Merge tag 'smp-core-2023-06-26' of ssh://gitolite.kernel.org/pub/scm/…

…linux/kernel/git/tip/tip

Pull SMP updates from Thomas Gleixner:
 "A large update for SMP management:

   - Parallel CPU bringup

     The reason why people are interested in parallel bringup is to
     shorten the (kexec) reboot time of cloud servers to reduce the
     downtime of the VM tenants.

     The current fully serialized bringup does the following per AP:

       1) Prepare callbacks (allocate, intialize, create threads)
       2) Kick the AP alive (e.g. INIT/SIPI on x86)
       3) Wait for the AP to report alive state
       4) Let the AP continue through the atomic bringup
       5) Let the AP run the threaded bringup to full online state

     There are two significant delays:

       #3 The time for an AP to report alive state in start_secondary()
          on x86 has been measured in the range between 350us and 3.5ms
          depending on vendor and CPU type, BIOS microcode size etc.

       #4 The atomic bringup does the microcode update. This has been
          measured to take up to ~8ms on the primary threads depending
          on the microcode patch size to apply.

     On a two socket SKL server with 56 cores (112 threads) the boot CPU
     spends on current mainline about 800ms busy waiting for the APs to
     come up and apply microcode. That's more than 80% of the actual
     onlining procedure.

     This can be reduced significantly by splitting the bringup
     mechanism into two parts:

       1) Run the prepare callbacks and kick the AP alive for each AP
          which needs to be brought up.

          The APs wake up, do their firmware initialization and run the
          low level kernel startup code including microcode loading in
          parallel up to the first synchronization point. (#1 and #2
          above)

       2) Run the rest of the bringup code strictly serialized per CPU
          (#3 - #5 above) as it's done today.

          Parallelizing that stage of the CPU bringup might be possible
          in theory, but it's questionable whether required surgery
          would be justified for a pretty small gain.

     If the system is large enough the first AP is already waiting at
     the first synchronization point when the boot CPU finished the
     wake-up of the last AP. That reduces the AP bringup time on that
     SKL from ~800ms to ~80ms, i.e. by a factor ~10x.

     The actual gain varies wildly depending on the system, CPU,
     microcode patch size and other factors. There are some
     opportunities to reduce the overhead further, but that needs some
     deep surgery in the x86 CPU bringup code.

     For now this is only enabled on x86, but the core functionality
     obviously works for all SMP capable architectures.

   - Enhancements for SMP function call tracing so it is possible to
     locate the scheduling and the actual execution points. That allows
     to measure IPI delivery time precisely"

* tag 'smp-core-2023-06-26' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/tip/tip: (45 commits)
  trace,smp: Add tracepoints for scheduling remotelly called functions
  trace,smp: Add tracepoints around remotelly called functions
  MAINTAINERS: Add CPU HOTPLUG entry
  x86/smpboot: Fix the parallel bringup decision
  x86/realmode: Make stack lock work in trampoline_compat()
  x86/smp: Initialize cpu_primary_thread_mask late
  cpu/hotplug: Fix off by one in cpuhp_bringup_mask()
  x86/apic: Fix use of X{,2}APIC_ENABLE in asm with older binutils
  x86/smpboot/64: Implement arch_cpuhp_init_parallel_bringup() and enable it
  x86/smpboot: Support parallel startup of secondary CPUs
  x86/smpboot: Implement a bit spinlock to protect the realmode stack
  x86/apic: Save the APIC virtual base address
  cpu/hotplug: Allow "parallel" bringup up to CPUHP_BP_KICK_AP_STATE
  x86/apic: Provide cpu_primary_thread mask
  x86/smpboot: Enable split CPU startup
  cpu/hotplug: Provide a split up CPUHP_BRINGUP mechanism
  cpu/hotplug: Reset task stack state in _cpu_up()
  cpu/hotplug: Remove unused state functions
  riscv: Switch to hotplug core state synchronization
  parisc: Switch to hotplug core state synchronization
  ...

Documentation/admin-guide/kernel-parameters.txt

@@ -818,20 +818,6 @@
 			Format:
 			<first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]
 
-	cpu0_hotplug	[X86] Turn on CPU0 hotplug feature when
-			CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off.
-			Some features depend on CPU0. Known dependencies are:
-			1. Resume from suspend/hibernate depends on CPU0.
-			Suspend/hibernate will fail if CPU0 is offline and you
-			need to online CPU0 before suspend/hibernate.
-			2. PIC interrupts also depend on CPU0. CPU0 can't be
-			removed if a PIC interrupt is detected.
-			It's said poweroff/reboot may depend on CPU0 on some
-			machines although I haven't seen such issues so far
-			after CPU0 is offline on a few tested machines.
-			If the dependencies are under your control, you can
-			turn on cpu0_hotplug.
-
 	cpuidle.off=1	[CPU_IDLE]
 			disable the cpuidle sub-system
 
@@ -852,6 +838,12 @@
 			on every CPU online, such as boot, and resume from suspend.
 			Default: 10000
 
+	cpuhp.parallel=
+			[SMP] Enable/disable parallel bringup of secondary CPUs
+			Format: <bool>
+			Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise
+			the parameter has no effect.
+
 	crash_kexec_post_notifiers
 			Run kdump after running panic-notifiers and dumping
 			kmsg. This only for the users who doubt kdump always

Documentation/core-api/cpu_hotplug.rst

@@ -127,17 +127,8 @@ bring CPU4 back online::
  $ echo 1 > /sys/devices/system/cpu/cpu4/online
  smpboot: Booting Node 0 Processor 4 APIC 0x1
 
-The CPU is usable again. This should work on all CPUs. CPU0 is often special
-and excluded from CPU hotplug. On X86 the kernel option
-*CONFIG_BOOTPARAM_HOTPLUG_CPU0* has to be enabled in order to be able to
-shutdown CPU0. Alternatively the kernel command option *cpu0_hotplug* can be
-used. Some known dependencies of CPU0:
-
-* Resume from hibernate/suspend. Hibernate/suspend will fail if CPU0 is offline.
-* PIC interrupts. CPU0 can't be removed if a PIC interrupt is detected.
-
-Please let Fenghua Yu <[email protected]> know if you find any dependencies
-on CPU0.
+The CPU is usable again. This should work on all CPUs, but CPU0 is often special
+and excluded from CPU hotplug.
 
 The CPU hotplug coordination
 ============================

MAINTAINERS

@@ -5344,6 +5344,18 @@ F:	include/linux/sched/cpufreq.h
 F:	kernel/sched/cpufreq*.c
 F:	tools/testing/selftests/cpufreq/
 
+CPU HOTPLUG
+M:	Thomas Gleixner <[email protected]>
+M:	Peter Zijlstra <[email protected]>
+L:	[email protected]
+S:	Maintained
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git smp/core
+F:	kernel/cpu.c
+F:	kernel/smpboot.*
+F:	include/linux/cpu.h
+F:	include/linux/cpuhotplug.h
+F:	include/linux/smpboot.h
+
 CPU IDLE TIME MANAGEMENT FRAMEWORK
 M:	"Rafael J. Wysocki" <[email protected]>
 M:	Daniel Lezcano <[email protected]>

arch/Kconfig

@@ -34,6 +34,29 @@ config ARCH_HAS_SUBPAGE_FAULTS
 config HOTPLUG_SMT
 	bool
 
+# Selected by HOTPLUG_CORE_SYNC_DEAD or HOTPLUG_CORE_SYNC_FULL
+config HOTPLUG_CORE_SYNC
+	bool
+
+# Basic CPU dead synchronization selected by architecture
+config HOTPLUG_CORE_SYNC_DEAD
+	bool
+	select HOTPLUG_CORE_SYNC
+
+# Full CPU synchronization with alive state selected by architecture
+config HOTPLUG_CORE_SYNC_FULL
+	bool
+	select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
+	select HOTPLUG_CORE_SYNC
+
+config HOTPLUG_SPLIT_STARTUP
+	bool
+	select HOTPLUG_CORE_SYNC_FULL
+
+config HOTPLUG_PARALLEL
+	bool
+	select HOTPLUG_SPLIT_STARTUP
+
 config GENERIC_ENTRY
 	bool
 

arch/arm/Kconfig

@@ -125,6 +125,7 @@ config ARM
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_UID16
 	select HAVE_VIRT_CPU_ACCOUNTING_GEN
+	select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
 	select IRQ_FORCED_THREADING
 	select MODULES_USE_ELF_REL
 	select NEED_DMA_MAP_STATE

arch/arm/include/asm/smp.h

@@ -64,7 +64,7 @@ extern void secondary_startup_arm(void);
 
 extern int __cpu_disable(void);
 
-extern void __cpu_die(unsigned int cpu);
+static inline void __cpu_die(unsigned int cpu) { }
 
 extern void arch_send_call_function_single_ipi(int cpu);
 extern void arch_send_call_function_ipi_mask(const struct cpumask *mask);

arch/arm/kernel/smp.c

@@ -288,15 +288,11 @@ int __cpu_disable(void)
 }
 
 /*
- * called on the thread which is asking for a CPU to be shutdown -
- * waits until shutdown has completed, or it is timed out.
+ * called on the thread which is asking for a CPU to be shutdown after the
+ * shutdown completed.
  */
-void __cpu_die(unsigned int cpu)
+void arch_cpuhp_cleanup_dead_cpu(unsigned int cpu)
 {
-	if (!cpu_wait_death(cpu, 5)) {
-		pr_err("CPU%u: cpu didn't die\n", cpu);
-		return;
-	}
 	pr_debug("CPU%u: shutdown\n", cpu);
 
 	clear_tasks_mm_cpumask(cpu);
@@ -336,11 +332,11 @@ void __noreturn arch_cpu_idle_dead(void)
 	flush_cache_louis();
 
 	/*
-	 * Tell __cpu_die() that this CPU is now safe to dispose of.  Once
-	 * this returns, power and/or clocks can be removed at any point
-	 * from this CPU and its cache by platform_cpu_kill().
+	 * Tell cpuhp_bp_sync_dead() that this CPU is now safe to dispose
+	 * of. Once this returns, power and/or clocks can be removed at
+	 * any point from this CPU and its cache by platform_cpu_kill().
 	 */
-	(void)cpu_report_death();
+	cpuhp_ap_report_dead();
 
 	/*
 	 * Ensure that the cache lines associated with that completion are

arch/arm64/Kconfig

@@ -222,6 +222,7 @@ config ARM64
 	select HAVE_KPROBES
 	select HAVE_KRETPROBES
 	select HAVE_GENERIC_VDSO
+	select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
 	select IRQ_DOMAIN
 	select IRQ_FORCED_THREADING
 	select KASAN_VMALLOC if KASAN

arch/arm64/include/asm/smp.h

@@ -99,7 +99,7 @@ static inline void arch_send_wakeup_ipi_mask(const struct cpumask *mask)
 
 extern int __cpu_disable(void);
 
-extern void __cpu_die(unsigned int cpu);
+static inline void __cpu_die(unsigned int cpu) { }
 extern void __noreturn cpu_die(void);
 extern void __noreturn cpu_die_early(void);
 

arch/arm64/kernel/smp.c

@@ -332,17 +332,13 @@ static int op_cpu_kill(unsigned int cpu)
 }
 
 /*
- * called on the thread which is asking for a CPU to be shutdown -
- * waits until shutdown has completed, or it is timed out.
+ * Called on the thread which is asking for a CPU to be shutdown after the
+ * shutdown completed.
  */
-void __cpu_die(unsigned int cpu)
+void arch_cpuhp_cleanup_dead_cpu(unsigned int cpu)
 {
 	int err;
 
-	if (!cpu_wait_death(cpu, 5)) {
-		pr_crit("CPU%u: cpu didn't die\n", cpu);
-		return;
-	}
 	pr_debug("CPU%u: shutdown\n", cpu);
 
 	/*
@@ -369,8 +365,8 @@ void __noreturn cpu_die(void)
 
 	local_daif_mask();
 
-	/* Tell __cpu_die() that this CPU is now safe to dispose of */
-	(void)cpu_report_death();
+	/* Tell cpuhp_bp_sync_dead() that this CPU is now safe to dispose of */
+	cpuhp_ap_report_dead();
 
 	/*
 	 * Actually shutdown the CPU. This must never fail. The specific hotplug

arch/csky/Kconfig

@@ -96,6 +96,7 @@ config CSKY
 	select HAVE_REGS_AND_STACK_ACCESS_API
 	select HAVE_STACKPROTECTOR
 	select HAVE_SYSCALL_TRACEPOINTS
+	select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
 	select MAY_HAVE_SPARSE_IRQ
 	select MODULES_USE_ELF_RELA if MODULES
 	select OF

arch/csky/include/asm/smp.h

@@ -23,7 +23,7 @@ void __init set_send_ipi(void (*func)(const struct cpumask *mask), int irq);
 
 int __cpu_disable(void);
 
-void __cpu_die(unsigned int cpu);
+static inline void __cpu_die(unsigned int cpu) { }
 
 #endif /* CONFIG_SMP */
 

arch/csky/kernel/smp.c

@@ -291,20 +291,16 @@ int __cpu_disable(void)
 	return 0;
 }
 
-void __cpu_die(unsigned int cpu)
+void arch_cpuhp_cleanup_dead_cpu(unsigned int cpu)
 {
-	if (!cpu_wait_death(cpu, 5)) {
-		pr_crit("CPU%u: shutdown failed\n", cpu);
-		return;
-	}
 	pr_notice("CPU%u: shutdown\n", cpu);
 }
 
 void __noreturn arch_cpu_idle_dead(void)
 {
 	idle_task_exit();
 
-	cpu_report_death();
+	cpuhp_ap_report_dead();
 
 	while (!secondary_stack)
 		arch_cpu_idle();

arch/mips/Kconfig

@@ -2287,6 +2287,7 @@ config MIPS_CPS
 	select MIPS_CM
 	select MIPS_CPS_PM if HOTPLUG_CPU
 	select SMP
+	select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
 	select SYNC_R4K if (CEVT_R4K || CSRC_R4K)
 	select SYS_SUPPORTS_HOTPLUG_CPU
 	select SYS_SUPPORTS_SCHED_SMT if CPU_MIPSR6

arch/mips/cavium-octeon/smp.c

@@ -345,6 +345,7 @@ void play_dead(void)
 	int cpu = cpu_number_map(cvmx_get_core_num());
 
 	idle_task_exit();
+	cpuhp_ap_report_dead();
 	octeon_processor_boot = 0xff;
 	per_cpu(cpu_state, cpu) = CPU_DEAD;
 

arch/mips/include/asm/smp-ops.h

@@ -33,6 +33,7 @@ struct plat_smp_ops {
 #ifdef CONFIG_HOTPLUG_CPU
 	int (*cpu_disable)(void);
 	void (*cpu_die)(unsigned int cpu);
+	void (*cleanup_dead_cpu)(unsigned cpu);
 #endif
 #ifdef CONFIG_KEXEC
 	void (*kexec_nonboot_cpu)(void);

arch/mips/kernel/smp-bmips.c

@@ -392,6 +392,7 @@ static void bmips_cpu_die(unsigned int cpu)
 void __ref play_dead(void)
 {
 	idle_task_exit();
+	cpuhp_ap_report_dead();
 
 	/* flush data cache */
 	_dma_cache_wback_inv(0, ~0);

arch/mips/kernel/smp-cps.c

@@ -503,8 +503,7 @@ void play_dead(void)
 		}
 	}
 
-	/* This CPU has chosen its way out */
-	(void)cpu_report_death();
+	cpuhp_ap_report_dead();
 
 	cps_shutdown_this_cpu(cpu_death);
 
@@ -527,20 +526,16 @@ static void wait_for_sibling_halt(void *ptr_cpu)
 	} while (!(halted & TCHALT_H));
 }
 
-static void cps_cpu_die(unsigned int cpu)
+static void cps_cpu_die(unsigned int cpu) { }
+
+static void cps_cleanup_dead_cpu(unsigned cpu)
 {
 	unsigned core = cpu_core(&cpu_data[cpu]);
 	unsigned int vpe_id = cpu_vpe_id(&cpu_data[cpu]);
 	ktime_t fail_time;
 	unsigned stat;
 	int err;
 
-	/* Wait for the cpu to choose its way out */
-	if (!cpu_wait_death(cpu, 5)) {
-		pr_err("CPU%u: didn't offline\n", cpu);
-		return;
-	}
-
 	/*
 	 * Now wait for the CPU to actually offline. Without doing this that
 	 * offlining may race with one or more of:
@@ -624,6 +619,7 @@ static const struct plat_smp_ops cps_smp_ops = {
 #ifdef CONFIG_HOTPLUG_CPU
 	.cpu_disable		= cps_cpu_disable,
 	.cpu_die		= cps_cpu_die,
+	.cleanup_dead_cpu	= cps_cleanup_dead_cpu,
 #endif
 #ifdef CONFIG_KEXEC
 	.kexec_nonboot_cpu	= cps_kexec_nonboot_cpu,

arch/mips/kernel/smp.c

@@ -690,6 +690,14 @@ void flush_tlb_one(unsigned long vaddr)
 EXPORT_SYMBOL(flush_tlb_page);
 EXPORT_SYMBOL(flush_tlb_one);
 
+#ifdef CONFIG_HOTPLUG_CORE_SYNC_DEAD
+void arch_cpuhp_cleanup_dead_cpu(unsigned int cpu)
+{
+	if (mp_ops->cleanup_dead_cpu)
+		mp_ops->cleanup_dead_cpu(cpu);
+}
+#endif
+
 #ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
 
 static void tick_broadcast_callee(void *info)

arch/mips/loongson64/smp.c

@@ -775,6 +775,7 @@ void play_dead(void)
 	void (*play_dead_at_ckseg1)(int *);
 
 	idle_task_exit();
+	cpuhp_ap_report_dead();
 
 	prid_imp = read_c0_prid() & PRID_IMP_MASK;
 	prid_rev = read_c0_prid() & PRID_REV_MASK;

arch/parisc/Kconfig

@@ -57,6 +57,7 @@ config PARISC
 	select HAVE_ARCH_SECCOMP_FILTER
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_REGS_AND_STACK_ACCESS_API
+	select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
 	select GENERIC_SCHED_CLOCK
 	select GENERIC_IRQ_MIGRATION if SMP
 	select HAVE_UNSTABLE_SCHED_CLOCK if SMP

arch/parisc/kernel/process.c

@@ -171,8 +171,8 @@ void __noreturn arch_cpu_idle_dead(void)
 
 	local_irq_disable();
 
-	/* Tell __cpu_die() that this CPU is now safe to dispose of. */
-	(void)cpu_report_death();
+	/* Tell the core that this CPU is now safe to dispose of. */
+	cpuhp_ap_report_dead();
 
 	/* Ensure that the cache lines are written out. */
 	flush_cache_all_local();

arch/parisc/kernel/smp.c

@@ -500,11 +500,10 @@ int __cpu_disable(void)
 void __cpu_die(unsigned int cpu)
 {
 	pdc_cpu_rendezvous_lock();
+}
 
-	if (!cpu_wait_death(cpu, 5)) {
-		pr_crit("CPU%u: cpu didn't die\n", cpu);
-		return;
-	}
+void arch_cpuhp_cleanup_dead_cpu(unsigned int cpu)
+{
 	pr_info("CPU%u: is shutting down\n", cpu);
 
 	/* set task's state to interruptible sleep */

arch/riscv/Kconfig

@@ -123,6 +123,7 @@ config RISCV
 	select HAVE_RSEQ
 	select HAVE_STACKPROTECTOR
 	select HAVE_SYSCALL_TRACEPOINTS
+	select HOTPLUG_CORE_SYNC_DEAD if HOTPLUG_CPU
 	select IRQ_DOMAIN
 	select IRQ_FORCED_THREADING
 	select KASAN_VMALLOC if KASAN

arch/riscv/include/asm/smp.h

@@ -70,7 +70,7 @@ asmlinkage void smp_callin(void);
 
 #if defined CONFIG_HOTPLUG_CPU
 int __cpu_disable(void);
-void __cpu_die(unsigned int cpu);
+static inline void __cpu_die(unsigned int cpu) { }
 #endif /* CONFIG_HOTPLUG_CPU */
 
 #else