linux-kernel - Re: [PATCH MANUALSEL 5.10 1/2] KVM: x86: lapic: don't touch irr_pending in kvm_apic_update

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 for Android: free password hash cracker in your pocket

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

Date:   Mon, 25 Jul 2022 23:49:21 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     Zenghui Yu <yuzenghui@...wei.com>
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        Sasha Levin <sashal@...nel.org>, linux-kernel@...r.kernel.org,
        stable@...r.kernel.org, Maxim Levitsky <mlevitsk@...hat.com>,
        tglx@...utronix.de, mingo@...hat.com, bp@...en8.de,
        dave.hansen@...ux.intel.com, x86@...nel.org, kvm@...r.kernel.org
Subject: Re: [PATCH MANUALSEL 5.10 1/2] KVM: x86: lapic: don't touch
 irr_pending in kvm_apic_update_apicv when inhibiting it

On Mon, Jul 25, 2022, Zenghui Yu wrote:
> Hi,
> 
> On 2022/3/2 1:10, Paolo Bonzini wrote:
> > On 2/22/22 15:05, Sasha Levin wrote:
> > > From: Maxim Levitsky <mlevitsk@...hat.com>
> > > 
> > > [ Upstream commit 755c2bf878607dbddb1423df9abf16b82205896f ]
> > 
> > Acked-by: Paolo Bonzini <pbonzini@...hat.com>
> 
> What prevented it to be accepted into 5.10-stable? It can still be
> applied cleanly on top of linux-5.10.y.

KVM opts out of the AUTOSEL logic and instead uses MANUALSEL.  The basic idea is
the same, use scripts/magic to determine what commits that _aren't_ tagged with an
explicit "Cc: stable@...r.kernel.org" should be backported to stable trees, the
difference being that MANUALSEL requires an explicit Acked-by from the maintainer.

Many (most?) of the automatically selected patches for KVM are good stable candidates,
but there are enough selected patches that we _don't_ want backported that the "apply
unless there's an objection" model of AUTOSEL is a bit too risky for us.