[PATCH 0/4] KVM: Introduce kvm_memory_slot::arch

[PATCH 0/4] KVM: Introduce kvm_memory_slot::arch

[Takuya Yoshikawa]

This is the first step to separate the architecture specific members.
The rmap and dirty_bitmap can be treated later based on this.

	Takuya

 arch/ia64/include/asm/kvm_host.h    |    3 +
 arch/ia64/kvm/kvm-ia64.c            |   10 ++++
 arch/powerpc/include/asm/kvm_host.h |    3 +
 arch/powerpc/kvm/powerpc.c          |   10 ++++
 arch/s390/include/asm/kvm_host.h    |    3 +
 arch/s390/kvm/kvm-s390.c            |   10 ++++
 arch/x86/include/asm/kvm_host.h     |    9 ++++
 arch/x86/kvm/mmu.c                  |    5 +-
 arch/x86/kvm/x86.c                  |   62 +++++++++++++++++++++++++
 include/linux/kvm_host.h            |   18 +++++---
 virt/kvm/kvm_main.c                 |   86 ++++++++---------------------------
 11 files changed, 144 insertions(+), 75 deletions(-)

-- 
1.7.5.4

[PATCH 1/4] KVM: Introduce gfn_to_index() which returns the index for a given level

[Takuya Yoshikawa]

From: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>

This patch cleans up the code and removes the "(void)level;" warning
suppressor.

Note that we can also use this for PT_PAGE_TABLE_LEVEL to treat every
level uniformly later.

Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
---
 arch/x86/kvm/mmu.c       |    3 +--
 include/linux/kvm_host.h |    7 +++++++
 virt/kvm/kvm_main.c      |    7 +------
 3 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index ae76cc3..37e7f10 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -688,8 +688,7 @@ static struct kvm_lpage_info *lpage_info_slot(gfn_t gfn,
 {
 	unsigned long idx;
 
-	idx = (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
-	      (slot->base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
+	idx = gfn_to_index(gfn, slot->base_gfn, level);
 	return &slot->lpage_info[level - 2][idx];
 }
 
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index eada8e6..06d4e41 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -656,6 +656,13 @@ static inline int memslot_id(struct kvm *kvm, gfn_t gfn)
 	return gfn_to_memslot(kvm, gfn)->id;
 }
 
+static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level)
+{
+	/* KVM_HPAGE_GFN_SHIFT(PT_PAGE_TABLE_LEVEL) must be 0. */
+	return (gfn >> KVM_HPAGE_GFN_SHIFT(level)) -
+		(base_gfn >> KVM_HPAGE_GFN_SHIFT(level));
+}
+
 static inline unsigned long gfn_to_hva_memslot(struct kvm_memory_slot *slot,
 					       gfn_t gfn)
 {
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 9f32bff..e483ae4 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -797,15 +797,10 @@ int __kvm_set_memory_region(struct kvm *kvm,
 		int lpages;
 		int level = i + 2;
 
-		/* Avoid unused variable warning if no large pages */
-		(void)level;
-
 		if (new.lpage_info[i])
 			continue;
 
-		lpages = 1 + ((base_gfn + npages - 1)
-			     >> KVM_HPAGE_GFN_SHIFT(level));
-		lpages -= base_gfn >> KVM_HPAGE_GFN_SHIFT(level);
+		lpages = gfn_to_index(base_gfn + npages - 1, base_gfn, level) + 1;
 
 		new.lpage_info[i] = vzalloc(lpages * sizeof(*new.lpage_info[i]));
 
-- 
1.7.5.4

[PATCH 2/4] KVM: Split lpage_info creation out from __kvm_set_memory_region()

[Takuya Yoshikawa]

From: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>

This makes it easy to make lpage_info architecture specific.

Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
---
 virt/kvm/kvm_main.c |   83 ++++++++++++++++++++++++++++++++-------------------
 1 files changed, 52 insertions(+), 31 deletions(-)

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index e483ae4..2ef9f2d 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -698,6 +698,56 @@ void update_memslots(struct kvm_memslots *slots, struct kvm_memory_slot *new)
 	slots->generation++;
 }
 
+#ifndef CONFIG_S390
+static int create_lpage_info(struct kvm_memory_slot *slot, unsigned long npages)
+{
+	int i;
+
+	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
+		unsigned long ugfn;
+		int lpages;
+		int level = i + 2;
+
+		if (slot->lpage_info[i])
+			continue;
+
+		lpages = gfn_to_index(slot->base_gfn + npages - 1,
+				      slot->base_gfn, level) + 1;
+
+		slot->lpage_info[i] = vzalloc(lpages * sizeof(*slot->lpage_info[i]));
+		if (!slot->lpage_info[i])
+			goto out_free;
+
+		if (slot->base_gfn & (KVM_PAGES_PER_HPAGE(level) - 1))
+			slot->lpage_info[i][0].write_count = 1;
+		if ((slot->base_gfn + npages) & (KVM_PAGES_PER_HPAGE(level) - 1))
+			slot->lpage_info[i][lpages - 1].write_count = 1;
+		ugfn = slot->userspace_addr >> PAGE_SHIFT;
+		/*
+		 * If the gfn and userspace address are not aligned wrt each
+		 * other, or if explicitly asked to, disable large page
+		 * support for this slot
+		 */
+		if ((slot->base_gfn ^ ugfn) & (KVM_PAGES_PER_HPAGE(level) - 1) ||
+		    !largepages_enabled) {
+			unsigned long j;
+
+			for (j = 0; j < lpages; ++j)
+				slot->lpage_info[i][j].write_count = 1;
+		}
+	}
+
+	return 0;
+
+out_free:
+	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
+		vfree(slot->lpage_info[i]);
+		slot->lpage_info[i] = NULL;
+	}
+	return -ENOMEM;
+}
+#endif /* not defined CONFIG_S390 */
+
 /*
  * Allocate some memory and give it an address in the guest physical address
  * space.
@@ -791,37 +841,8 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	if (!npages)
 		goto skip_lpage;
 
-	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
-		unsigned long ugfn;
-		unsigned long j;
-		int lpages;
-		int level = i + 2;
-
-		if (new.lpage_info[i])
-			continue;
-
-		lpages = gfn_to_index(base_gfn + npages - 1, base_gfn, level) + 1;
-
-		new.lpage_info[i] = vzalloc(lpages * sizeof(*new.lpage_info[i]));
-
-		if (!new.lpage_info[i])
-			goto out_free;
-
-		if (base_gfn & (KVM_PAGES_PER_HPAGE(level) - 1))
-			new.lpage_info[i][0].write_count = 1;
-		if ((base_gfn+npages) & (KVM_PAGES_PER_HPAGE(level) - 1))
-			new.lpage_info[i][lpages - 1].write_count = 1;
-		ugfn = new.userspace_addr >> PAGE_SHIFT;
-		/*
-		 * If the gfn and userspace address are not aligned wrt each
-		 * other, or if explicitly asked to, disable large page
-		 * support for this slot
-		 */
-		if ((base_gfn ^ ugfn) & (KVM_PAGES_PER_HPAGE(level) - 1) ||
-		    !largepages_enabled)
-			for (j = 0; j < lpages; ++j)
-				new.lpage_info[i][j].write_count = 1;
-	}
+	if (create_lpage_info(&new, npages))
+		goto out_free;
 
 skip_lpage:
 
-- 
1.7.5.4

[PATCH 3/4] KVM: Simplify ifndef conditional usage in __kvm_set_memory_region()

[Takuya Yoshikawa]

From: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>

Narrow down the controlled text inside the conditional so that it will
include lpage_info and rmap stuff only.

For this we change the way we check whether the slot is being created
from "if (npages && !new.rmap)" to "if (npages && !old.npages)".

Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
---
 virt/kvm/kvm_main.c |   34 +++++++++++++---------------------
 1 files changed, 13 insertions(+), 21 deletions(-)

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 2ef9f2d..ee233ce 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -616,7 +616,6 @@ static int kvm_vm_release(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-#ifndef CONFIG_S390
 /*
  * Allocation size is twice as large as the actual dirty bitmap size.
  * This makes it possible to do double buffering: see x86's
@@ -624,6 +623,7 @@ static int kvm_vm_release(struct inode *inode, struct file *filp)
  */
 static int kvm_create_dirty_bitmap(struct kvm_memory_slot *memslot)
 {
+#ifndef CONFIG_S390
 	unsigned long dirty_bytes = 2 * kvm_dirty_bitmap_bytes(memslot);
 
 	if (dirty_bytes > PAGE_SIZE)
@@ -636,9 +636,9 @@ static int kvm_create_dirty_bitmap(struct kvm_memory_slot *memslot)
 
 	memslot->dirty_bitmap_head = memslot->dirty_bitmap;
 	memslot->nr_dirty_pages = 0;
+#endif /* !CONFIG_S390 */
 	return 0;
 }
-#endif /* !CONFIG_S390 */
 
 static struct kvm_memory_slot *
 search_memslots(struct kvm_memslots *slots, gfn_t gfn)
@@ -828,23 +828,20 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	r = -ENOMEM;
 
 	/* Allocate if a slot is being created */
+	if (npages && !old.npages) {
+		new.user_alloc = user_alloc;
+		if (user_alloc)
+			new.userspace_addr = mem->userspace_addr;
 #ifndef CONFIG_S390
-	if (npages && !new.rmap) {
-		new.rmap = vzalloc(npages * sizeof(*new.rmap));
-
-		if (!new.rmap)
+		if (!new.rmap) {
+			new.rmap = vzalloc(npages * sizeof(*new.rmap));
+			if (!new.rmap)
+				goto out_free;
+		}
+		if (create_lpage_info(&new, npages))
 			goto out_free;
-
-		new.user_alloc = user_alloc;
-		new.userspace_addr = mem->userspace_addr;
+#endif /* not defined CONFIG_S390 */
 	}
-	if (!npages)
-		goto skip_lpage;
-
-	if (create_lpage_info(&new, npages))
-		goto out_free;
-
-skip_lpage:
 
 	/* Allocate page dirty bitmap if needed */
 	if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) {
@@ -852,11 +849,6 @@ skip_lpage:
 			goto out_free;
 		/* destroy any largepage mappings for dirty tracking */
 	}
-#else  /* not defined CONFIG_S390 */
-	new.user_alloc = user_alloc;
-	if (user_alloc)
-		new.userspace_addr = mem->userspace_addr;
-#endif /* not defined CONFIG_S390 */
 
 	if (!npages) {
 		struct kvm_memory_slot *slot;
-- 
1.7.5.4

[PATCH 4/4] KVM: Introduce kvm_memory_slot::arch and move lpage_info into it

[Takuya Yoshikawa]

From: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>

Some members added recently to kvm_memory_slot are not used by every
architectures.

This patch is the first step to make this difference clear by
introducing kvm_memory_slot::arch;  lpage_info is moved into it.

Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
---
 arch/ia64/include/asm/kvm_host.h    |    3 +
 arch/ia64/kvm/kvm-ia64.c            |   10 +++++
 arch/powerpc/include/asm/kvm_host.h |    3 +
 arch/powerpc/kvm/powerpc.c          |   10 +++++
 arch/s390/include/asm/kvm_host.h    |    3 +
 arch/s390/kvm/kvm-s390.c            |   10 +++++
 arch/x86/include/asm/kvm_host.h     |    9 ++++
 arch/x86/kvm/mmu.c                  |    2 +-
 arch/x86/kvm/x86.c                  |   62 +++++++++++++++++++++++++++++++
 include/linux/kvm_host.h            |   11 ++---
 virt/kvm/kvm_main.c                 |   70 ++++-------------------------------
 11 files changed, 124 insertions(+), 69 deletions(-)

diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index 2689ee5..e35b3a8 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -459,6 +459,9 @@ struct kvm_sal_data {
 	unsigned long boot_gp;
 };
 
+struct kvm_arch_memory_slot {
+};
+
 struct kvm_arch {
 	spinlock_t dirty_log_lock;
 
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 8ca7261..d8ddbba 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1571,6 +1571,16 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
+void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+			   struct kvm_memory_slot *dont)
+{
+}
+
+int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+{
+	return 0;
+}
+
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 		struct kvm_memory_slot *memslot,
 		struct kvm_memory_slot old,
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index af438b1..b9188aa 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -212,6 +212,9 @@ struct revmap_entry {
 #define KVMPPC_PAGE_WRITETHRU	HPTE_R_W	/* 0x40 */
 #define KVMPPC_GOT_PAGE		0x80
 
+struct kvm_arch_memory_slot {
+};
+
 struct kvm_arch {
 #ifdef CONFIG_KVM_BOOK3S_64_HV
 	unsigned long hpt_virt;
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 0e21d15..00d7e34 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -281,6 +281,16 @@ long kvm_arch_dev_ioctl(struct file *filp,
 	return -EINVAL;
 }
 
+void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+			   struct kvm_memory_slot *dont)
+{
+}
+
+int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+{
+	return 0;
+}
+
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
                                    struct kvm_memory_slot *memslot,
                                    struct kvm_memory_slot old,
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index e630426..7343872 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -245,6 +245,9 @@ struct kvm_vm_stat {
 	u32 remote_tlb_flush;
 };
 
+struct kvm_arch_memory_slot {
+};
+
 struct kvm_arch{
 	struct sca_block *sca;
 	debug_info_t *dbf;
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 0b91679..418a69c 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -807,6 +807,16 @@ int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf)
 	return VM_FAULT_SIGBUS;
 }
 
+void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+			   struct kvm_memory_slot *dont)
+{
+}
+
+int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+{
+	return 0;
+}
+
 /* Section: memory related */
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				   struct kvm_memory_slot *memslot,
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 4610166..de3aa43 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -479,6 +479,15 @@ struct kvm_vcpu_arch {
 	} osvw;
 };
 
+struct kvm_lpage_info {
+	unsigned long rmap_pde;
+	int write_count;
+};
+
+struct kvm_arch_memory_slot {
+	struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1];
+};
+
 struct kvm_arch {
 	unsigned int n_used_mmu_pages;
 	unsigned int n_requested_mmu_pages;
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 37e7f10..ff053ca 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -689,7 +689,7 @@ static struct kvm_lpage_info *lpage_info_slot(gfn_t gfn,
 	unsigned long idx;
 
 	idx = gfn_to_index(gfn, slot->base_gfn, level);
-	return &slot->lpage_info[level - 2][idx];
+	return &slot->arch.lpage_info[level - 2][idx];
 }
 
 static void account_shadowed(struct kvm *kvm, gfn_t gfn)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 2bd77a3..bbb267b 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -6120,6 +6120,68 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 		put_page(kvm->arch.ept_identity_pagetable);
 }
 
+void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+			   struct kvm_memory_slot *dont)
+{
+	int i;
+
+	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
+		if (!dont || free->arch.lpage_info[i] != dont->arch.lpage_info[i]) {
+			vfree(free->arch.lpage_info[i]);
+			free->arch.lpage_info[i] = NULL;
+		}
+	}
+}
+
+int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages)
+{
+	int i;
+
+	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
+		unsigned long ugfn;
+		int lpages;
+		int level = i + 2;
+
+		if (slot->arch.lpage_info[i])
+			continue;
+
+		lpages = gfn_to_index(slot->base_gfn + npages - 1,
+				      slot->base_gfn, level) + 1;
+
+		slot->arch.lpage_info[i] =
+			vzalloc(lpages * sizeof(*slot->arch.lpage_info[i]));
+		if (!slot->arch.lpage_info[i])
+			goto out_free;
+
+		if (slot->base_gfn & (KVM_PAGES_PER_HPAGE(level) - 1))
+			slot->arch.lpage_info[i][0].write_count = 1;
+		if ((slot->base_gfn + npages) & (KVM_PAGES_PER_HPAGE(level) - 1))
+			slot->arch.lpage_info[i][lpages - 1].write_count = 1;
+		ugfn = slot->userspace_addr >> PAGE_SHIFT;
+		/*
+		 * If the gfn and userspace address are not aligned wrt each
+		 * other, or if explicitly asked to, disable large page
+		 * support for this slot
+		 */
+		if ((slot->base_gfn ^ ugfn) & (KVM_PAGES_PER_HPAGE(level) - 1) ||
+		    !kvm_largepages_enabled()) {
+			unsigned long j;
+
+			for (j = 0; j < lpages; ++j)
+				slot->arch.lpage_info[i][j].write_count = 1;
+		}
+	}
+
+	return 0;
+
+out_free:
+	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
+		vfree(slot->arch.lpage_info[i]);
+		slot->arch.lpage_info[i] = NULL;
+	}
+	return -ENOMEM;
+}
+
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				struct kvm_memory_slot *memslot,
 				struct kvm_memory_slot old,
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 06d4e41..1d833d7 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -171,11 +171,6 @@ static inline int kvm_vcpu_exiting_guest_mode(struct kvm_vcpu *vcpu)
  */
 #define KVM_MEM_MAX_NR_PAGES ((1UL << 31) - 1)
 
-struct kvm_lpage_info {
-	unsigned long rmap_pde;
-	int write_count;
-};
-
 struct kvm_memory_slot {
 	gfn_t base_gfn;
 	unsigned long npages;
@@ -184,7 +179,7 @@ struct kvm_memory_slot {
 	unsigned long *dirty_bitmap;
 	unsigned long *dirty_bitmap_head;
 	unsigned long nr_dirty_pages;
-	struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1];
+	struct kvm_arch_memory_slot arch;
 	unsigned long userspace_addr;
 	int user_alloc;
 	int id;
@@ -376,6 +371,9 @@ int kvm_set_memory_region(struct kvm *kvm,
 int __kvm_set_memory_region(struct kvm *kvm,
 			    struct kvm_userspace_memory_region *mem,
 			    int user_alloc);
+void kvm_arch_free_memslot(struct kvm_memory_slot *free,
+			   struct kvm_memory_slot *dont);
+int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages);
 int kvm_arch_prepare_memory_region(struct kvm *kvm,
 				struct kvm_memory_slot *memslot,
 				struct kvm_memory_slot old,
@@ -385,6 +383,7 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
 				struct kvm_userspace_memory_region *mem,
 				struct kvm_memory_slot old,
 				int user_alloc);
+bool kvm_largepages_enabled(void);
 void kvm_disable_largepages(void);
 void kvm_arch_flush_shadow(struct kvm *kvm);
 
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index ee233ce..ad14e1c 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -535,21 +535,13 @@ static void kvm_destroy_dirty_bitmap(struct kvm_memory_slot *memslot)
 static void kvm_free_physmem_slot(struct kvm_memory_slot *free,
 				  struct kvm_memory_slot *dont)
 {
-	int i;
-
 	if (!dont || free->rmap != dont->rmap)
 		vfree(free->rmap);
 
 	if (!dont || free->dirty_bitmap != dont->dirty_bitmap)
 		kvm_destroy_dirty_bitmap(free);
 
-
-	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
-		if (!dont || free->lpage_info[i] != dont->lpage_info[i]) {
-			vfree(free->lpage_info[i]);
-			free->lpage_info[i] = NULL;
-		}
-	}
+	kvm_arch_free_memslot(free, dont);
 
 	free->npages = 0;
 	free->rmap = NULL;
@@ -698,56 +690,6 @@ void update_memslots(struct kvm_memslots *slots, struct kvm_memory_slot *new)
 	slots->generation++;
 }
 
-#ifndef CONFIG_S390
-static int create_lpage_info(struct kvm_memory_slot *slot, unsigned long npages)
-{
-	int i;
-
-	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
-		unsigned long ugfn;
-		int lpages;
-		int level = i + 2;
-
-		if (slot->lpage_info[i])
-			continue;
-
-		lpages = gfn_to_index(slot->base_gfn + npages - 1,
-				      slot->base_gfn, level) + 1;
-
-		slot->lpage_info[i] = vzalloc(lpages * sizeof(*slot->lpage_info[i]));
-		if (!slot->lpage_info[i])
-			goto out_free;
-
-		if (slot->base_gfn & (KVM_PAGES_PER_HPAGE(level) - 1))
-			slot->lpage_info[i][0].write_count = 1;
-		if ((slot->base_gfn + npages) & (KVM_PAGES_PER_HPAGE(level) - 1))
-			slot->lpage_info[i][lpages - 1].write_count = 1;
-		ugfn = slot->userspace_addr >> PAGE_SHIFT;
-		/*
-		 * If the gfn and userspace address are not aligned wrt each
-		 * other, or if explicitly asked to, disable large page
-		 * support for this slot
-		 */
-		if ((slot->base_gfn ^ ugfn) & (KVM_PAGES_PER_HPAGE(level) - 1) ||
-		    !largepages_enabled) {
-			unsigned long j;
-
-			for (j = 0; j < lpages; ++j)
-				slot->lpage_info[i][j].write_count = 1;
-		}
-	}
-
-	return 0;
-
-out_free:
-	for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) {
-		vfree(slot->lpage_info[i]);
-		slot->lpage_info[i] = NULL;
-	}
-	return -ENOMEM;
-}
-#endif /* not defined CONFIG_S390 */
-
 /*
  * Allocate some memory and give it an address in the guest physical address
  * space.
@@ -838,7 +780,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 			if (!new.rmap)
 				goto out_free;
 		}
-		if (create_lpage_info(&new, npages))
+		if (kvm_arch_create_memslot(&new, npages))
 			goto out_free;
 #endif /* not defined CONFIG_S390 */
 	}
@@ -898,8 +840,7 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	if (!npages) {
 		new.rmap = NULL;
 		new.dirty_bitmap = NULL;
-		for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i)
-			new.lpage_info[i] = NULL;
+		memset(&new.arch, 0, sizeof(new.arch));
 	}
 
 	update_memslots(slots, &new);
@@ -986,6 +927,11 @@ out:
 	return r;
 }
 
+bool kvm_largepages_enabled(void)
+{
+	return largepages_enabled;
+}
+
 void kvm_disable_largepages(void)
 {
 	largepages_enabled = false;
-- 
1.7.5.4

Re: [PATCH 3/4] KVM: Simplify ifndef conditional usage in __kvm_set_memory_region()

[Avi Kivity]

On 01/29/2012 02:30 PM, Takuya Yoshikawa wrote:
> From: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
>
> Narrow down the controlled text inside the conditional so that it will
> include lpage_info and rmap stuff only.
>
> For this we change the way we check whether the slot is being created
> from "if (npages && !new.rmap)" to "if (npages && !old.npages)".
>  
>  static struct kvm_memory_slot *
>  search_memslots(struct kvm_memslots *slots, gfn_t gfn)
> @@ -828,23 +828,20 @@ int __kvm_set_memory_region(struct kvm *kvm,
>  	r = -ENOMEM;
>  
>  	/* Allocate if a slot is being created */
> +	if (npages && !old.npages) {
> +		new.user_alloc = user_alloc;
> +		if (user_alloc)
> +			new.userspace_addr = mem->userspace_addr;

This if () is new, is it not? ah, I see it was in the S390 path.  Better
to remove it.

>  #ifndef CONFIG_S390
> -	if (npages && !new.rmap) {
> -		new.rmap = vzalloc(npages * sizeof(*new.rmap));
> -
> -		if (!new.rmap)
> +		if (!new.rmap) {

Why this new test?  Can we have a new slot that does have an rmap?

> +			new.rmap = vzalloc(npages * sizeof(*new.rmap));
> +			if (!new.rmap)
> +				goto out_free;
> +		}
> +		if (create_lpage_info(&new, npages))
>  			goto out_free;
> -
> -		new.user_alloc = user_alloc;
> -		new.userspace_addr = mem->userspace_addr;
> +#endif /* not defined CONFIG_S390 */
>  	}
> -	if (!npages)
> -		goto skip_lpage;
> -
> -	if (create_lpage_info(&new, npages))
> -		goto out_free;
> -
> -skip_lpage:
>  
>  	/* Allocate page dirty bitmap if needed */
>  	if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) {
> @@ -852,11 +849,6 @@ skip_lpage:
>  			goto out_free;
>  		/* destroy any largepage mappings for dirty tracking */
>  	}
> -#else  /* not defined CONFIG_S390 */
> -	new.user_alloc = user_alloc;
> -	if (user_alloc)
> -		new.userspace_addr = mem->userspace_addr;
> -#endif /* not defined CONFIG_S390 */
>  
>  	if (!npages) {
>  		struct kvm_memory_slot *slot;


-- 
error compiling committee.c: too many arguments to function

[PATCH 3/4] KVM: Simplify ifndef conditional usage in __kvm_set_memory_region()

[Takuya Yoshikawa]

Narrow down the controlled text inside the conditional so that it will
include lpage_info and rmap stuff only.

For this we change the way we check whether the slot is being created
from "if (npages && !new.rmap)" to "if (npages && !old.npages)".

We also stop checking if lpage_info is NULL when we create lpage_info
because we do it from inside the slot creation code block.

Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
---
 virt/kvm/kvm_main.c |   29 ++++++++---------------------
 1 files changed, 8 insertions(+), 21 deletions(-)

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 7adaa20..a30447c 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -616,7 +616,6 @@ static int kvm_vm_release(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-#ifndef CONFIG_S390
 /*
  * Allocation size is twice as large as the actual dirty bitmap size.
  * This makes it possible to do double buffering: see x86's
@@ -624,6 +623,7 @@ static int kvm_vm_release(struct inode *inode, struct file *filp)
  */
 static int kvm_create_dirty_bitmap(struct kvm_memory_slot *memslot)
 {
+#ifndef CONFIG_S390
 	unsigned long dirty_bytes = 2 * kvm_dirty_bitmap_bytes(memslot);
 
 	if (dirty_bytes > PAGE_SIZE)
@@ -636,9 +636,9 @@ static int kvm_create_dirty_bitmap(struct kvm_memory_slot *memslot)
 
 	memslot->dirty_bitmap_head = memslot->dirty_bitmap;
 	memslot->nr_dirty_pages = 0;
+#endif /* !CONFIG_S390 */
 	return 0;
 }
-#endif /* !CONFIG_S390 */
 
 static int cmp_memslot(const void *slot1, const void *slot2)
 {
@@ -695,9 +695,6 @@ static int create_lpage_info(struct kvm_memory_slot *slot, unsigned long npages)
 		int lpages;
 		int level = i + 2;
 
-		if (slot->lpage_info[i])
-			continue;
-
 		lpages = gfn_to_index(slot->base_gfn + npages - 1,
 				      slot->base_gfn, level) + 1;
 
@@ -815,23 +812,18 @@ int __kvm_set_memory_region(struct kvm *kvm,
 	r = -ENOMEM;
 
 	/* Allocate if a slot is being created */
+	if (npages && !old.npages) {
+		new.user_alloc = user_alloc;
+		new.userspace_addr = mem->userspace_addr;
 #ifndef CONFIG_S390
-	if (npages && !new.rmap) {
 		new.rmap = vzalloc(npages * sizeof(*new.rmap));
-
 		if (!new.rmap)
 			goto out_free;
 
-		new.user_alloc = user_alloc;
-		new.userspace_addr = mem->userspace_addr;
+		if (create_lpage_info(&new, npages))
+			goto out_free;
+#endif /* not defined CONFIG_S390 */
 	}
-	if (!npages)
-		goto skip_lpage;
-
-	if (create_lpage_info(&new, npages))
-		goto out_free;
-
-skip_lpage:
 
 	/* Allocate page dirty bitmap if needed */
 	if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) {
@@ -839,11 +831,6 @@ skip_lpage:
 			goto out_free;
 		/* destroy any largepage mappings for dirty tracking */
 	}
-#else  /* not defined CONFIG_S390 */
-	new.user_alloc = user_alloc;
-	if (user_alloc)
-		new.userspace_addr = mem->userspace_addr;
-#endif /* not defined CONFIG_S390 */
 
 	if (!npages) {
 		struct kvm_memory_slot *slot;
-- 
1.7.5.4