Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1247643 > unrolled thread

[PATCH 0/5] KVM: x86: MMU: Eliminate extra memory slot searches in page fault handlers

Back to article view | Back to linux.kernel

Contents

  [PATCH 0/5] KVM: x86: MMU: Eliminate extra memory slot searches in  page fault handlers Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp> - 2015-10-15 12:30 +0200
    [PATCH 2/5] KVM: x86: MMU: Simplify force_pt_level calculation code  in FNAME(page_fault)() Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp> - 2015-10-15 12:30 +0200
    [PATCH 4/5] KVM: x86: MMU: Remove mapping_level_dirty_bitmap() Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp> - 2015-10-15 12:40 +0200
      Re: [PATCH 4/5] KVM: x86: MMU: Remove mapping_level_dirty_bitmap() Paolo Bonzini <pbonzini@redhat.com> - 2015-10-15 17:40 +0200
    [PATCH 5/5] KVM: x86: MMU: Eliminate an extra memory slot search in  mapping_level() Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp> - 2015-10-15 12:40 +0200

#1247643 — [PATCH 0/5] KVM: x86: MMU: Eliminate extra memory slot searches in page fault handlers

In page fault handlers, both mapping_level_dirty_bitmap() and mapping_level()
do a memory slot search, binary search, through kvm_vcpu_gfn_to_memslot(), which
may not be negligible especially for virtual machines with many memory slots.

With a bit of cleanup effort, the patch set reduces this overhead.

  [PATCH 1/5] KVM: x86: MMU: Make force_pt_level bool
  [PATCH 2/5] KVM: x86: MMU: Simplify force_pt_level calculation code in FNAME(page_fault)()
  [PATCH 3/5] KVM: x86: MMU: Merge mapping_level_dirty_bitmap() into mapping_level()
  [PATCH 4/5] KVM: x86: MMU: Remove mapping_level_dirty_bitmap()
  [PATCH 5/5] KVM: x86: MMU: Eliminate an extra memory slot search in mapping_level()

	Takuya
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1247647 — [PATCH 2/5] KVM: x86: MMU: Simplify force_pt_level calculation code in FNAME(page_fault)()

As a bonus, an extra memory slot search can be eliminated when
is_self_change_mapping is true.

Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
---
 arch/x86/kvm/paging_tmpl.h |   15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 07f1a4e..8ebc3a5 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -743,15 +743,14 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
 	is_self_change_mapping = FNAME(is_self_change_mapping)(vcpu,
 	      &walker, user_fault, &vcpu->arch.write_fault_to_shadow_pgtable);
 
-	if (walker.level >= PT_DIRECTORY_LEVEL)
-		force_pt_level = mapping_level_dirty_bitmap(vcpu, walker.gfn)
-		   || is_self_change_mapping;
-	else
+	if (walker.level >= PT_DIRECTORY_LEVEL && !is_self_change_mapping) {
+		force_pt_level = mapping_level_dirty_bitmap(vcpu, walker.gfn);
+		if (!force_pt_level) {
+			level = min(walker.level, mapping_level(vcpu, walker.gfn));
+			walker.gfn = walker.gfn & ~(KVM_PAGES_PER_HPAGE(level) - 1);
+		}
+	} else
 		force_pt_level = true;
-	if (!force_pt_level) {
-		level = min(walker.level, mapping_level(vcpu, walker.gfn));
-		walker.gfn = walker.gfn & ~(KVM_PAGES_PER_HPAGE(level) - 1);
-	}
 
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
-- 
1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1247651 — [PATCH 4/5] KVM: x86: MMU: Remove mapping_level_dirty_bitmap()

Now that it has only one caller, and its name is not so helpful for
readers, just remove it.

Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
---
 arch/x86/kvm/mmu.c |   21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 890cd69..78a3d08 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -851,6 +851,14 @@ static int host_mapping_level(struct kvm *kvm, gfn_t gfn)
 	return ret;
 }
 
+static inline bool memslot_invalid(struct kvm_memory_slot *slot)
+{
+	if (!slot || slot->flags & KVM_MEMSLOT_INVALID)
+		return true;
+
+	return false;
+}
+
 static struct kvm_memory_slot *
 gfn_to_memslot_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t gfn,
 			    bool no_dirty_log)
@@ -858,25 +866,22 @@ gfn_to_memslot_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t gfn,
 	struct kvm_memory_slot *slot;
 
 	slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
-	if (!slot || slot->flags & KVM_MEMSLOT_INVALID ||
-	      (no_dirty_log && slot->dirty_bitmap))
+	if (memslot_invalid(slot) || (no_dirty_log && slot->dirty_bitmap))
 		slot = NULL;
 
 	return slot;
 }
 
-static bool mapping_level_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t large_gfn)
-{
-	return !gfn_to_memslot_dirty_bitmap(vcpu, large_gfn, true);
-}
-
 static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn,
 			 bool *force_pt_level)
 {
 	int host_level, level, max_level;
+	struct kvm_memory_slot *slot;
+
+	slot = kvm_vcpu_gfn_to_memslot(vcpu, large_gfn);
 
 	if (likely(!*force_pt_level))
-		*force_pt_level = mapping_level_dirty_bitmap(vcpu, large_gfn);
+		*force_pt_level = memslot_invalid(slot) || slot->dirty_bitmap;
 	if (unlikely(*force_pt_level))
 		return PT_PAGE_TABLE_LEVEL;
 
-- 
1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1247908 — Re: [PATCH 4/5] KVM: x86: MMU: Remove mapping_level_dirty_bitmap()

On 15/10/2015 12:43, Takuya Yoshikawa wrote:
> +static inline bool memslot_invalid(struct kvm_memory_slot *slot)

Can you make this function memslot_valid_for_gpte(struct kvm_memory_slot
*slot, bool no_dirty_log), and have it return

	slot && !(slot->flags & KVM_MEMSLOT_INVALID) &&
	      (!no_dirty_log || !slot->dirty_bitmap)

?  If gfn_to_memslot_dirty_bitmap and mapping_level call the same
function, it helps highlighting the similarity between them.  Your
optimization loses that similarity in the name, but I think we can bring
it back somehow.

Otherwise, the patches are great.  Thanks!

Paolo


> +{
> +	if (!slot || slot->flags & KVM_MEMSLOT_INVALID)
> +		return true;
> +
> +	return false;
> +}
> +
>  static struct kvm_memory_slot *
>  gfn_to_memslot_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t gfn,
>  			    bool no_dirty_log)
> @@ -858,25 +866,22 @@ gfn_to_memslot_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t gfn,
>  	struct kvm_memory_slot *slot;
>  
>  	slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
> -	if (!slot || slot->flags & KVM_MEMSLOT_INVALID ||
> -	      (no_dirty_log && slot->dirty_bitmap))
> +	if (memslot_invalid(slot) || (no_dirty_log && slot->dirty_bitmap))
>  		slot = NULL;
>  
>  	return slot;
>  }
>  
> -static bool mapping_level_dirty_bitmap(struct kvm_vcpu *vcpu, gfn_t large_gfn)
> -{
> -	return !gfn_to_memslot_dirty_bitmap(vcpu, large_gfn, true);
> -}
> -
>  static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn,
>  			 bool *force_pt_level)
>  {
>  	int host_level, level, max_level;
> +	struct kvm_memory_slot *slot;
> +
> +	slot = kvm_vcpu_gfn_to_memslot(vcpu, large_gfn);
>  
>  	if (likely(!*force_pt_level))
> -		*force_pt_level = mapping_level_dirty_bitmap(vcpu, large_gfn);
> +		*force_pt_level = memslot_invalid(slot) || slot->dirty_bitmap;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1247659 — [PATCH 5/5] KVM: x86: MMU: Eliminate an extra memory slot search in mapping_level()

Calling kvm_vcpu_gfn_to_memslot() twice in mapping_level() should be
avoided since getting a slot by binary search may not be negligible,
especially for virtual machines with many memory slots.

Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
---
 arch/x86/kvm/mmu.c |   17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 78a3d08..8d285dc 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -818,14 +818,11 @@ static void unaccount_shadowed(struct kvm *kvm, struct kvm_mmu_page *sp)
 	kvm->arch.indirect_shadow_pages--;
 }
 
-static int has_wrprotected_page(struct kvm_vcpu *vcpu,
-				gfn_t gfn,
-				int level)
+static int __has_wrprotected_page(gfn_t gfn, int level,
+				  struct kvm_memory_slot *slot)
 {
-	struct kvm_memory_slot *slot;
 	struct kvm_lpage_info *linfo;
 
-	slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
 	if (slot) {
 		linfo = lpage_info_slot(gfn, slot, level);
 		return linfo->write_count;
@@ -834,6 +831,14 @@ static int has_wrprotected_page(struct kvm_vcpu *vcpu,
 	return 1;
 }
 
+static int has_wrprotected_page(struct kvm_vcpu *vcpu, gfn_t gfn, int level)
+{
+	struct kvm_memory_slot *slot;
+
+	slot = kvm_vcpu_gfn_to_memslot(vcpu, gfn);
+	return __has_wrprotected_page(gfn, level, slot);
+}
+
 static int host_mapping_level(struct kvm *kvm, gfn_t gfn)
 {
 	unsigned long page_size;
@@ -893,7 +898,7 @@ static int mapping_level(struct kvm_vcpu *vcpu, gfn_t large_gfn,
 	max_level = min(kvm_x86_ops->get_lpage_level(), host_level);
 
 	for (level = PT_DIRECTORY_LEVEL; level <= max_level; ++level)
-		if (has_wrprotected_page(vcpu, large_gfn, level))
+		if (__has_wrprotected_page(large_gfn, level, slot))
 			break;
 
 	return level - 1;
-- 
1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web