From: Oliver Upton <oliver.upton@linux.dev>
To: Marc Zyngier <maz@kernel.org>
Cc: kvmarm@lists.linux.dev, linux-arm-kernel@lists.infradead.org,
Joey Gouly <joey.gouly@arm.com>,
Suzuki K Poulose <suzuki.poulose@arm.com>,
Zenghui Yu <yuzenghui@huawei.com>,
Sascha Bischoff <sascha.bischoff@arm.com>
Subject: Re: [PATCH] KVM: arm64: Don't access ICC_SRE_EL2 if GICv3 doesn't support v2 compatibility
Date: Wed, 17 Sep 2025 09:28:29 -0700 [thread overview]
Message-ID: <aMrhrU8Khn0rfbMh@linux.dev> (raw)
In-Reply-To: <20250917161935.1630908-1-maz@kernel.org>
On Wed, Sep 17, 2025 at 05:19:35PM +0100, Marc Zyngier wrote:
> We currently access ICC_SRE_EL2 at each load/put on VHE, and on each
> entry/exit on nVHE. Both are quite onerous on NV, as this register
> always traps.
>
> We do this to make sure the EL1 guest doesn't flip between v2 and v3
> behind our back. But all modern implementations have dropped v2,
> and this is just overhead.
>
> At the same time, the GICv5 spec has been fixed to allow access to
> ICC_SRE_EL2 in legacy mode. Use this opportunity to replace the
> GICv5 checks for v2 compat checks, with an ad-hoc static key.
>
> Co-developed-by: Sascha Bischoff <sascha.bischoff@arm.com>
> Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
> Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Thanks,
Oliver
> ---
> arch/arm64/kernel/image-vars.h | 3 +++
> arch/arm64/kvm/hyp/vgic-v3-sr.c | 25 +++++++++----------------
> arch/arm64/kvm/vgic/vgic-v3.c | 8 ++++++++
> include/kvm/arm_vgic.h | 1 +
> 4 files changed, 21 insertions(+), 16 deletions(-)
>
> diff --git a/arch/arm64/kernel/image-vars.h b/arch/arm64/kernel/image-vars.h
> index 714b0b5ec5ac4..5369763606e71 100644
> --- a/arch/arm64/kernel/image-vars.h
> +++ b/arch/arm64/kernel/image-vars.h
> @@ -105,6 +105,9 @@ KVM_NVHE_ALIAS(__hyp_stub_vectors);
> KVM_NVHE_ALIAS(vgic_v2_cpuif_trap);
> KVM_NVHE_ALIAS(vgic_v3_cpuif_trap);
>
> +/* Static key indicating whether GICv3 has GICv2 compatibility */
> +KVM_NVHE_ALIAS(vgic_v3_has_v2_compat);
> +
> /* Static key which is set if CNTVOFF_EL2 is unusable */
> KVM_NVHE_ALIAS(broken_cntvoff_key);
>
> diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c b/arch/arm64/kvm/hyp/vgic-v3-sr.c
> index d81275790e69b..acd909b7f2257 100644
> --- a/arch/arm64/kvm/hyp/vgic-v3-sr.c
> +++ b/arch/arm64/kvm/hyp/vgic-v3-sr.c
> @@ -295,12 +295,8 @@ void __vgic_v3_activate_traps(struct vgic_v3_cpu_if *cpu_if)
> }
> }
>
> - /*
> - * GICv5 BET0 FEAT_GCIE_LEGACY doesn't include ICC_SRE_EL2. This is due
> - * to be relaxed in a future spec release, at which point this in
> - * condition can be dropped.
> - */
> - if (!cpus_have_final_cap(ARM64_HAS_GICV5_CPUIF)) {
> + /* Only disable SRE if the host implements the GICv2 interface */
> + if (static_branch_unlikely(&vgic_v3_has_v2_compat)) {
> /*
> * Prevent the guest from touching the ICC_SRE_EL1 system
> * register. Note that this may not have any effect, as
> @@ -329,19 +325,16 @@ void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if)
> cpu_if->vgic_vmcr = read_gicreg(ICH_VMCR_EL2);
> }
>
> - /*
> - * Can be dropped in the future when GICv5 spec is relaxed. See comment
> - * above.
> - */
> - if (!cpus_have_final_cap(ARM64_HAS_GICV5_CPUIF)) {
> + /* Only restore SRE if the host implements the GICv2 interface */
> + if (static_branch_unlikely(&vgic_v3_has_v2_compat)) {
> val = read_gicreg(ICC_SRE_EL2);
> write_gicreg(val | ICC_SRE_EL2_ENABLE, ICC_SRE_EL2);
> - }
>
> - if (!cpu_if->vgic_sre) {
> - /* Make sure ENABLE is set at EL2 before setting SRE at EL1 */
> - isb();
> - write_gicreg(1, ICC_SRE_EL1);
> + if (!cpu_if->vgic_sre) {
> + /* Make sure ENABLE is set at EL2 before setting SRE at EL1 */
> + isb();
> + write_gicreg(1, ICC_SRE_EL1);
> + }
> }
>
> /*
> diff --git a/arch/arm64/kvm/vgic/vgic-v3.c b/arch/arm64/kvm/vgic/vgic-v3.c
> index b9ad7c42c5b01..f1c153106c563 100644
> --- a/arch/arm64/kvm/vgic/vgic-v3.c
> +++ b/arch/arm64/kvm/vgic/vgic-v3.c
> @@ -588,6 +588,7 @@ int vgic_v3_map_resources(struct kvm *kvm)
> }
>
> DEFINE_STATIC_KEY_FALSE(vgic_v3_cpuif_trap);
> +DEFINE_STATIC_KEY_FALSE(vgic_v3_has_v2_compat);
>
> static int __init early_group0_trap_cfg(char *buf)
> {
> @@ -697,6 +698,13 @@ int vgic_v3_probe(const struct gic_kvm_info *info)
> if (kvm_vgic_global_state.vcpu_base == 0)
> kvm_info("disabling GICv2 emulation\n");
>
> + /*
> + * Flip the static branch if the HW supports v2, even if we're
> + * not using it (such as in protected mode).
> + */
> + if (has_v2)
> + static_branch_enable(&vgic_v3_has_v2_compat);
> +
> if (cpus_have_final_cap(ARM64_WORKAROUND_CAVIUM_30115)) {
> group0_trap = true;
> group1_trap = true;
> diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
> index 404883c7af6e8..9a6340d9c91e5 100644
> --- a/include/kvm/arm_vgic.h
> +++ b/include/kvm/arm_vgic.h
> @@ -375,6 +375,7 @@ struct vgic_cpu {
>
> extern struct static_key_false vgic_v2_cpuif_trap;
> extern struct static_key_false vgic_v3_cpuif_trap;
> +extern struct static_key_false vgic_v3_has_v2_compat;
>
> int kvm_set_legacy_vgic_v2_addr(struct kvm *kvm, struct kvm_arm_device_addr *dev_addr);
> void kvm_vgic_early_init(struct kvm *kvm);
> --
> 2.39.2
>
prev parent reply other threads:[~2025-09-17 16:28 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-17 16:19 [PATCH] KVM: arm64: Don't access ICC_SRE_EL2 if GICv3 doesn't support v2 compatibility Marc Zyngier
2025-09-17 16:28 ` Oliver Upton [this message]
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=aMrhrU8Khn0rfbMh@linux.dev \
--to=oliver.upton@linux.dev \
--cc=joey.gouly@arm.com \
--cc=kvmarm@lists.linux.dev \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=maz@kernel.org \
--cc=sascha.bischoff@arm.com \
--cc=suzuki.poulose@arm.com \
--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