Merge branch 'queue' into next

Merge patches queued during the run-up to the merge window.

* queue: (25 commits)
  KVM: Choose better candidate for directed yield
  KVM: Note down when cpu relax intercepted or pause loop exited
  KVM: Add config to support ple or cpu relax optimzation
  KVM: switch to symbolic name for irq_states size
  KVM: x86: Fix typos in pmu.c
  KVM: x86: Fix typos in lapic.c
  KVM: x86: Fix typos in cpuid.c
  KVM: x86: Fix typos in emulate.c
  KVM: x86: Fix typos in x86.c
  KVM: SVM: Fix typos
  KVM: VMX: Fix typos
  KVM: remove the unused parameter of gfn_to_pfn_memslot
  KVM: remove is_error_hpa
  KVM: make bad_pfn static to kvm_main.c
  KVM: using get_fault_pfn to get the fault pfn
  KVM: MMU: track the refcount when unmap the page
  KVM: x86: remove unnecessary mark_page_dirty
  KVM: MMU: Avoid handling same rmap_pde in kvm_handle_hva_range()
  KVM: MMU: Push trace_kvm_age_page() into kvm_age_rmapp()
  KVM: MMU: Add memslot parameter to hva handlers
  ...

Signed-off-by: Avi Kivity <avi@redhat.com>

arch/powerpc/include/asm/kvm_host.h

@@ -52,6 +52,8 @@
 
 struct kvm;
 extern int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
+extern int kvm_unmap_hva_range(struct kvm *kvm,
+			       unsigned long start, unsigned long end);
 extern int kvm_age_hva(struct kvm *kvm, unsigned long hva);
 extern int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 extern void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);

arch/powerpc/kvm/book3s_64_mmu_hv.c

@@ -756,8 +756,11 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	goto out_put;
 }
 
-static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
+static int kvm_handle_hva_range(struct kvm *kvm,
-			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+				unsigned long start,
+				unsigned long end,
+				int (*handler)(struct kvm *kvm,
+					       unsigned long *rmapp,
 					       unsigned long gfn))
 {
 	int ret;
@@ -767,15 +770,25 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 
 	slots = kvm_memslots(kvm);
 	kvm_for_each_memslot(memslot, slots) {
-		unsigned long start = memslot->userspace_addr;
+		unsigned long hva_start, hva_end;
-		unsigned long end;
+		gfn_t gfn, gfn_end;
 
-		end = start + (memslot->npages << PAGE_SHIFT);
+		hva_start = max(start, memslot->userspace_addr);
-		if (hva >= start && hva < end) {
+		hva_end = min(end, memslot->userspace_addr +
-			gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT;
+					(memslot->npages << PAGE_SHIFT));
+		if (hva_start >= hva_end)
+			continue;
+		/*
+		 * {gfn(page) | page intersects with [hva_start, hva_end)} =
+		 * {gfn, gfn+1, ..., gfn_end-1}.
+		 */
+		gfn = hva_to_gfn_memslot(hva_start, memslot);
+		gfn_end = hva_to_gfn_memslot(hva_end + PAGE_SIZE - 1, memslot);
 
-			ret = handler(kvm, &memslot->rmap[gfn_offset],
+		for (; gfn < gfn_end; ++gfn) {
-				      memslot->base_gfn + gfn_offset);
+			gfn_t gfn_offset = gfn - memslot->base_gfn;
+
+			ret = handler(kvm, &memslot->rmap[gfn_offset], gfn);
 			retval |= ret;
 		}
 	}
@@ -783,6 +796,13 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
 	return retval;
 }
 
+static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
+			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+					 unsigned long gfn))
+{
+	return kvm_handle_hva_range(kvm, hva, hva + 1, handler);
+}
+
 static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
 			   unsigned long gfn)
 {
@@ -850,6 +870,13 @@ int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
 	return 0;
 }
 
+int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end)
+{
+	if (kvm->arch.using_mmu_notifiers)
+		kvm_handle_hva_range(kvm, start, end, kvm_unmap_rmapp);
+	return 0;
+}
+
 static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
 			 unsigned long gfn)
 {

arch/powerpc/kvm/e500_tlb.c

@@ -520,7 +520,7 @@ static inline void kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500,
 
 	if (likely(!pfnmap)) {
 		unsigned long tsize_pages = 1 << (tsize + 10 - PAGE_SHIFT);
-		pfn = gfn_to_pfn_memslot(vcpu_e500->vcpu.kvm, slot, gfn);
+		pfn = gfn_to_pfn_memslot(slot, gfn);
 		if (is_error_pfn(pfn)) {
 			printk(KERN_ERR "Couldn't get real page for gfn %lx!\n",
 					(long)gfn);

arch/s390/kvm/Kconfig

@@ -21,6 +21,7 @@ config KVM
 	depends on HAVE_KVM && EXPERIMENTAL
 	select PREEMPT_NOTIFIERS
 	select ANON_INODES
+	select HAVE_KVM_CPU_RELAX_INTERCEPT
 	---help---
 	  Support hosting paravirtualized guest machines using the SIE
 	  virtualization capability on the mainframe. This should work

arch/x86/include/asm/kvm_host.h

@@ -500,11 +500,11 @@ struct kvm_vcpu_arch {
 };
 
 struct kvm_lpage_info {
-	unsigned long rmap_pde;
 	int write_count;
 };
 
 struct kvm_arch_memory_slot {
+	unsigned long *rmap_pde[KVM_NR_PAGE_SIZES - 1];
 	struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1];
 };
 
@@ -957,6 +957,7 @@ extern bool kvm_rebooting;
 
 #define KVM_ARCH_WANT_MMU_NOTIFIER
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva);
+int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end);
 int kvm_age_hva(struct kvm *kvm, unsigned long hva);
 int kvm_test_age_hva(struct kvm *kvm, unsigned long hva);
 void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte);

arch/x86/kvm/Kconfig

@@ -37,6 +37,7 @@ config KVM
 	select TASK_DELAY_ACCT
 	select PERF_EVENTS
 	select HAVE_KVM_MSI
+	select HAVE_KVM_CPU_RELAX_INTERCEPT
 	---help---
 	  Support hosting fully virtualized guest machines using hardware
 	  virtualization extensions.  You will need a fairly recent

arch/x86/kvm/cpuid.c

@@ -316,7 +316,7 @@ static int do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 	}
 	case 7: {
 		entry->flags |= KVM_CPUID_FLAG_SIGNIFCANT_INDEX;
-		/* Mask ebx against host capbability word 9 */
+		/* Mask ebx against host capability word 9 */
 		if (index == 0) {
 			entry->ebx &= kvm_supported_word9_x86_features;
 			cpuid_mask(&entry->ebx, 9);

arch/x86/kvm/emulate.c

@@ -642,7 +642,7 @@ static int __linearize(struct x86_emulate_ctxt *ctxt,
 			if (addr.ea > lim || (u32)(addr.ea + size - 1) > lim)
 				goto bad;
 		} else {
-			/* exapand-down segment */
+			/* expand-down segment */
 			if (addr.ea <= lim || (u32)(addr.ea + size - 1) <= lim)
 				goto bad;
 			lim = desc.d ? 0xffffffff : 0xffff;
@@ -1383,7 +1383,7 @@ static int load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 	err_code = selector & 0xfffc;
 	err_vec = GP_VECTOR;
 
-	/* can't load system descriptor into segment selecor */
+	/* can't load system descriptor into segment selector */
 	if (seg <= VCPU_SREG_GS && !seg_desc.s)
 		goto exception;
 
@@ -2398,7 +2398,7 @@ static int load_state_from_tss16(struct x86_emulate_ctxt *ctxt,
 	set_segment_selector(ctxt, tss->ds, VCPU_SREG_DS);
 
 	/*
-	 * Now load segment descriptors. If fault happenes at this stage
+	 * Now load segment descriptors. If fault happens at this stage
 	 * it is handled in a context of new task
 	 */
 	ret = load_segment_descriptor(ctxt, tss->ldt, VCPU_SREG_LDTR);
@@ -2640,7 +2640,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
 	 *
 	 * 1. jmp/call/int to task gate: Check against DPL of the task gate
 	 * 2. Exception/IRQ/iret: No check is performed
-	 * 3. jmp/call to TSS: Check agains DPL of the TSS
+	 * 3. jmp/call to TSS: Check against DPL of the TSS
 	 */
 	if (reason == TASK_SWITCH_GATE) {
 		if (idt_index != -1) {
@@ -2681,7 +2681,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt,
 		ctxt->eflags = ctxt->eflags & ~X86_EFLAGS_NT;
 
 	/* set back link to prev task only if NT bit is set in eflags
-	   note that old_tss_sel is not used afetr this point */
+	   note that old_tss_sel is not used after this point */
 	if (reason != TASK_SWITCH_CALL && reason != TASK_SWITCH_GATE)
 		old_tss_sel = 0xffff;
 

arch/x86/kvm/irq.h

@@ -70,7 +70,7 @@ struct kvm_pic {
 	struct kvm_io_device dev_slave;
 	struct kvm_io_device dev_eclr;
 	void (*ack_notifier)(void *opaque, int irq);
-	unsigned long irq_states[16];
+	unsigned long irq_states[PIC_NUM_PINS];
 };
 
 struct kvm_pic *kvm_create_pic(struct kvm *kvm);

arch/x86/kvm/lapic.c

@@ -719,7 +719,7 @@ static int apic_reg_read(struct kvm_lapic *apic, u32 offset, int len,
 {
 	unsigned char alignment = offset & 0xf;
 	u32 result;
-	/* this bitmask has a bit cleared for each reserver register */
+	/* this bitmask has a bit cleared for each reserved register */
 	static const u64 rmask = 0x43ff01ffffffe70cULL;
 
 	if ((alignment + len) > 4) {
@@ -792,7 +792,7 @@ static void start_apic_timer(struct kvm_lapic *apic)
 	atomic_set(&apic->lapic_timer.pending, 0);
 
 	if (apic_lvtt_period(apic) || apic_lvtt_oneshot(apic)) {
-		/* lapic timer in oneshot or peroidic mode */
+		/* lapic timer in oneshot or periodic mode */
 		now = apic->lapic_timer.timer.base->get_time();
 		apic->lapic_timer.period = (u64)apic_get_reg(apic, APIC_TMICT)
 			    * APIC_BUS_CYCLE_NS * apic->divide_count;

arch/x86/kvm/mmu.c

@@ -556,6 +556,14 @@ static int mmu_spte_clear_track_bits(u64 *sptep)
 		return 0;
 
 	pfn = spte_to_pfn(old_spte);
+
+	/*
+	 * KVM does not hold the refcount of the page used by
+	 * kvm mmu, before reclaiming the page, we should
+	 * unmap it from mmu first.
+	 */
+	WARN_ON(!kvm_is_mmio_pfn(pfn) && !page_count(pfn_to_page(pfn)));
+
 	if (!shadow_accessed_mask || old_spte & shadow_accessed_mask)
 		kvm_set_pfn_accessed(pfn);
 	if (!shadow_dirty_mask || (old_spte & shadow_dirty_mask))
@@ -960,13 +968,13 @@ static void pte_list_walk(unsigned long *pte_list, pte_list_walk_fn fn)
 static unsigned long *__gfn_to_rmap(gfn_t gfn, int level,
 				    struct kvm_memory_slot *slot)
 {
-	struct kvm_lpage_info *linfo;
+	unsigned long idx;
 
 	if (likely(level == PT_PAGE_TABLE_LEVEL))
 		return &slot->rmap[gfn - slot->base_gfn];
 
-	linfo = lpage_info_slot(gfn, slot, level);
+	idx = gfn_to_index(gfn, slot->base_gfn, level);
-	return &linfo->rmap_pde;
+	return &slot->arch.rmap_pde[level - PT_DIRECTORY_LEVEL][idx];
 }
 
 /*
@@ -1200,7 +1208,7 @@ static bool rmap_write_protect(struct kvm *kvm, u64 gfn)
 }
 
 static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			   unsigned long data)
+			   struct kvm_memory_slot *slot, unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator iter;
@@ -1218,7 +1226,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, unsigned long *rmapp,
 }
 
 static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			     unsigned long data)
+			     struct kvm_memory_slot *slot, unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator iter;
@@ -1259,43 +1267,67 @@ static int kvm_set_pte_rmapp(struct kvm *kvm, unsigned long *rmapp,
 	return 0;
 }
 
-static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
+static int kvm_handle_hva_range(struct kvm *kvm,
+				unsigned long start,
+				unsigned long end,
 				unsigned long data,
-			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+				int (*handler)(struct kvm *kvm,
+					       unsigned long *rmapp,
+					       struct kvm_memory_slot *slot,
 					       unsigned long data))
 {
 	int j;
-	int ret;
+	int ret = 0;
-	int retval = 0;
 	struct kvm_memslots *slots;
 	struct kvm_memory_slot *memslot;
 
 	slots = kvm_memslots(kvm);
 
 	kvm_for_each_memslot(memslot, slots) {
-		unsigned long start = memslot->userspace_addr;
+		unsigned long hva_start, hva_end;
-		unsigned long end;
+		gfn_t gfn_start, gfn_end;
 
-		end = start + (memslot->npages << PAGE_SHIFT);
+		hva_start = max(start, memslot->userspace_addr);
-		if (hva >= start && hva < end) {
+		hva_end = min(end, memslot->userspace_addr +
-			gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT;
+					(memslot->npages << PAGE_SHIFT));
-			gfn_t gfn = memslot->base_gfn + gfn_offset;
+		if (hva_start >= hva_end)
+			continue;
+		/*
+		 * {gfn(page) | page intersects with [hva_start, hva_end)} =
+		 * {gfn_start, gfn_start+1, ..., gfn_end-1}.
+		 */
+		gfn_start = hva_to_gfn_memslot(hva_start, memslot);
+		gfn_end = hva_to_gfn_memslot(hva_end + PAGE_SIZE - 1, memslot);
 
-			ret = handler(kvm, &memslot->rmap[gfn_offset], data);
+		for (j = PT_PAGE_TABLE_LEVEL;
+		     j < PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES; ++j) {
+			unsigned long idx, idx_end;
+			unsigned long *rmapp;
 
-			for (j = 0; j < KVM_NR_PAGE_SIZES - 1; ++j) {
+			/*
-				struct kvm_lpage_info *linfo;
+			 * {idx(page_j) | page_j intersects with
+			 *  [hva_start, hva_end)} = {idx, idx+1, ..., idx_end}.
+			 */
+			idx = gfn_to_index(gfn_start, memslot->base_gfn, j);
+			idx_end = gfn_to_index(gfn_end - 1, memslot->base_gfn, j);
 
-				linfo = lpage_info_slot(gfn, memslot,
+			rmapp = __gfn_to_rmap(gfn_start, j, memslot);
-							PT_DIRECTORY_LEVEL + j);
+
-				ret |= handler(kvm, &linfo->rmap_pde, data);
+			for (; idx <= idx_end; ++idx)
-			}
+				ret |= handler(kvm, rmapp++, memslot, data);
-			trace_kvm_age_page(hva, memslot, ret);
-			retval |= ret;
 		}
 	}
 
-	return retval;
+	return ret;
+}
+
+static int kvm_handle_hva(struct kvm *kvm, unsigned long hva,
+			  unsigned long data,
+			  int (*handler)(struct kvm *kvm, unsigned long *rmapp,
+					 struct kvm_memory_slot *slot,
+					 unsigned long data))
+{
+	return kvm_handle_hva_range(kvm, hva, hva + 1, data, handler);
 }
 
 int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
@@ -1303,13 +1335,18 @@ int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
 	return kvm_handle_hva(kvm, hva, 0, kvm_unmap_rmapp);
 }
 
+int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end)
+{
+	return kvm_handle_hva_range(kvm, start, end, 0, kvm_unmap_rmapp);
+}
+
 void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
 {
 	kvm_handle_hva(kvm, hva, (unsigned long)&pte, kvm_set_pte_rmapp);
 }
 
 static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			 unsigned long data)
+			 struct kvm_memory_slot *slot, unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator uninitialized_var(iter);
@@ -1323,8 +1360,10 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
 	 * This has some overhead, but not as much as the cost of swapping
 	 * out actively used pages or breaking up actively used hugepages.
 	 */
-	if (!shadow_accessed_mask)
+	if (!shadow_accessed_mask) {
-		return kvm_unmap_rmapp(kvm, rmapp, data);
+		young = kvm_unmap_rmapp(kvm, rmapp, slot, data);
+		goto out;
+	}
 
 	for (sptep = rmap_get_first(*rmapp, &iter); sptep;
 	     sptep = rmap_get_next(&iter)) {
@@ -1336,12 +1375,14 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
 				 (unsigned long *)sptep);
 		}
 	}
-
+out:
+	/* @data has hva passed to kvm_age_hva(). */
+	trace_kvm_age_page(data, slot, young);
 	return young;
 }
 
 static int kvm_test_age_rmapp(struct kvm *kvm, unsigned long *rmapp,
-			      unsigned long data)
+			      struct kvm_memory_slot *slot, unsigned long data)
 {
 	u64 *sptep;
 	struct rmap_iterator iter;
@@ -1379,13 +1420,13 @@ static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn)
 
 	rmapp = gfn_to_rmap(vcpu->kvm, gfn, sp->role.level);
 
-	kvm_unmap_rmapp(vcpu->kvm, rmapp, 0);
+	kvm_unmap_rmapp(vcpu->kvm, rmapp, NULL, 0);
 	kvm_flush_remote_tlbs(vcpu->kvm);
 }
 
 int kvm_age_hva(struct kvm *kvm, unsigned long hva)
 {
-	return kvm_handle_hva(kvm, hva, 0, kvm_age_rmapp);
+	return kvm_handle_hva(kvm, hva, hva, kvm_age_rmapp);
 }
 
 int kvm_test_age_hva(struct kvm *kvm, unsigned long hva)
@@ -2472,14 +2513,12 @@ static pfn_t pte_prefetch_gfn_to_pfn(struct kvm_vcpu *vcpu, gfn_t gfn,
 	unsigned long hva;
 
 	slot = gfn_to_memslot_dirty_bitmap(vcpu, gfn, no_dirty_log);
-	if (!slot) {
+	if (!slot)
-		get_page(fault_page);
+		return get_fault_pfn();
-		return page_to_pfn(fault_page);
-	}
 
 	hva = gfn_to_hva_memslot(slot, gfn);
 
-	return hva_to_pfn_atomic(vcpu->kvm, hva);
+	return hva_to_pfn_atomic(hva);
 }
 
 static int direct_pte_prefetch_many(struct kvm_vcpu *vcpu,

arch/x86/kvm/pmu.c

@@ -1,5 +1,5 @@
 /*
- * Kernel-based Virtual Machine -- Performane Monitoring Unit support
+ * Kernel-based Virtual Machine -- Performance Monitoring Unit support
  *
  * Copyright 2011 Red Hat, Inc. and/or its affiliates.
  *

arch/x86/kvm/svm.c

@@ -2063,7 +2063,7 @@ static inline bool nested_svm_intr(struct vcpu_svm *svm)
 	if (svm->nested.intercept & 1ULL) {
 		/*
 		 * The #vmexit can't be emulated here directly because this
-		 * code path runs with irqs and preemtion disabled. A
+		 * code path runs with irqs and preemption disabled. A
 		 * #vmexit emulation might sleep. Only signal request for
 		 * the #vmexit here.
 		 */
@@ -2409,7 +2409,7 @@ static bool nested_svm_vmrun_msrpm(struct vcpu_svm *svm)
 {
 	/*
 	 * This function merges the msr permission bitmaps of kvm and the
-	 * nested vmcb. It is omptimized in that it only merges the parts where
+	 * nested vmcb. It is optimized in that it only merges the parts where
 	 * the kvm msr permission bitmap may contain zero bits
 	 */
 	int i;

arch/x86/kvm/vmx.c

@@ -1343,7 +1343,7 @@ static bool update_transition_efer(struct vcpu_vmx *vmx, int efer_offset)
 	guest_efer = vmx->vcpu.arch.efer;
 
 	/*
-	 * NX is emulated; LMA and LME handled by hardware; SCE meaninless
+	 * NX is emulated; LMA and LME handled by hardware; SCE meaningless
 	 * outside long mode
 	 */
 	ignore_bits = EFER_NX | EFER_SCE;
@@ -3261,7 +3261,7 @@ static void vmx_set_segment(struct kvm_vcpu *vcpu,
 	 * qemu binaries.
 	 *   IA32 arch specifies that at the time of processor reset the
 	 * "Accessed" bit in the AR field of segment registers is 1. And qemu
-	 * is setting it to 0 in the usedland code. This causes invalid guest
+	 * is setting it to 0 in the userland code. This causes invalid guest
 	 * state vmexit when "unrestricted guest" mode is turned on.
 	 *    Fix for this setup issue in cpu_reset is being pushed in the qemu
 	 * tree. Newer qemu binaries with that qemu fix would not need this
@@ -4446,7 +4446,7 @@ vmx_patch_hypercall(struct kvm_vcpu *vcpu, unsigned char *hypercall)
 	hypercall[2] = 0xc1;
 }
 
-/* called to set cr0 as approriate for a mov-to-cr0 exit. */
+/* called to set cr0 as appropriate for a mov-to-cr0 exit. */
 static int handle_set_cr0(struct kvm_vcpu *vcpu, unsigned long val)
 {
 	if (to_vmx(vcpu)->nested.vmxon &&

arch/x86/kvm/x86.c

@@ -1093,7 +1093,7 @@ void kvm_write_tsc(struct kvm_vcpu *vcpu, u64 data)
 		 * For each generation, we track the original measured
 		 * nanosecond time, offset, and write, so if TSCs are in
 		 * sync, we can match exact offset, and if not, we can match
-		 * exact software computaion in compute_guest_tsc()
+		 * exact software computation in compute_guest_tsc()
 		 *
 		 * These values are tracked in kvm->arch.cur_xxx variables.
 		 */
@@ -1500,7 +1500,7 @@ static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data)
 {
 	gpa_t gpa = data & ~0x3f;
 
-	/* Bits 2:5 are resrved, Should be zero */
+	/* Bits 2:5 are reserved, Should be zero */
 	if (data & 0x3c)
 		return 1;
 
@@ -1723,7 +1723,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data)
 		 * Ignore all writes to this no longer documented MSR.
 		 * Writes are only relevant for old K7 processors,
 		 * all pre-dating SVM, but a recommended workaround from
-		 * AMD for these chips. It is possible to speicify the
+		 * AMD for these chips. It is possible to specify the
 		 * affected processor models on the command line, hence
 		 * the need to ignore the workaround.
 		 */
@@ -2632,7 +2632,6 @@ static int kvm_set_guest_paused(struct kvm_vcpu *vcpu)
 	if (!vcpu->arch.time_page)
 		return -EINVAL;
 	src->flags |= PVCLOCK_GUEST_STOPPED;
-	mark_page_dirty(vcpu->kvm, vcpu->arch.time >> PAGE_SHIFT);
 	kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
 	return 0;
 }
@@ -4492,7 +4491,7 @@ static bool reexecute_instruction(struct kvm_vcpu *vcpu, gva_t gva)
 
 	/*
 	 * if emulation was due to access to shadowed page table
-	 * and it failed try to unshadow page and re-entetr the
+	 * and it failed try to unshadow page and re-enter the
 	 * guest to let CPU execute the instruction.
 	 */
 	if (kvm_mmu_unprotect_page_virt(vcpu, gva))
@@ -5588,7 +5587,7 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 		/*
 		 * We are here if userspace calls get_regs() in the middle of
 		 * instruction emulation. Registers state needs to be copied
-		 * back from emulation context to vcpu. Usrapace shouldn't do
+		 * back from emulation context to vcpu. Userspace shouldn't do
 		 * that usually, but some bad designed PV devices (vmware
 		 * backdoor interface) need this to work
 		 */
@@ -6117,7 +6116,7 @@ int kvm_arch_hardware_enable(void *garbage)
 	 * as we reset last_host_tsc on all VCPUs to stop this from being
 	 * called multiple times (one for each physical CPU bringup).
 	 *
-	 * Platforms with unnreliable TSCs don't have to deal with this, they
+	 * Platforms with unreliable TSCs don't have to deal with this, they
 	 * will be compensated by the logic in vcpu_load, which sets the TSC to
 	 * catchup mode.  This will catchup all VCPUs to real time, but cannot
 	 * guarantee that they stay in perfect synchronization.
@@ -6314,6 +6313,10 @@ void kvm_arch_free_memslot(struct kvm_memory_slot *free,
 	int i;
 
 	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
+		if (!dont || free->arch.rmap_pde[i] != dont->arch.rmap_pde[i]) {
+			kvm_kvfree(free->arch.rmap_pde[i]);
+			free->arch.rmap_pde[i] = NULL;
+		}
 		if (!dont || free->arch.lpage_info[i] != dont->arch.lpage_info[i]) {
 			kvm_kvfree(free->arch.lpage_info[i]);
 			free->arch.lpage_info[i] = NULL;
@@ -6333,6 +6336,11 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
 		lpages = gfn_to_index(slot->base_gfn + npages - 1,
 				      slot->base_gfn, level) + 1;
 
+		slot->arch.rmap_pde[i] =
+			kvm_kvzalloc(lpages * sizeof(*slot->arch.rmap_pde[i]));
+		if (!slot->arch.rmap_pde[i])
+			goto out_free;
+
 		slot->arch.lpage_info[i] =
 			kvm_kvzalloc(lpages * sizeof(*slot->arch.lpage_info[i]));
 		if (!slot->arch.lpage_info[i])
@@ -6361,7 +6369,9 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
 
 out_free:
 	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
+		kvm_kvfree(slot->arch.rmap_pde[i]);
 		kvm_kvfree(slot->arch.lpage_info[i]);
+		slot->arch.rmap_pde[i] = NULL;
 		slot->arch.lpage_info[i] = NULL;
 	}
 	return -ENOMEM;
@@ -6381,7 +6391,7 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 		map_flags = MAP_SHARED | MAP_ANONYMOUS;
 
 	/*To keep backward compatibility with older userspace,
-	 *x86 needs to hanlde !user_alloc case.
+	 *x86 needs to handle !user_alloc case.
 	 */
 	if (!user_alloc) {
 		if (npages && !old.rmap) {

include/linux/kvm_host.h

@@ -183,6 +183,18 @@ struct kvm_vcpu {
 	} async_pf;
 #endif
 
+#ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
+	/*
+	 * Cpu relax intercept or pause loop exit optimization
+	 * in_spin_loop: set when a vcpu does a pause loop exit
+	 *  or cpu relax intercepted.
+	 * dy_eligible: indicates whether vcpu is eligible for directed yield.
+	 */
+	struct {
+		bool in_spin_loop;
+		bool dy_eligible;
+	} spin_loop;
+#endif
 	struct kvm_vcpu_arch arch;
 };
 
@@ -378,20 +390,11 @@ id_to_memslot(struct kvm_memslots *slots, int id)
 	return slot;
 }
 
-#define HPA_MSB ((sizeof(hpa_t) * 8) - 1)
-#define HPA_ERR_MASK ((hpa_t)1 << HPA_MSB)
-static inline int is_error_hpa(hpa_t hpa) { return hpa >> HPA_MSB; }
-
 extern struct page *bad_page;
-extern struct page *fault_page;
-
-extern pfn_t bad_pfn;
-extern pfn_t fault_pfn;
 
 int is_error_page(struct page *page);
 int is_error_pfn(pfn_t pfn);
 int is_hwpoison_pfn(pfn_t pfn);
-int is_fault_pfn(pfn_t pfn);
 int is_noslot_pfn(pfn_t pfn);
 int is_invalid_pfn(pfn_t pfn);
 int kvm_is_error_hva(unsigned long addr);
@@ -427,20 +430,20 @@ void kvm_release_page_dirty(struct page *page);
 void kvm_set_page_dirty(struct page *page);
 void kvm_set_page_accessed(struct page *page);
 
-pfn_t hva_to_pfn_atomic(struct kvm *kvm, unsigned long addr);
+pfn_t hva_to_pfn_atomic(unsigned long addr);
 pfn_t gfn_to_pfn_atomic(struct kvm *kvm, gfn_t gfn);
 pfn_t gfn_to_pfn_async(struct kvm *kvm, gfn_t gfn, bool *async,
 		       bool write_fault, bool *writable);
 pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn);
 pfn_t gfn_to_pfn_prot(struct kvm *kvm, gfn_t gfn, bool write_fault,
 		      bool *writable);
-pfn_t gfn_to_pfn_memslot(struct kvm *kvm,
+pfn_t gfn_to_pfn_memslot(struct kvm_memory_slot *slot, gfn_t gfn);
-			 struct kvm_memory_slot *slot, gfn_t gfn);
 void kvm_release_pfn_dirty(pfn_t);
 void kvm_release_pfn_clean(pfn_t pfn);
 void kvm_set_pfn_dirty(pfn_t pfn);
 void kvm_set_pfn_accessed(pfn_t pfn);
 void kvm_get_pfn(pfn_t pfn);
+pfn_t get_fault_pfn(void);
 
 int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset,
 			int len);
@@ -740,6 +743,14 @@ static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
 		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
 }
 
+static inline gfn_t
+hva_to_gfn_memslot(unsigned long hva, struct kvm_memory_slot *slot)
+{
+	gfn_t gfn_offset = (hva - slot->userspace_addr) >> PAGE_SHIFT;
+
+	return slot->base_gfn + gfn_offset;
+}
+
 static inline unsigned long gfn_to_hva_memslot(struct kvm_memory_slot *slot,
 					       gfn_t gfn)
 {
@@ -899,5 +910,32 @@ static inline bool kvm_check_request(int req, struct kvm_vcpu *vcpu)
 	}
 }
 
+#ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
+
+static inline void kvm_vcpu_set_in_spin_loop(struct kvm_vcpu *vcpu, bool val)
+{
+	vcpu->spin_loop.in_spin_loop = val;
+}
+static inline void kvm_vcpu_set_dy_eligible(struct kvm_vcpu *vcpu, bool val)
+{
+	vcpu->spin_loop.dy_eligible = val;
+}
+
+#else /* !CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT */
+
+static inline void kvm_vcpu_set_in_spin_loop(struct kvm_vcpu *vcpu, bool val)
+{
+}
+
+static inline void kvm_vcpu_set_dy_eligible(struct kvm_vcpu *vcpu, bool val)
+{
+}
+
+static inline bool kvm_vcpu_eligible_for_directed_yield(struct kvm_vcpu *vcpu)
+{
+	return true;
+}
+
+#endif /* CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT */
 #endif
 

virt/kvm/Kconfig

@@ -21,3 +21,6 @@ config KVM_ASYNC_PF
 
 config HAVE_KVM_MSI
        bool
+
+config HAVE_KVM_CPU_RELAX_INTERCEPT
+       bool

virt/kvm/iommu.c

@@ -42,13 +42,13 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm);
 static void kvm_iommu_put_pages(struct kvm *kvm,
 				gfn_t base_gfn, unsigned long npages);
 
-static pfn_t kvm_pin_pages(struct kvm *kvm, struct kvm_memory_slot *slot,
+static pfn_t kvm_pin_pages(struct kvm_memory_slot *slot, gfn_t gfn,
-			   gfn_t gfn, unsigned long size)
+			   unsigned long size)
 {
 	gfn_t end_gfn;
 	pfn_t pfn;
 
-	pfn     = gfn_to_pfn_memslot(kvm, slot, gfn);
+	pfn     = gfn_to_pfn_memslot(slot, gfn);
 	end_gfn = gfn + (size >> PAGE_SHIFT);
 	gfn    += 1;
 
@@ -56,7 +56,7 @@ static pfn_t kvm_pin_pages(struct kvm *kvm, struct kvm_memory_slot *slot,
 		return pfn;
 
 	while (gfn < end_gfn)
-		gfn_to_pfn_memslot(kvm, slot, gfn++);
+		gfn_to_pfn_memslot(slot, gfn++);
 
 	return pfn;
 }
@@ -105,7 +105,7 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
 		 * Pin all pages we are about to map in memory. This is
 		 * important because we unmap and unpin in 4kb steps later.
 		 */
-		pfn = kvm_pin_pages(kvm, slot, gfn, page_size);
+		pfn = kvm_pin_pages(slot, gfn, page_size);
 		if (is_error_pfn(pfn)) {
 			gfn += 1;
 			continue;

virt/kvm/irq_comm.c

@@ -321,11 +321,11 @@ static int setup_routing_entry(struct kvm_irq_routing_table *rt,
 		switch (ue->u.irqchip.irqchip) {
 		case KVM_IRQCHIP_PIC_MASTER:
 			e->set = kvm_set_pic_irq;
-			max_pin = 16;
+			max_pin = PIC_NUM_PINS;
 			break;
 		case KVM_IRQCHIP_PIC_SLAVE:
 			e->set = kvm_set_pic_irq;
-			max_pin = 16;
+			max_pin = PIC_NUM_PINS;
 			delta = 8;
 			break;
 		case KVM_IRQCHIP_IOAPIC:

virt/kvm/kvm_main.c

@@ -100,11 +100,14 @@ EXPORT_SYMBOL_GPL(kvm_rebooting);
 
 static bool largepages_enabled = true;
 
+struct page *bad_page;
+static pfn_t bad_pfn;
+
 static struct page *hwpoison_page;
 static pfn_t hwpoison_pfn;
 
-struct page *fault_page;
+static struct page *fault_page;
-pfn_t fault_pfn;
+static pfn_t fault_pfn;
 
 inline int kvm_is_mmio_pfn(pfn_t pfn)
 {
@@ -236,6 +239,9 @@ int kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
 	}
 	vcpu->run = page_address(page);
 
+	kvm_vcpu_set_in_spin_loop(vcpu, false);
+	kvm_vcpu_set_dy_eligible(vcpu, false);
+
 	r = kvm_arch_vcpu_init(vcpu);
 	if (r < 0)
 		goto fail_free_run;
@@ -332,8 +338,7 @@ static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 	 * count is also read inside the mmu_lock critical section.
 	 */
 	kvm->mmu_notifier_count++;
-	for (; start < end; start += PAGE_SIZE)
+	need_tlb_flush = kvm_unmap_hva_range(kvm, start, end);
-		need_tlb_flush |= kvm_unmap_hva(kvm, start);
 	need_tlb_flush |= kvm->tlbs_dirty;
 	/* we've to flush the tlb before the pages can be freed */
 	if (need_tlb_flush)
@@ -950,12 +955,6 @@ int is_hwpoison_pfn(pfn_t pfn)
 }
 EXPORT_SYMBOL_GPL(is_hwpoison_pfn);
 
-int is_fault_pfn(pfn_t pfn)
-{
-	return pfn == fault_pfn;
-}
-EXPORT_SYMBOL_GPL(is_fault_pfn);
-
 int is_noslot_pfn(pfn_t pfn)
 {
 	return pfn == bad_pfn;
@@ -1039,11 +1038,12 @@ unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn)
 }
 EXPORT_SYMBOL_GPL(gfn_to_hva);
 
-static pfn_t get_fault_pfn(void)
+pfn_t get_fault_pfn(void)
 {
 	get_page(fault_page);
 	return fault_pfn;
 }
+EXPORT_SYMBOL_GPL(get_fault_pfn);
 
 int get_user_page_nowait(struct task_struct *tsk, struct mm_struct *mm,
 	unsigned long start, int write, struct page **page)
@@ -1065,8 +1065,8 @@ static inline int check_user_page_hwpoison(unsigned long addr)
 	return rc == -EHWPOISON;
 }
 
-static pfn_t hva_to_pfn(struct kvm *kvm, unsigned long addr, bool atomic,
+static pfn_t hva_to_pfn(unsigned long addr, bool atomic, bool *async,
-			bool *async, bool write_fault, bool *writable)
+			bool write_fault, bool *writable)
 {
 	struct page *page[1];
 	int npages = 0;
@@ -1146,9 +1146,9 @@ static pfn_t hva_to_pfn(struct kvm *kvm, unsigned long addr, bool atomic,
 	return pfn;
 }
 
-pfn_t hva_to_pfn_atomic(struct kvm *kvm, unsigned long addr)
+pfn_t hva_to_pfn_atomic(unsigned long addr)
 {
-	return hva_to_pfn(kvm, addr, true, NULL, true, NULL);
+	return hva_to_pfn(addr, true, NULL, true, NULL);
 }
 EXPORT_SYMBOL_GPL(hva_to_pfn_atomic);
 
@@ -1166,7 +1166,7 @@ static pfn_t __gfn_to_pfn(struct kvm *kvm, gfn_t gfn, bool atomic, bool *async,
 		return page_to_pfn(bad_page);
 	}
 
-	return hva_to_pfn(kvm, addr, atomic, async, write_fault, writable);
+	return hva_to_pfn(addr, atomic, async, write_fault, writable);
 }
 
 pfn_t gfn_to_pfn_atomic(struct kvm *kvm, gfn_t gfn)
@@ -1195,11 +1195,10 @@ pfn_t gfn_to_pfn_prot(struct kvm *kvm, gfn_t gfn, bool write_fault,
 }
 EXPORT_SYMBOL_GPL(gfn_to_pfn_prot);
 
-pfn_t gfn_to_pfn_memslot(struct kvm *kvm,
+pfn_t gfn_to_pfn_memslot(struct kvm_memory_slot *slot, gfn_t gfn)
-			 struct kvm_memory_slot *slot, gfn_t gfn)
 {
 	unsigned long addr = gfn_to_hva_memslot(slot, gfn);
-	return hva_to_pfn(kvm, addr, false, NULL, true, NULL);
+	return hva_to_pfn(addr, false, NULL, true, NULL);
 }
 
 int gfn_to_page_many_atomic(struct kvm *kvm, gfn_t gfn, struct page **pages,
@@ -1580,6 +1579,43 @@ bool kvm_vcpu_yield_to(struct kvm_vcpu *target)
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_yield_to);
 
+#ifdef CONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT
+/*
+ * Helper that checks whether a VCPU is eligible for directed yield.
+ * Most eligible candidate to yield is decided by following heuristics:
+ *
+ *  (a) VCPU which has not done pl-exit or cpu relax intercepted recently
+ *  (preempted lock holder), indicated by @in_spin_loop.
+ *  Set at the beiginning and cleared at the end of interception/PLE handler.
+ *
+ *  (b) VCPU which has done pl-exit/ cpu relax intercepted but did not get
+ *  chance last time (mostly it has become eligible now since we have probably
+ *  yielded to lockholder in last iteration. This is done by toggling
+ *  @dy_eligible each time a VCPU checked for eligibility.)
+ *
+ *  Yielding to a recently pl-exited/cpu relax intercepted VCPU before yielding
+ *  to preempted lock-holder could result in wrong VCPU selection and CPU
+ *  burning. Giving priority for a potential lock-holder increases lock
+ *  progress.
+ *
+ *  Since algorithm is based on heuristics, accessing another VCPU data without
+ *  locking does not harm. It may result in trying to yield to  same VCPU, fail
+ *  and continue with next VCPU and so on.
+ */
+bool kvm_vcpu_eligible_for_directed_yield(struct kvm_vcpu *vcpu)
+{
+	bool eligible;
+
+	eligible = !vcpu->spin_loop.in_spin_loop ||
+			(vcpu->spin_loop.in_spin_loop &&
+			 vcpu->spin_loop.dy_eligible);
+
+	if (vcpu->spin_loop.in_spin_loop)
+		kvm_vcpu_set_dy_eligible(vcpu, !vcpu->spin_loop.dy_eligible);
+
+	return eligible;
+}
+#endif
 void kvm_vcpu_on_spin(struct kvm_vcpu *me)
 {
 	struct kvm *kvm = me->kvm;
@@ -1589,6 +1625,7 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me)
 	int pass;
 	int i;
 
+	kvm_vcpu_set_in_spin_loop(me, true);
 	/*
 	 * We boost the priority of a VCPU that is runnable but not
 	 * currently running, because it got preempted by something
@@ -1607,6 +1644,8 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me)
 				continue;
 			if (waitqueue_active(&vcpu->wq))
 				continue;
+			if (!kvm_vcpu_eligible_for_directed_yield(vcpu))
+				continue;
 			if (kvm_vcpu_yield_to(vcpu)) {
 				kvm->last_boosted_vcpu = i;
 				yielded = 1;
@@ -1614,6 +1653,10 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me)
 			}
 		}
 	}
+	kvm_vcpu_set_in_spin_loop(me, false);
+
+	/* Ensure vcpu is not eligible during next spinloop */
+	kvm_vcpu_set_dy_eligible(me, false);
 }
 EXPORT_SYMBOL_GPL(kvm_vcpu_on_spin);
 
@@ -2697,9 +2740,6 @@ static struct syscore_ops kvm_syscore_ops = {
 	.resume = kvm_resume,
 };
 
-struct page *bad_page;
-pfn_t bad_pfn;
-
 static inline
 struct kvm_vcpu *preempt_notifier_to_vcpu(struct preempt_notifier *pn)
 {