Merge tag 'x86-cleanups-2024-03-11' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 cleanups from Ingo Molnar:
 "Misc cleanups, including a large series from Thomas Gleixner to cure
  sparse warnings"

* tag 'x86-cleanups-2024-03-11' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/nmi: Drop unused declaration of proc_nmi_enabled()
  x86/callthunks: Use EXPORT_PER_CPU_SYMBOL_GPL() for per CPU variables
  x86/cpu: Provide a declaration for itlb_multihit_kvm_mitigation
  x86/cpu: Use EXPORT_PER_CPU_SYMBOL_GPL() for x86_spec_ctrl_current
  x86/uaccess: Add missing __force to casts in __access_ok() and valid_user_address()
  x86/percpu: Cure per CPU madness on UP
  smp: Consolidate smp_prepare_boot_cpu()
  x86/msr: Add missing __percpu annotations
  x86/msr: Prepare for including <linux/percpu.h> into <asm/msr.h>
  perf/x86/amd/uncore: Fix __percpu annotation
  x86/nmi: Remove an unnecessary IS_ENABLED(CONFIG_SMP)
  x86/apm_32: Remove dead function apm_get_battery_status()
  x86/insn-eval: Fix function param name in get_eff_addr_sib()

arch/alpha/kernel/smp.c

@@ -467,11 +467,6 @@ smp_prepare_cpus(unsigned int max_cpus)
 	smp_num_cpus = smp_num_probed;
 }
 
-void
-smp_prepare_boot_cpu(void)
-{
-}
-
 int
 __cpu_up(unsigned int cpu, struct task_struct *tidle)
 {

arch/arc/kernel/smp.c

@@ -39,11 +39,6 @@ struct plat_smp_ops  __weak plat_smp_ops;
 /* XXX: per cpu ? Only needed once in early secondary boot */
 struct task_struct *secondary_idle_tsk;
 
-/* Called from start_kernel */
-void __init smp_prepare_boot_cpu(void)
-{
-}
-
 static int __init arc_get_cpu_map(const char *name, struct cpumask *cpumask)
 {
 	unsigned long dt_root = of_get_flat_dt_root();

arch/csky/kernel/smp.c

@@ -152,10 +152,6 @@ void arch_irq_work_raise(void)
 }
 #endif
 
-void __init smp_prepare_boot_cpu(void)
-{
-}
-
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 }

arch/hexagon/kernel/smp.c

@@ -114,10 +114,6 @@ void send_ipi(const struct cpumask *cpumask, enum ipi_message_type msg)
 	local_irq_restore(flags);
 }
 
-void __init smp_prepare_boot_cpu(void)
-{
-}
-
 /*
  * interrupts should already be disabled from the VM
  * SP should already be correct; need to set THREADINFO_REG

arch/openrisc/kernel/smp.c

@@ -57,10 +57,6 @@ static void boot_secondary(unsigned int cpu, struct task_struct *idle)
 	spin_unlock(&boot_lock);
 }
 
-void __init smp_prepare_boot_cpu(void)
-{
-}
-
 void __init smp_init_cpus(void)
 {
 	struct device_node *cpu;

arch/riscv/kernel/smpboot.c

@@ -42,10 +42,6 @@
 
 static DECLARE_COMPLETION(cpu_running);
 
-void __init smp_prepare_boot_cpu(void)
-{
-}
-
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 	int cpuid;

arch/sparc/kernel/smp_64.c

@@ -1206,10 +1206,6 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 }
 
-void smp_prepare_boot_cpu(void)
-{
-}
-
 void __init smp_setup_processor_id(void)
 {
 	if (tlb_type == spitfire)

arch/x86/events/amd/uncore.c

@@ -71,7 +71,7 @@ union amd_uncore_info {
 };
 
 struct amd_uncore {
-	union amd_uncore_info * __percpu info;
+	union amd_uncore_info  __percpu *info;
 	struct amd_uncore_pmu *pmus;
 	unsigned int num_pmus;
 	bool init_done;

arch/x86/events/intel/core.c

@@ -17,6 +17,7 @@
 #include <linux/kvm_host.h>
 
 #include <asm/cpufeature.h>
+#include <asm/debugreg.h>
 #include <asm/hardirq.h>
 #include <asm/intel-family.h>
 #include <asm/intel_pt.h>

arch/x86/events/intel/ds.c

@@ -5,6 +5,7 @@
 #include <linux/sched/clock.h>
 
 #include <asm/cpu_entry_area.h>
+#include <asm/debugreg.h>
 #include <asm/perf_event.h>
 #include <asm/tlbflush.h>
 #include <asm/insn.h>

arch/x86/include/asm/debugreg.h

@@ -5,7 +5,9 @@
 #include <linux/bug.h>
 #include <linux/percpu.h>
 #include <uapi/asm/debugreg.h>
+
 #include <asm/cpufeature.h>
+#include <asm/msr.h>
 
 DECLARE_PER_CPU(unsigned long, cpu_dr7);
 
@@ -159,4 +161,26 @@ static inline unsigned long amd_get_dr_addr_mask(unsigned int dr)
 }
 #endif
 
+static inline unsigned long get_debugctlmsr(void)
+{
+	unsigned long debugctlmsr = 0;
+
+#ifndef CONFIG_X86_DEBUGCTLMSR
+	if (boot_cpu_data.x86 < 6)
+		return 0;
+#endif
+	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr);
+
+	return debugctlmsr;
+}
+
+static inline void update_debugctlmsr(unsigned long debugctlmsr)
+{
+#ifndef CONFIG_X86_DEBUGCTLMSR
+	if (boot_cpu_data.x86 < 6)
+		return;
+#endif
+	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr);
+}
+
 #endif /* _ASM_X86_DEBUGREG_H */

arch/x86/include/asm/fsgsbase.h

@@ -6,7 +6,7 @@
 
 #ifdef CONFIG_X86_64
 
-#include <asm/msr-index.h>
+#include <asm/msr.h>
 
 /*
  * Read/write a task's FSBASE or GSBASE. This returns the value that

arch/x86/include/asm/msr.h

@@ -12,11 +12,13 @@
 #include <uapi/asm/msr.h>
 #include <asm/shared/msr.h>
 
+#include <linux/percpu.h>
+
 struct msr_info {
-	u32 msr_no;
-	struct msr reg;
-	struct msr *msrs;
-	int err;
+	u32			msr_no;
+	struct msr		reg;
+	struct msr __percpu	*msrs;
+	int			err;
 };
 
 struct msr_regs_info {
@@ -323,8 +325,8 @@ static inline int wrmsrl_safe(u32 msr, u64 val)
 	return wrmsr_safe(msr, (u32)val,  (u32)(val >> 32));
 }
 
-struct msr *msrs_alloc(void);
-void msrs_free(struct msr *msrs);
+struct msr __percpu *msrs_alloc(void);
+void msrs_free(struct msr __percpu *msrs);
 int msr_set_bit(u32 msr, u8 bit);
 int msr_clear_bit(u32 msr, u8 bit);
 
@@ -333,8 +335,8 @@ int rdmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
 int wrmsr_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
 int rdmsrl_on_cpu(unsigned int cpu, u32 msr_no, u64 *q);
 int wrmsrl_on_cpu(unsigned int cpu, u32 msr_no, u64 q);
-void rdmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr *msrs);
-void wrmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr *msrs);
+void rdmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr __percpu *msrs);
+void wrmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr __percpu *msrs);
 int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 *l, u32 *h);
 int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h);
 int rdmsrl_safe_on_cpu(unsigned int cpu, u32 msr_no, u64 *q);
@@ -363,14 +365,14 @@ static inline int wrmsrl_on_cpu(unsigned int cpu, u32 msr_no, u64 q)
 	return 0;
 }
 static inline void rdmsr_on_cpus(const struct cpumask *m, u32 msr_no,
-				struct msr *msrs)
+				struct msr __percpu *msrs)
 {
-	rdmsr_on_cpu(0, msr_no, &(msrs[0].l), &(msrs[0].h));
+	rdmsr_on_cpu(0, msr_no, raw_cpu_ptr(&msrs->l), raw_cpu_ptr(&msrs->h));
 }
 static inline void wrmsr_on_cpus(const struct cpumask *m, u32 msr_no,
-				struct msr *msrs)
+				struct msr __percpu *msrs)
 {
-	wrmsr_on_cpu(0, msr_no, msrs[0].l, msrs[0].h);
+	wrmsr_on_cpu(0, msr_no, raw_cpu_read(msrs->l), raw_cpu_read(msrs->h));
 }
 static inline int rdmsr_safe_on_cpu(unsigned int cpu, u32 msr_no,
 				    u32 *l, u32 *h)

arch/x86/include/asm/nmi.h

@@ -14,9 +14,6 @@ extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
-struct ctl_table;
-extern int proc_nmi_enabled(struct ctl_table *, int ,
-			void __user *, size_t *, loff_t *);
 extern int unknown_nmi_panic;
 
 #endif /* CONFIG_X86_LOCAL_APIC */

arch/x86/include/asm/processor.h

@@ -20,7 +20,6 @@ struct vm86;
 #include <asm/page.h>
 #include <asm/pgtable_types.h>
 #include <asm/percpu.h>
-#include <asm/msr.h>
 #include <asm/desc_defs.h>
 #include <asm/nops.h>
 #include <asm/special_insns.h>
@@ -185,13 +184,8 @@ extern struct cpuinfo_x86	new_cpu_data;
 extern __u32			cpu_caps_cleared[NCAPINTS + NBUGINTS];
 extern __u32			cpu_caps_set[NCAPINTS + NBUGINTS];
 
-#ifdef CONFIG_SMP
 DECLARE_PER_CPU_READ_MOSTLY(struct cpuinfo_x86, cpu_info);
 #define cpu_data(cpu)		per_cpu(cpu_info, cpu)
-#else
-#define cpu_info		boot_cpu_data
-#define cpu_data(cpu)		boot_cpu_data
-#endif
 
 extern const struct seq_operations cpuinfo_op;
 
@@ -575,28 +569,6 @@ extern void cpu_init(void);
 extern void cpu_init_exception_handling(void);
 extern void cr4_init(void);
 
-static inline unsigned long get_debugctlmsr(void)
-{
-	unsigned long debugctlmsr = 0;
-
-#ifndef CONFIG_X86_DEBUGCTLMSR
-	if (boot_cpu_data.x86 < 6)
-		return 0;
-#endif
-	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr);
-
-	return debugctlmsr;
-}
-
-static inline void update_debugctlmsr(unsigned long debugctlmsr)
-{
-#ifndef CONFIG_X86_DEBUGCTLMSR
-	if (boot_cpu_data.x86 < 6)
-		return;
-#endif
-	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctlmsr);
-}
-
 extern void set_task_blockstep(struct task_struct *task, bool on);
 
 /* Boot loader type from the setup header: */

arch/x86/include/asm/smp.h

@@ -56,11 +56,6 @@ static inline void stop_other_cpus(void)
 	smp_ops.stop_other_cpus(1);
 }
 
-static inline void smp_prepare_boot_cpu(void)
-{
-	smp_ops.smp_prepare_boot_cpu();
-}
-
 static inline void smp_prepare_cpus(unsigned int max_cpus)
 {
 	smp_ops.smp_prepare_cpus(max_cpus);

arch/x86/include/asm/spec-ctrl.h

@@ -96,4 +96,6 @@ static inline void speculative_store_bypass_ht_init(void) { }
 extern void speculation_ctrl_update(unsigned long tif);
 extern void speculation_ctrl_update_current(void);
 
+extern bool itlb_multihit_kvm_mitigation;
+
 #endif

arch/x86/include/asm/special_insns.h

@@ -2,11 +2,11 @@
 #ifndef _ASM_X86_SPECIAL_INSNS_H
 #define _ASM_X86_SPECIAL_INSNS_H
 
-
 #ifdef __KERNEL__
-
 #include <asm/nops.h>
 #include <asm/processor-flags.h>
+
+#include <linux/errno.h>
 #include <linux/irqflags.h>
 #include <linux/jump_label.h>
 

arch/x86/include/asm/tsc.h

@@ -5,8 +5,9 @@
 #ifndef _ASM_X86_TSC_H
 #define _ASM_X86_TSC_H
 
-#include <asm/processor.h>
 #include <asm/cpufeature.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
 
 /*
  * Standard way to access the cycle counter.

arch/x86/include/asm/uaccess_64.h

@@ -54,7 +54,7 @@ static inline unsigned long __untagged_addr_remote(struct mm_struct *mm,
  * half and a user half.  When cast to a signed type, user pointers
  * are positive and kernel pointers are negative.
  */
-#define valid_user_address(x) ((long)(x) >= 0)
+#define valid_user_address(x) ((__force long)(x) >= 0)
 
 /*
  * User pointers can have tag bits on x86-64.  This scheme tolerates
@@ -87,8 +87,9 @@ static inline bool __access_ok(const void __user *ptr, unsigned long size)
 	if (__builtin_constant_p(size <= PAGE_SIZE) && size <= PAGE_SIZE) {
 		return valid_user_address(ptr);
 	} else {
-		unsigned long sum = size + (unsigned long)ptr;
-		return valid_user_address(sum) && sum >= (unsigned long)ptr;
+		unsigned long sum = size + (__force unsigned long)ptr;
+
+		return valid_user_address(sum) && sum >= (__force unsigned long)ptr;
 	}
 }
 #define __access_ok __access_ok

arch/x86/kernel/apm_32.c

@@ -1055,35 +1055,6 @@ static int apm_get_power_status(u_short *status, u_short *bat, u_short *life)
 	return APM_SUCCESS;
 }
 
-#if 0
-static int apm_get_battery_status(u_short which, u_short *status,
-				  u_short *bat, u_short *life, u_short *nbat)
-{
-	u32 eax;
-	u32 ebx;
-	u32 ecx;
-	u32 edx;
-	u32 esi;
-
-	if (apm_info.connection_version < 0x0102) {
-		/* pretend we only have one battery. */
-		if (which != 1)
-			return APM_BAD_DEVICE;
-		*nbat = 1;
-		return apm_get_power_status(status, bat, life);
-	}
-
-	if (apm_bios_call(APM_FUNC_GET_STATUS, (0x8000 | (which)), 0, &eax,
-			  &ebx, &ecx, &edx, &esi))
-		return (eax >> 8) & 0xff;
-	*status = ebx;
-	*bat = ecx;
-	*life = edx;
-	*nbat = esi;
-	return APM_SUCCESS;
-}
-#endif
-
 /**
  *	apm_engage_power_management	-	enable PM on a device
  *	@device: identity of device

arch/x86/kernel/callthunks.c

@@ -42,8 +42,8 @@ DEFINE_PER_CPU(u64, __x86_call_count);
 DEFINE_PER_CPU(u64, __x86_ret_count);
 DEFINE_PER_CPU(u64, __x86_stuffs_count);
 DEFINE_PER_CPU(u64, __x86_ctxsw_count);
-EXPORT_SYMBOL_GPL(__x86_ctxsw_count);
-EXPORT_SYMBOL_GPL(__x86_call_count);
+EXPORT_PER_CPU_SYMBOL_GPL(__x86_ctxsw_count);
+EXPORT_PER_CPU_SYMBOL_GPL(__x86_call_count);
 #endif
 
 extern s32 __call_sites[], __call_sites_end[];

arch/x86/kernel/cpu/bugs.c

@@ -56,7 +56,7 @@ EXPORT_SYMBOL_GPL(x86_spec_ctrl_base);
 
 /* The current value of the SPEC_CTRL MSR with task-specific bits set */
 DEFINE_PER_CPU(u64, x86_spec_ctrl_current);
-EXPORT_SYMBOL_GPL(x86_spec_ctrl_current);
+EXPORT_PER_CPU_SYMBOL_GPL(x86_spec_ctrl_current);
 
 u64 x86_pred_cmd __ro_after_init = PRED_CMD_IBPB;
 EXPORT_SYMBOL_GPL(x86_pred_cmd);

arch/x86/kernel/cpu/common.c

@@ -71,6 +71,9 @@
 
 #include "cpu.h"
 
+DEFINE_PER_CPU_READ_MOSTLY(struct cpuinfo_x86, cpu_info);
+EXPORT_PER_CPU_SYMBOL(cpu_info);
+
 u32 elf_hwcap2 __read_mostly;
 
 /* Number of siblings per CPU package */

arch/x86/kernel/cpu/intel_pconfig.c

@@ -7,6 +7,8 @@
  * Author:
  *	Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
  */
+#include <linux/bug.h>
+#include <linux/limits.h>
 
 #include <asm/cpufeature.h>
 #include <asm/intel_pconfig.h>

arch/x86/kernel/cpu/rdrand.c

@@ -6,6 +6,7 @@
  * Authors: Fenghua Yu <fenghua.yu@intel.com>,
  *          H. Peter Anvin <hpa@linux.intel.com>
  */
+#include <linux/printk.h>
 
 #include <asm/processor.h>
 #include <asm/archrandom.h>

arch/x86/kernel/fpu/bugs.c

@@ -2,6 +2,8 @@
 /*
  * x86 FPU bug checks:
  */
+#include <linux/printk.h>
+
 #include <asm/cpufeature.h>
 #include <asm/fpu/api.h>
 

arch/x86/kernel/nmi.c

@@ -503,7 +503,7 @@ DEFINE_IDTENTRY_RAW(exc_nmi)
 	if (IS_ENABLED(CONFIG_NMI_CHECK_CPU))
 		raw_atomic_long_inc(&nsp->idt_calls);
 
-	if (IS_ENABLED(CONFIG_SMP) && arch_cpu_is_offline(smp_processor_id())) {
+	if (arch_cpu_is_offline(smp_processor_id())) {
 		if (microcode_nmi_handler_enabled())
 			microcode_offline_nmi_handler();
 		return;

arch/x86/kernel/setup.c

@@ -1206,6 +1206,16 @@ void __init i386_reserve_resources(void)
 
 #endif /* CONFIG_X86_32 */
 
+#ifndef CONFIG_SMP
+void __init smp_prepare_boot_cpu(void)
+{
+	struct cpuinfo_x86 *c = &cpu_data(0);
+
+	*c = boot_cpu_data;
+	c->initialized = true;
+}
+#endif
+
 static struct notifier_block kernel_offset_notifier = {
 	.notifier_call = dump_kernel_offset
 };

arch/x86/kernel/smpboot.c

@@ -101,10 +101,6 @@ EXPORT_PER_CPU_SYMBOL(cpu_core_map);
 DEFINE_PER_CPU_READ_MOSTLY(cpumask_var_t, cpu_die_map);
 EXPORT_PER_CPU_SYMBOL(cpu_die_map);
 
-/* Per CPU bogomips and other parameters */
-DEFINE_PER_CPU_READ_MOSTLY(struct cpuinfo_x86, cpu_info);
-EXPORT_PER_CPU_SYMBOL(cpu_info);
-
 /* CPUs which are the primary SMT threads */
 struct cpumask __cpu_primary_thread_mask __read_mostly;
 
@@ -1078,6 +1074,11 @@ void __init smp_prepare_cpus_common(void)
 	set_cpu_sibling_map(0);
 }
 
+void __init smp_prepare_boot_cpu(void)
+{
+	smp_ops.smp_prepare_boot_cpu();
+}
+
 #ifdef CONFIG_X86_64
 /* Establish whether parallel bringup can be supported. */
 bool __init arch_cpuhp_init_parallel_bringup(void)

arch/x86/kernel/step.c

@@ -6,7 +6,9 @@
 #include <linux/sched/task_stack.h>
 #include <linux/mm.h>
 #include <linux/ptrace.h>
+
 #include <asm/desc.h>
+#include <asm/debugreg.h>
 #include <asm/mmu_context.h>
 
 unsigned long convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs)

arch/x86/kvm/mmu/mmu.c

@@ -53,12 +53,11 @@
 #include <asm/cmpxchg.h>
 #include <asm/io.h>
 #include <asm/set_memory.h>
+#include <asm/spec-ctrl.h>
 #include <asm/vmx.h>
 
 #include "trace.h"
 
-extern bool itlb_multihit_kvm_mitigation;
-
 static bool nx_hugepage_mitigation_hard_disabled;
 
 int __read_mostly nx_huge_pages = -1;

arch/x86/lib/insn-eval.c

@@ -1129,15 +1129,15 @@ static int get_eff_addr_modrm_16(struct insn *insn, struct pt_regs *regs,
  * get_eff_addr_sib() - Obtain referenced effective address via SIB
  * @insn:	Instruction. Must be valid.
  * @regs:	Register values as seen when entering kernel mode
- * @regoff:	Obtained operand offset, in pt_regs, associated with segment
+ * @base_offset: Obtained operand offset, in pt_regs, associated with segment
  * @eff_addr:	Obtained effective address
  *
  * Obtain the effective address referenced by the SIB byte of @insn. After
  * identifying the registers involved in the indexed, register-indirect memory
  * reference, its value is obtained from the operands in @regs. The computed
  * address is stored @eff_addr. Also, the register operand that indicates the
- * associated segment is stored in @regoff, this parameter can later be used to
- * determine such segment.
+ * associated segment is stored in @base_offset; this parameter can later be
+ * used to determine such segment.
  *
  * Returns:
  *

arch/x86/lib/msr-smp.c

@@ -9,10 +9,9 @@ static void __rdmsr_on_cpu(void *info)
 {
 	struct msr_info *rv = info;
 	struct msr *reg;
-	int this_cpu = raw_smp_processor_id();
 
 	if (rv->msrs)
-		reg = per_cpu_ptr(rv->msrs, this_cpu);
+		reg = this_cpu_ptr(rv->msrs);
 	else
 		reg = &rv->reg;
 
@@ -23,10 +22,9 @@ static void __wrmsr_on_cpu(void *info)
 {
 	struct msr_info *rv = info;
 	struct msr *reg;
-	int this_cpu = raw_smp_processor_id();
 
 	if (rv->msrs)
-		reg = per_cpu_ptr(rv->msrs, this_cpu);
+		reg = this_cpu_ptr(rv->msrs);
 	else
 		reg = &rv->reg;
 
@@ -97,7 +95,7 @@ int wrmsrl_on_cpu(unsigned int cpu, u32 msr_no, u64 q)
 EXPORT_SYMBOL(wrmsrl_on_cpu);
 
 static void __rwmsr_on_cpus(const struct cpumask *mask, u32 msr_no,
-			    struct msr *msrs,
+			    struct msr __percpu *msrs,
 			    void (*msr_func) (void *info))
 {
 	struct msr_info rv;
@@ -124,7 +122,7 @@ static void __rwmsr_on_cpus(const struct cpumask *mask, u32 msr_no,
  * @msrs:       array of MSR values
  *
  */
-void rdmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr *msrs)
+void rdmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr __percpu *msrs)
 {
 	__rwmsr_on_cpus(mask, msr_no, msrs, __rdmsr_on_cpu);
 }
@@ -138,7 +136,7 @@ EXPORT_SYMBOL(rdmsr_on_cpus);
  * @msrs:       array of MSR values
  *
  */
-void wrmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr *msrs)
+void wrmsr_on_cpus(const struct cpumask *mask, u32 msr_no, struct msr __percpu *msrs)
 {
 	__rwmsr_on_cpus(mask, msr_no, msrs, __wrmsr_on_cpu);
 }

arch/x86/lib/msr.c

@@ -6,9 +6,9 @@
 #define CREATE_TRACE_POINTS
 #include <asm/msr-trace.h>
 
-struct msr *msrs_alloc(void)
+struct msr __percpu *msrs_alloc(void)
 {
-	struct msr *msrs = NULL;
+	struct msr __percpu *msrs = NULL;
 
 	msrs = alloc_percpu(struct msr);
 	if (!msrs) {
@@ -20,7 +20,7 @@ struct msr *msrs_alloc(void)
 }
 EXPORT_SYMBOL(msrs_alloc);
 
-void msrs_free(struct msr *msrs)
+void msrs_free(struct msr __percpu *msrs)
 {
 	free_percpu(msrs);
 }

include/linux/smp.h

@@ -105,6 +105,12 @@ static inline void on_each_cpu_cond(smp_cond_func_t cond_func,
 	on_each_cpu_cond_mask(cond_func, func, info, wait, cpu_online_mask);
 }
 
+/*
+ * Architecture specific boot CPU setup.  Defined as empty weak function in
+ * init/main.c. Architectures can override it.
+ */
+void smp_prepare_boot_cpu(void);
+
 #ifdef CONFIG_SMP
 
 #include <linux/preempt.h>
@@ -171,12 +177,6 @@ void generic_smp_call_function_single_interrupt(void);
 #define generic_smp_call_function_interrupt \
 	generic_smp_call_function_single_interrupt
 
-/*
- * Mark the boot cpu "online" so that it can call console drivers in
- * printk() and can access its per-cpu storage.
- */
-void smp_prepare_boot_cpu(void);
-
 extern unsigned int setup_max_cpus;
 extern void __init setup_nr_cpu_ids(void);
 extern void __init smp_init(void);
@@ -203,7 +203,6 @@ static inline void up_smp_call_function(smp_call_func_t func, void *info)
 			(up_smp_call_function(func, info))
 
 static inline void smp_send_reschedule(int cpu) { }
-#define smp_prepare_boot_cpu()			do {} while (0)
 #define smp_call_function_many(mask, func, info, wait) \
 			(up_smp_call_function(func, info))
 static inline void call_function_init(void) { }

init/main.c

@@ -776,6 +776,10 @@ void __init __weak smp_setup_processor_id(void)
 {
 }
 
+void __init __weak smp_prepare_boot_cpu(void)
+{
+}
+
 # if THREAD_SIZE >= PAGE_SIZE
 void __init __weak thread_stack_cache_init(void)
 {