Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 21 Nov 2018 19:21:31 +0100 (CET)
From:   Jiri Kosina <jikos@kernel.org>
To:     Tim Chen <tim.c.chen@linux.intel.com>
cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        linux-kernel@vger.kernel.org, stable@vger.kernel.org,
        Thomas Gleixner <tglx@linutronix.de>,
        Peter Zijlstra <peterz@infradead.org>,
        Josh Poimboeuf <jpoimboe@redhat.com>,
        Andrea Arcangeli <aarcange@redhat.com>,
        WoodhouseDavid <dwmw@amazon.co.uk>,
        Andi Kleen <ak@linux.intel.com>,
        SchauflerCasey <casey.schaufler@intel.com>,
        Dave Stewart <david.c.stewart@intel.com>
Subject: Re: [PATCH 4.19 041/361] x86/speculation: Enable cross-hyperthread
 spectre v2 STIBP mitigation
In-Reply-To: <41234d20-294f-d0f4-336c-64bf9fcb92a7@linux.intel.com>
Message-ID: <nycvar.YFH.7.76.1811211920570.21108@cbobk.fhfr.pm>
References: <20181111221619.915519183@linuxfoundation.org> <20181111221625.598298393@linuxfoundation.org> <alpine.LRH.2.00.1811211420130.18805@gjva.wvxbf.pm> <20181121135611.GA26411@kroah.com> <41234d20-294f-d0f4-336c-64bf9fcb92a7@linux.intel.com>
User-Agent: Alpine 2.21 (LSU 202 2017-01-01)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, 21 Nov 2018, Tim Chen wrote:

> > Is it reverted in Linus's tree?  If not, then anything that comes "later
> > on" will not apply here, right?
> > 
> > I see the thread asking about this, but I got really conflicting
> > messages here, and now it's in all of the latest releases, and no
> > testing seems to have uncovered issues.  Is it just a "slow down"
> > problem?
> 
> Greg,
> 
> It could be a big slow down in excess of 20% for some applications.
> And cross sibling Spectre v2 attack is quite hard to pull off.
> 
> So till we have the accompanying patchset that only apply STIBP on processes
> that really need it instead of universally, it should be withheld from
> stable.

Agreed; it will be trivially reintroduced with the rest, once it's ready. 
It's being built on top of that patch.

Thanks,

-- 
Jiri Kosina
SUSE Labs