Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 21 Oct 2021 10:53:58 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Josh Poimboeuf <jpoimboe@redhat.com>
Cc:     Alexei Starovoitov <alexei.starovoitov@gmail.com>, x86@kernel.org,
        andrew.cooper3@citrix.com, linux-kernel@vger.kernel.org,
        ndesaulniers@google.com, bpf@vger.kernel.org, daniel@iogearbox.net
Subject: Re: [PATCH v2 14/14] bpf,x86: Respect X86_FEATURE_RETPOLINE*
Message-ID: <YXEqpuKa5UiRNEfc@hirez.programming.kicks-ass.net>
References: <20211020104442.021802560@infradead.org>
 <20211020105843.345016338@infradead.org>
 <20211021000753.kdxtjl3nzre2zshb@ast-mbp.dhcp.thefacebook.com>
 <20211021001808.3ng6qeggi5lfwx7k@treble>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20211021001808.3ng6qeggi5lfwx7k@treble>
Precedence: bulk

On Wed, Oct 20, 2021 at 05:18:08PM -0700, Josh Poimboeuf wrote:
> On Wed, Oct 20, 2021 at 05:07:53PM -0700, Alexei Starovoitov wrote:
> > On Wed, Oct 20, 2021 at 12:44:56PM +0200, Peter Zijlstra wrote:
> > > +
> > > +	if (cpu_feature_enabled(X86_FEATURE_RETPOLINE_AMD)) {
> > > +		EMIT_LFENCE();
> > > +		EMIT2(0xFF, 0xE0 + reg);
> > > +	} else if (cpu_feature_enabled(X86_FEATURE_RETPOLINE)) {
> > > +		emit_jump(&prog, reg_thunk[reg], ip);
> > > +	} else
> > 
> > One more question.
> > What's a deal with AMD? I thought the retpoline is effective on it as well.
> > lfence is an optimization or retpoline turned out to be not enough
> > in some cases?
> 
> Yes, it's basically an optimization.  AMD recommends it presumably
> because it's quite a bit faster than a retpoline.
> 
> According to AMD it shrinks the speculative execution window enough so
> that Spectre v2 isn't a threat.

Right, also note that we've been using alternatives to patch the thunk
to lfence;jmp for AMD pretty much forever.

Inlining it is better tho; just a shame clang seems to insist on r11,
which means we cannot fit it in the thunk call site for them :/