non-zero TSC value for socket 0 may be valid then the default
+	 * adjusted value cannot assumed to be zero either.
+	 */
+	if (tsc_async_resets)
+		cur->adjusted = bootval;
+
+	/*
 	 * Check whether this CPU is the first in a package to come up. In
 	 * this case do not check the boot value against another package
 	 * because the new package might have been physically hotplugged,
@@ -139,10 +178,9 @@ bool tsc_store_and_check_tsc_adjust(bool bootcpu)
 	 * Compare the boot value and complain if it differs in the
 	 * package.
 	 */
-	if (bootval != ref->bootval) {
-		pr_warn(FW_BUG "TSC ADJUST differs: Reference CPU%u: %lld CPU%u: %lld\n",
-			refcpu, ref->bootval, cpu, bootval);
-	}
+	if (bootval != ref->bootval)
+		printk_once(FW_BUG "TSC ADJUST differs within socket(s), fixing all errors\n");
+
 	/*
 	 * The TSC_ADJUST values in a package must be the same. If the boot
 	 * value on this newly upcoming CPU differs from the adjustment
@@ -150,8 +188,6 @@ bool tsc_store_and_check_tsc_adjust(bool bootcpu)
 	 * adjusted value.
 	 */
 	if (bootval != ref->adjusted) {
-		pr_warn("TSC ADJUST synchronize: Reference CPU%u: %lld CPU%u: %lld\n",
-			refcpu, ref->adjusted, cpu, bootval);
 		cur->adjusted = ref->adjusted;
 		wrmsrl(MSR_IA32_TSC_ADJUST, ref->adjusted);
 	}