[PATCH v5 1/3] KVM: MMU: rename 'no_apf' to 'prefault'

[PATCH v5 1/3] KVM: MMU: rename 'no_apf' to 'prefault'

[Xiao Guangrong]

It's the speculative path if 'no_apf = 1' and we will specially handle this
speculative path in the later patch, so 'prefault' is better to fit the sense

Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
---
 arch/x86/include/asm/kvm_host.h |    3 ++-
 arch/x86/kvm/mmu.c              |   18 +++++++++---------
 arch/x86/kvm/paging_tmpl.h      |    4 ++--
 3 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index cfbcbfa..6dae67e 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -241,7 +241,8 @@ struct kvm_mmu {
 	void (*new_cr3)(struct kvm_vcpu *vcpu);
 	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
 	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
-	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err, bool no_apf);
+	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err,
+			  bool prefoult);
 	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
 				  struct x86_exception *fault);
 	void (*free)(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index d75ba1e..4954de9 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2284,11 +2284,11 @@ static int kvm_handle_bad_page(struct kvm *kvm, gfn_t gfn, pfn_t pfn)
 	return 1;
 }
 
-static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
+static bool try_async_pf(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn,
 			 gva_t gva, pfn_t *pfn, bool write, bool *writable);
 
 static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn,
-			 bool no_apf)
+			 bool prefault)
 {
 	int r;
 	int level;
@@ -2310,7 +2310,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn,
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (try_async_pf(vcpu, no_apf, gfn, v, &pfn, write, &map_writable))
+	if (try_async_pf(vcpu, prefault, gfn, v, &pfn, write, &map_writable))
 		return 0;
 
 	/* mmio */
@@ -2583,7 +2583,7 @@ static gpa_t nonpaging_gva_to_gpa_nested(struct kvm_vcpu *vcpu, gva_t vaddr,
 }
 
 static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
-				u32 error_code, bool no_apf)
+				u32 error_code, bool prefault)
 {
 	gfn_t gfn;
 	int r;
@@ -2599,7 +2599,7 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
 	gfn = gva >> PAGE_SHIFT;
 
 	return nonpaging_map(vcpu, gva & PAGE_MASK,
-			     error_code & PFERR_WRITE_MASK, gfn, no_apf);
+			     error_code & PFERR_WRITE_MASK, gfn, prefault);
 }
 
 static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn)
@@ -2621,7 +2621,7 @@ static bool can_do_async_pf(struct kvm_vcpu *vcpu)
 	return kvm_x86_ops->interrupt_allowed(vcpu);
 }
 
-static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
+static bool try_async_pf(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn,
 			 gva_t gva, pfn_t *pfn, bool write, bool *writable)
 {
 	bool async;
@@ -2633,7 +2633,7 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
 
 	put_page(pfn_to_page(*pfn));
 
-	if (!no_apf && can_do_async_pf(vcpu)) {
+	if (!prefault && can_do_async_pf(vcpu)) {
 		trace_kvm_try_async_get_page(gva, gfn);
 		if (kvm_find_async_pf_gfn(vcpu, gfn)) {
 			trace_kvm_async_pf_doublefault(gva, gfn);
@@ -2649,7 +2649,7 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
 }
 
 static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
-			  bool no_apf)
+			  bool prefault)
 {
 	pfn_t pfn;
 	int r;
@@ -2673,7 +2673,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (try_async_pf(vcpu, no_apf, gfn, gpa, &pfn, write, &map_writable))
+	if (try_async_pf(vcpu, prefault, gfn, gpa, &pfn, write, &map_writable))
 		return 0;
 
 	/* mmio */
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index d5a0a11..52b3e91 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -539,7 +539,7 @@ out_gpte_changed:
  *           a negative value on error.
  */
 static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
-			     bool no_apf)
+			     bool prefault)
 {
 	int write_fault = error_code & PFERR_WRITE_MASK;
 	int user_fault = error_code & PFERR_USER_MASK;
@@ -581,7 +581,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (try_async_pf(vcpu, no_apf, walker.gfn, addr, &pfn, write_fault,
+	if (try_async_pf(vcpu, prefault, walker.gfn, addr, &pfn, write_fault,
 			 &map_writable))
 		return 0;
 
-- 
1.7.0.4

[PATCH v5 2/3] KVM: MMU: fix accessed bit set on prefault path

[Xiao Guangrong]

Retry #PF is the speculative path, so don't set the accessed bit

Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
---
 arch/x86/kvm/mmu.c |   10 ++++++----
 1 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 4954de9..04f9033 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2214,7 +2214,8 @@ static void direct_pte_prefetch(struct kvm_vcpu *vcpu, u64 *sptep)
 }
 
 static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
-			int map_writable, int level, gfn_t gfn, pfn_t pfn)
+			int map_writable, int level, gfn_t gfn, pfn_t pfn,
+			bool prefault)
 {
 	struct kvm_shadow_walk_iterator iterator;
 	struct kvm_mmu_page *sp;
@@ -2229,7 +2230,7 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write,
 				pte_access &= ~ACC_WRITE_MASK;
 			mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, pte_access,
 				     0, write, 1, &pt_write,
-				     level, gfn, pfn, false, map_writable);
+				     level, gfn, pfn, prefault, map_writable);
 			direct_pte_prefetch(vcpu, iterator.sptep);
 			++vcpu->stat.pf_fixed;
 			break;
@@ -2321,7 +2322,8 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn,
 	if (mmu_notifier_retry(vcpu, mmu_seq))
 		goto out_unlock;
 	kvm_mmu_free_some_pages(vcpu);
-	r = __direct_map(vcpu, v, write, map_writable, level, gfn, pfn);
+	r = __direct_map(vcpu, v, write, map_writable, level, gfn, pfn,
+			 prefault);
 	spin_unlock(&vcpu->kvm->mmu_lock);
 
 
@@ -2684,7 +2686,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
 		goto out_unlock;
 	kvm_mmu_free_some_pages(vcpu);
 	r = __direct_map(vcpu, gpa, write, map_writable,
-			 level, gfn, pfn);
+			 level, gfn, pfn, prefault);
 	spin_unlock(&vcpu->kvm->mmu_lock);
 
 	return r;
-- 
1.7.0.4

[PATCH v5 3/3] KVM: MMU: retry #PF for softmmu

[Xiao Guangrong]

Retry #PF for softmmu only when the current vcpu has the same cr3 as the time
when #PF occurs

Changelog:
  Just compare cr3 value since It's harmless to instantiate an spte for an
  unused translation from Marcelo's comment

Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
---
 arch/x86/include/asm/kvm_host.h |    1 +
 arch/x86/kvm/mmu.c              |    2 ++
 arch/x86/kvm/paging_tmpl.h      |   14 +++++++++-----
 arch/x86/kvm/x86.c              |    6 +++++-
 4 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 6dae67e..be49043 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -593,6 +593,7 @@ struct kvm_x86_ops {
 struct kvm_arch_async_pf {
 	u32 token;
 	gfn_t gfn;
+	unsigned long cr3;
 	bool direct_map;
 };
 
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 04f9033..1a953ac 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2607,9 +2607,11 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
 static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn)
 {
 	struct kvm_arch_async_pf arch;
+
 	arch.token = (vcpu->arch.apf.id++ << 12) | vcpu->vcpu_id;
 	arch.gfn = gfn;
 	arch.direct_map = vcpu->arch.mmu.direct_map;
+	arch.cr3 = vcpu->arch.mmu.get_cr3(vcpu);
 
 	return kvm_setup_async_pf(vcpu, gva, gfn, &arch);
 }
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 52b3e91..146b681 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -438,7 +438,8 @@ static void FNAME(pte_prefetch)(struct kvm_vcpu *vcpu, struct guest_walker *gw,
 static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 			 struct guest_walker *gw,
 			 int user_fault, int write_fault, int hlevel,
-			 int *ptwrite, pfn_t pfn, bool map_writable)
+			 int *ptwrite, pfn_t pfn, bool map_writable,
+			 bool prefault)
 {
 	unsigned access = gw->pt_access;
 	struct kvm_mmu_page *sp = NULL;
@@ -512,7 +513,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 
 	mmu_set_spte(vcpu, it.sptep, access, gw->pte_access & access,
 		     user_fault, write_fault, dirty, ptwrite, it.level,
-		     gw->gfn, pfn, false, map_writable);
+		     gw->gfn, pfn, prefault, map_writable);
 	FNAME(pte_prefetch)(vcpu, gw, it.sptep);
 
 	return it.sptep;
@@ -568,8 +569,11 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
 	 */
 	if (!r) {
 		pgprintk("%s: guest page fault\n", __func__);
-		inject_page_fault(vcpu, &walker.fault);
-		vcpu->arch.last_pt_write_count = 0; /* reset fork detector */
+		if (!prefault) {
+			inject_page_fault(vcpu, &walker.fault);
+			/* reset fork detector */
+			vcpu->arch.last_pt_write_count = 0;
+		}
 		return 0;
 	}
 
@@ -599,7 +603,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
 	trace_kvm_mmu_audit(vcpu, AUDIT_PRE_PAGE_FAULT);
 	kvm_mmu_free_some_pages(vcpu);
 	sptep = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault,
-			     level, &write_pt, pfn, map_writable);
+			     level, &write_pt, pfn, map_writable, prefault);
 	(void)sptep;
 	pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __func__,
 		 sptep, *sptep, write_pt);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index ed373ba..018bb70 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -6183,7 +6183,7 @@ void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work)
 {
 	int r;
 
-	if (!vcpu->arch.mmu.direct_map || !work->arch.direct_map ||
+	if ((vcpu->arch.mmu.direct_map != work->arch.direct_map) ||
 	      is_error_page(work->page))
 		return;
 
@@ -6191,6 +6191,10 @@ void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work)
 	if (unlikely(r))
 		return;
 
+	if (!vcpu->arch.mmu.direct_map &&
+	      work->arch.cr3 != vcpu->arch.mmu.get_cr3(vcpu))
+		return;
+
 	vcpu->arch.mmu.page_fault(vcpu, work->gva, 0, true);
 }
 
-- 
1.7.0.4

Re: [PATCH v5 1/3] KVM: MMU: rename 'no_apf' to 'prefault'

[Xiao Guangrong]

On 12/07/2010 10:33 AM, Xiao Guangrong wrote:
> It's the speculative path if 'no_apf = 1' and we will specially handle this
> speculative path in the later patch, so 'prefault' is better to fit the sense
> 
> Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
> ---
>  arch/x86/include/asm/kvm_host.h |    3 ++-
>  arch/x86/kvm/mmu.c              |   18 +++++++++---------
>  arch/x86/kvm/paging_tmpl.h      |    4 ++--
>  3 files changed, 13 insertions(+), 12 deletions(-)
> 
> diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> index cfbcbfa..6dae67e 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -241,7 +241,8 @@ struct kvm_mmu {
>  	void (*new_cr3)(struct kvm_vcpu *vcpu);
>  	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
>  	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
> -	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err, bool no_apf);
> +	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err,
> +			  bool prefoult);

Sorry for the noise, there is a typo, resent it.

Subject: [PATCH v5 1/3] KVM: MMU: rename 'no_apf' to 'prefault'

It's the speculative path if 'no_apf = 1' and we will specially handle this
speculative path in the later patch, so 'prefault' is better to fit the sense

Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
---
 arch/x86/include/asm/kvm_host.h |    3 ++-
 arch/x86/kvm/mmu.c              |   18 +++++++++---------
 arch/x86/kvm/paging_tmpl.h      |    4 ++--
 3 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index cfbcbfa..6dae67e 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -241,7 +241,8 @@ struct kvm_mmu {
 	void (*new_cr3)(struct kvm_vcpu *vcpu);
 	void (*set_cr3)(struct kvm_vcpu *vcpu, unsigned long root);
 	unsigned long (*get_cr3)(struct kvm_vcpu *vcpu);
-	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err, bool no_apf);
+	int (*page_fault)(struct kvm_vcpu *vcpu, gva_t gva, u32 err,
+			  bool prefault);
 	void (*inject_page_fault)(struct kvm_vcpu *vcpu,
 				  struct x86_exception *fault);
 	void (*free)(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index d75ba1e..4954de9 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2284,11 +2284,11 @@ static int kvm_handle_bad_page(struct kvm *kvm, gfn_t gfn, pfn_t pfn)
 	return 1;
 }
 
-static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
+static bool try_async_pf(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn,
 			 gva_t gva, pfn_t *pfn, bool write, bool *writable);
 
 static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn,
-			 bool no_apf)
+			 bool prefault)
 {
 	int r;
 	int level;
@@ -2310,7 +2310,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn,
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (try_async_pf(vcpu, no_apf, gfn, v, &pfn, write, &map_writable))
+	if (try_async_pf(vcpu, prefault, gfn, v, &pfn, write, &map_writable))
 		return 0;
 
 	/* mmio */
@@ -2583,7 +2583,7 @@ static gpa_t nonpaging_gva_to_gpa_nested(struct kvm_vcpu *vcpu, gva_t vaddr,
 }
 
 static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
-				u32 error_code, bool no_apf)
+				u32 error_code, bool prefault)
 {
 	gfn_t gfn;
 	int r;
@@ -2599,7 +2599,7 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva,
 	gfn = gva >> PAGE_SHIFT;
 
 	return nonpaging_map(vcpu, gva & PAGE_MASK,
-			     error_code & PFERR_WRITE_MASK, gfn, no_apf);
+			     error_code & PFERR_WRITE_MASK, gfn, prefault);
 }
 
 static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn)
@@ -2621,7 +2621,7 @@ static bool can_do_async_pf(struct kvm_vcpu *vcpu)
 	return kvm_x86_ops->interrupt_allowed(vcpu);
 }
 
-static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
+static bool try_async_pf(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn,
 			 gva_t gva, pfn_t *pfn, bool write, bool *writable)
 {
 	bool async;
@@ -2633,7 +2633,7 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
 
 	put_page(pfn_to_page(*pfn));
 
-	if (!no_apf && can_do_async_pf(vcpu)) {
+	if (!prefault && can_do_async_pf(vcpu)) {
 		trace_kvm_try_async_get_page(gva, gfn);
 		if (kvm_find_async_pf_gfn(vcpu, gfn)) {
 			trace_kvm_async_pf_doublefault(gva, gfn);
@@ -2649,7 +2649,7 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn,
 }
 
 static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
-			  bool no_apf)
+			  bool prefault)
 {
 	pfn_t pfn;
 	int r;
@@ -2673,7 +2673,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code,
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (try_async_pf(vcpu, no_apf, gfn, gpa, &pfn, write, &map_writable))
+	if (try_async_pf(vcpu, prefault, gfn, gpa, &pfn, write, &map_writable))
 		return 0;
 
 	/* mmio */
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index d5a0a11..52b3e91 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -539,7 +539,7 @@ out_gpte_changed:
  *           a negative value on error.
  */
 static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
-			     bool no_apf)
+			     bool prefault)
 {
 	int write_fault = error_code & PFERR_WRITE_MASK;
 	int user_fault = error_code & PFERR_USER_MASK;
@@ -581,7 +581,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code,
 	mmu_seq = vcpu->kvm->mmu_notifier_seq;
 	smp_rmb();
 
-	if (try_async_pf(vcpu, no_apf, walker.gfn, addr, &pfn, write_fault,
+	if (try_async_pf(vcpu, prefault, walker.gfn, addr, &pfn, write_fault,
 			 &map_writable))
 		return 0;
 
-- 
1.7.0.4

Re: [PATCH v5 1/3] KVM: MMU: rename 'no_apf' to 'prefault'

[Avi Kivity]

On 12/07/2010 04:33 AM, Xiao Guangrong wrote:
> It's the speculative path if 'no_apf = 1' and we will specially handle this
> speculative path in the later patch, so 'prefault' is better to fit the sense

Applied all, thanks.

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