All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marc Zyngier <maz@kernel.org>
To: Zenghui Yu <yuzenghui@huawei.com>
Cc: <kvmarm@lists.linux.dev>, <linux-arm-kernel@lists.infradead.org>,
	<kvm@vger.kernel.org>, James Morse <james.morse@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Oliver Upton <oliver.upton@linux.dev>,
	Joey Gouly <joey.gouly@arm.com>,
	Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>,
	Xu Zhao <zhaoxu.35@bytedance.com>,
	Eric Auger <eric.auger@redhat.com>
Subject: Re: [PATCH v2 02/11] KVM: arm64: vgic-its: Treat the collection target address as a vcpu_id
Date: Thu, 21 Sep 2023 12:46:51 +0100	[thread overview]
Message-ID: <861qerpx3o.wl-maz@kernel.org> (raw)
In-Reply-To: <d1d5ba61-1fac-a701-38db-b8bd5dcebeb8@huawei.com>

On Thu, 21 Sep 2023 10:14:55 +0100,
Zenghui Yu <yuzenghui@huawei.com> wrote:
> 
> On 2023/9/21 2:17, Marc Zyngier wrote:
> > Since our emulated ITS advertises GITS_TYPER.PTA=0, the target
> > address associated to a collection is a PE number and not
> > an address. So far, so good. However, the PE number is what userspace
> > has provided given us (aka the vcpu_id), and not the internal vcpu
> > index.
> > 
> > Make sure we consistently retrieve the vcpu by ID rather than
> > by index, adding a helper that deals with most of the cases.
> > 
> > Signed-off-by: Marc Zyngier <maz@kernel.org>
> 
> Looks good, with 2 more points:
> 
> - Like patch#1, we should have a go at all
>  'target_addr >= kvm->online_vcpus' comparisons in vgic-its.c
> - There is still a remaining kvm_get_vcpu() in vgic_its_restore_ite()
>  which needs to be fixed

Yup, well spotted. I have this additional hack which I plan to put on
top.

Thanks,

	M.

diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c
index 4aadcd24f6f6..6ec9dd970cbb 100644
--- a/arch/arm64/kvm/vgic/vgic-its.c
+++ b/arch/arm64/kvm/vgic/vgic-its.c
@@ -1248,21 +1248,22 @@ static int vgic_its_cmd_handle_mapc(struct kvm *kvm, struct vgic_its *its,
 				    u64 *its_cmd)
 {
 	u16 coll_id;
-	u32 target_addr;
 	struct its_collection *collection;
 	bool valid;
 
 	valid = its_cmd_get_validbit(its_cmd);
 	coll_id = its_cmd_get_collection(its_cmd);
-	target_addr = its_cmd_get_target_addr(its_cmd);
-
-	if (target_addr >= atomic_read(&kvm->online_vcpus))
-		return E_ITS_MAPC_PROCNUM_OOR;
 
 	if (!valid) {
 		vgic_its_free_collection(its, coll_id);
 		vgic_its_invalidate_cache(kvm);
 	} else {
+		struct kvm_vcpu *vcpu;
+
+		vcpu = kvm_get_vcpu_by_id(kvm, its_cmd_get_target_addr(its_cmd));
+		if (!vcpu)
+			return E_ITS_MAPC_PROCNUM_OOR;
+
 		collection = find_collection(its, coll_id);
 
 		if (!collection) {
@@ -1276,9 +1277,9 @@ static int vgic_its_cmd_handle_mapc(struct kvm *kvm, struct vgic_its *its,
 							coll_id);
 			if (ret)
 				return ret;
-			collection->target_addr = target_addr;
+			collection->target_addr = vcpu->vcpu_id;
 		} else {
-			collection->target_addr = target_addr;
+			collection->target_addr = vcpu->vcpu_id;
 			update_affinity_collection(kvm, its, collection);
 		}
 	}
@@ -1405,24 +1406,21 @@ static int vgic_its_cmd_handle_invall(struct kvm *kvm, struct vgic_its *its,
 static int vgic_its_cmd_handle_movall(struct kvm *kvm, struct vgic_its *its,
 				      u64 *its_cmd)
 {
-	u32 target1_addr = its_cmd_get_target_addr(its_cmd);
-	u32 target2_addr = its_cmd_mask_field(its_cmd, 3, 16, 32);
 	struct kvm_vcpu *vcpu1, *vcpu2;
 	struct vgic_irq *irq;
 	u32 *intids;
 	int irq_count, i;
 
-	if (target1_addr >= atomic_read(&kvm->online_vcpus) ||
-	    target2_addr >= atomic_read(&kvm->online_vcpus))
+	/* We advertise GITS_TYPER.PTA==0, making the address the vcpu ID */
+	vcpu1 = kvm_get_vcpu_by_id(kvm, its_cmd_get_target_addr(its_cmd));
+	vcpu2 = kvm_get_vcpu_by_id(kvm, its_cmd_mask_field(its_cmd, 3, 16, 32));
+
+	if (!vcpu1 || !vcpu2)
 		return E_ITS_MOVALL_PROCNUM_OOR;
 
-	if (target1_addr == target2_addr)
+	if (vcpu1 == vcpu2)
 		return 0;
 
-	/* We advertise GITS_TYPER.PTA==0, making the address the vcpu ID */
-	vcpu1 = kvm_get_vcpu_by_id(kvm, target1_addr);
-	vcpu2 = kvm_get_vcpu_by_id(kvm, target2_addr);
-
 	irq_count = vgic_copy_lpi_list(kvm, vcpu1, &intids);
 	if (irq_count < 0)
 		return irq_count;
@@ -2265,7 +2263,7 @@ static int vgic_its_restore_ite(struct vgic_its *its, u32 event_id,
 		return PTR_ERR(ite);
 
 	if (its_is_collection_mapped(collection))
-		vcpu = kvm_get_vcpu(kvm, collection->target_addr);
+		vcpu = kvm_get_vcpu_by_id(kvm, collection->target_addr);
 
 	irq = vgic_add_lpi(kvm, lpi_id, vcpu);
 	if (IS_ERR(irq)) {
@@ -2580,7 +2578,7 @@ static int vgic_its_restore_cte(struct vgic_its *its, gpa_t gpa, int esz)
 	coll_id = val & KVM_ITS_CTE_ICID_MASK;
 
 	if (target_addr != COLLECTION_NOT_MAPPED &&
-	    target_addr >= atomic_read(&kvm->online_vcpus))
+	    !kvm_get_vcpu_by_id(kvm, target_addr))
 		return -EINVAL;
 
 	collection = find_collection(its, coll_id);

-- 
Without deviation from the norm, progress is not possible.

WARNING: multiple messages have this Message-ID (diff)
From: Marc Zyngier <maz@kernel.org>
To: Zenghui Yu <yuzenghui@huawei.com>
Cc: <kvmarm@lists.linux.dev>, <linux-arm-kernel@lists.infradead.org>,
	<kvm@vger.kernel.org>, James Morse <james.morse@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Oliver Upton <oliver.upton@linux.dev>,
	Joey Gouly <joey.gouly@arm.com>,
	Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>,
	Xu Zhao <zhaoxu.35@bytedance.com>,
	Eric Auger <eric.auger@redhat.com>
Subject: Re: [PATCH v2 02/11] KVM: arm64: vgic-its: Treat the collection target address as a vcpu_id
Date: Thu, 21 Sep 2023 12:46:51 +0100	[thread overview]
Message-ID: <861qerpx3o.wl-maz@kernel.org> (raw)
In-Reply-To: <d1d5ba61-1fac-a701-38db-b8bd5dcebeb8@huawei.com>

On Thu, 21 Sep 2023 10:14:55 +0100,
Zenghui Yu <yuzenghui@huawei.com> wrote:
> 
> On 2023/9/21 2:17, Marc Zyngier wrote:
> > Since our emulated ITS advertises GITS_TYPER.PTA=0, the target
> > address associated to a collection is a PE number and not
> > an address. So far, so good. However, the PE number is what userspace
> > has provided given us (aka the vcpu_id), and not the internal vcpu
> > index.
> > 
> > Make sure we consistently retrieve the vcpu by ID rather than
> > by index, adding a helper that deals with most of the cases.
> > 
> > Signed-off-by: Marc Zyngier <maz@kernel.org>
> 
> Looks good, with 2 more points:
> 
> - Like patch#1, we should have a go at all
>  'target_addr >= kvm->online_vcpus' comparisons in vgic-its.c
> - There is still a remaining kvm_get_vcpu() in vgic_its_restore_ite()
>  which needs to be fixed

Yup, well spotted. I have this additional hack which I plan to put on
top.

Thanks,

	M.

diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c
index 4aadcd24f6f6..6ec9dd970cbb 100644
--- a/arch/arm64/kvm/vgic/vgic-its.c
+++ b/arch/arm64/kvm/vgic/vgic-its.c
@@ -1248,21 +1248,22 @@ static int vgic_its_cmd_handle_mapc(struct kvm *kvm, struct vgic_its *its,
 				    u64 *its_cmd)
 {
 	u16 coll_id;
-	u32 target_addr;
 	struct its_collection *collection;
 	bool valid;
 
 	valid = its_cmd_get_validbit(its_cmd);
 	coll_id = its_cmd_get_collection(its_cmd);
-	target_addr = its_cmd_get_target_addr(its_cmd);
-
-	if (target_addr >= atomic_read(&kvm->online_vcpus))
-		return E_ITS_MAPC_PROCNUM_OOR;
 
 	if (!valid) {
 		vgic_its_free_collection(its, coll_id);
 		vgic_its_invalidate_cache(kvm);
 	} else {
+		struct kvm_vcpu *vcpu;
+
+		vcpu = kvm_get_vcpu_by_id(kvm, its_cmd_get_target_addr(its_cmd));
+		if (!vcpu)
+			return E_ITS_MAPC_PROCNUM_OOR;
+
 		collection = find_collection(its, coll_id);
 
 		if (!collection) {
@@ -1276,9 +1277,9 @@ static int vgic_its_cmd_handle_mapc(struct kvm *kvm, struct vgic_its *its,
 							coll_id);
 			if (ret)
 				return ret;
-			collection->target_addr = target_addr;
+			collection->target_addr = vcpu->vcpu_id;
 		} else {
-			collection->target_addr = target_addr;
+			collection->target_addr = vcpu->vcpu_id;
 			update_affinity_collection(kvm, its, collection);
 		}
 	}
@@ -1405,24 +1406,21 @@ static int vgic_its_cmd_handle_invall(struct kvm *kvm, struct vgic_its *its,
 static int vgic_its_cmd_handle_movall(struct kvm *kvm, struct vgic_its *its,
 				      u64 *its_cmd)
 {
-	u32 target1_addr = its_cmd_get_target_addr(its_cmd);
-	u32 target2_addr = its_cmd_mask_field(its_cmd, 3, 16, 32);
 	struct kvm_vcpu *vcpu1, *vcpu2;
 	struct vgic_irq *irq;
 	u32 *intids;
 	int irq_count, i;
 
-	if (target1_addr >= atomic_read(&kvm->online_vcpus) ||
-	    target2_addr >= atomic_read(&kvm->online_vcpus))
+	/* We advertise GITS_TYPER.PTA==0, making the address the vcpu ID */
+	vcpu1 = kvm_get_vcpu_by_id(kvm, its_cmd_get_target_addr(its_cmd));
+	vcpu2 = kvm_get_vcpu_by_id(kvm, its_cmd_mask_field(its_cmd, 3, 16, 32));
+
+	if (!vcpu1 || !vcpu2)
 		return E_ITS_MOVALL_PROCNUM_OOR;
 
-	if (target1_addr == target2_addr)
+	if (vcpu1 == vcpu2)
 		return 0;
 
-	/* We advertise GITS_TYPER.PTA==0, making the address the vcpu ID */
-	vcpu1 = kvm_get_vcpu_by_id(kvm, target1_addr);
-	vcpu2 = kvm_get_vcpu_by_id(kvm, target2_addr);
-
 	irq_count = vgic_copy_lpi_list(kvm, vcpu1, &intids);
 	if (irq_count < 0)
 		return irq_count;
@@ -2265,7 +2263,7 @@ static int vgic_its_restore_ite(struct vgic_its *its, u32 event_id,
 		return PTR_ERR(ite);
 
 	if (its_is_collection_mapped(collection))
-		vcpu = kvm_get_vcpu(kvm, collection->target_addr);
+		vcpu = kvm_get_vcpu_by_id(kvm, collection->target_addr);
 
 	irq = vgic_add_lpi(kvm, lpi_id, vcpu);
 	if (IS_ERR(irq)) {
@@ -2580,7 +2578,7 @@ static int vgic_its_restore_cte(struct vgic_its *its, gpa_t gpa, int esz)
 	coll_id = val & KVM_ITS_CTE_ICID_MASK;
 
 	if (target_addr != COLLECTION_NOT_MAPPED &&
-	    target_addr >= atomic_read(&kvm->online_vcpus))
+	    !kvm_get_vcpu_by_id(kvm, target_addr))
 		return -EINVAL;
 
 	collection = find_collection(its, coll_id);

-- 
Without deviation from the norm, progress is not possible.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2023-09-21 11:46 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-20 18:17 [PATCH v2 00/11] KVM: arm64: Accelerate lookup of vcpus by MPIDR values (and other fixes) Marc Zyngier
2023-09-20 18:17 ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 01/11] KVM: arm64: vgic: Make kvm_vgic_inject_irq() take a vcpu pointer Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-21  9:11   ` Zenghui Yu
2023-09-21  9:11     ` Zenghui Yu
2023-09-21 11:20     ` Marc Zyngier
2023-09-21 11:20       ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 02/11] KVM: arm64: vgic-its: Treat the collection target address as a vcpu_id Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-21  9:14   ` Zenghui Yu
2023-09-21  9:14     ` Zenghui Yu
2023-09-21 11:46     ` Marc Zyngier [this message]
2023-09-21 11:46       ` Marc Zyngier
2023-09-21 13:12       ` Zenghui Yu
2023-09-21 13:12         ` Zenghui Yu
2023-09-20 18:17 ` [PATCH v2 03/11] KVM: arm64: vgic-v3: Refactor GICv3 SGI generation Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-21  9:42   ` Joey Gouly
2023-09-21  9:42     ` Joey Gouly
2023-09-21 11:13     ` Marc Zyngier
2023-09-21 11:13       ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 04/11] KVM: arm64: vgic-v2: Use cpuid from userspace as vcpu_id Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 05/11] KVM: arm64: vgic: Use vcpu_idx for the debug information Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 06/11] KVM: arm64: Use vcpu_idx for invalidation tracking Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-21  9:16   ` Zenghui Yu
2023-09-21  9:16     ` Zenghui Yu
2023-09-21 12:58     ` Marc Zyngier
2023-09-21 12:58       ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 07/11] KVM: arm64: Simplify kvm_vcpu_get_mpidr_aff() Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 08/11] KVM: arm64: Build MPIDR to vcpu index cache at runtime Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 09/11] KVM: arm64: Fast-track kvm_mpidr_to_vcpu() when mpidr_data is available Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-20 18:17 ` [PATCH v2 10/11] KVM: arm64: vgic-v3: Optimize affinity-based SGI injection Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-21  9:32   ` Zenghui Yu
2023-09-21  9:32     ` Zenghui Yu
2023-09-20 18:17 ` [PATCH v2 11/11] KVM: arm64: Clarify the ordering requirements for vcpu/RD creation Marc Zyngier
2023-09-20 18:17   ` Marc Zyngier
2023-09-21  2:11 ` [PATCH v2 00/11] KVM: arm64: Accelerate lookup of vcpus by MPIDR values (and other fixes) Oliver Upton
2023-09-21  2:11   ` Oliver Upton
2023-09-21  9:39 ` Zenghui Yu
2023-09-21  9:39   ` Zenghui Yu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=861qerpx3o.wl-maz@kernel.org \
    --to=maz@kernel.org \
    --cc=eric.auger@redhat.com \
    --cc=james.morse@arm.com \
    --cc=joey.gouly@arm.com \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=oliver.upton@linux.dev \
    --cc=shameerali.kolothum.thodi@huawei.com \
    --cc=suzuki.poulose@arm.com \
    --cc=yuzenghui@huawei.com \
    --cc=zhaoxu.35@bytedance.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.