linux-kernel - Re: [PATCH v3] KVM: x86: Add x2APIC "features" to control EOI broadcast suppression

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <994FA1C8-7AC2-4FD2-958D-373C0AA1553A@nutanix.com>
Date: Wed, 3 Dec 2025 07:45:34 +0000
From: Khushit Shah <khushit.shah@...anix.com>
To: Sean Christopherson <seanjc@...gle.com>
CC: David Woodhouse <dwmw2@...radead.org>,
        "pbonzini@...hat.com"
	<pbonzini@...hat.com>,
        "kai.huang@...el.com" <kai.huang@...el.com>,
        "mingo@...hat.com" <mingo@...hat.com>,
        "x86@...nel.org" <x86@...nel.org>, "bp@...en8.de" <bp@...en8.de>,
        "hpa@...or.com" <hpa@...or.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "kvm@...r.kernel.org" <kvm@...r.kernel.org>,
        "dave.hansen@...ux.intel.com"
	<dave.hansen@...ux.intel.com>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        Jon
 Kohler <jon@...anix.com>,
        Shaju Abraham <shaju.abraham@...anix.com>,
        "stable@...r.kernel.org" <stable@...r.kernel.org>
Subject: Re: [PATCH v3] KVM: x86: Add x2APIC "features" to control EOI
 broadcast suppression



> On 2 Dec 2025, at 10:06 PM, Sean Christopherson <seanjc@...gle.com> wrote:
> 
> + if (cap->args[0] & KVM_X2APIC_ENABLE_SUPPRESS_EOI_BROADCAST)
> + kvm->arch.suppress_eoi_broadcast = KVM_SUPPRESS_EOI_ENABLED;
> + if (cap->args[0] & KVM_X2APIC_DISABLE_SUPPRESS_EOI_BROADCAST)
> + kvm->arch.suppress_eoi_broadcast = KVM_SUPPRESS_EOI_DISABLED;

Should this also update existing vCPUs by calling kvm_apic_set_version() 
for all already-created vCPUs? or is the intent that userspace must set these flags
before any vCPUs are created? If so, should this warn if called after vcpus are 
created?

QEMU currently creates vCPUs before initializing the I/O APIC, so deciding KVM’s
SEOIB behavior at vCPU initialization time is just a bit awkward. It’s manageable
either way, I just want to confirm the uAPI semantics.

For reference, the initial QEMU side patch:
https://patchew.org/QEMU/20251126093742.2110483-1-khushit.shah@nutanix.com/