linux-kernel - Re: [RFC PATCH 3/8] KVM: x86/mmu: Introduce initialier macro for struct kvm_page

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Mon, 11 Mar 2024 15:56:11 -0700
From: Isaku Yamahata <isaku.yamahata@...el.com>
To: Sean Christopherson <seanjc@...gle.com>
Cc: isaku.yamahata@...el.com, kvm@...r.kernel.org, isaku.yamahata@...il.com,
	linux-kernel@...r.kernel.org, Paolo Bonzini <pbonzini@...hat.com>,
	Michael Roth <michael.roth@....com>,
	David Matlack <dmatlack@...gle.com>,
	Federico Parola <federico.parola@...ito.it>,
	isaku.yamahata@...ux.intel.com
Subject: Re: [RFC PATCH 3/8] KVM: x86/mmu: Introduce initialier macro for
 struct kvm_page_fault

On Mon, Mar 11, 2024 at 10:24:50AM -0700,
Sean Christopherson <seanjc@...gle.com> wrote:

> On Fri, Mar 01, 2024, isaku.yamahata@...el.com wrote:
> > From: Isaku Yamahata <isaku.yamahata@...el.com>
> > 
> > Another function will initialize struct kvm_page_fault.  Add initializer
> > macro to unify the big struct initialization.
> > 
> > No functional change intended.
> > 
> > Signed-off-by: Isaku Yamahata <isaku.yamahata@...el.com>
> > ---
> >  arch/x86/kvm/mmu/mmu_internal.h | 44 +++++++++++++++++++--------------
> >  1 file changed, 26 insertions(+), 18 deletions(-)
> > 
> > diff --git a/arch/x86/kvm/mmu/mmu_internal.h b/arch/x86/kvm/mmu/mmu_internal.h
> > index 0669a8a668ca..72ef09fc9322 100644
> > --- a/arch/x86/kvm/mmu/mmu_internal.h
> > +++ b/arch/x86/kvm/mmu/mmu_internal.h
> > @@ -279,27 +279,35 @@ enum {
> >  	RET_PF_SPURIOUS,
> >  };
> >  
> > +#define KVM_PAGE_FAULT_INIT(_vcpu, _cr2_or_gpa, _err, _prefetch, _max_level) {	\
> > +	.addr = (_cr2_or_gpa),							\
> > +	.error_code = (_err),							\
> > +	.exec = (_err) & PFERR_FETCH_MASK,					\
> > +	.write = (_err) & PFERR_WRITE_MASK,					\
> > +	.present = (_err) & PFERR_PRESENT_MASK,					\
> > +	.rsvd = (_err) & PFERR_RSVD_MASK,					\
> > +	.user = (_err) & PFERR_USER_MASK,					\
> > +	.prefetch = (_prefetch),						\
> > +	.is_tdp =								\
> > +	likely((_vcpu)->arch.mmu->page_fault == kvm_tdp_page_fault),		\
> > +	.nx_huge_page_workaround_enabled =					\
> > +	is_nx_huge_page_enabled((_vcpu)->kvm),					\
> > +										\
> > +	.max_level = (_max_level),						\
> > +	.req_level = PG_LEVEL_4K,						\
> > +	.goal_level = PG_LEVEL_4K,						\
> > +	.is_private =								\
> > +	kvm_mem_is_private((_vcpu)->kvm, (_cr2_or_gpa) >> PAGE_SHIFT),		\
> > +										\
> > +	.pfn = KVM_PFN_ERR_FAULT,						\
> > +	.hva = KVM_HVA_ERR_BAD, }
> > +
> 
> Oof, no.  I would much rather refactor kvm_mmu_do_page_fault() as needed than
> have to maintain a macro like this.

Ok, I updated it as follows.

diff --git a/arch/x86/kvm/mmu/mmu_internal.h b/arch/x86/kvm/mmu/mmu_internal.h
index 0669a8a668ca..e57cc3c56a6d 100644
--- a/arch/x86/kvm/mmu/mmu_internal.h
+++ b/arch/x86/kvm/mmu/mmu_internal.h
@@ -279,8 +279,8 @@ enum {
 	RET_PF_SPURIOUS,
 };
 
-static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
-					u32 err, bool prefetch, int *emulation_type)
+static inline int __kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
+					  u32 err, bool prefetch, int *emulation_type)
 {
 	struct kvm_page_fault fault = {
 		.addr = cr2_or_gpa,
@@ -307,6 +307,21 @@ static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
 		fault.slot = kvm_vcpu_gfn_to_memslot(vcpu, fault.gfn);
 	}
 
+	if (IS_ENABLED(CONFIG_RETPOLINE) && fault.is_tdp)
+		r = kvm_tdp_page_fault(vcpu, &fault);
+	else
+		r = vcpu->arch.mmu->page_fault(vcpu, &fault);
+
+	if (fault.write_fault_to_shadow_pgtable && emulation_type)
+		*emulation_type |= EMULTYPE_WRITE_PF_TO_SP;
+	return r;
+}
+
+static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
+					u32 err, bool prefetch, int *emulation_type)
+{
+	int r;
+
 	/*
 	 * Async #PF "faults", a.k.a. prefetch faults, are not faults from the
 	 * guest perspective and have already been counted at the time of the
@@ -315,13 +330,7 @@ static inline int kvm_mmu_do_page_fault(struct kvm_vcpu *vcpu, gpa_t cr2_or_gpa,
 	if (!prefetch)
 		vcpu->stat.pf_taken++;
 
-	if (IS_ENABLED(CONFIG_RETPOLINE) && fault.is_tdp)
-		r = kvm_tdp_page_fault(vcpu, &fault);
-	else
-		r = vcpu->arch.mmu->page_fault(vcpu, &fault);
-
-	if (fault.write_fault_to_shadow_pgtable && emulation_type)
-		*emulation_type |= EMULTYPE_WRITE_PF_TO_SP;
+	r = __kvm_mmu_do_page_fault(vcpu, cr2_or_gpa, err, prefetch, emulation_type);
 
 	/*
 	 * Similar to above, prefetch faults aren't truly spurious, and the
-- 
2.43.2

-- 
Isaku Yamahata <isaku.yamahata@...el.com>