Kernel KVM virtualization development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Marc Zyngier" <maz@kernel.org>
Cc: kvmarm@lists.linux.dev, "Marc Zyngier" <maz@kernel.org>,
	kvm@vger.kernel.org, "Oliver Upton" <oupton@kernel.org>
Subject: Re: [PATCH v3 12/28] arm64: Add FEAT_NV2p1 detection
Date: Wed, 22 Jul 2026 08:13:05 +0000	[thread overview]
Message-ID: <20260722081305.F28D91F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260722074234.3330999-13-maz@kernel.org>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Missing isb() immediately after writing to control-plane system registers.
--

--- Patch [12]: [PATCH v3 12/28] arm64: Add FEAT_NV2p1 detection ---
commit 5488b94ad37f884fedfaca076980f395c30a72bf
Author: Marc Zyngier <maz@kernel.org>

arm64: Add FEAT_NV2p1 detection
    
This commit adds the necessary NV2p1 probing to the cpufeature
infrastructure.

> diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
> index 9a22df0c5120f..c9c124b0ccc8e 100644
> --- a/arch/arm64/kernel/cpufeature.c
> +++ b/arch/arm64/kernel/cpufeature.c

[ ... ]

> @@ -2620,6 +2620,13 @@ static const struct arm64_cpu_capabilities arm64_features[] = {
>  			{ /* Sentinel */ }
>  		},
>  	},
> +	{
> +		.desc = "FEAT_NV2p1",
> +		.capability = ARM64_HAS_NV2P1,
> +		.type = ARM64_CPUCAP_SYSTEM_FEATURE,
> +		.matches = has_cpuid_feature,
> +		ARM64_CPUID_FIELDS(ID_AA64MMFR4_EL1, NV_frac, NV2P1)
> +	},

[Severity: High]
Are we missing an isb() after writing to control-plane system registers when
this capability is enabled?

Looking at arch/arm64/kvm/arch_timer.c in timer_set_traps(), there is a
missing isb() after the system register writes:

arch/arm64/kvm/arch_timer.c:timer_set_traps() {
    ...
	if (!cpus_have_final_cap(ARM64_HAS_NV2P1))
		sysreg_clear_set(cnthctl_el2, clr, set);
	else
		sysreg_clear_set(cntkctl_el1, clr, set);
}

The sysreg_clear_set macro evaluates to a conditional write but lacks an
immediate isb(), and the function returns shortly after.

Similarly, in arch/arm64/kvm/sys_regs.c, are we missing synchronization before
the break?

arch/arm64/kvm/sys_regs.c:vcpu_write_sys_reg() {
    ...
		case CNTHCTL_EL2:
			write_sysreg_el1(val, SYS_CNTKCTL);
			break;
		case CPTR_EL2:
			write_sysreg_el1(val, SYS_CPACR);
			break;
    ...
}

The writes to SYS_CPACR and SYS_CNTKCTL are followed by a break and a
subsequent function return, rather than an immediate isb().

Without immediate synchronization, the CPU might operate under a stale
configuration for several cycles, which could cause unexpected traps or break
memory safety semantics.

>  	{
>  		.capability = ARM64_HAS_32BIT_EL0_DO_NOT_USE,
>  		.type = ARM64_CPUCAP_SYSTEM_FEATURE,

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260722074234.3330999-1-maz@kernel.org?part=12

  reply	other threads:[~2026-07-22  8:13 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-22  7:42 [PATCH v3 00/28] KVM: arm64: Add support for FEAT_NV2p1 and FEAT_NV3 Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 01/28] arm64: sysreg: Emit RESx/UNKN values for Mapping/Fields definitions Marc Zyngier
2026-07-22  7:55   ` sashiko-bot
2026-07-22  9:05     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 02/28] arm64: Update ID_AA64MMFR4_EL1 description to 2026-03 JSON release Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 03/28] KVM: arm64: Merge guest's HCRX_EL2 using NV_HCRX_GUEST_EXCLUDE Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 04/28] KVM: arm64: Drop __HCRX_EL2_* masks Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 05/28] KVM: arm64: Plumb HCRX_EL2.SRMASKEn in HCRX_EL2 sanitisation Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 06/28] KVM: arm64: Classify CPTR_EL2 as a SR_LOC_SPECIAL register Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 07/28] KVM: arm64: Don't evaluate HCR_EL2.NV nor HFGITR_EL2.ERET on ERET fast path Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 08/28] arm64: Add ARM64_HAS_NV2P1 capability Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 09/28] KVM: arm64: Relax CPTR_EL2 handling when FEAT_NV2p1 is present Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 10/28] KVM: arm64: Relax CNTHCTL_EL2 " Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 11/28] KVM: arm64: Expose FEAT_NV2p1 to NV guests Marc Zyngier
2026-07-22  8:32   ` sashiko-bot
2026-07-22  9:01     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 12/28] arm64: Add FEAT_NV2p1 detection Marc Zyngier
2026-07-22  8:13   ` sashiko-bot [this message]
2026-07-22  8:57     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 13/28] arm64: sysreg: Add NVHCR_EL2 description as a mirror of HCR_EL2 Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 14/28] arm64: sysreg: Add HCRX_EL2 bits related to FEAT_NV3 Marc Zyngier
2026-07-22  8:11   ` sashiko-bot
2026-07-22  8:56     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 15/28] arm64: Add ARM64_HAS_NV3 capability Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 16/28] KVM: arm64: Split NV-specific exit fixups from the non-NV handling Marc Zyngier
2026-07-22  8:23   ` sashiko-bot
2026-07-22  9:08     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 17/28] KVM: arm64: Add NV3 control bits to HCRX_EL2 sanitisation Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 18/28] KVM: arm64: Add kvm_has_nv{2,3}() predicates Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 19/28] KVM: arm64: Make HCR_EL2 a non-VNCR register Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 20/28] KVM: arm64: Add sanitisation for NVHCR_EL2 Marc Zyngier
2026-07-22  8:38   ` sashiko-bot
2026-07-22  9:11     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 21/28] KVM: arm64: Add NVHCR_EL2 handling to the sysreg array Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 22/28] KVM: arm64: Add routing for NVHCR_EL2 trap Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 23/28] KVM: arm64: Add NVHCR_EL2 context switching Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 24/28] KVM: arm64: Engage NV3 ERET trap elision Marc Zyngier
2026-07-22  8:50   ` sashiko-bot
2026-07-22  9:16     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 25/28] KVM: arm64: Engage NV3 TLBI " Marc Zyngier
2026-07-22  8:57   ` sashiko-bot
2026-07-22  9:04     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 26/28] KVM: arm64: Add FEAT_NV3 detection Marc Zyngier
2026-07-22  8:53   ` sashiko-bot
2026-07-22  9:04     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 27/28] KVM: arm64: Expose FEAT_NV3 to guests Marc Zyngier
2026-07-22  9:06   ` sashiko-bot
2026-07-22  9:18     ` Marc Zyngier
2026-07-22  7:42 ` [PATCH v3 28/28] arm64: Add override for ID_AA64MMFR4_EL1.NV_frac Marc Zyngier

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=20260722081305.F28D91F000E9@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=maz@kernel.org \
    --cc=oupton@kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    /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