[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230826154922.mdmjzwqwdgufb3y4@treble>
Date: Sat, 26 Aug 2023 08:49:22 -0700
From: Josh Poimboeuf <jpoimboe@...nel.org>
To: Sean Christopherson <seanjc@...gle.com>
Cc: x86@...nel.org, linux-kernel@...r.kernel.org,
Borislav Petkov <bp@...en8.de>,
Peter Zijlstra <peterz@...radead.org>,
Babu Moger <babu.moger@....com>,
Paolo Bonzini <pbonzini@...hat.com>, David.Kaplan@....com,
Andrew Cooper <andrew.cooper3@...rix.com>,
Nikolay Borisov <nik.borisov@...e.com>,
gregkh@...uxfoundation.org, Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH 04/23] KVM: x86: Add IBPB_BRTYPE support
On Fri, Aug 25, 2023 at 11:15:49AM -0700, Sean Christopherson wrote:
> On Fri, Aug 25, 2023, Josh Poimboeuf wrote:
> > Add support for the IBPB_BRTYPE CPUID flag, which indicates that IBPB
> > includes branch type prediction flushing.
>
> Please add:
>
> Note, like SRSO_NO, advertise support for IBPB_BRTYPE even if it's not
> enumerated by in the raw CPUID, i.e. bypass the cpuid_count() in
> __kvm_cpu_cap_mask(). Some CPUs that gained support via a uCode patch
> don't report IBPB_BRTYPE via CPUID (the kernel forces the flag).
>
> Opportunistically use kvm_cpu_cap_check_and_set() for SRSS_NO instead
"SRSO_NO"
> of manually querying host support (cpu_feature_enabled() and
> boot_cpu_has() yield the same end result in this case).
Sounds good.
> Feel free to take this through tip if this is urgent enough to go into 6.6,
> otherwise I'll grab it for 6.7.
Ingo grabbed all the patches except for the two KVM ones, so I think
he's expecting you to take them.
> Acked-by: Sean Christopherson <seanjc@...gle.com>
Thanks!
--
Josh
Powered by blists - more mailing lists