Skip to contentSkip to conversation

KVM: arm64: nv: Even more VNCR fixes

PATCH RESEND Superseded Oliver Upton 3 participants 10 messages Jun 9 – Jun 23

Subsystems (3)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Oliver Upton

author maintainer
#0 lore raw

v2 cover KVM: arm64: nv: Even more VNCR fixes

Argh, resending. Forgot to clean out my staging directory first. Sorry!

Yes, even more.

On top of the first pile:

- KVM needs to fault in the PFN for write before mapping the L1 VNCR with write permissions in the host stage-1

- Hack to prevent KVM from mapping a non-memory PFN with cacheable mappings. This doesn't align perfectly with the semantics at stage-2, boo hoo.

- Bonus change to break VNCR abort loops in case by bugging the VM if KVM doesn't know how to handle the fault

Tested by running an L3 VM with kvmarm/next

v1: https://lore.kernel.org/kvmarm/[email protected]/

Oliver Upton (5):
KVM: arm64: nv: Respect read-only PFN when mapping L1 VNCR
KVM: arm64: nv: Inject SEA if kvm_translate_vncr() can't resolve PFN
KVM: arm64: nv: Re-translate VNCR before injecting abort
KVM: arm64: nv: Inject SEA if guest VNCR isn't normal memory
KVM: arm64: nv: Mark VM as bugged for unexpected VNCR abort

 arch/arm64/include/asm/kvm_nested.h |   8 ++
 arch/arm64/kvm/at.c                 |   8 --
 arch/arm64/kvm/nested.c             | 164 ++++++++++++++--------------
 3 files changed, 92 insertions(+), 88 deletions(-)

base-commit: 406f0c31f47877db036e885f15830106b89ca950

Signature · 1 line
2.47.3
author maintainer
#1 lore raw

PATCH RESEND v2 1/5 KVM: arm64: nv: Respect read-only PFN when mapping L1 VNCR

KVM currently maps the L1 VNCR into the host stage-1 by relying entirely on the permissions of the guest stage-1. At the same time, it is entirely possible that the backing PFN is read-only (e.g. RO memslot), meaning that the L1 VNCR should use at most a read-only mapping.

Cache the writability of the PFN in the VNCR TLB and use it to constrain the resulting fixmap permissions. Promote VNCR permission faults to an SEA in the case where the guest attempts to write to a read-only endpoint. Conveniently, this also plugs a page leak found by Sashiko [*] resulting from the early return for a read-only PFN.

Fixes 2a359e072596 ("KVM: arm64: nv: Handle mapping of VNCR_EL2 at EL2")
---
 arch/arm64/kvm/nested.c | 36 ++++++++++++++++++++++++++----------
 1 file changed, 26 insertions(+), 10 deletions(-)
arch/arm64/kvm/nested.c +26 -10
index 257813289c76..84b3bd528e11 100644
@@ -24,6 +24,7 @@ struct vncr_tlb {
 	struct s1_walk_result	wr;
 
 	u64			hpa;
+	bool			hpa_writable;
 
 	/* -1 when not mapped on a CPU */
 	int			cpu;
@@ -1395,7 +1396,7 @@ static int kvm_translate_vncr(struct kvm_vcpu *vcpu, bool *is_gmem)
 	if (!*is_gmem) {
 		pfn = __kvm_faultin_pfn(memslot, gfn, write_fault ? FOLL_WRITE : 0,
 					&writable, &page);
-		if (is_error_noslot_pfn(pfn) || (write_fault && !writable))
+		if (is_error_noslot_pfn(pfn))
 			return -EFAULT;
 	} else {
 		ret = kvm_gmem_get_pfn(vcpu->kvm, memslot, gfn, &pfn, &page, NULL);
@@ -1404,6 +1405,8 @@ static int kvm_translate_vncr(struct kvm_vcpu *vcpu, bool *is_gmem)
 					      write_fault, false, false);
 			return ret;
 		}
+
+		writable = !(memslot->flags & KVM_MEM_READONLY);
 	}
 
 	scoped_guard(write_lock, &vcpu->kvm->mmu_lock) {
@@ -1414,28 +1417,41 @@ static int kvm_translate_vncr(struct kvm_vcpu *vcpu, bool *is_gmem)
 
 		vt->gva = va;
 		vt->hpa = pfn << PAGE_SHIFT;
+		vt->hpa_writable = writable;
 		vt->valid = true;
 		vt->cpu = -1;
 
 		kvm_make_request(KVM_REQ_MAP_L1_VNCR_EL2, vcpu);
-		kvm_release_faultin_page(vcpu->kvm, page, false, vt->wr.pw);
+		kvm_release_faultin_page(vcpu->kvm, page, false, vt->wr.pw && vt->hpa_writable);
 	}
 
-	if (vt->wr.pw)
+	if (vt->wr.pw && vt->hpa_writable)
 		mark_page_dirty(vcpu->kvm, gfn);
 
 	return 0;
 }
 
-static void inject_vncr_perm(struct kvm_vcpu *vcpu)
+static void handle_vncr_perm(struct kvm_vcpu *vcpu)
 {
 	struct vncr_tlb *vt = vcpu->arch.vncr_tlb;
 	u64 esr = kvm_vcpu_get_esr(vcpu);
+	u64 fsc;
+
+	/*
+	 * Promote to an external abort if the stage-1 permits writes but the
+	 * HPA is read-only (e.g. RO memslot).
+	 */
+	if (kvm_is_write_fault(vcpu) && vt->wr.pw && !vt->hpa_writable)
+		fsc = ESR_ELx_FSC_EXTABT;
+	/*
+	 * Otherwise, inject a permission fault using the guest's translation
+	 * level rather than the host's.
+	 */
+	else
+		fsc = ESR_ELx_FSC_PERM_L(vt->wr.level);
 
-	/* Adjust the fault level to reflect that of the guest's */
 	esr &= ~ESR_ELx_FSC;
-	esr |= FIELD_PREP(ESR_ELx_FSC,
-			  ESR_ELx_FSC_PERM_L(vt->wr.level));
+	esr |= FIELD_PREP(ESR_ELx_FSC, fsc);
 
 	kvm_inject_nested_sync(vcpu, esr);
 }
@@ -1469,7 +1485,7 @@ int kvm_handle_vncr_abort(struct kvm_vcpu *vcpu)
 		return kvm_handle_guest_sea(vcpu);
 
 	if (esr_fsc_is_permission_fault(esr)) {
-		inject_vncr_perm(vcpu);
+		handle_vncr_perm(vcpu);
 	} else if (esr_fsc_is_translation_fault(esr)) {
 		bool valid, is_gmem = false;
 		int ret;
@@ -1517,7 +1533,7 @@ int kvm_handle_vncr_abort(struct kvm_vcpu *vcpu)
 			break;
 		case -EPERM:
 			/* Hack to deal with POE until we get kernel support */
-			inject_vncr_perm(vcpu);
+			handle_vncr_perm(vcpu);
 			break;
 		case 0:
 			break;
@@ -1561,7 +1577,7 @@ static void kvm_map_l1_vncr(struct kvm_vcpu *vcpu)
 
 	vt->cpu = smp_processor_id();
 
-	if (vt->wr.pw && vt->wr.pr)
+	if (vt->hpa_writable && vt->wr.pw && vt->wr.pr)
 		prot = PAGE_KERNEL;
 	else if (vt->wr.pr)
 		prot = PAGE_KERNEL_RO;
Signature · 1 line
2.47.3
author maintainer
#2 lore raw

PATCH RESEND v2 2/5 KVM: arm64: nv: Inject SEA if kvm_translate_vncr() can't resolve PFN

In reply to Oliver Upton

kvm_handle_vncr_abort() assumes that s1_walk_result conveys an abort when kvm_translate_vncr() returns -EFAULT. This is not always the case as it's possible to encounter 'late' failures on the output of S1 translation, e.g. a GFN outside of the memslots.

Fix it by preparing an external abort before returning from kvm_translate_vncr().

Fixes 2a359e072596 ("KVM: arm64: nv: Handle mapping of VNCR_EL2 at EL2")
---
 arch/arm64/include/asm/kvm_nested.h | 8 ++++++++
 arch/arm64/kvm/at.c                 | 8 --------
 arch/arm64/kvm/nested.c             | 8 ++++++--
 3 files changed, 14 insertions(+), 10 deletions(-)
arch/arm64/include/asm/kvm_nested.h +8 -0
index dc2957662ff2..cbdaaa2a2903 100644
@@ -388,6 +388,14 @@ struct s1_walk_result {
 	bool	failed;
 };
 
+static inline void fail_s1_walk(struct s1_walk_result *wr, u8 fst, bool s1ptw)
+{
+	wr->fst		= fst;
+	wr->ptw		= s1ptw;
+	wr->s2		= s1ptw;
+	wr->failed	= true;
+}
+
 int __kvm_translate_va(struct kvm_vcpu *vcpu, struct s1_walk_info *wi,
 		       struct s1_walk_result *wr, u64 va);
 int __kvm_find_s1_desc_level(struct kvm_vcpu *vcpu, u64 va, u64 ipa,
arch/arm64/kvm/at.c +0 -8
index 30e6fa8ac07c..8263c648207b 100644
@@ -11,14 +11,6 @@
 #include <asm/kvm_mmu.h>
 #include <asm/lsui.h>
 
-static void fail_s1_walk(struct s1_walk_result *wr, u8 fst, bool s1ptw)
-{
-	wr->fst		= fst;
-	wr->ptw		= s1ptw;
-	wr->s2		= s1ptw;
-	wr->failed	= true;
-}
-
 #define S1_MMU_DISABLED		(-127)
 
 static int get_ia_size(struct s1_walk_info *wi)
arch/arm64/kvm/nested.c +6 -2
index 84b3bd528e11..ebd7ccfeee99 100644
@@ -1389,15 +1389,19 @@ static int kvm_translate_vncr(struct kvm_vcpu *vcpu, bool *is_gmem)
 
 	gfn = vt->wr.pa >> PAGE_SHIFT;
 	memslot = gfn_to_memslot(vcpu->kvm, gfn);
-	if (!memslot)
+	if (!memslot) {
+		fail_s1_walk(&vt->wr, ESR_ELx_FSC_EXTABT, false);
 		return -EFAULT;
+	}
 
 	*is_gmem = kvm_slot_has_gmem(memslot);
 	if (!*is_gmem) {
 		pfn = __kvm_faultin_pfn(memslot, gfn, write_fault ? FOLL_WRITE : 0,
 					&writable, &page);
-		if (is_error_noslot_pfn(pfn))
+		if (is_error_noslot_pfn(pfn)) {
+			fail_s1_walk(&vt->wr, ESR_ELx_FSC_EXTABT, false);
 			return -EFAULT;
+		}
 	} else {
 		ret = kvm_gmem_get_pfn(vcpu->kvm, memslot, gfn, &pfn, &page, NULL);
 		if (ret) {
Signature · 1 line
2.47.3
author maintainer
#3 lore raw

PATCH RESEND v2 3/5 KVM: arm64: nv: Re-translate VNCR before injecting abort

In reply to Oliver Upton

KVM faults in the VNCR page with FOLL_WRITE whenver the guest aborts for a write, similar to how a regular stage-2 mapping is handled. It is entirely possible that the guest reads from the VNCR before writing to it, in which case the PFN could only be read-only.

Invalidate the VNCR TLB and re-fetch the translation upon taking a VNCR abort, allowing the host mapping to be faulted in for write the second time around. Interestingly enough, this also satisfies the ordering requirements of FEAT_ETS2/3 between descriptor updates and MMU faults.

Fixes 2a359e072596 ("KVM: arm64: nv: Handle mapping of VNCR_EL2 at EL2")
---
 arch/arm64/kvm/nested.c | 115 +++++++++++++++-------------------------
 1 file changed, 44 insertions(+), 71 deletions(-)
arch/arm64/kvm/nested.c +44 -71
index ebd7ccfeee99..d5c4b57123a9 100644
@@ -1460,92 +1460,65 @@ static void handle_vncr_perm(struct kvm_vcpu *vcpu)
 	kvm_inject_nested_sync(vcpu, esr);
 }
 
-static bool kvm_vncr_tlb_lookup(struct kvm_vcpu *vcpu)
-{
-	struct vncr_tlb *vt = vcpu->arch.vncr_tlb;
-
-	lockdep_assert_held_read(&vcpu->kvm->mmu_lock);
-
-	if (!vt->valid)
-		return false;
-
-	if (read_vncr_el2(vcpu) != vt->gva)
-		return false;
-
-	if (vt->wr.nG)
-		return get_asid_by_regime(vcpu, TR_EL20) == vt->wr.asid;
-
-	return true;
-}
-
 int kvm_handle_vncr_abort(struct kvm_vcpu *vcpu)
 {
 	struct vncr_tlb *vt = vcpu->arch.vncr_tlb;
 	u64 esr = kvm_vcpu_get_esr(vcpu);
+	bool is_gmem, perm;
+	int ret;
 
 	WARN_ON_ONCE(!(esr & ESR_ELx_VNCR));
 
 	if (kvm_vcpu_abt_issea(vcpu))
 		return kvm_handle_guest_sea(vcpu);
 
-	if (esr_fsc_is_permission_fault(esr)) {
-		handle_vncr_perm(vcpu);
-	} else if (esr_fsc_is_translation_fault(esr)) {
-		bool valid, is_gmem = false;
-		int ret;
-
-		scoped_guard(read_lock, &vcpu->kvm->mmu_lock)
-			valid = kvm_vncr_tlb_lookup(vcpu);
-
-		if (!valid)
-			ret = kvm_translate_vncr(vcpu, &is_gmem);
-		else
-			ret = -EPERM;
+	if (!esr_fsc_is_translation_fault(esr) && !esr_fsc_is_permission_fault(esr)) {
+		WARN_ONCE(1, "Unhandled VNCR abort, ESR=%llx\n", esr);
+		return 1;
+	}
 
-		switch (ret) {
-		case -EAGAIN:
-			/* Let's try again... */
-			break;
-		case -ENOMEM:
-			/*
-			 * For guest_memfd, this indicates that it failed to
-			 * create a folio to back the memory. Inform userspace.
-			 */
-			if (is_gmem)
-				return 0;
-			/* Otherwise, let's try again... */
-			break;
-		case -EFAULT:
-		case -EIO:
-		case -EHWPOISON:
-			if (is_gmem)
-				return 0;
-			fallthrough;
-		case -EINVAL:
-		case -ENOENT:
-		case -EACCES:
-			/*
-			 * Translation failed, inject the corresponding
-			 * exception back to EL2.
-			 */
-			BUG_ON(!vt->wr.failed);
+	ret = kvm_translate_vncr(vcpu, &is_gmem);
+	switch (ret) {
+	case -EAGAIN:
+		/* Let's try again... */
+		break;
+	case -ENOMEM:
+		/*
+		 * For guest_memfd, this indicates that it failed to
+		 * create a folio to back the memory. Inform userspace.
+		 */
+		if (is_gmem)
+			return 0;
+		/* Otherwise, let's try again... */
+		break;
+	case -EFAULT:
+	case -EIO:
+	case -EHWPOISON:
+		if (is_gmem)
+			return 0;
+		fallthrough;
+	case -EINVAL:
+	case -ENOENT:
+	case -EACCES:
+		/*
+		 * Translation failed, inject the corresponding
+		 * exception back to EL2.
+		 */
+		BUG_ON(!vt->wr.failed);
 
-			esr &= ~ESR_ELx_FSC;
-			esr |= FIELD_PREP(ESR_ELx_FSC, vt->wr.fst);
+		esr &= ~ESR_ELx_FSC;
+		esr |= FIELD_PREP(ESR_ELx_FSC, vt->wr.fst);
 
-			kvm_inject_nested_sync(vcpu, esr);
-			break;
-		case -EPERM:
-			/* Hack to deal with POE until we get kernel support */
-			handle_vncr_perm(vcpu);
-			break;
-		case 0:
-			break;
-		}
-	} else {
-		WARN_ONCE(1, "Unhandled VNCR abort, ESR=%llx\n", esr);
+		kvm_inject_nested_sync(vcpu, esr);
+		break;
+	case 0:
+		break;
 	}
 
+	perm = kvm_is_write_fault(vcpu) ? vt->wr.pw && vt->hpa_writable : vt->wr.pr;
+	if (!perm)
+		handle_vncr_perm(vcpu);
+
 	return 1;
 }
 
Signature · 1 line
2.47.3
maintainer
#4 lore raw

Re: [PATCH RESEND v2 3/5] KVM: arm64: nv: Re-translate VNCR before injecting abort

Quoted 120 lines · Oliver Upton

KVM faults in the VNCR page with FOLL_WRITE whenver the guest aborts for a write, similar to how a regular stage-2 mapping is handled. It is entirely possible that the guest reads from the VNCR before writing to it, in which case the PFN could only be read-only.

Invalidate the VNCR TLB and re-fetch the translation upon taking a VNCR abort, allowing the host mapping to be faulted in for write the second time around. Interestingly enough, this also satisfies the ordering requirements of FEAT_ETS2/3 between descriptor updates and MMU faults.

Fixes 2a359e072596 ("KVM: arm64: nv: Handle mapping of VNCR_EL2 at EL2")
---
 arch/arm64/kvm/nested.c | 115 +++++++++++++++-------------------------
 1 file changed, 44 insertions(+), 71 deletions(-)
arch/arm64/kvm/nested.c +24 -58
index ebd7ccfeee99..d5c4b57123a9 100644
@@ -1460,92 +1460,65 @@ static void handle_vncr_perm(struct kvm_vcpu *vcpu)
 	kvm_inject_nested_sync(vcpu, esr);
 }
 
-static bool kvm_vncr_tlb_lookup(struct kvm_vcpu *vcpu)
-{
-	struct vncr_tlb *vt = vcpu->arch.vncr_tlb;
-
-	lockdep_assert_held_read(&vcpu->kvm->mmu_lock);
-
-	if (!vt->valid)
-		return false;
-
-	if (read_vncr_el2(vcpu) != vt->gva)
-		return false;
-
-	if (vt->wr.nG)
-		return get_asid_by_regime(vcpu, TR_EL20) == vt->wr.asid;
-
-	return true;
-}
-
 int kvm_handle_vncr_abort(struct kvm_vcpu *vcpu)
 {
 	struct vncr_tlb *vt = vcpu->arch.vncr_tlb;
 	u64 esr = kvm_vcpu_get_esr(vcpu);
+	bool is_gmem, perm;
+	int ret;
 
 	WARN_ON_ONCE(!(esr & ESR_ELx_VNCR));
 
 	if (kvm_vcpu_abt_issea(vcpu))
 		return kvm_handle_guest_sea(vcpu);
 
-	if (esr_fsc_is_permission_fault(esr)) {
-		handle_vncr_perm(vcpu);
-	} else if (esr_fsc_is_translation_fault(esr)) {
-		bool valid, is_gmem = false;
-		int ret;
-
-		scoped_guard(read_lock, &vcpu->kvm->mmu_lock)
-			valid = kvm_vncr_tlb_lookup(vcpu);
-
-		if (!valid)
-			ret = kvm_translate_vncr(vcpu, &is_gmem);
-		else
-			ret = -EPERM;
+	if (!esr_fsc_is_translation_fault(esr) && !esr_fsc_is_permission_fault(esr)) {
+		WARN_ONCE(1, "Unhandled VNCR abort, ESR=%llx\n", esr);
+		return 1;
+	}
 
-		switch (ret) {
-		case -EAGAIN:
-			/* Let's try again... */
-			break;
-		case -ENOMEM:
-			/*
-			 * For guest_memfd, this indicates that it failed to
-			 * create a folio to back the memory. Inform userspace.
-			 */
-			if (is_gmem)
-				return 0;
-			/* Otherwise, let's try again... */
-			break;
-		case -EFAULT:
-		case -EIO:
-		case -EHWPOISON:
-			if (is_gmem)
-				return 0;
-			fallthrough;
-		case -EINVAL:
-		case -ENOENT:
-		case -EACCES:
-			/*
-			 * Translation failed, inject the corresponding
-			 * exception back to EL2.
-			 */
-			BUG_ON(!vt->wr.failed);
+	ret = kvm_translate_vncr(vcpu, &is_gmem);
+	switch (ret) {
+	case -EAGAIN:
+		/* Let's try again... */
+		break;
+	case -ENOMEM:
+		/*
+		 * For guest_memfd, this indicates that it failed to
+		 * create a folio to back the memory. Inform userspace.
+		 */
+		if (is_gmem)
+			return 0;
+		/* Otherwise, let's try again... */
+		break;
+	case -EFAULT:
+	case -EIO:
+	case -EHWPOISON:
+		if (is_gmem)

arch/arm64/kvm/nested.c static void handle_vncr_perm(struct kvm_vcpu *vcpu) · line 1507

Quoted 10 lines
arch/arm64/kvm/nested.c +10 -0
+			return 0;
+		fallthrough;
+	case -EINVAL:
+	case -ENOENT:
+	case -EACCES:
+		/*
+		 * Translation failed, inject the corresponding
+		 * exception back to EL2.
+		 */
+		BUG_ON(!vt->wr.failed);

Maybe we can lose this BUG_ON(). We rely on it being correct anyway.

Quoted 17 lines
arch/arm64/kvm/nested.c +2 -13
-			esr &= ~ESR_ELx_FSC;
-			esr |= FIELD_PREP(ESR_ELx_FSC, vt->wr.fst);
+		esr &= ~ESR_ELx_FSC;
+		esr |= FIELD_PREP(ESR_ELx_FSC, vt->wr.fst);
 
-			kvm_inject_nested_sync(vcpu, esr);
-			break;
-		case -EPERM:
-			/* Hack to deal with POE until we get kernel support */
-			handle_vncr_perm(vcpu);
-			break;
-		case 0:
-			break;
-		}
-	} else {
-		WARN_ONCE(1, "Unhandled VNCR abort, ESR=%llx\n", esr);

arch/arm64/kvm/nested.c static void handle_vncr_perm(struct kvm_vcpu *vcpu) · line 1520

Quoted 10 lines
arch/arm64/kvm/nested.c +8 -0
+		kvm_inject_nested_sync(vcpu, esr);
+		break;
+	case 0:
+		break;
 	}
 
+	perm = kvm_is_write_fault(vcpu) ? vt->wr.pw && vt->hpa_writable : vt->wr.pr;
+	if (!perm)
+		handle_vncr_perm(vcpu);
+

Isn't there a problem here, where anything that doesn't perform an early return ends up evaluating the permissions, even if the translation has failed? My hunch is that you'd want something like this:

arch/arm64/kvm/nested.c +3 -4
index 736fdd6e99cdc..0cefc73d97199 100644
@@ -1460,13 +1460,12 @@ int kvm_handle_vncr_abort(struct kvm_vcpu *vcpu)
 		kvm_inject_nested_sync(vcpu, esr);
 		break;
 	case 0:
+		perm = kvm_is_write_fault(vcpu) ? vt->wr.pw && vt->hpa_writable : vt->wr.pr;
+		if (!perm)
+			handle_vncr_perm(vcpu);
 		break;
 	}
 
-	perm = kvm_is_write_fault(vcpu) ? vt->wr.pw && vt->hpa_writable : vt->wr.pr;
-	if (!perm)
-		handle_vncr_perm(vcpu);
-
 	return 1;
 }
 

Thoughts?

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
#5 lore raw

Re: [PATCH RESEND v2 3/5] KVM: arm64: nv: Re-translate VNCR before injecting abort

In reply to Oliver Upton

Hi Oliver and Marc,

Quoted 148 lines · Oliver Upton

KVM faults in the VNCR page with FOLL_WRITE whenver the guest aborts for a write, similar to how a regular stage-2 mapping is handled. It is entirely possible that the guest reads from the VNCR before writing to it, in which case the PFN could only be read-only.

Invalidate the VNCR TLB and re-fetch the translation upon taking a VNCR abort, allowing the host mapping to be faulted in for write the second time around. Interestingly enough, this also satisfies the ordering requirements of FEAT_ETS2/3 between descriptor updates and MMU faults.

Fixes 2a359e072596 ("KVM: arm64: nv: Handle mapping of VNCR_EL2 at EL2")
---
 arch/arm64/kvm/nested.c | 115 +++++++++++++++-------------------------
 1 file changed, 44 insertions(+), 71 deletions(-)
arch/arm64/kvm/nested.c +38 -71
index ebd7ccfeee99..d5c4b57123a9 100644
@@ -1460,92 +1460,65 @@ static void handle_vncr_perm(struct kvm_vcpu *vcpu)
 	kvm_inject_nested_sync(vcpu, esr);
 }
 
-static bool kvm_vncr_tlb_lookup(struct kvm_vcpu *vcpu)
-{
-	struct vncr_tlb *vt = vcpu->arch.vncr_tlb;
-
-	lockdep_assert_held_read(&vcpu->kvm->mmu_lock);
-
-	if (!vt->valid)
-		return false;
-
-	if (read_vncr_el2(vcpu) != vt->gva)
-		return false;
-
-	if (vt->wr.nG)
-		return get_asid_by_regime(vcpu, TR_EL20) == vt->wr.asid;
-
-	return true;
-}
-
 int kvm_handle_vncr_abort(struct kvm_vcpu *vcpu)
 {
 	struct vncr_tlb *vt = vcpu->arch.vncr_tlb;
 	u64 esr = kvm_vcpu_get_esr(vcpu);
+	bool is_gmem, perm;
+	int ret;
 
 	WARN_ON_ONCE(!(esr & ESR_ELx_VNCR));
 
 	if (kvm_vcpu_abt_issea(vcpu))
 		return kvm_handle_guest_sea(vcpu);
 
-	if (esr_fsc_is_permission_fault(esr)) {
-		handle_vncr_perm(vcpu);
-	} else if (esr_fsc_is_translation_fault(esr)) {
-		bool valid, is_gmem = false;
-		int ret;
-
-		scoped_guard(read_lock, &vcpu->kvm->mmu_lock)
-			valid = kvm_vncr_tlb_lookup(vcpu);
-
-		if (!valid)
-			ret = kvm_translate_vncr(vcpu, &is_gmem);
-		else
-			ret = -EPERM;
+	if (!esr_fsc_is_translation_fault(esr) && !esr_fsc_is_permission_fault(esr)) {
+		WARN_ONCE(1, "Unhandled VNCR abort, ESR=%llx\n", esr);
+		return 1;
+	}
 
-		switch (ret) {
-		case -EAGAIN:
-			/* Let's try again... */
-			break;
-		case -ENOMEM:
-			/*
-			 * For guest_memfd, this indicates that it failed to
-			 * create a folio to back the memory. Inform userspace.
-			 */
-			if (is_gmem)
-				return 0;
-			/* Otherwise, let's try again... */
-			break;
-		case -EFAULT:
-		case -EIO:
-		case -EHWPOISON:
-			if (is_gmem)
-				return 0;
-			fallthrough;
-		case -EINVAL:
-		case -ENOENT:
-		case -EACCES:
-			/*
-			 * Translation failed, inject the corresponding
-			 * exception back to EL2.
-			 */
-			BUG_ON(!vt->wr.failed);
+	ret = kvm_translate_vncr(vcpu, &is_gmem);
+	switch (ret) {
+	case -EAGAIN:
+		/* Let's try again... */
+		break;
+	case -ENOMEM:
+		/*
+		 * For guest_memfd, this indicates that it failed to
+		 * create a folio to back the memory. Inform userspace.
+		 */
+		if (is_gmem)
+			return 0;
+		/* Otherwise, let's try again... */
+		break;
+	case -EFAULT:
+	case -EIO:
+	case -EHWPOISON:
+		if (is_gmem)
+			return 0;
+		fallthrough;
+	case -EINVAL:
+	case -ENOENT:
+	case -EACCES:
+		/*
+		 * Translation failed, inject the corresponding
+		 * exception back to EL2.
+		 */
+		BUG_ON(!vt->wr.failed);
 
-			esr &= ~ESR_ELx_FSC;
-			esr |= FIELD_PREP(ESR_ELx_FSC, vt->wr.fst);
+		esr &= ~ESR_ELx_FSC;
+		esr |= FIELD_PREP(ESR_ELx_FSC, vt->wr.fst);
 
-			kvm_inject_nested_sync(vcpu, esr);
-			break;
-		case -EPERM:
-			/* Hack to deal with POE until we get kernel support */
-			handle_vncr_perm(vcpu);
-			break;
-		case 0:
-			break;
-		}
-	} else {
-		WARN_ONCE(1, "Unhandled VNCR abort, ESR=%llx\n", esr);
+		kvm_inject_nested_sync(vcpu, esr);
+		break;

arch/arm64/kvm/nested.c static void handle_vncr_perm(struct kvm_vcpu *vcpu) · line 1523

Quoted 10 lines
arch/arm64/kvm/nested.c +6 -0
+	case 0:
+		break;
 	}
 
+	perm = kvm_is_write_fault(vcpu) ? vt->wr.pw && vt->hpa_writable : vt->wr.pr;
+	if (!perm)
+		handle_vncr_perm(vcpu);
+
 	return 1;
 }

Just a comment using this thread:

While reading this, I found this part of the code (not this patch in particular) a little bit difficult to reason about. I think it's because kvm_translate_vncr() is doing many things, and there are multiple potential failure reasons e.g. s1 walk fault, no memslot, gmem/user mem faultin errors, MMU notifier check, etc., and they are all mux'ed into an error code with some context visible by the caller.

So in kvm_handle_vncr_abort() we demux the error code and handle the errors with the help of the context (vt, is_gmem). We essentially have to keep track of what error codes correspond to what error reasons.

Do you think it is better if we refactor and handle the errors when they occur? Like inject the exception back to vEL2 right after getting the results of __kvm_translate_va(), and finish up the abort handling there. Same for other cases.

I can try it out and make it concrete if you also think this is reasonable. Probably after this series gets applied when the comments from Marc & Sashiko are addressed. (I reviewed and don't have additional comments though.)

Thanks,
Wei-Lin Chang

Quoted 4 lines
Signature · 2 lines
-- 
2.47.3
author maintainer
#6 lore raw

Re: [PATCH RESEND v2 3/5] KVM: arm64: nv: Re-translate VNCR before injecting abort

Hey Wei-Lin,

Sorry for the latency on my end.

Quoted 18 lines · Wei-Lin Chang

Just a comment using this thread:

While reading this, I found this part of the code (not this patch in particular) a little bit difficult to reason about. I think it's because kvm_translate_vncr() is doing many things, and there are multiple potential failure reasons e.g. s1 walk fault, no memslot, gmem/user mem faultin errors, MMU notifier check, etc., and they are all mux'ed into an error code with some context visible by the caller.

So in kvm_handle_vncr_abort() we demux the error code and handle the errors with the help of the context (vt, is_gmem). We essentially have to keep track of what error codes correspond to what error reasons.

Do you think it is better if we refactor and handle the errors when they occur? Like inject the exception back to vEL2 right after getting the results of __kvm_translate_va(), and finish up the abort handling there. Same for other cases.

Quoted 4 lines

I can try it out and make it concrete if you also think this is reasonable. Probably after this series gets applied when the comments from Marc & Sashiko are addressed. (I reviewed and don't have additional comments though.)

Yeah, returning error codes for 'normal' behavior is extremely difficult to work with. TBH, I'd rather we go a step further and rework the whole software PTW to only return errors in the case that we have to report it to userspace.

Otherwise, aborted walks due to guest behavior should ideally return 1 and and inspect the walk result to differentiate between a successful / aborted translation.

Thanks,
Oliver

#7 lore raw

Re: [PATCH RESEND v2 3/5] KVM: arm64: nv: Re-translate VNCR before injecting abort

Quoted 28 lines · Oliver Upton

Hey Wei-Lin,

Sorry for the latency on my end.

Quoted 18 lines · Wei-Lin Chang

Just a comment using this thread:

While reading this, I found this part of the code (not this patch in particular) a little bit difficult to reason about. I think it's because kvm_translate_vncr() is doing many things, and there are multiple potential failure reasons e.g. s1 walk fault, no memslot, gmem/user mem faultin errors, MMU notifier check, etc., and they are all mux'ed into an error code with some context visible by the caller.

So in kvm_handle_vncr_abort() we demux the error code and handle the errors with the help of the context (vt, is_gmem). We essentially have to keep track of what error codes correspond to what error reasons.

Do you think it is better if we refactor and handle the errors when they occur? Like inject the exception back to vEL2 right after getting the results of __kvm_translate_va(), and finish up the abort handling there. Same for other cases.

Quoted 4 lines

I can try it out and make it concrete if you also think this is reasonable. Probably after this series gets applied when the comments from Marc & Sashiko are addressed. (I reviewed and don't have additional comments though.)

Quoted 8 lines

Yeah, returning error codes for 'normal' behavior is extremely difficult to work with. TBH, I'd rather we go a step further and rework the whole software PTW to only return errors in the case that we have to report it to userspace.

Otherwise, aborted walks due to guest behavior should ideally return 1 and and inspect the walk result to differentiate between a successful / aborted translation.

Thanks for your suggestion!
I'll put this in the "potential improvements" list for now, there are other fixes I need to work on.

Thanks,
Wei-Lin Chang

Quoted 3 lines

Thanks,
Oliver

author maintainer
#8 lore raw

PATCH RESEND v2 4/5 KVM: arm64: nv: Inject SEA if guest VNCR isn't normal memory

In reply to Oliver Upton

When constructing an L1 VNCR mapping, KVM unconditionally uses cacheable memory attributes, even if the underlying PFN isn't memory. This gets particularly hairy if the endpoint doesn't support cacheable memory attributes, potentially throwing an SError on writeback...

While KVM does permit cacheable memory attributes on certain PFNMAP VMAs, kvm_translate_vncr() isn't currently grabbing the VMA. So do the simpler thing for now and just reject everything that isn't memory.

Fixes 2a359e072596 ("KVM: arm64: nv: Handle mapping of VNCR_EL2 at EL2")
---
 arch/arm64/kvm/nested.c | 11 +++++++++++
 1 file changed, 11 insertions(+)
arch/arm64/kvm/nested.c +11 -0
index d5c4b57123a9..a6bd60856fc3 100644
@@ -1413,6 +1413,17 @@ static int kvm_translate_vncr(struct kvm_vcpu *vcpu, bool *is_gmem)
 		writable = !(memslot->flags & KVM_MEM_READONLY);
 	}
 
+	/*
+	 * FIXME: This check is too restrictive as KVM allows cacheable memory
+	 * attributes for PFNMAP VMAs that have cacheable attributes in host
+	 * stage-1.
+	 */
+	if (!pfn_is_map_memory(pfn)) {
+		kvm_release_faultin_page(vcpu->kvm, page, true, false);
+		fail_s1_walk(&vt->wr, ESR_ELx_FSC_EXTABT, false);
+		return -EFAULT;
+	}
+
 	scoped_guard(write_lock, &vcpu->kvm->mmu_lock) {
 		if (mmu_invalidate_retry(vcpu->kvm, mmu_seq)) {
 			kvm_release_faultin_page(vcpu->kvm, page, true, false);
Signature · 1 line
2.47.3
author maintainer
#9 lore raw

PATCH RESEND v2 5/5 KVM: arm64: nv: Mark VM as bugged for unexpected VNCR abort

In reply to Oliver Upton

KVM is unlikely to resolve an unexpected VNCR abort, meaning that returning to the guest will likely leave the vCPU stuck in an abort loop. Bug the VM and exit to userspace instead.

---
 arch/arm64/kvm/nested.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/kvm/nested.c +2 -2
index a6bd60856fc3..e9bd4991f7a0 100644
@@ -1484,8 +1484,8 @@ int kvm_handle_vncr_abort(struct kvm_vcpu *vcpu)
 		return kvm_handle_guest_sea(vcpu);
 
 	if (!esr_fsc_is_translation_fault(esr) && !esr_fsc_is_permission_fault(esr)) {
-		WARN_ONCE(1, "Unhandled VNCR abort, ESR=%llx\n", esr);
-		return 1;
+		KVM_BUG(1, vcpu->kvm, "Unhandled VNCR abort, ESR=%llx\n", esr);
+		return -EIO;
 	}
 
 	ret = kvm_translate_vncr(vcpu, &is_gmem);
Signature · 1 line
2.47.3