Kernel KVM virtualization development
 help / color / mirror / Atom feed
From: Sascha Bischoff <Sascha.Bischoff@arm.com>
To: "linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"kvmarm@lists.linux.dev" <kvmarm@lists.linux.dev>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Cc: nd <nd@arm.com>, "maz@kernel.org" <maz@kernel.org>,
	"oliver.upton@linux.dev" <oliver.upton@linux.dev>,
	Joey Gouly <Joey.Gouly@arm.com>,
	Suzuki Poulose <Suzuki.Poulose@arm.com>,
	"yuzenghui@huawei.com" <yuzenghui@huawei.com>,
	"peter.maydell@linaro.org" <peter.maydell@linaro.org>,
	"lpieralisi@kernel.org" <lpieralisi@kernel.org>,
	Timothy Hayes <Timothy.Hayes@arm.com>,
	"fuad.tabba@linux.dev" <fuad.tabba@linux.dev>
Subject: [PATCH v5 30/49] KVM: arm64: gic-v5: Mask per-vCPU PPI state in vgic_v5_finalize_ppi_state()
Date: Fri, 7 Aug 2026 11:27:52 +0000	[thread overview]
Message-ID: <20260807111159.429128-31-sascha.bischoff@arm.com> (raw)
In-Reply-To: <20260807111159.429128-1-sascha.bischoff@arm.com>

Only a subset of the possible PPIs are exposed to a guest when running
with a vGICv5. First of all, only the architected PPIs are considered
by KVM. Secondly, only a set of those is exposed to a guest: those
corresponding to devices that KVM emulates, such as the timers and
PMU, and the GICv5 SW_PPI.

The finalisation of exposed PPIs happens on first vCPU run, as this is
the first time when the full set of exposed devices is known. At this
stage a mask is calculated, and this mask is applied both to hide
non-exposed PPI state from the guest and to reduce overhead when
iterating over the PPIs.

While preparing userspace access to the GICv5 system registers, it
became apparent that restoring the GICv5 PPI registers can result in a
mismatch between the state supplied by userspace and the state KVM
intends to expose. Userspace can provide Enable, Active, and Pending
state for PPIs that KVM has chosen to hide from the guest.

Userspace must restore PPI state before any vCPU runs. The userspace
access path added subsequently enforces this ordering. Rework
vgic_v5_finalize_ppi_state() to calculate the mask of exposed PPIs and
clear any state belonging to non-exposed PPIs. This ensures that only
the state KVM intends to expose is visible to the guest.

Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
---
 arch/arm64/kvm/arm.c          |  2 +-
 arch/arm64/kvm/vgic/vgic-v5.c | 71 +++++++++++++++++++++++++----------
 include/kvm/arm_vgic.h        |  2 +-
 3 files changed, 53 insertions(+), 22 deletions(-)

diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index 908a058314117..c56e2a007b2e7 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -974,7 +974,7 @@ int kvm_arch_vcpu_run_pid_change(struct kvm_vcpu *vcpu)
 			return ret;
 	}
 
-	ret = vgic_v5_finalize_ppi_state(kvm);
+	ret = vgic_v5_finalize_ppi_state(vcpu);
 	if (ret)
 		return ret;
 
diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index 62e069c21367a..beabc5980b2c1 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -781,9 +781,10 @@ int vgic_v5_map_resources(struct kvm *kvm)
 	return 0;
 }
 
-int vgic_v5_finalize_ppi_state(struct kvm *kvm)
+int vgic_v5_finalize_ppi_state(struct kvm_vcpu *vcpu)
 {
-	struct kvm_vcpu *vcpu0;
+	struct kvm *kvm	= vcpu->kvm;
+	struct vgic_v5_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v5;
 	int i;
 
 	if (!vgic_is_v5(kvm))
@@ -792,35 +793,65 @@ int vgic_v5_finalize_ppi_state(struct kvm *kvm)
 	guard(mutex)(&kvm->arch.config_lock);
 
 	/*
-	 * If SW_PPI has been advertised, then we know we already
-	 * initialised the whole thing, and we can return early. Yes,
-	 * this is pretty hackish as far as state tracking goes...
+	 * Discover the set of PPIs that are exposed to the guest once per VM.
+	 * Once known, apply that mask to each VCPU's restored PPI state as the
+	 * VCPUs are first run.
 	 */
-	if (test_bit(GICV5_ARCH_PPI_SW_PPI, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask))
-		return 0;
-
-	/* The PPI state for all VCPUs should be the same. Pick the first. */
-	vcpu0 = kvm_get_vcpu(kvm, 0);
+	if (!test_bit(GICV5_ARCH_PPI_SW_PPI, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask)) {
+		bitmap_zero(kvm->arch.vgic.gicv5_vm.vgic_ppi_mask,
+			    VGIC_V5_NR_PRIVATE_IRQS);
+		bitmap_zero(kvm->arch.vgic.gicv5_vm.vgic_ppi_hmr,
+			    VGIC_V5_NR_PRIVATE_IRQS);
+
+		for_each_set_bit(i, ppi_caps.impl_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS) {
+			const u32 intid = vgic_v5_make_ppi(i);
+			struct vgic_irq *irq;
+
+			irq = vgic_get_vcpu_irq(vcpu, intid);
+
+			/* Expose PPIs with an owner or the SW_PPI, only */
+			scoped_guard(raw_spinlock_irqsave, &irq->irq_lock) {
+				if (irq->owner || i == GICV5_ARCH_PPI_SW_PPI) {
+					__set_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask);
+					__assign_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_hmr,
+						     irq->config == VGIC_CONFIG_LEVEL);
+				}
+			}
 
-	bitmap_zero(kvm->arch.vgic.gicv5_vm.vgic_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS);
-	bitmap_zero(kvm->arch.vgic.gicv5_vm.vgic_ppi_hmr, VGIC_V5_NR_PRIVATE_IRQS);
+			vgic_put_irq(kvm, irq);
+		}
+	}
 
-	for_each_set_bit(i, ppi_caps.impl_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS) {
+	/*
+	 * Apply the mask to Enable, Active. Skip pending as that's calculated
+	 * on guest entry.
+	 */
+	bitmap_and(cpu_if->vgic_ppi_enabler, cpu_if->vgic_ppi_enabler,
+		   kvm->arch.vgic.gicv5_vm.vgic_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS);
+	bitmap_and(cpu_if->vgic_ppi_activer, cpu_if->vgic_ppi_activer,
+		   kvm->arch.vgic.gicv5_vm.vgic_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS);
+
+	/* Also update the vgic_irqs */
+	for (i = 0; i < VGIC_V5_NR_PRIVATE_IRQS; i++) {
+		bool visible = test_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask);
 		const u32 intid = vgic_v5_make_ppi(i);
 		struct vgic_irq *irq;
 
-		irq = vgic_get_vcpu_irq(vcpu0, intid);
+		irq = vgic_get_vcpu_irq(vcpu, intid);
 
-		/* Expose PPIs with an owner or the SW_PPI, only */
 		scoped_guard(raw_spinlock_irqsave, &irq->irq_lock) {
-			if (irq->owner || i == GICV5_ARCH_PPI_SW_PPI) {
-				__set_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask);
-				__assign_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_hmr,
-					     irq->config == VGIC_CONFIG_LEVEL);
+			if (!visible) {
+				irq->enabled = false;
+				irq->active = false;
+				irq->pending_latch = false;
+				irq->line_level = false;
+			} else {
+				irq->enabled = test_bit(i, cpu_if->vgic_ppi_enabler);
+				irq->active = test_bit(i, cpu_if->vgic_ppi_activer);
 			}
 		}
 
-		vgic_put_irq(vcpu0->kvm, irq);
+		vgic_put_irq(kvm, irq);
 	}
 
 	return 0;
diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
index 9f16ee63d5763..b6526c678a5a2 100644
--- a/include/kvm/arm_vgic.h
+++ b/include/kvm/arm_vgic.h
@@ -823,7 +823,7 @@ static inline u32 vgic_v5_irs_max_vpes(const struct vgic_v5_irs_caps *caps)
 	return BIT(FIELD_GET(GICV5_IRS_IDR4_VPE_ID_BITS, caps->idr4) + 1);
 }
 
-int vgic_v5_finalize_ppi_state(struct kvm *kvm);
+int vgic_v5_finalize_ppi_state(struct kvm_vcpu *vcpu);
 bool vgic_v5_ppi_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq,
 				  unsigned long flags);
 void vgic_v5_set_ppi_dvi(struct kvm_vcpu *vcpu, struct vgic_irq *irq, bool dvi);
-- 
2.34.1

  parent reply	other threads:[~2026-08-07 11:28 UTC|newest]

Thread overview: 84+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-07 11:12 [PATCH v5 00/49] KVM: arm64: Add GICv5 IRS support Sascha Bischoff
2026-08-07 11:13 ` [PATCH v5 01/49] irqchip/gic-v5: Allow KVM setup without a maintenance IRQ Sascha Bischoff
2026-08-07 11:13 ` [PATCH v5 02/49] irqchip/gic-v5: Provide OF IRS config frame attrs to KVM Sascha Bischoff
2026-08-07 11:53   ` sashiko-bot
2026-08-07 11:14 ` [PATCH v5 03/49] irqchip/gic-v5: Set up gic_kvm_info on ACPI hosts Sascha Bischoff
2026-08-07 12:01   ` sashiko-bot
2026-08-07 13:44   ` Lorenzo Pieralisi
2026-08-07 11:14 ` [PATCH v5 04/49] KVM: arm64: gic-v5: Define remaining IRS MMIO registers Sascha Bischoff
2026-08-07 12:05   ` sashiko-bot
2026-08-07 11:15 ` [PATCH v5 05/49] arm64/sysreg: Add GICv5 GIC VDPEND encoding Sascha Bischoff
2026-08-07 11:15 ` [PATCH v5 06/49] arm64/sysreg: Update ICC_CR0_EL1 with LINK and LINK_IDLE fields Sascha Bischoff
2026-08-07 12:17   ` sashiko-bot
2026-08-07 11:16 ` [PATCH v5 07/49] KVM: arm64: gic-v5: Cache host IRS ID registers Sascha Bischoff
2026-08-07 12:27   ` sashiko-bot
2026-08-07 11:16 ` [PATCH v5 08/49] KVM: arm64: gic-v5: Add VPE doorbell domain Sascha Bischoff
2026-08-07 12:45   ` sashiko-bot
2026-08-07 11:17 ` [PATCH v5 09/49] KVM: arm64: gic-v5: Create and manage VM and VPE tables Sascha Bischoff
2026-08-07 12:50   ` sashiko-bot
2026-08-07 11:17 ` [PATCH v5 10/49] KVM: arm64: gic-v5: Introduce guest IST alloc and management Sascha Bischoff
2026-08-07 13:07   ` sashiko-bot
2026-08-07 11:18 ` [PATCH v5 11/49] KVM: arm64: gic-v5: Implement VMT/vIST IRS MMIO Ops Sascha Bischoff
2026-08-07 13:13   ` sashiko-bot
2026-08-07 11:18 ` [PATCH v5 12/49] KVM: arm64: gic-v5: Keep GICv5 vCPU limit model-specific Sascha Bischoff
2026-08-07 13:30   ` sashiko-bot
2026-08-07 11:19 ` [PATCH v5 13/49] KVM: arm64: gic-v5: Implement VPE IRS MMIO Ops Sascha Bischoff
2026-08-07 11:19 ` [PATCH v5 14/49] KVM: arm64: gic-v5: Set up VMTEs and VPE doorbells Sascha Bischoff
2026-08-07 13:42   ` sashiko-bot
2026-08-07 11:20 ` [PATCH v5 15/49] KVM: arm64: gic-v5: Add resident/non-resident hyp calls Sascha Bischoff
2026-08-07 11:20 ` [PATCH v5 16/49] KVM: arm64: gic-v5: Request doorbells when VPEs enter WFI Sascha Bischoff
2026-08-07 14:17   ` sashiko-bot
2026-08-07 11:21 ` [PATCH v5 17/49] KVM: arm64: gic-v5: Introduce struct vgic_v5_irs and IRS base address Sascha Bischoff
2026-08-07 11:21 ` [PATCH v5 18/49] KVM: arm64: gic-v5: Add IRS IODEV support to MMIO handlers Sascha Bischoff
2026-08-07 11:22 ` [PATCH v5 19/49] KVM: arm64: gic-v5: Add KVM_VGIC_V5_ADDR_TYPE_IRS to UAPI Sascha Bischoff
2026-08-07 14:27   ` sashiko-bot
2026-08-07 11:22 ` [PATCH v5 20/49] KVM: arm64: gic-v5: Add GICv5 IRS IODEV and MMIO emulation Sascha Bischoff
2026-08-07 14:34   ` sashiko-bot
2026-08-07 11:23 ` [PATCH v5 21/49] KVM: arm64: gic-v5: Initialise per-VM IRS state Sascha Bischoff
2026-08-07 14:49   ` sashiko-bot
2026-08-07 11:23 ` [PATCH v5 22/49] KVM: arm64: gic-v5: Register the IRS IODEV Sascha Bischoff
2026-08-07 14:52   ` sashiko-bot
2026-08-07 11:24 ` [PATCH v5 23/49] KVM: arm64: gic-v5: Set IRICHPPIDIS based on IRS enable state Sascha Bischoff
2026-08-07 11:24 ` [PATCH v5 24/49] KVM: arm64: selftests: Update vGICv5 selftest to set IRS address Sascha Bischoff
2026-08-07 15:04   ` sashiko-bot
2026-08-07 11:25 ` [PATCH v5 25/49] KVM: arm64: gic-v5: Add GIC VDPEND hyp call Sascha Bischoff
2026-08-07 11:25 ` [PATCH v5 26/49] KVM: arm64: gic: Introduce set_pending_state() to irq_ops Sascha Bischoff
2026-08-07 15:14   ` sashiko-bot
2026-08-07 11:26 ` [PATCH v5 27/49] KVM: arm64: gic-v5: Support SPI injection Sascha Bischoff
2026-08-07 15:23   ` sashiko-bot
2026-08-07 11:26 ` [PATCH v5 28/49] Documentation: KVM: Extend VGICv5 device attribute docs Sascha Bischoff
2026-08-07 15:29   ` sashiko-bot
2026-08-07 11:27 ` [PATCH v5 29/49] KVM: arm64: gic-v5: Add GICv5 SPI injection to irqfd Sascha Bischoff
2026-08-07 15:40   ` sashiko-bot
2026-08-07 11:27 ` Sascha Bischoff [this message]
2026-08-07 11:28 ` [PATCH v5 31/49] KVM: arm64: gic-v5: Add GICv5 EL1 sysreg userspace accessors Sascha Bischoff
2026-08-07 16:27   ` sashiko-bot
2026-08-07 11:28 ` [PATCH v5 32/49] KVM: arm64: gic-v5: Handle userspace accesses to IRS MMIO region Sascha Bischoff
2026-08-07 16:20   ` sashiko-bot
2026-08-07 11:29 ` [PATCH v5 33/49] KVM: arm64: gic-v5: Add CoreSight MMIO regs to IRS Sascha Bischoff
2026-08-07 11:29 ` [PATCH v5 34/49] KVM: arm64: gic-v5: Add VGICv5 IST save/restore UAPI Sascha Bischoff
2026-08-07 16:30   ` sashiko-bot
2026-08-07 11:30 ` [PATCH v5 35/49] KVM: arm64: gic-v5: Implement save/restore mechanisms for ISTs Sascha Bischoff
2026-08-07 16:48   ` sashiko-bot
2026-08-07 11:30 ` [PATCH v5 36/49] Documentation: KVM: Document KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS for VGICv5 Sascha Bischoff
2026-08-07 16:55   ` sashiko-bot
2026-08-07 11:31 ` [PATCH v5 37/49] Documentation: KVM: Add KVM_DEV_ARM_VGIC_GRP_IRS_REGS to VGICv5 docs Sascha Bischoff
2026-08-07 16:52   ` sashiko-bot
2026-08-07 11:31 ` [PATCH v5 38/49] Documentation: KVM: Add docs for KVM_DEV_ARM_VGIC_GRP_IST Sascha Bischoff
2026-08-07 11:32 ` [PATCH v5 39/49] Documentation: KVM: Add the VGICv5 IRS save/restore sequences Sascha Bischoff
2026-08-07 11:32 ` [PATCH v5 40/49] KVM: selftests: Add VGICv5 IRS address attribute tests Sascha Bischoff
2026-08-07 11:33 ` [PATCH v5 41/49] KVM: selftests: Add VGICv5 NR_IRQS " Sascha Bischoff
2026-08-07 17:12   ` sashiko-bot
2026-08-07 11:33 ` [PATCH v5 42/49] KVM: selftests: Add VGICv5 IRS_REGS " Sascha Bischoff
2026-08-07 17:17   ` sashiko-bot
2026-08-07 11:34 ` [PATCH v5 43/49] KVM: selftests: Add VGICv5 IST " Sascha Bischoff
2026-08-07 17:21   ` sashiko-bot
2026-08-07 11:35 ` [PATCH v5 44/49] KVM: selftests: Add VGICv5 USERSPACE_PPIS tests Sascha Bischoff
2026-08-07 11:35 ` [PATCH v5 45/49] KVM: selftests: Add VGICv5 CPU sysreg attribute tests Sascha Bischoff
2026-08-07 11:36 ` [PATCH v5 46/49] KVM: selftests: Add VGICv5 SPI injection tests Sascha Bischoff
2026-08-07 11:36 ` [PATCH v5 47/49] KVM: selftests: Add VGICv5 LPI delivery tests Sascha Bischoff
2026-08-07 17:39   ` sashiko-bot
2026-08-07 11:37 ` [PATCH v5 48/49] KVM: selftests: Add VGICv5 IST save/restore coverage Sascha Bischoff
2026-08-07 17:50   ` sashiko-bot
2026-08-07 11:37 ` [PATCH v5 49/49] KVM: selftests: Add VGICv5 sparse vCPU IDs test Sascha Bischoff
2026-08-07 17:56   ` sashiko-bot

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=20260807111159.429128-31-sascha.bischoff@arm.com \
    --to=sascha.bischoff@arm.com \
    --cc=Joey.Gouly@arm.com \
    --cc=Suzuki.Poulose@arm.com \
    --cc=Timothy.Hayes@arm.com \
    --cc=fuad.tabba@linux.dev \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=lpieralisi@kernel.org \
    --cc=maz@kernel.org \
    --cc=nd@arm.com \
    --cc=oliver.upton@linux.dev \
    --cc=peter.maydell@linaro.org \
    --cc=yuzenghui@huawei.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox