@@ -8831,6 +8831,8 @@ F:	kernel/sched/membarrier.c
 F:	include/uapi/linux/membarrier.h
 F:	arch/powerpc/kernel/membarrier.c
 F:	arch/powerpc/include/asm/membarrier.h
+F:	arch/x86/kernel/membarrier.c
+F:	arch/x86/include/asm/membarrier.h
 
 MEMORY MANAGEMENT
 L:	linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org

@@ -21,6 +21,12 @@ static inline void membarrier_arch_switch_mm(struct mm_struct *prev,
 	 */
 	smp_mb();
 }
-void membarrier_arch_register_private_expedited(struct task_struct *t);
+
+static inline void membarrier_arch_mm_sync_core(void)
+{
+}
+
+void membarrier_arch_register_private_expedited(struct task_struct *t,
+		int flags);
 
 #endif /* _ASM_POWERPC_MEMBARRIER_H */

@@ -21,7 +21,8 @@
 #include <linux/rcupdate.h>
 #include <linux/atomic.h>
 
-void membarrier_arch_register_private_expedited(struct task_struct *p)
+void membarrier_arch_register_private_expedited(struct task_struct *p,
+		int flags)
 {
 	struct mm_struct *mm = p->mm;
 

@@ -54,6 +54,8 @@ config X86
 	select ARCH_HAS_FORTIFY_SOURCE
 	select ARCH_HAS_GCOV_PROFILE_ALL
 	select ARCH_HAS_KCOV			if X86_64
+	select ARCH_HAS_MEMBARRIER_HOOKS
+	select ARCH_HAS_MEMBARRIER_SYNC_CORE
 	select ARCH_HAS_PMEM_API		if X86_64
 	# Causing hangs/crashes, see the commit that added this change for details.
 	select ARCH_HAS_REFCOUNT		if BROKEN

@@ -553,6 +553,11 @@ restore_all:
 .Lrestore_nocheck:
 	RESTORE_REGS 4				# skip orig_eax/error_code
 .Lirq_return:
+	/*
+	 * ARCH_HAS_MEMBARRIER_SYNC_CORE rely on iret core serialization
+	 * when returning from IPI handler and when returning from
+	 * scheduler to user-space.
+	 */
 	INTERRUPT_RETURN
 
 .section .fixup, "ax"

@@ -642,6 +642,10 @@ GLOBAL(restore_regs_and_iret)
 restore_c_regs_and_iret:
 	RESTORE_C_REGS
 	REMOVE_PT_GPREGS_FROM_STACK 8
+	/*
+	 * ARCH_HAS_MEMBARRIER_SYNC_CORE rely on iret core serialization
+	 * when returning from IPI handler.
+	 */
 	INTERRUPT_RETURN
 
 ENTRY(native_iret)

@@ -1122,6 +1126,10 @@ paranoid_exit_restore:
 	RESTORE_EXTRA_REGS
 	RESTORE_C_REGS
 	REMOVE_PT_GPREGS_FROM_STACK 8
+	/*
+	 * ARCH_HAS_MEMBARRIER_SYNC_CORE rely on iret core serialization
+	 * when returning from IPI handler.
+	 */
 	INTERRUPT_RETURN
 END(paranoid_exit)
 

@@ -0,0 +1,36 @@
+#ifndef _ASM_X86_MEMBARRIER_H
+#define _ASM_X86_MEMBARRIER_H
+
+#include <asm/processor.h>
+
+static inline void membarrier_arch_switch_mm(struct mm_struct *prev,
+		struct mm_struct *next, struct task_struct *tsk)
+{
+}
+
+#ifdef CONFIG_X86_32
+static inline void membarrier_arch_mm_sync_core(struct mm_struct *mm)
+{
+}
+static inline
+void membarrier_arch_register_private_expedited(struct task_struct *t,
+		int flags);
+#else
+/*
+ * x86-64 implements return to user-space through sysret, which is not a
+ * core-serializing instruction. Therefore, we need an explicit core
+ * serializing instruction after going from kernel thread back to
+ * user-space thread (active_mm moved back to current mm).
+ */
+static inline void membarrier_arch_mm_sync_core(struct mm_struct *mm)
+{
+	if (likely(!(atomic_read(&mm->membarrier_state) &
+			MEMBARRIER_STATE_SYNC_CORE)))
+		return;
+	sync_core();
+}
+void membarrier_arch_register_private_expedited(struct task_struct *t,
+		int flags);
+#endif
+
+#endif /* _ASM_X86_MEMBARRIER_H */

@@ -111,6 +111,7 @@ obj-$(CONFIG_PARAVIRT)		+= paravirt.o paravirt_patch_$(BITS).o
 obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= paravirt-spinlocks.o
 obj-$(CONFIG_PARAVIRT_CLOCK)	+= pvclock.o
 obj-$(CONFIG_X86_PMEM_LEGACY_DEVICE) += pmem.o
+obj-$(CONFIG_X86_64)		+= membarrier.o
 
 obj-$(CONFIG_EISA)		+= eisa.o
 obj-$(CONFIG_PCSPKR_PLATFORM)	+= pcspeaker.o

@@ -0,0 +1,39 @@
+/*
+ * Copyright (C) 2010-2017 Mathieu Desnoyers <mathieu.desnoyers-vg+e7yoeK/dWk0Htik3J/w@public.gmane.org>
+ *
+ * membarrier system call - x86 architecture code
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/sched/mm.h>
+#include <linux/sched/signal.h>
+#include <linux/thread_info.h>
+#include <linux/spinlock.h>
+#include <linux/rcupdate.h>
+#include <linux/atomic.h>
+
+void membarrier_arch_register_private_expedited(struct task_struct *p,
+		int flags)
+{
+	struct mm_struct *mm = p->mm;
+
+	if (!(flags & MEMBARRIER_FLAG_SYNC_CORE))
+		return;
+	atomic_or(MEMBARRIER_STATE_SYNC_CORE, &mm->membarrier_state);
+	if (atomic_read(&mm->mm_users) == 1 && get_nr_threads(p) == 1)
+		return;
+	/*
+	 * Ensure all future scheduler executions will observe the new
+	 * thread flag state for this process.
+	 */
+	synchronize_sched();
+}

@@ -147,9 +147,10 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 	this_cpu_write(cpu_tlbstate.is_lazy, false);
 
 	/*
-	 * The membarrier system call requires a full memory barrier
-	 * before returning to user-space, after storing to rq->curr.
-	 * Writing to CR3 provides that full memory barrier.
+	 * The membarrier system call requires a full memory barrier and
+	 * core serialization before returning to user-space, after
+	 * storing to rq->curr. Writing to CR3 provides that full
+	 * memory barrier and core serializing instruction.
 	 */
 	if (real_prev == next) {
 		VM_WARN_ON(this_cpu_read(cpu_tlbstate.ctxs[prev_asid].ctx_id) !=

@@ -222,6 +222,7 @@ enum {
 	MEMBARRIER_STATE_PRIVATE_EXPEDITED_READY		= (1U << 0),
 	MEMBARRIER_STATE_SWITCH_MM				= (1U << 1),
 	MEMBARRIER_STATE_PRIVATE_EXPEDITED_SYNC_CORE_READY	= (1U << 2),
+	MEMBARRIER_STATE_SYNC_CORE				= (1U << 3),
 };
 
 enum {

@@ -232,7 +233,10 @@ enum {
 #include <asm/membarrier.h>
 #else
 static inline void membarrier_arch_register_private_expedited(
-		struct task_struct *p)
+		struct task_struct *p, int flags)
+{
+}
+static inline void membarrier_arch_mm_sync_core(struct mm_struct *mm)
 {
 }
 #endif

@@ -251,6 +255,9 @@ static inline void membarrier_arch_switch_mm(struct mm_struct *prev,
 static inline void membarrier_execve(struct task_struct *t)
 {
 }
+static inline void membarrier_arch_mm_sync_core(struct mm_struct *mm)
+{
+}
 #endif
 
 #endif /* _LINUX_SCHED_MM_H */

@@ -2653,9 +2653,13 @@ static struct rq *finish_task_switch(struct task_struct *prev)
 	 * thread, mmdrop()'s implicit full barrier is required by the
 	 * membarrier system call, because the current active_mm can
 	 * become the current mm without going through switch_mm().
+	 * membarrier also requires a core serializing instruction
+	 * before going back to user-space after storing to rq->curr.
 	 */
-	if (mm)
+	if (mm) {
 		mmdrop(mm);
+		membarrier_arch_mm_sync_core(mm);
+	}
 	if (unlikely(prev_state == TASK_DEAD)) {
 		if (prev->sched_class->task_dead)
 			prev->sched_class->task_dead(prev);