Merge tag 'timers-core-2023-02-20' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull timer updates from Thomas Gleixner:
 "Updates for timekeeping, timers and clockevent/source drivers:

  Core:

   - Yet another round of improvements to make the clocksource watchdog
     more robust:

       - Relax the clocksource-watchdog skew criteria to match the NTP
         criteria.

       - Temporarily skip the watchdog when high memory latencies are
         detected which can lead to false-positives.

       - Provide an option to enable TSC skew detection even on systems
         where TSC is marked as reliable.

     Sigh!

   - Initialize the restart block in the nanosleep syscalls to be
     directed to the no restart function instead of doing a partial
     setup on entry.

     This prevents an erroneous restart_syscall() invocation from
     corrupting user space data. While such a situation is clearly a
     user space bug, preventing this is a correctness issue and caters
     to the least suprise principle.

   - Ignore the hrtimer slack for realtime tasks in schedule_hrtimeout()
     to align it with the nanosleep semantics.

  Drivers:

   - The obligatory new driver bindings for Mediatek, Rockchip and
     RISC-V variants.

   - Add support for the C3STOP misfeature to the RISC-V timer to handle
     the case where the timer stops in deeper idle state.

   - Set up a static key in the RISC-V timer correctly before first use.

   - The usual small improvements and fixes all over the place"

* tag 'timers-core-2023-02-20' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (30 commits)
  clocksource/drivers/timer-sun4i: Add CLOCK_EVT_FEAT_DYNIRQ
  clocksource/drivers/em_sti: Mark driver as non-removable
  clocksource/drivers/sh_tmu: Mark driver as non-removable
  clocksource/drivers/riscv: Patch riscv_clock_next_event() jump before first use
  clocksource/drivers/timer-microchip-pit64b: Add delay timer
  clocksource/drivers/timer-microchip-pit64b: Select driver only on ARM
  dt-bindings: timer: sifive,clint: add comaptibles for T-Head's C9xx
  dt-bindings: timer: mediatek,mtk-timer: add MT8365
  clocksource/drivers/riscv: Get rid of clocksource_arch_init() callback
  clocksource/drivers/sh_cmt: Mark driver as non-removable
  clocksource/drivers/timer-microchip-pit64b: Drop obsolete dependency on COMPILE_TEST
  clocksource/drivers/riscv: Increase the clock source rating
  clocksource/drivers/timer-riscv: Set CLOCK_EVT_FEAT_C3STOP based on DT
  dt-bindings: timer: Add bindings for the RISC-V timer device
  RISC-V: time: initialize hrtimer based broadcast clock event device
  dt-bindings: timer: rk-timer: Add rktimer for rv1126
  time/debug: Fix memory leak with using debugfs_lookup()
  clocksource: Enable TSC watchdog checking of HPET and PMTMR only when requested
  posix-timers: Use atomic64_try_cmpxchg() in __update_gt_cputime()
  clocksource: Verify HPET and PMTMR when TSC unverified
  ...

Documentation/admin-guide/kernel-parameters.txt

@@ -6369,6 +6369,16 @@
 			in situations with strict latency requirements (where
 			interruptions from clocksource watchdog are not
 			acceptable).
+			[x86] recalibrate: force recalibration against a HW timer
+			(HPET or PM timer) on systems whose TSC frequency was
+			obtained from HW or FW using either an MSR or CPUID(0x15).
+			Warn if the difference is more than 500 ppm.
+			[x86] watchdog: Use TSC as the watchdog clocksource with
+			which to check other HW timers (HPET or PM timer), but
+			only on systems where TSC has been deemed trustworthy.
+			This will be suppressed by an earlier tsc=nowatchdog and
+			can be overridden by a later tsc=nowatchdog.  A console
+			message will flag any such suppression or overriding.
 
 	tsc_early_khz=  [X86] Skip early TSC calibration and use the given
 			value instead. Useful when the early TSC frequency discovery

Documentation/devicetree/bindings/timer/mediatek,mtk-timer.txt

@@ -33,6 +33,7 @@ Required properties:
 
 	For those SoCs that use CPUX
 	* "mediatek,mt6795-systimer" for MT6795 compatible timers (CPUX)
+	* "mediatek,mt8365-systimer" for MT8365 compatible timers (CPUX)
 
 - reg: Should contain location and length for timer register.
 - clocks: Should contain system clock.

Documentation/devicetree/bindings/timer/riscv,timer.yaml

+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/timer/riscv,timer.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: RISC-V timer
+
+maintainers:
+  - Anup Patel <anup@brainfault.org>
+
+description: |+
+  RISC-V platforms always have a RISC-V timer device for the supervisor-mode
+  based on the time CSR defined by the RISC-V privileged specification. The
+  timer interrupts of this device are configured using the RISC-V SBI Time
+  extension or the RISC-V Sstc extension.
+
+  The clock frequency of RISC-V timer device is specified via the
+  "timebase-frequency" DT property of "/cpus" DT node which is described
+  in Documentation/devicetree/bindings/riscv/cpus.yaml
+
+properties:
+  compatible:
+    enum:
+      - riscv,timer
+
+  interrupts-extended:
+    minItems: 1
+    maxItems: 4096   # Should be enough?
+
+  riscv,timer-cannot-wake-cpu:
+    type: boolean
+    description:
+      If present, the timer interrupt cannot wake up the CPU from one or
+      more suspend/idle states.
+
+additionalProperties: false
+
+required:
+  - compatible
+  - interrupts-extended
+
+examples:
+  - |
+    timer {
+      compatible = "riscv,timer";
+      interrupts-extended = <&cpu1intc 5>,
+                            <&cpu2intc 5>,
+                            <&cpu3intc 5>,
+                            <&cpu4intc 5>;
+    };
+...

Documentation/devicetree/bindings/timer/rockchip,rk-timer.yaml

@@ -17,6 +17,7 @@ properties:
       - items:
           - enum:
               - rockchip,rv1108-timer
+              - rockchip,rv1126-timer
               - rockchip,rk3036-timer
               - rockchip,rk3128-timer
               - rockchip,rk3188-timer

Documentation/devicetree/bindings/timer/sifive,clint.yaml

@@ -20,6 +20,10 @@ description:
   property of "/cpus" DT node. The "timebase-frequency" DT property is
   described in Documentation/devicetree/bindings/riscv/cpus.yaml
 
+  T-Head C906/C910 CPU cores include an implementation of CLINT too, however
+  their implementation lacks a memory-mapped MTIME register, thus not
+  compatible with SiFive ones.
+
 properties:
   compatible:
     oneOf:
@@ -29,6 +33,10 @@ properties:
               - starfive,jh7100-clint
               - canaan,k210-clint
           - const: sifive,clint0
+      - items:
+          - enum:
+              - allwinner,sun20i-d1-clint
+          - const: thead,c900-clint
       - items:
           - const: sifive,clint0
           - const: riscv,clint0

arch/riscv/Kconfig

@@ -12,7 +12,6 @@ config 32BIT
 
 config RISCV
 	def_bool y
-	select ARCH_CLOCKSOURCE_INIT
 	select ARCH_ENABLE_HUGEPAGE_MIGRATION if HUGETLB_PAGE && MIGRATION
 	select ARCH_ENABLE_SPLIT_PMD_PTLOCK if PGTABLE_LEVELS > 2
 	select ARCH_HAS_BINFMT_FLAT

arch/riscv/kernel/time.c

@@ -5,6 +5,7 @@
  */
 
 #include <linux/of_clk.h>
+#include <linux/clockchips.h>
 #include <linux/clocksource.h>
 #include <linux/delay.h>
 #include <asm/sbi.h>
@@ -29,13 +30,6 @@ void __init time_init(void)
 
 	of_clk_init(NULL);
 	timer_probe();
-}
 
-void clocksource_arch_init(struct clocksource *cs)
-{
-#ifdef CONFIG_GENERIC_GETTIMEOFDAY
-	cs->vdso_clock_mode = VDSO_CLOCKMODE_ARCHTIMER;
-#else
-	cs->vdso_clock_mode = VDSO_CLOCKMODE_NONE;
-#endif
+	tick_setup_hrtimer_broadcast();
 }

arch/x86/include/asm/time.h

@@ -8,6 +8,7 @@
 extern void hpet_time_init(void);
 extern void time_init(void);
 extern bool pit_timer_init(void);
+extern bool tsc_clocksource_watchdog_disabled(void);
 
 extern struct clock_event_device *global_clock_event;
 

arch/x86/kernel/hpet.c

@@ -1091,6 +1091,8 @@ int __init hpet_enable(void)
 	if (!hpet_counting())
 		goto out_nohpet;
 
+	if (tsc_clocksource_watchdog_disabled())
+		clocksource_hpet.flags |= CLOCK_SOURCE_MUST_VERIFY;
 	clocksource_register_hz(&clocksource_hpet, (u32)hpet_freq);
 
 	if (id & HPET_ID_LEGSUP) {

arch/x86/kernel/tsc.c

@@ -48,6 +48,8 @@ static DEFINE_STATIC_KEY_FALSE(__use_tsc);
 
 int tsc_clocksource_reliable;
 
+static int __read_mostly tsc_force_recalibrate;
+
 static u32 art_to_tsc_numerator;
 static u32 art_to_tsc_denominator;
 static u64 art_to_tsc_offset;
@@ -291,6 +293,7 @@ __setup("notsc", notsc_setup);
 
 static int no_sched_irq_time;
 static int no_tsc_watchdog;
+static int tsc_as_watchdog;
 
 static int __init tsc_setup(char *str)
 {
@@ -300,8 +303,22 @@ static int __init tsc_setup(char *str)
 		no_sched_irq_time = 1;
 	if (!strcmp(str, "unstable"))
 		mark_tsc_unstable("boot parameter");
-	if (!strcmp(str, "nowatchdog"))
+	if (!strcmp(str, "nowatchdog")) {
 		no_tsc_watchdog = 1;
+		if (tsc_as_watchdog)
+			pr_alert("%s: Overriding earlier tsc=watchdog with tsc=nowatchdog\n",
+				 __func__);
+		tsc_as_watchdog = 0;
+	}
+	if (!strcmp(str, "recalibrate"))
+		tsc_force_recalibrate = 1;
+	if (!strcmp(str, "watchdog")) {
+		if (no_tsc_watchdog)
+			pr_alert("%s: tsc=watchdog overridden by earlier tsc=nowatchdog\n",
+				 __func__);
+		else
+			tsc_as_watchdog = 1;
+	}
 	return 1;
 }
 
@@ -1184,6 +1201,12 @@ static void __init tsc_disable_clocksource_watchdog(void)
 	clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY;
 }
 
+bool tsc_clocksource_watchdog_disabled(void)
+{
+	return !(clocksource_tsc.flags & CLOCK_SOURCE_MUST_VERIFY) &&
+	       tsc_as_watchdog && !no_tsc_watchdog;
+}
+
 static void __init check_system_tsc_reliable(void)
 {
 #if defined(CONFIG_MGEODEGX1) || defined(CONFIG_MGEODE_LX) || defined(CONFIG_X86_GENERIC)
@@ -1372,6 +1395,25 @@ static void tsc_refine_calibration_work(struct work_struct *work)
 	else
 		freq = calc_pmtimer_ref(delta, ref_start, ref_stop);
 
+	/* Will hit this only if tsc_force_recalibrate has been set */
+	if (boot_cpu_has(X86_FEATURE_TSC_KNOWN_FREQ)) {
+
+		/* Warn if the deviation exceeds 500 ppm */
+		if (abs(tsc_khz - freq) > (tsc_khz >> 11)) {
+			pr_warn("Warning: TSC freq calibrated by CPUID/MSR differs from what is calibrated by HW timer, please check with vendor!!\n");
+			pr_info("Previous calibrated TSC freq:\t %lu.%03lu MHz\n",
+				(unsigned long)tsc_khz / 1000,
+				(unsigned long)tsc_khz % 1000);
+		}
+
+		pr_info("TSC freq recalibrated by [%s]:\t %lu.%03lu MHz\n",
+			hpet ? "HPET" : "PM_TIMER",
+			(unsigned long)freq / 1000,
+			(unsigned long)freq % 1000);
+
+		return;
+	}
+
 	/* Make sure we're within 1% */
 	if (abs(tsc_khz - freq) > tsc_khz/100)
 		goto out;
@@ -1405,8 +1447,10 @@ static int __init init_tsc_clocksource(void)
 	if (!boot_cpu_has(X86_FEATURE_TSC) || !tsc_khz)
 		return 0;
 
-	if (tsc_unstable)
-		goto unreg;
+	if (tsc_unstable) {
+		clocksource_unregister(&clocksource_tsc_early);
+		return 0;
+	}
 
 	if (boot_cpu_has(X86_FEATURE_NONSTOP_TSC_S3))
 		clocksource_tsc.flags |= CLOCK_SOURCE_SUSPEND_NONSTOP;
@@ -1419,8 +1463,9 @@ static int __init init_tsc_clocksource(void)
 		if (boot_cpu_has(X86_FEATURE_ART))
 			art_related_clocksource = &clocksource_tsc;
 		clocksource_register_khz(&clocksource_tsc, tsc_khz);
-unreg:
 		clocksource_unregister(&clocksource_tsc_early);
+
+		if (!tsc_force_recalibrate)
 			return 0;
 	}
 

drivers/clocksource/Kconfig

@@ -706,7 +706,7 @@ config INGENIC_OST
 
 config MICROCHIP_PIT64B
 	bool "Microchip PIT64B support"
-	depends on OF || COMPILE_TEST
+	depends on OF && ARM
 	select TIMER_OF
 	help
 	  This option enables Microchip PIT64B timer for Atmel

drivers/clocksource/acpi_pm.c

@@ -23,6 +23,7 @@
 #include <linux/pci.h>
 #include <linux/delay.h>
 #include <asm/io.h>
+#include <asm/time.h>
 
 /*
  * The I/O port the PMTMR resides at.
@@ -210,8 +211,9 @@ static int __init init_acpi_pm_clocksource(void)
 		return -ENODEV;
 	}
 
-	return clocksource_register_hz(&clocksource_acpi_pm,
-						PMTMR_TICKS_PER_SEC);
+	if (tsc_clocksource_watchdog_disabled())
+		clocksource_acpi_pm.flags |= CLOCK_SOURCE_MUST_VERIFY;
+	return clocksource_register_hz(&clocksource_acpi_pm, PMTMR_TICKS_PER_SEC);
 }
 
 /* We use fs_initcall because we want the PCI fixups to have run

drivers/clocksource/em_sti.c

@@ -333,11 +333,6 @@ static int em_sti_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int em_sti_remove(struct platform_device *pdev)
-{
-	return -EBUSY; /* cannot unregister clockevent and clocksource */
-}
-
 static const struct of_device_id em_sti_dt_ids[] = {
 	{ .compatible = "renesas,em-sti", },
 	{},
@@ -346,10 +341,10 @@ MODULE_DEVICE_TABLE(of, em_sti_dt_ids);
 
 static struct platform_driver em_sti_device_driver = {
 	.probe		= em_sti_probe,
-	.remove		= em_sti_remove,
 	.driver		= {
 		.name	= "em_sti",
 		.of_match_table = em_sti_dt_ids,
+		.suppress_bind_attrs = true,
 	}
 };
 

drivers/clocksource/sh_cmt.c

@@ -1145,17 +1145,12 @@ static int sh_cmt_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int sh_cmt_remove(struct platform_device *pdev)
-{
-	return -EBUSY; /* cannot unregister clockevent and clocksource */
-}
-
 static struct platform_driver sh_cmt_device_driver = {
 	.probe		= sh_cmt_probe,
-	.remove		= sh_cmt_remove,
 	.driver		= {
 		.name	= "sh_cmt",
 		.of_match_table = of_match_ptr(sh_cmt_of_table),
+		.suppress_bind_attrs = true,
 	},
 	.id_table	= sh_cmt_id_table,
 };

drivers/clocksource/sh_tmu.c

@@ -632,11 +632,6 @@ static int sh_tmu_probe(struct platform_device *pdev)
 	return 0;
 }
 
-static int sh_tmu_remove(struct platform_device *pdev)
-{
-	return -EBUSY; /* cannot unregister clockevent and clocksource */
-}
-
 static const struct platform_device_id sh_tmu_id_table[] = {
 	{ "sh-tmu", SH_TMU },
 	{ "sh-tmu-sh3", SH_TMU_SH3 },
@@ -652,10 +647,10 @@ MODULE_DEVICE_TABLE(of, sh_tmu_of_table);
 
 static struct platform_driver sh_tmu_device_driver = {
 	.probe		= sh_tmu_probe,
-	.remove		= sh_tmu_remove,
 	.driver		= {
 		.name	= "sh_tmu",
 		.of_match_table = of_match_ptr(sh_tmu_of_table),
+		.suppress_bind_attrs = true,
 	},
 	.id_table	= sh_tmu_id_table,
 };

drivers/clocksource/timer-microchip-pit64b.c

@@ -9,6 +9,7 @@
 
 #include <linux/clk.h>
 #include <linux/clockchips.h>
+#include <linux/delay.h>
 #include <linux/interrupt.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
@@ -92,6 +93,8 @@ struct mchp_pit64b_clksrc {
 static void __iomem *mchp_pit64b_cs_base;
 /* Default cycles for clockevent timer. */
 static u64 mchp_pit64b_ce_cycles;
+/* Delay timer. */
+static struct delay_timer mchp_pit64b_dt;
 
 static inline u64 mchp_pit64b_cnt_read(void __iomem *base)
 {
@@ -169,6 +172,11 @@ static u64 notrace mchp_pit64b_sched_read_clk(void)
 	return mchp_pit64b_cnt_read(mchp_pit64b_cs_base);
 }
 
+static unsigned long notrace mchp_pit64b_dt_read(void)
+{
+	return mchp_pit64b_cnt_read(mchp_pit64b_cs_base);
+}
+
 static int mchp_pit64b_clkevt_shutdown(struct clock_event_device *cedev)
 {
 	struct mchp_pit64b_timer *timer = clkevt_to_mchp_pit64b_timer(cedev);
@@ -376,6 +384,10 @@ static int __init mchp_pit64b_init_clksrc(struct mchp_pit64b_timer *timer,
 
 	sched_clock_register(mchp_pit64b_sched_read_clk, 64, clk_rate);
 
+	mchp_pit64b_dt.read_current_timer = mchp_pit64b_dt_read;
+	mchp_pit64b_dt.freq = clk_rate;
+	register_current_timer_delay(&mchp_pit64b_dt);
+
 	return 0;
 }
 

drivers/clocksource/timer-riscv.c

@@ -28,6 +28,7 @@
 #include <asm/timex.h>
 
 static DEFINE_STATIC_KEY_FALSE(riscv_sstc_available);
+static bool riscv_timer_cannot_wake_cpu;
 
 static int riscv_clock_next_event(unsigned long delta,
 		struct clock_event_device *ce)
@@ -73,10 +74,15 @@ static u64 notrace riscv_sched_clock(void)
 
 static struct clocksource riscv_clocksource = {
 	.name		= "riscv_clocksource",
-	.rating		= 300,
+	.rating		= 400,
 	.mask		= CLOCKSOURCE_MASK(64),
 	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
 	.read		= riscv_clocksource_rdtime,
+#if IS_ENABLED(CONFIG_GENERIC_GETTIMEOFDAY)
+	.vdso_clock_mode = VDSO_CLOCKMODE_ARCHTIMER,
+#else
+	.vdso_clock_mode = VDSO_CLOCKMODE_NONE,
+#endif
 };
 
 static int riscv_timer_starting_cpu(unsigned int cpu)
@@ -85,6 +91,8 @@ static int riscv_timer_starting_cpu(unsigned int cpu)
 
 	ce->cpumask = cpumask_of(cpu);
 	ce->irq = riscv_clock_event_irq;
+	if (riscv_timer_cannot_wake_cpu)
+		ce->features |= CLOCK_EVT_FEAT_C3STOP;
 	clockevents_config_and_register(ce, riscv_timebase, 100, 0x7fffffff);
 
 	enable_percpu_irq(riscv_clock_event_irq,
@@ -139,6 +147,13 @@ static int __init riscv_timer_init_dt(struct device_node *n)
 	if (cpuid != smp_processor_id())
 		return 0;
 
+	child = of_find_compatible_node(NULL, NULL, "riscv,timer");
+	if (child) {
+		riscv_timer_cannot_wake_cpu = of_property_read_bool(child,
+					"riscv,timer-cannot-wake-cpu");
+		of_node_put(child);
+	}
+
 	domain = NULL;
 	child = of_get_compatible_child(n, "riscv,cpu-intc");
 	if (!child) {
@@ -177,6 +192,11 @@ static int __init riscv_timer_init_dt(struct device_node *n)
 		return error;
 	}
 
+	if (riscv_isa_extension_available(NULL, SSTC)) {
+		pr_info("Timer interrupt in S-mode is available via sstc extension\n");
+		static_branch_enable(&riscv_sstc_available);
+	}
+
 	error = cpuhp_setup_state(CPUHP_AP_RISCV_TIMER_STARTING,
 			 "clockevents/riscv/timer:starting",
 			 riscv_timer_starting_cpu, riscv_timer_dying_cpu);
@@ -184,11 +204,6 @@ static int __init riscv_timer_init_dt(struct device_node *n)
 		pr_err("cpu hp setup state failed for RISCV timer [%d]\n",
 		       error);
 
-	if (riscv_isa_extension_available(NULL, SSTC)) {
-		pr_info("Timer interrupt in S-mode is available via sstc extension\n");
-		static_branch_enable(&riscv_sstc_available);
-	}
-
 	return error;
 }
 

drivers/clocksource/timer-sun4i.c

@@ -144,7 +144,8 @@ static struct timer_of to = {
 	.clkevt = {
 		.name = "sun4i_tick",
 		.rating = 350,
-		.features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
+		.features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT |
+				CLOCK_EVT_FEAT_DYNIRQ,
 		.set_state_shutdown = sun4i_clkevt_shutdown,
 		.set_state_periodic = sun4i_clkevt_set_periodic,
 		.set_state_oneshot = sun4i_clkevt_set_oneshot,

include/linux/bits.h

@@ -6,7 +6,6 @@
 #include <vdso/bits.h>
 #include <asm/bitsperlong.h>
 
-#define BIT_ULL(nr)		(ULL(1) << (nr))
 #define BIT_MASK(nr)		(UL(1) << ((nr) % BITS_PER_LONG))
 #define BIT_WORD(nr)		((nr) / BITS_PER_LONG)
 #define BIT_ULL_MASK(nr)	(ULL(1) << ((nr) % BITS_PER_LONG_LONG))

include/vdso/bits.h

@@ -5,5 +5,6 @@
 #include <vdso/const.h>
 
 #define BIT(nr)			(UL(1) << (nr))
+#define BIT_ULL(nr)		(ULL(1) << (nr))
 
 #endif	/* __VDSO_BITS_H */

kernel/time/Kconfig

@@ -200,10 +200,14 @@ config CLOCKSOURCE_WATCHDOG_MAX_SKEW_US
 	int "Clocksource watchdog maximum allowable skew (in μs)"
 	depends on CLOCKSOURCE_WATCHDOG
 	range 50 1000
-	default 100
+	default 125
 	help
 	  Specify the maximum amount of allowable watchdog skew in
 	  microseconds before reporting the clocksource to be unstable.
+	  The default is based on a half-second clocksource watchdog
+	  interval and NTP's maximum frequency drift of 500 parts
+	  per million.	If the clocksource is good enough for NTP,
+	  it is good enough for the clocksource watchdog!
 
 endmenu
 endif

kernel/time/clocksource.c

@@ -95,6 +95,11 @@ static char override_name[CS_NAME_LEN];
 static int finished_booting;
 static u64 suspend_start;
 
+/*
+ * Interval: 0.5sec.
+ */
+#define WATCHDOG_INTERVAL (HZ >> 1)
+
 /*
  * Threshold: 0.0312s, when doubled: 0.0625s.
  * Also a default for cs->uncertainty_margin when registering clocks.
@@ -106,11 +111,14 @@ static u64 suspend_start;
  * clocksource surrounding a read of the clocksource being validated.
  * This delay could be due to SMIs, NMIs, or to VCPU preemptions.  Used as
  * a lower bound for cs->uncertainty_margin values when registering clocks.
+ *
+ * The default of 500 parts per million is based on NTP's limits.
+ * If a clocksource is good enough for NTP, it is good enough for us!
  */
 #ifdef CONFIG_CLOCKSOURCE_WATCHDOG_MAX_SKEW_US
 #define MAX_SKEW_USEC	CONFIG_CLOCKSOURCE_WATCHDOG_MAX_SKEW_US
 #else
-#define MAX_SKEW_USEC	100
+#define MAX_SKEW_USEC	(125 * WATCHDOG_INTERVAL / HZ)
 #endif
 
 #define WATCHDOG_MAX_SKEW (MAX_SKEW_USEC * NSEC_PER_USEC)
@@ -140,11 +148,6 @@ static inline void clocksource_watchdog_unlock(unsigned long *flags)
 static int clocksource_watchdog_kthread(void *data);
 static void __clocksource_change_rating(struct clocksource *cs, int rating);
 
-/*
- * Interval: 0.5sec.
- */
-#define WATCHDOG_INTERVAL (HZ >> 1)
-
 static void clocksource_watchdog_work(struct work_struct *work)
 {
 	/*
@@ -257,8 +260,8 @@ static enum wd_read_status cs_watchdog_read(struct clocksource *cs, u64 *csnow,
 			goto skip_test;
 	}
 
-	pr_warn("timekeeping watchdog on CPU%d: %s read-back delay of %lldns, attempt %d, marking unstable\n",
-		smp_processor_id(), watchdog->name, wd_delay, nretries);
+	pr_warn("timekeeping watchdog on CPU%d: wd-%s-wd excessive read-back delay of %lldns vs. limit of %ldns, wd-wd read-back delay only %lldns, attempt %d, marking %s unstable\n",
+		smp_processor_id(), cs->name, wd_delay, WATCHDOG_MAX_SKEW, wd_seq_delay, nretries, cs->name);
 	return WD_READ_UNSTABLE;
 
 skip_test:
@@ -384,6 +387,15 @@ void clocksource_verify_percpu(struct clocksource *cs)
 }
 EXPORT_SYMBOL_GPL(clocksource_verify_percpu);
 
+static inline void clocksource_reset_watchdog(void)
+{
+	struct clocksource *cs;
+
+	list_for_each_entry(cs, &watchdog_list, wd_list)
+		cs->flags &= ~CLOCK_SOURCE_WATCHDOG;
+}
+
+
 static void clocksource_watchdog(struct timer_list *unused)
 {
 	u64 csnow, wdnow, cslast, wdlast, delta;
@@ -391,6 +403,7 @@ static void clocksource_watchdog(struct timer_list *unused)
 	int64_t wd_nsec, cs_nsec;
 	struct clocksource *cs;
 	enum wd_read_status read_ret;
+	unsigned long extra_wait = 0;
 	u32 md;
 
 	spin_lock(&watchdog_lock);
@@ -410,13 +423,30 @@ static void clocksource_watchdog(struct timer_list *unused)
 
 		read_ret = cs_watchdog_read(cs, &csnow, &wdnow);
 
-		if (read_ret != WD_READ_SUCCESS) {
-			if (read_ret == WD_READ_UNSTABLE)
+		if (read_ret == WD_READ_UNSTABLE) {
 			/* Clock readout unreliable, so give it up. */
 			__clocksource_unstable(cs);
 			continue;
 		}
 
+		/*
+		 * When WD_READ_SKIP is returned, it means the system is likely
+		 * under very heavy load, where the latency of reading
+		 * watchdog/clocksource is very big, and affect the accuracy of
+		 * watchdog check. So give system some space and suspend the
+		 * watchdog check for 5 minutes.
+		 */
+		if (read_ret == WD_READ_SKIP) {
+			/*
+			 * As the watchdog timer will be suspended, and
+			 * cs->last could keep unchanged for 5 minutes, reset
+			 * the counters.
+			 */
+			clocksource_reset_watchdog();
+			extra_wait = HZ * 300;
+			break;
+		}
+
 		/* Clocksource initialized ? */
 		if (!(cs->flags & CLOCK_SOURCE_WATCHDOG) ||
 		    atomic_read(&watchdog_reset_pending)) {
@@ -443,12 +473,20 @@ static void clocksource_watchdog(struct timer_list *unused)
 		/* Check the deviation from the watchdog clocksource. */
 		md = cs->uncertainty_margin + watchdog->uncertainty_margin;
 		if (abs(cs_nsec - wd_nsec) > md) {
+			u64 cs_wd_msec;
+			u64 wd_msec;
+			u32 wd_rem;
+
 			pr_warn("timekeeping watchdog on CPU%d: Marking clocksource '%s' as unstable because the skew is too large:\n",
 				smp_processor_id(), cs->name);
 			pr_warn("                      '%s' wd_nsec: %lld wd_now: %llx wd_last: %llx mask: %llx\n",
 				watchdog->name, wd_nsec, wdnow, wdlast, watchdog->mask);
 			pr_warn("                      '%s' cs_nsec: %lld cs_now: %llx cs_last: %llx mask: %llx\n",
 				cs->name, cs_nsec, csnow, cslast, cs->mask);
+			cs_wd_msec = div_u64_rem(cs_nsec - wd_nsec, 1000U * 1000U, &wd_rem);
+			wd_msec = div_u64_rem(wd_nsec, 1000U * 1000U, &wd_rem);
+			pr_warn("                      Clocksource '%s' skewed %lld ns (%lld ms) over watchdog '%s' interval of %lld ns (%lld ms)\n",
+				cs->name, cs_nsec - wd_nsec, cs_wd_msec, watchdog->name, wd_nsec, wd_msec);
 			if (curr_clocksource == cs)
 				pr_warn("                      '%s' is current clocksource.\n", cs->name);
 			else if (curr_clocksource)
@@ -512,7 +550,7 @@ static void clocksource_watchdog(struct timer_list *unused)
 	 * pair clocksource_stop_watchdog() clocksource_start_watchdog().
 	 */
 	if (!timer_pending(&watchdog_timer)) {
-		watchdog_timer.expires += WATCHDOG_INTERVAL;
+		watchdog_timer.expires += WATCHDOG_INTERVAL + extra_wait;
 		add_timer_on(&watchdog_timer, next_cpu);
 	}
 out:
@@ -537,14 +575,6 @@ static inline void clocksource_stop_watchdog(void)
 	watchdog_running = 0;
 }
 
-static inline void clocksource_reset_watchdog(void)
-{
-	struct clocksource *cs;
-
-	list_for_each_entry(cs, &watchdog_list, wd_list)
-		cs->flags &= ~CLOCK_SOURCE_WATCHDOG;
-}
-
 static void clocksource_resume_watchdog(void)
 {
 	atomic_inc(&watchdog_reset_pending);

kernel/time/hrtimer.c

@@ -2089,7 +2089,7 @@ long hrtimer_nanosleep(ktime_t rqtp, const enum hrtimer_mode mode,
 	u64 slack;
 
 	slack = current->timer_slack_ns;
-	if (dl_task(current) || rt_task(current))
+	if (rt_task(current))
 		slack = 0;
 
 	hrtimer_init_sleeper_on_stack(&t, clockid, mode);
@@ -2126,6 +2126,7 @@ SYSCALL_DEFINE2(nanosleep, struct __kernel_timespec __user *, rqtp,
 	if (!timespec64_valid(&tu))
 		return -EINVAL;
 
+	current->restart_block.fn = do_no_restart_syscall;
 	current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE;
 	current->restart_block.nanosleep.rmtp = rmtp;
 	return hrtimer_nanosleep(timespec64_to_ktime(tu), HRTIMER_MODE_REL,
@@ -2147,6 +2148,7 @@ SYSCALL_DEFINE2(nanosleep_time32, struct old_timespec32 __user *, rqtp,
 	if (!timespec64_valid(&tu))
 		return -EINVAL;
 
+	current->restart_block.fn = do_no_restart_syscall;
 	current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE;
 	current->restart_block.nanosleep.compat_rmtp = rmtp;
 	return hrtimer_nanosleep(timespec64_to_ktime(tu), HRTIMER_MODE_REL,
@@ -2270,7 +2272,7 @@ void __init hrtimers_init(void)
 /**
  * schedule_hrtimeout_range_clock - sleep until timeout
  * @expires:	timeout value (ktime_t)
- * @delta:	slack in expires timeout (ktime_t)
+ * @delta:	slack in expires timeout (ktime_t) for SCHED_OTHER tasks
  * @mode:	timer mode
  * @clock_id:	timer clock to be used
  */
@@ -2297,6 +2299,13 @@ schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta,
 		return -EINTR;
 	}
 
+	/*
+	 * Override any slack passed by the user if under
+	 * rt contraints.
+	 */
+	if (rt_task(current))
+		delta = 0;
+
 	hrtimer_init_sleeper_on_stack(&t, clock_id, mode);
 	hrtimer_set_expires_range_ns(&t.timer, *expires, delta);
 	hrtimer_sleeper_start_expires(&t, mode);
@@ -2316,7 +2325,7 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock);
 /**
  * schedule_hrtimeout_range - sleep until timeout
  * @expires:	timeout value (ktime_t)
- * @delta:	slack in expires timeout (ktime_t)
+ * @delta:	slack in expires timeout (ktime_t) for SCHED_OTHER tasks
  * @mode:	timer mode
  *
  * Make the current task sleep until the given expiry time has
@@ -2324,7 +2333,8 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock);
  * the current task state has been set (see set_current_state()).
  *
  * The @delta argument gives the kernel the freedom to schedule the
- * actual wakeup to a time that is both power and performance friendly.
+ * actual wakeup to a time that is both power and performance friendly
+ * for regular (non RT/DL) tasks.
  * The kernel give the normal best effort behavior for "@expires+@delta",
  * but may decide to fire the timer earlier, but no earlier than @expires.
  *

kernel/time/posix-cpu-timers.c

@@ -243,13 +243,12 @@ static void proc_sample_cputime_atomic(struct task_cputime_atomic *at,
  */
 static inline void __update_gt_cputime(atomic64_t *cputime, u64 sum_cputime)
 {
-	u64 curr_cputime;
-retry:
-	curr_cputime = atomic64_read(cputime);
-	if (sum_cputime > curr_cputime) {
-		if (atomic64_cmpxchg(cputime, curr_cputime, sum_cputime) != curr_cputime)
-			goto retry;
-	}
+	u64 curr_cputime = atomic64_read(cputime);
+
+	do {
+		if (sum_cputime <= curr_cputime)
+			return;
+	} while (!atomic64_try_cmpxchg(cputime, &curr_cputime, sum_cputime));
 }
 
 static void update_gt_cputime(struct task_cputime_atomic *cputime_atomic,

kernel/time/posix-stubs.c

@@ -147,6 +147,7 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags,
 		return -EINVAL;
 	if (flags & TIMER_ABSTIME)
 		rmtp = NULL;
+	current->restart_block.fn = do_no_restart_syscall;
 	current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE;
 	current->restart_block.nanosleep.rmtp = rmtp;
 	texp = timespec64_to_ktime(t);
@@ -240,6 +241,7 @@ SYSCALL_DEFINE4(clock_nanosleep_time32, clockid_t, which_clock, int, flags,
 		return -EINVAL;
 	if (flags & TIMER_ABSTIME)
 		rmtp = NULL;
+	current->restart_block.fn = do_no_restart_syscall;
 	current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE;
 	current->restart_block.nanosleep.compat_rmtp = rmtp;
 	texp = timespec64_to_ktime(t);

kernel/time/posix-timers.c

@@ -1270,6 +1270,7 @@ SYSCALL_DEFINE4(clock_nanosleep, const clockid_t, which_clock, int, flags,
 		return -EINVAL;
 	if (flags & TIMER_ABSTIME)
 		rmtp = NULL;
+	current->restart_block.fn = do_no_restart_syscall;
 	current->restart_block.nanosleep.type = rmtp ? TT_NATIVE : TT_NONE;
 	current->restart_block.nanosleep.rmtp = rmtp;
 
@@ -1297,6 +1298,7 @@ SYSCALL_DEFINE4(clock_nanosleep_time32, clockid_t, which_clock, int, flags,
 		return -EINVAL;
 	if (flags & TIMER_ABSTIME)
 		rmtp = NULL;
+	current->restart_block.fn = do_no_restart_syscall;
 	current->restart_block.nanosleep.type = rmtp ? TT_COMPAT : TT_NONE;
 	current->restart_block.nanosleep.compat_rmtp = rmtp;
 

kernel/time/test_udelay.c

@@ -149,7 +149,7 @@ module_init(udelay_test_init);
 static void __exit udelay_test_exit(void)
 {
 	mutex_lock(&udelay_test_lock);
-	debugfs_remove(debugfs_lookup(DEBUGFS_FILENAME, NULL));
+	debugfs_lookup_and_remove(DEBUGFS_FILENAME, NULL);
 	mutex_unlock(&udelay_test_lock);
 }