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

Pull x86 fixes from Ingo Molnar:
 "A landry list of fixes:

   - fix reboot breakage on some PCID-enabled system

   - fix crashes/hangs on some PCID-enabled systems

   - fix microcode loading on certain older CPUs

   - various unwinder fixes

   - extend an APIC quirk to more hardware systems and disable APIC
     related warning on virtualized systems

   - various Hyper-V fixes

   - a macro definition robustness fix

   - remove jprobes IRQ disabling

   - various mem-encryption fixes"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/microcode: Do the family check first
  x86/mm: Flush more aggressively in lazy TLB mode
  x86/apic: Update TSC_DEADLINE quirk with additional SKX stepping
  x86/apic: Silence "FW_BUG TSC_DEADLINE disabled due to Errata" on hypervisors
  x86/mm: Disable various instrumentations of mm/mem_encrypt.c and mm/tlb.c
  x86/hyperv: Fix hypercalls with extended CPU ranges for TLB flushing
  x86/hyperv: Don't use percpu areas for pcpu_flush/pcpu_flush_ex structures
  x86/hyperv: Clear vCPU banks between calls to avoid flushing unneeded vCPUs
  x86/unwind: Disable unwinder warnings on 32-bit
  x86/unwind: Align stack pointer in unwinder dump
  x86/unwind: Use MSB for frame pointer encoding on 32-bit
  x86/unwind: Fix dereference of untrusted pointer
  x86/alternatives: Fix alt_max_short macro to really be a max()
  x86/mm/64: Fix reboot interaction with CR4.PCIDE
  kprobes/x86: Remove IRQ disabling from jprobe handlers
  kprobes/x86: Set up frame pointer in kprobe trampoline

arch/x86/entry/entry_32.S

@@ -176,7 +176,7 @@
 /*
  * This is a sneaky trick to help the unwinder find pt_regs on the stack.  The
  * frame pointer is replaced with an encoded pointer to pt_regs.  The encoding
- * is just setting the LSB, which makes it an invalid stack address and is also
+ * is just clearing the MSB, which makes it an invalid stack address and is also
  * a signal to the unwinder that it's a pt_regs pointer in disguise.
  *
  * NOTE: This macro must be used *after* SAVE_ALL because it corrupts the
@@ -185,7 +185,7 @@
 .macro ENCODE_FRAME_POINTER
 #ifdef CONFIG_FRAME_POINTER
 	mov %esp, %ebp
-	orl $0x1, %ebp
+	andl $0x7fffffff, %ebp
 #endif
 .endm
 

arch/x86/hyperv/hv_init.c

@@ -85,6 +85,8 @@ EXPORT_SYMBOL_GPL(hyperv_cs);
 u32 *hv_vp_index;
 EXPORT_SYMBOL_GPL(hv_vp_index);
 
+u32 hv_max_vp_index;
+
 static int hv_cpu_init(unsigned int cpu)
 {
 	u64 msr_vp_index;
@@ -93,6 +95,9 @@ static int hv_cpu_init(unsigned int cpu)
 
 	hv_vp_index[smp_processor_id()] = msr_vp_index;
 
+	if (msr_vp_index > hv_max_vp_index)
+		hv_max_vp_index = msr_vp_index;
+
 	return 0;
 }
 

arch/x86/hyperv/mmu.c

@@ -36,9 +36,9 @@ struct hv_flush_pcpu_ex {
 /* Each gva in gva_list encodes up to 4096 pages to flush */
 #define HV_TLB_FLUSH_UNIT (4096 * PAGE_SIZE)
 
-static struct hv_flush_pcpu __percpu *pcpu_flush;
+static struct hv_flush_pcpu __percpu **pcpu_flush;
 
-static struct hv_flush_pcpu_ex __percpu *pcpu_flush_ex;
+static struct hv_flush_pcpu_ex __percpu **pcpu_flush_ex;
 
 /*
  * Fills in gva_list starting from offset. Returns the number of items added.
@@ -76,6 +76,18 @@ static inline int cpumask_to_vp_set(struct hv_flush_pcpu_ex *flush,
 {
 	int cpu, vcpu, vcpu_bank, vcpu_offset, nr_bank = 1;
 
+	/* valid_bank_mask can represent up to 64 banks */
+	if (hv_max_vp_index / 64 >= 64)
+		return 0;
+
+	/*
+	 * Clear all banks up to the maximum possible bank as hv_flush_pcpu_ex
+	 * structs are not cleared between calls, we risk flushing unneeded
+	 * vCPUs otherwise.
+	 */
+	for (vcpu_bank = 0; vcpu_bank <= hv_max_vp_index / 64; vcpu_bank++)
+		flush->hv_vp_set.bank_contents[vcpu_bank] = 0;
+
 	/*
 	 * Some banks may end up being empty but this is acceptable.
 	 */
@@ -83,11 +95,6 @@ static inline int cpumask_to_vp_set(struct hv_flush_pcpu_ex *flush,
 		vcpu = hv_cpu_number_to_vp_number(cpu);
 		vcpu_bank = vcpu / 64;
 		vcpu_offset = vcpu % 64;
-
-		/* valid_bank_mask can represent up to 64 banks */
-		if (vcpu_bank >= 64)
-			return 0;
-
 		__set_bit(vcpu_offset, (unsigned long *)
 			  &flush->hv_vp_set.bank_contents[vcpu_bank]);
 		if (vcpu_bank >= nr_bank)
@@ -102,6 +109,7 @@ static void hyperv_flush_tlb_others(const struct cpumask *cpus,
 				    const struct flush_tlb_info *info)
 {
 	int cpu, vcpu, gva_n, max_gvas;
+	struct hv_flush_pcpu **flush_pcpu;
 	struct hv_flush_pcpu *flush;
 	u64 status = U64_MAX;
 	unsigned long flags;
@@ -116,7 +124,17 @@ static void hyperv_flush_tlb_others(const struct cpumask *cpus,
 
 	local_irq_save(flags);
 
-	flush = this_cpu_ptr(pcpu_flush);
+	flush_pcpu = this_cpu_ptr(pcpu_flush);
+
+	if (unlikely(!*flush_pcpu))
+		*flush_pcpu = page_address(alloc_page(GFP_ATOMIC));
+
+	flush = *flush_pcpu;
+
+	if (unlikely(!flush)) {
+		local_irq_restore(flags);
+		goto do_native;
+	}
 
 	if (info->mm) {
 		flush->address_space = virt_to_phys(info->mm->pgd);
@@ -173,6 +191,7 @@ static void hyperv_flush_tlb_others_ex(const struct cpumask *cpus,
 				       const struct flush_tlb_info *info)
 {
 	int nr_bank = 0, max_gvas, gva_n;
+	struct hv_flush_pcpu_ex **flush_pcpu;
 	struct hv_flush_pcpu_ex *flush;
 	u64 status = U64_MAX;
 	unsigned long flags;
@@ -187,7 +206,17 @@ static void hyperv_flush_tlb_others_ex(const struct cpumask *cpus,
 
 	local_irq_save(flags);
 
-	flush = this_cpu_ptr(pcpu_flush_ex);
+	flush_pcpu = this_cpu_ptr(pcpu_flush_ex);
+
+	if (unlikely(!*flush_pcpu))
+		*flush_pcpu = page_address(alloc_page(GFP_ATOMIC));
+
+	flush = *flush_pcpu;
+
+	if (unlikely(!flush)) {
+		local_irq_restore(flags);
+		goto do_native;
+	}
 
 	if (info->mm) {
 		flush->address_space = virt_to_phys(info->mm->pgd);
@@ -222,18 +251,18 @@ static void hyperv_flush_tlb_others_ex(const struct cpumask *cpus,
 		flush->flags |= HV_FLUSH_NON_GLOBAL_MAPPINGS_ONLY;
 		status = hv_do_rep_hypercall(
 			HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX,
-			0, nr_bank + 2, flush, NULL);
+			0, nr_bank, flush, NULL);
 	} else if (info->end &&
 		   ((info->end - info->start)/HV_TLB_FLUSH_UNIT) > max_gvas) {
 		status = hv_do_rep_hypercall(
 			HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX,
-			0, nr_bank + 2, flush, NULL);
+			0, nr_bank, flush, NULL);
 	} else {
 		gva_n = fill_gva_list(flush->gva_list, nr_bank,
 				      info->start, info->end);
 		status = hv_do_rep_hypercall(
 			HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST_EX,
-			gva_n, nr_bank + 2, flush, NULL);
+			gva_n, nr_bank, flush, NULL);
 	}
 
 	local_irq_restore(flags);
@@ -266,7 +295,7 @@ void hyper_alloc_mmu(void)
 		return;
 
 	if (!(ms_hyperv.hints & HV_X64_EX_PROCESSOR_MASKS_RECOMMENDED))
-		pcpu_flush = __alloc_percpu(PAGE_SIZE, PAGE_SIZE);
+		pcpu_flush = alloc_percpu(struct hv_flush_pcpu *);
 	else
-		pcpu_flush_ex = __alloc_percpu(PAGE_SIZE, PAGE_SIZE);
+		pcpu_flush_ex = alloc_percpu(struct hv_flush_pcpu_ex *);
 }

arch/x86/include/asm/alternative-asm.h

@@ -62,8 +62,10 @@
 #define new_len2		145f-144f
 
 /*
- * max without conditionals. Idea adapted from:
+ * gas compatible max based on the idea from:
  * http://graphics.stanford.edu/~seander/bithacks.html#IntegerMinOrMax
+ *
+ * The additional "-" is needed because gas uses a "true" value of -1.
  */
 #define alt_max_short(a, b)	((a) ^ (((a) ^ (b)) & -(-((a) < (b)))))
 

arch/x86/include/asm/alternative.h

@@ -103,12 +103,12 @@ static inline int alternatives_text_reserved(void *start, void *end)
 	alt_end_marker ":\n"
 
 /*
- * max without conditionals. Idea adapted from:
+ * gas compatible max based on the idea from:
  * http://graphics.stanford.edu/~seander/bithacks.html#IntegerMinOrMax
  *
- * The additional "-" is needed because gas works with s32s.
+ * The additional "-" is needed because gas uses a "true" value of -1.
  */
-#define alt_max_short(a, b)	"((" a ") ^ (((" a ") ^ (" b ")) & -(-((" a ") - (" b ")))))"
+#define alt_max_short(a, b)	"((" a ") ^ (((" a ") ^ (" b ")) & -(-((" a ") < (" b ")))))"
 
 /*
  * Pad the second replacement alternative with additional NOPs if it is

arch/x86/include/asm/mmu_context.h

@@ -126,13 +126,7 @@ static inline void switch_ldt(struct mm_struct *prev, struct mm_struct *next)
 	DEBUG_LOCKS_WARN_ON(preemptible());
 }
 
-static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
-{
-	int cpu = smp_processor_id();
-
-	if (cpumask_test_cpu(cpu, mm_cpumask(mm)))
-		cpumask_clear_cpu(cpu, mm_cpumask(mm));
-}
+void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk);
 
 static inline int init_new_context(struct task_struct *tsk,
 				   struct mm_struct *mm)

arch/x86/include/asm/mshyperv.h

@@ -289,6 +289,7 @@ static inline u64 hv_do_rep_hypercall(u16 code, u16 rep_count, u16 varhead_size,
  * to this information.
  */
 extern u32 *hv_vp_index;
+extern u32 hv_max_vp_index;
 
 /**
  * hv_cpu_number_to_vp_number() - Map CPU to VP.

arch/x86/include/asm/tlbflush.h

@@ -82,6 +82,13 @@ static inline u64 inc_mm_tlb_gen(struct mm_struct *mm)
 #define __flush_tlb_single(addr) __native_flush_tlb_single(addr)
 #endif
 
+/*
+ * If tlb_use_lazy_mode is true, then we try to avoid switching CR3 to point
+ * to init_mm when we switch to a kernel thread (e.g. the idle thread).  If
+ * it's false, then we immediately switch CR3 when entering a kernel thread.
+ */
+DECLARE_STATIC_KEY_TRUE(tlb_use_lazy_mode);
+
 /*
  * 6 because 6 should be plenty and struct tlb_state will fit in
  * two cache lines.
@@ -104,6 +111,23 @@ struct tlb_state {
 	u16 loaded_mm_asid;
 	u16 next_asid;
 
+	/*
+	 * We can be in one of several states:
+	 *
+	 *  - Actively using an mm.  Our CPU's bit will be set in
+	 *    mm_cpumask(loaded_mm) and is_lazy == false;
+	 *
+	 *  - Not using a real mm.  loaded_mm == &init_mm.  Our CPU's bit
+	 *    will not be set in mm_cpumask(&init_mm) and is_lazy == false.
+	 *
+	 *  - Lazily using a real mm.  loaded_mm != &init_mm, our bit
+	 *    is set in mm_cpumask(loaded_mm), but is_lazy == true.
+	 *    We're heuristically guessing that the CR3 load we
+	 *    skipped more than makes up for the overhead added by
+	 *    lazy mode.
+	 */
+	bool is_lazy;
+
 	/*
 	 * Access to this CR4 shadow and to H/W CR4 is protected by
 	 * disabling interrupts when modifying either one.

arch/x86/kernel/apic/apic.c

@@ -573,11 +573,21 @@ static u32 bdx_deadline_rev(void)
 	return ~0U;
 }
 
+static u32 skx_deadline_rev(void)
+{
+	switch (boot_cpu_data.x86_mask) {
+	case 0x03: return 0x01000136;
+	case 0x04: return 0x02000014;
+	}
+
+	return ~0U;
+}
+
 static const struct x86_cpu_id deadline_match[] = {
 	DEADLINE_MODEL_MATCH_FUNC( INTEL_FAM6_HASWELL_X,	hsx_deadline_rev),
 	DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_BROADWELL_X,	0x0b000020),
 	DEADLINE_MODEL_MATCH_FUNC( INTEL_FAM6_BROADWELL_XEON_D,	bdx_deadline_rev),
-	DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_SKYLAKE_X,	0x02000014),
+	DEADLINE_MODEL_MATCH_FUNC( INTEL_FAM6_SKYLAKE_X,	skx_deadline_rev),
 
 	DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_HASWELL_CORE,	0x22),
 	DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_HASWELL_ULT,	0x20),
@@ -600,7 +610,8 @@ static void apic_check_deadline_errata(void)
 	const struct x86_cpu_id *m;
 	u32 rev;
 
-	if (!boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+	if (!boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER) ||
+	    boot_cpu_has(X86_FEATURE_HYPERVISOR))
 		return;
 
 	m = x86_match_cpu(deadline_match);

arch/x86/kernel/cpu/microcode/core.c

@@ -122,9 +122,6 @@ static bool __init check_loader_disabled_bsp(void)
 	bool *res = &dis_ucode_ldr;
 #endif
 
-	if (!have_cpuid_p())
-		return *res;
-
 	/*
 	 * CPUID(1).ECX[31]: reserved for hypervisor use. This is still not
 	 * completely accurate as xen pv guests don't see that CPUID bit set but
@@ -166,24 +163,36 @@ bool get_builtin_firmware(struct cpio_data *cd, const char *name)
 void __init load_ucode_bsp(void)
 {
 	unsigned int cpuid_1_eax;
+	bool intel = true;
 
-	if (check_loader_disabled_bsp())
+	if (!have_cpuid_p())
 		return;
 
 	cpuid_1_eax = native_cpuid_eax(1);
 
 	switch (x86_cpuid_vendor()) {
 	case X86_VENDOR_INTEL:
-		if (x86_family(cpuid_1_eax) >= 6)
-			load_ucode_intel_bsp();
+		if (x86_family(cpuid_1_eax) < 6)
+			return;
 		break;
+
 	case X86_VENDOR_AMD:
-		if (x86_family(cpuid_1_eax) >= 0x10)
-			load_ucode_amd_bsp(cpuid_1_eax);
+		if (x86_family(cpuid_1_eax) < 0x10)
+			return;
+		intel = false;
 		break;
+
 	default:
-		break;
+		return;
 	}
+
+	if (check_loader_disabled_bsp())
+		return;
+
+	if (intel)
+		load_ucode_intel_bsp();
+	else
+		load_ucode_amd_bsp(cpuid_1_eax);
 }
 
 static bool check_loader_disabled_ap(void)

arch/x86/kernel/kprobes/common.h

@@ -3,6 +3,15 @@
 
 /* Kprobes and Optprobes common header */
 
+#include <asm/asm.h>
+
+#ifdef CONFIG_FRAME_POINTER
+# define SAVE_RBP_STRING "	push %" _ASM_BP "\n" \
+			 "	mov  %" _ASM_SP ", %" _ASM_BP "\n"
+#else
+# define SAVE_RBP_STRING "	push %" _ASM_BP "\n"
+#endif
+
 #ifdef CONFIG_X86_64
 #define SAVE_REGS_STRING			\
 	/* Skip cs, ip, orig_ax. */		\
@@ -17,7 +26,7 @@
 	"	pushq %r10\n"			\
 	"	pushq %r11\n"			\
 	"	pushq %rbx\n"			\
-	"	pushq %rbp\n"			\
+	SAVE_RBP_STRING				\
 	"	pushq %r12\n"			\
 	"	pushq %r13\n"			\
 	"	pushq %r14\n"			\
@@ -48,7 +57,7 @@
 	"	pushl %es\n"			\
 	"	pushl %ds\n"			\
 	"	pushl %eax\n"			\
-	"	pushl %ebp\n"			\
+	SAVE_RBP_STRING				\
 	"	pushl %edi\n"			\
 	"	pushl %esi\n"			\
 	"	pushl %edx\n"			\

arch/x86/kernel/kprobes/core.c

@@ -1080,8 +1080,6 @@ int setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
 	 * raw stack chunk with redzones:
 	 */
 	__memcpy(kcb->jprobes_stack, (kprobe_opcode_t *)addr, MIN_STACK_SIZE(addr));
-	regs->flags &= ~X86_EFLAGS_IF;
-	trace_hardirqs_off();
 	regs->ip = (unsigned long)(jp->entry);
 
 	/*

arch/x86/kernel/reboot.c

@@ -105,6 +105,10 @@ void __noreturn machine_real_restart(unsigned int type)
 	load_cr3(initial_page_table);
 #else
 	write_cr3(real_mode_header->trampoline_pgd);
+
+	/* Exiting long mode will fail if CR4.PCIDE is set. */
+	if (static_cpu_has(X86_FEATURE_PCID))
+		cr4_clear_bits(X86_CR4_PCIDE);
 #endif
 
 	/* Jump to the identity-mapped low memory code */

arch/x86/kernel/unwind_frame.c

@@ -44,7 +44,8 @@ static void unwind_dump(struct unwind_state *state)
 			state->stack_info.type, state->stack_info.next_sp,
 			state->stack_mask, state->graph_idx);
 
-	for (sp = state->orig_sp; sp; sp = PTR_ALIGN(stack_info.next_sp, sizeof(long))) {
+	for (sp = PTR_ALIGN(state->orig_sp, sizeof(long)); sp;
+	     sp = PTR_ALIGN(stack_info.next_sp, sizeof(long))) {
 		if (get_stack_info(sp, state->task, &stack_info, &visit_mask))
 			break;
 
@@ -174,6 +175,7 @@ static bool is_last_task_frame(struct unwind_state *state)
  * This determines if the frame pointer actually contains an encoded pointer to
  * pt_regs on the stack.  See ENCODE_FRAME_POINTER.
  */
+#ifdef CONFIG_X86_64
 static struct pt_regs *decode_frame_pointer(unsigned long *bp)
 {
 	unsigned long regs = (unsigned long)bp;
@@ -183,6 +185,23 @@ static struct pt_regs *decode_frame_pointer(unsigned long *bp)
 
 	return (struct pt_regs *)(regs & ~0x1);
 }
+#else
+static struct pt_regs *decode_frame_pointer(unsigned long *bp)
+{
+	unsigned long regs = (unsigned long)bp;
+
+	if (regs & 0x80000000)
+		return NULL;
+
+	return (struct pt_regs *)(regs | 0x80000000);
+}
+#endif
+
+#ifdef CONFIG_X86_32
+#define KERNEL_REGS_SIZE (sizeof(struct pt_regs) - 2*sizeof(long))
+#else
+#define KERNEL_REGS_SIZE (sizeof(struct pt_regs))
+#endif
 
 static bool update_stack_state(struct unwind_state *state,
 			       unsigned long *next_bp)
@@ -202,7 +221,7 @@ static bool update_stack_state(struct unwind_state *state,
 	regs = decode_frame_pointer(next_bp);
 	if (regs) {
 		frame = (unsigned long *)regs;
-		len = regs_size(regs);
+		len = KERNEL_REGS_SIZE;
 		state->got_irq = true;
 	} else {
 		frame = next_bp;
@@ -226,6 +245,14 @@ static bool update_stack_state(struct unwind_state *state,
 	    frame < prev_frame_end)
 		return false;
 
+	/*
+	 * On 32-bit with user mode regs, make sure the last two regs are safe
+	 * to access:
+	 */
+	if (IS_ENABLED(CONFIG_X86_32) && regs && user_mode(regs) &&
+	    !on_stack(info, frame, len + 2*sizeof(long)))
+		return false;
+
 	/* Move state to the next frame: */
 	if (regs) {
 		state->regs = regs;
@@ -328,6 +355,13 @@ bool unwind_next_frame(struct unwind_state *state)
 	    state->regs->sp < (unsigned long)task_pt_regs(state->task))
 		goto the_end;
 
+	/*
+	 * There are some known frame pointer issues on 32-bit.  Disable
+	 * unwinder warnings on 32-bit until it gets objtool support.
+	 */
+	if (IS_ENABLED(CONFIG_X86_32))
+		goto the_end;
+
 	if (state->regs) {
 		printk_deferred_once(KERN_WARNING
 			"WARNING: kernel stack regs at %p in %s:%d has bad 'bp' value %p\n",

arch/x86/mm/Makefile

-# Kernel does not boot with instrumentation of tlb.c.
-KCOV_INSTRUMENT_tlb.o	:= n
+# Kernel does not boot with instrumentation of tlb.c and mem_encrypt.c
+KCOV_INSTRUMENT_tlb.o		:= n
+KCOV_INSTRUMENT_mem_encrypt.o	:= n
+
+KASAN_SANITIZE_mem_encrypt.o	:= n
+
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_mem_encrypt.o	= -pg
+endif
 
 obj-y	:=  init.o init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
 	    pat.o pgtable.o physaddr.o setup_nx.o tlb.o

arch/x86/mm/tlb.c

@@ -30,6 +30,8 @@
 
 atomic64_t last_mm_ctx_id = ATOMIC64_INIT(1);
 
+DEFINE_STATIC_KEY_TRUE(tlb_use_lazy_mode);
+
 static void choose_new_asid(struct mm_struct *next, u64 next_tlb_gen,
 			    u16 *new_asid, bool *need_flush)
 {
@@ -80,7 +82,7 @@ void leave_mm(int cpu)
 		return;
 
 	/* Warn if we're not lazy. */
-	WARN_ON(cpumask_test_cpu(smp_processor_id(), mm_cpumask(loaded_mm)));
+	WARN_ON(!this_cpu_read(cpu_tlbstate.is_lazy));
 
 	switch_mm(NULL, &init_mm, NULL);
 }
@@ -142,45 +144,24 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 		__flush_tlb_all();
 	}
 #endif
+	this_cpu_write(cpu_tlbstate.is_lazy, false);
 
 	if (real_prev == next) {
 		VM_BUG_ON(this_cpu_read(cpu_tlbstate.ctxs[prev_asid].ctx_id) !=
 			  next->context.ctx_id);
 
-		if (cpumask_test_cpu(cpu, mm_cpumask(next))) {
-			/*
-			 * There's nothing to do: we weren't lazy, and we
-			 * aren't changing our mm.  We don't need to flush
-			 * anything, nor do we need to update CR3, CR4, or
-			 * LDTR.
-			 */
-			return;
-		}
-
-		/* Resume remote flushes and then read tlb_gen. */
-		cpumask_set_cpu(cpu, mm_cpumask(next));
-		next_tlb_gen = atomic64_read(&next->context.tlb_gen);
-
-		if (this_cpu_read(cpu_tlbstate.ctxs[prev_asid].tlb_gen) <
-		    next_tlb_gen) {
-			/*
-			 * Ideally, we'd have a flush_tlb() variant that
-			 * takes the known CR3 value as input.  This would
-			 * be faster on Xen PV and on hypothetical CPUs
-			 * on which INVPCID is fast.
-			 */
-			this_cpu_write(cpu_tlbstate.ctxs[prev_asid].tlb_gen,
-				       next_tlb_gen);
-			write_cr3(build_cr3(next, prev_asid));
-			trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH,
-					TLB_FLUSH_ALL);
-		}
-
 		/*
-		 * We just exited lazy mode, which means that CR4 and/or LDTR
-		 * may be stale.  (Changes to the required CR4 and LDTR states
-		 * are not reflected in tlb_gen.)
+		 * We don't currently support having a real mm loaded without
+		 * our cpu set in mm_cpumask().  We have all the bookkeeping
+		 * in place to figure out whether we would need to flush
+		 * if our cpu were cleared in mm_cpumask(), but we don't
+		 * currently use it.
 		 */
+		if (WARN_ON_ONCE(real_prev != &init_mm &&
+				 !cpumask_test_cpu(cpu, mm_cpumask(next))))
+			cpumask_set_cpu(cpu, mm_cpumask(next));
+
+		return;
 	} else {
 		u16 new_asid;
 		bool need_flush;
@@ -199,10 +180,9 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 		}
 
 		/* Stop remote flushes for the previous mm */
-		if (cpumask_test_cpu(cpu, mm_cpumask(real_prev)))
-			cpumask_clear_cpu(cpu, mm_cpumask(real_prev));
-
-		VM_WARN_ON_ONCE(cpumask_test_cpu(cpu, mm_cpumask(next)));
+		VM_WARN_ON_ONCE(!cpumask_test_cpu(cpu, mm_cpumask(real_prev)) &&
+				real_prev != &init_mm);
+		cpumask_clear_cpu(cpu, mm_cpumask(real_prev));
 
 		/*
 		 * Start remote flushes and then read tlb_gen.
@@ -232,6 +212,37 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 	switch_ldt(real_prev, next);
 }
 
+/*
+ * enter_lazy_tlb() is a hint from the scheduler that we are entering a
+ * kernel thread or other context without an mm.  Acceptable implementations
+ * include doing nothing whatsoever, switching to init_mm, or various clever
+ * lazy tricks to try to minimize TLB flushes.
+ *
+ * The scheduler reserves the right to call enter_lazy_tlb() several times
+ * in a row.  It will notify us that we're going back to a real mm by
+ * calling switch_mm_irqs_off().
+ */
+void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+	if (this_cpu_read(cpu_tlbstate.loaded_mm) == &init_mm)
+		return;
+
+	if (static_branch_unlikely(&tlb_use_lazy_mode)) {
+		/*
+		 * There's a significant optimization that may be possible
+		 * here.  We have accurate enough TLB flush tracking that we
+		 * don't need to maintain coherence of TLB per se when we're
+		 * lazy.  We do, however, need to maintain coherence of
+		 * paging-structure caches.  We could, in principle, leave our
+		 * old mm loaded and only switch to init_mm when
+		 * tlb_remove_page() happens.
+		 */
+		this_cpu_write(cpu_tlbstate.is_lazy, true);
+	} else {
+		switch_mm(NULL, &init_mm, NULL);
+	}
+}
+
 /*
  * Call this when reinitializing a CPU.  It fixes the following potential
  * problems:
@@ -303,16 +314,20 @@ static void flush_tlb_func_common(const struct flush_tlb_info *f,
 	/* This code cannot presently handle being reentered. */
 	VM_WARN_ON(!irqs_disabled());
 
+	if (unlikely(loaded_mm == &init_mm))
+		return;
+
 	VM_WARN_ON(this_cpu_read(cpu_tlbstate.ctxs[loaded_mm_asid].ctx_id) !=
 		   loaded_mm->context.ctx_id);
 
-	if (!cpumask_test_cpu(smp_processor_id(), mm_cpumask(loaded_mm))) {
+	if (this_cpu_read(cpu_tlbstate.is_lazy)) {
 		/*
-		 * We're in lazy mode -- don't flush.  We can get here on
-		 * remote flushes due to races and on local flushes if a
-		 * kernel thread coincidentally flushes the mm it's lazily
-		 * still using.
+		 * We're in lazy mode.  We need to at least flush our
+		 * paging-structure cache to avoid speculatively reading
+		 * garbage into our TLB.  Since switching to init_mm is barely
+		 * slower than a minimal flush, just switch to init_mm.
 		 */
+		switch_mm_irqs_off(NULL, &init_mm, NULL);
 		return;
 	}
 
@@ -611,3 +626,57 @@ static int __init create_tlb_single_page_flush_ceiling(void)
 	return 0;
 }
 late_initcall(create_tlb_single_page_flush_ceiling);
+
+static ssize_t tlblazy_read_file(struct file *file, char __user *user_buf,
+				 size_t count, loff_t *ppos)
+{
+	char buf[2];
+
+	buf[0] = static_branch_likely(&tlb_use_lazy_mode) ? '1' : '0';
+	buf[1] = '\n';
+
+	return simple_read_from_buffer(user_buf, count, ppos, buf, 2);
+}
+
+static ssize_t tlblazy_write_file(struct file *file,
+		 const char __user *user_buf, size_t count, loff_t *ppos)
+{
+	bool val;
+
+	if (kstrtobool_from_user(user_buf, count, &val))
+		return -EINVAL;
+
+	if (val)
+		static_branch_enable(&tlb_use_lazy_mode);
+	else
+		static_branch_disable(&tlb_use_lazy_mode);
+
+	return count;
+}
+
+static const struct file_operations fops_tlblazy = {
+	.read = tlblazy_read_file,
+	.write = tlblazy_write_file,
+	.llseek = default_llseek,
+};
+
+static int __init init_tlb_use_lazy_mode(void)
+{
+	if (boot_cpu_has(X86_FEATURE_PCID)) {
+		/*
+		 * Heuristic: with PCID on, switching to and from
+		 * init_mm is reasonably fast, but remote flush IPIs
+		 * as expensive as ever, so turn off lazy TLB mode.
+		 *
+		 * We can't do this in setup_pcid() because static keys
+		 * haven't been initialized yet, and it would blow up
+		 * badly.
+		 */
+		static_branch_disable(&tlb_use_lazy_mode);
+	}
+
+	debugfs_create_file("tlb_use_lazy_mode", S_IRUSR | S_IWUSR,
+			    arch_debugfs_dir, NULL, &fops_tlblazy);
+	return 0;
+}
+late_initcall(init_tlb_use_lazy_mode);