Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <Yg54nse5qNQO3sbW@zn.tnic> <20220218013209.2436006-1-juew@google.com>
 <Yg+2Hc78nfSRmh/j@zn.tnic> <CAPcxDJ4c3eGXTB9UPJmZ8dzyCNPW4Lv9s1QSeoCWq_LdNWTrJw@mail.gmail.com>
 <Yg/FyrvLWhZHB/UC@zn.tnic>
In-Reply-To: <Yg/FyrvLWhZHB/UC@zn.tnic>
From:   Jue Wang <juew@google.com>
Date:   Fri, 18 Feb 2022 08:21:36 -0800
Message-ID: <CAPcxDJ72dMOpbKXxyb__OeMaEyjYSPtsL_ubVsKOuRXefAQ3_w@mail.gmail.com>
Subject: Re: [PATCH v2] x86/mce: work around an erratum on fast string copy instructions.
To:     Borislav Petkov <bp@alien8.de>
Cc:     Tony Luck <tony.luck@intel.com>, x86@kernel.org,
        linux-kernel@vger.kernel.org, patches@lists.linux.dev
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Feb 18, 2022 at 8:14 AM Borislav Petkov <bp@alien8.de> wrote:
>
> On Fri, Feb 18, 2022 at 08:03:24AM -0800, Jue Wang wrote:
> > Since MSR_IA32_MISC_ENABLE is not a MCA register, I wonder if we want
> > to mix its read/write with the injected MCE code.
> >
> > I was a bit concerned about potential race with mce-inject and the
> > read/write to MSR_IA32_MISC_ENABLE.
>
> It won't inject anything:
>
>                 offset = msr_to_offset(msr);
>                 if (offset < 0)
>                         ret = 0;
>
Thanks.

My concern was that here returns 0 instead the value read from the msr.

Maybe this cannot happen?
>
> Besides, you need to use those routines due to EX_TYPE_{RD,WR}MSR_IN_MCE
>
> --
> Regards/Gruss,
>     Boris.
>
> https://people.kernel.org/tglx/notes-about-netiquette