Re: [PATCH v2 5/5] KVM: arm64: Disable privileged hypercalls after pKVM finalisation

On Thu, Oct 07, 2021 at 01:56:47PM +0100, Marc Zyngier wrote:

On Tue, 05 Oct 2021 12:37:21 +0100,
Will Deacon [off-list ref] wrote:

quoted

After pKVM has been 'finalised' using the __pkvm_prot_finalize hypercall,
the calling CPU will have a Stage-2 translation enabled to prevent access
to memory pages owned by EL2.

Although this forms a significant part of the process to deprivilege the
host kernel, we also need to ensure that the hypercall interface is
reduced so that the EL2 code cannot, for example, be re-initialised using
a new set of vectors.

Re-order the hypercalls so that only a suffix remains available after
finalisation of pKVM.

Cc: Marc Zyngier <maz@kernel.org>
Cc: Quentin Perret <redacted>
Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/include/asm/kvm_asm.h   | 43 ++++++++++++++++--------------
 arch/arm64/kvm/hyp/nvhe/hyp-main.c | 37 +++++++++++++++++--------
 2 files changed, 49 insertions(+), 31 deletions(-)

diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h
index e86045ac43ba..68630fd382c5 100644
--- a/arch/arm64/include/asm/kvm_asm.h
+++ b/arch/arm64/include/asm/kvm_asm.h

@@ -43,27 +43,30 @@
 
 #define KVM_HOST_SMCCC_FUNC(name) KVM_HOST_SMCCC_ID(__KVM_HOST_SMCCC_FUNC_##name)
 
+/* Hypercalls available only prior to pKVM finalisation */
 #define __KVM_HOST_SMCCC_FUNC___kvm_hyp_init			0
-#define __KVM_HOST_SMCCC_FUNC___kvm_vcpu_run			1
-#define __KVM_HOST_SMCCC_FUNC___kvm_flush_vm_context		2
-#define __KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid_ipa		3
-#define __KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid		4
-#define __KVM_HOST_SMCCC_FUNC___kvm_flush_cpu_context		5
-#define __KVM_HOST_SMCCC_FUNC___kvm_timer_set_cntvoff		6
-#define __KVM_HOST_SMCCC_FUNC___kvm_enable_ssbs			7
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_get_gic_config		8
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_read_vmcr		9
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_write_vmcr		10
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_init_lrs		11
-#define __KVM_HOST_SMCCC_FUNC___kvm_get_mdcr_el2		12
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_save_aprs		13
-#define __KVM_HOST_SMCCC_FUNC___vgic_v3_restore_aprs		14
-#define __KVM_HOST_SMCCC_FUNC___pkvm_init			15
-#define __KVM_HOST_SMCCC_FUNC___pkvm_host_share_hyp		16
-#define __KVM_HOST_SMCCC_FUNC___pkvm_create_private_mapping	17
-#define __KVM_HOST_SMCCC_FUNC___pkvm_cpu_set_vector		18
-#define __KVM_HOST_SMCCC_FUNC___pkvm_prot_finalize		19
-#define __KVM_HOST_SMCCC_FUNC___kvm_adjust_pc			20
+#define __KVM_HOST_SMCCC_FUNC___kvm_get_mdcr_el2		1
+#define __KVM_HOST_SMCCC_FUNC___pkvm_init			2
+#define __KVM_HOST_SMCCC_FUNC___pkvm_create_private_mapping	3
+#define __KVM_HOST_SMCCC_FUNC___pkvm_cpu_set_vector		4
+#define __KVM_HOST_SMCCC_FUNC___kvm_enable_ssbs			5
+#define __KVM_HOST_SMCCC_FUNC___vgic_v3_init_lrs		6
+#define __KVM_HOST_SMCCC_FUNC___vgic_v3_get_gic_config		7
+#define __KVM_HOST_SMCCC_FUNC___pkvm_prot_finalize		8
+
+/* Hypercalls available after pKVM finalisation */
+#define __KVM_HOST_SMCCC_FUNC___pkvm_host_share_hyp		9
+#define __KVM_HOST_SMCCC_FUNC___kvm_adjust_pc			10
+#define __KVM_HOST_SMCCC_FUNC___kvm_vcpu_run			11
+#define __KVM_HOST_SMCCC_FUNC___kvm_flush_vm_context		12
+#define __KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid_ipa		13
+#define __KVM_HOST_SMCCC_FUNC___kvm_tlb_flush_vmid		14
+#define __KVM_HOST_SMCCC_FUNC___kvm_flush_cpu_context		15
+#define __KVM_HOST_SMCCC_FUNC___kvm_timer_set_cntvoff		16
+#define __KVM_HOST_SMCCC_FUNC___vgic_v3_read_vmcr		17
+#define __KVM_HOST_SMCCC_FUNC___vgic_v3_write_vmcr		18
+#define __KVM_HOST_SMCCC_FUNC___vgic_v3_save_aprs		19
+#define __KVM_HOST_SMCCC_FUNC___vgic_v3_restore_aprs		20

This is no helping with the above, but can we *please* try to get rid
of this #define insanity before moving things around? I came up with
the following, which seems to build.

Thoughts?

	M.

From 8a50c98489220d2ebaf02d4ffdbef3cf0d6634ee Mon Sep 17 00:00:00 2001
From: Marc Zyngier <maz@kernel.org>
Date: Thu, 7 Oct 2021 13:18:29 +0100
Subject: [PATCH] KVM: arm64: Turn __KVM_HOST_SMCCC_FUNC_* into an enum
 (mostly)

__KVM_HOST_SMCCC_FUNC_* is a royal pain, as there is a fair amount
of churn around these #defines, and we avoid making it an enum
only for the sake of the early init, low level code that requires
__KVM_HOST_SMCCC_FUNC___kvm_hyp_init to be usable from assembly.

Let's be brave and turn everything but this symbol into an enum,
using a bit of arithmetic to avoid any overlap.

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/include/asm/kvm_asm.h | 44 +++++++++++++++++---------------
 1 file changed, 24 insertions(+), 20 deletions(-)

Looks fine to me:

Acked-by: Will Deacon <will@kernel.org>

Do you need me to respin my series on top of this?

Will

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