Date: Wed, 10 Jan 2018 00:43:11 +0100 (CET)
From: Thomas Gleixner <tglx@linutronix.de>
To: Tom Lendacky <thomas.lendacky@amd.com>
cc: x86@kernel.org, linux-kernel@vger.kernel.org,
        Rik van Riel <riel@redhat.com>,
        Andi Kleen <ak@linux.intel.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Jiri Kosina <jikos@kernel.org>,
        Dan Williams <dan.j.williams@intel.com>,
        Dave Hansen <dave.hansen@intel.com>,
        Borislav Petkov <bp@alien8.de>,
        Andy Lutomirski <luto@kernel.org>,
        Kees Cook <keescook@google.com>,
        Tim Chen <tim.c.chen@linux.intel.com>,
        Greg Kroah-Hartman <gregkh@linux-foundation.org>,
        David Woodhouse <dwmw@amazon.co.uk>,
        Paul Turner <pjt@google.com>
Subject: Re: [PATCH v1 1/2] x86/retpoline: Add a function to clear the
 RETPOLINE_AMD feature
In-Reply-To: <a366ccaf-736d-225a-5eb9-bcbcd6251d47@amd.com>
Message-ID: <alpine.DEB.2.20.1801100040240.2200@nanos>
References: <20180109223723.29322.59126.stgit@tlendack-t1.amdoffice.net> <20180109223736.29322.5436.stgit@tlendack-t1.amdoffice.net> <alpine.DEB.2.20.1801092345560.2200@nanos> <12147db9-e975-268f-55ee-247868e0e0d8@amd.com> <alpine.DEB.2.20.1801100004040.2200@nanos>
 <a366ccaf-736d-225a-5eb9-bcbcd6251d47@amd.com>
User-Agent: Alpine 2.20 (DEB 67 2015-01-07)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org

On Tue, 9 Jan 2018, Tom Lendacky wrote:
> On 1/9/2018 5:09 PM, Thomas Gleixner wrote:
> > On Tue, 9 Jan 2018, Tom Lendacky wrote:
> >> On 1/9/2018 4:46 PM, Thomas Gleixner wrote:
> >>> Urgh. That's an awful hack. why not do the obvious?
> >>
> >> My first attempt was very similar to your change below, but testing
> >> showed that spectre_v2_check_boottime_disable() is called before the
> >> X86_FEATURE_LFENCE_RDTSC can be set.  I can look at moving where the
> >> X86_FEATURE_LFENCE_RDTSC is set, maybe into early_init_amd() or such
> >> if you think that would be best.
> > 
> > Wait, we can move the selection _AFTER_ identify_boot_cpu().
> 
> Much cleaner.  Since it's just a single patch now, do you want me to
> re-submit this after I test it or will you just pick this up as is?
> 
> I did notice that the patch does change the behavior associated with
> the command line options, though.  Not sure if that was intentional.

Yes. I noticed while looking at your issue that when AMD is selected on the
command line then we have no support at all on intel. So I prefer to err
out and enable the generic version.

If you can just polish it up with a changelog and resubmit after testing
that would be appreciated as I'm steam blasting the IBRS stuff at the
moment.

Thanks,

	tglx