Re: [Patch v2 2/4] x86/speculation: Provide application property based STIBP protection

From: Thomas Gleixner <tglx@linutronix.de>
To: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Jiri Kosina <jikos@kernel.org>,
	Tom Lendacky <thomas.lendacky@amd.com>,
	Ingo Molnar <mingo@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Josh Poimboeuf <jpoimboe@redhat.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	David Woodhouse <dwmw@amazon.co.uk>,
	Andi Kleen <ak@linux.intel.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Casey Schaufler <casey.schaufler@intel.com>,
	Asit Mallick <asit.k.mallick@intel.com>,
	Arjan van de Ven <arjan@linux.intel.com>,
	Jon Masters <jcm@redhat.com>,
	linux-kernel@vger.kernel.org, x86@kernel.org
Subject: Re: [Patch v2 2/4] x86/speculation: Provide application property based STIBP protection
Date: Tue, 2 Oct 2018 21:10:36 +0200 (CEST)	[thread overview]
Message-ID: <alpine.DEB.2.21.1810022104480.1435@nanos.tec.linutronix.de> (raw)
In-Reply-To: <ce0454aa84320b34de5b00058a56daefdd817e22.1537920575.git.tim.c.chen@linux.intel.com>

On Tue, 25 Sep 2018, Tim Chen wrote:

> This patch provides an application property based spectre_v2

# git grep 'This patch' Documentation/process/

> protection with STIBP against attack from another app from

s/app/application/ please. This is not android.

> a sibling hyper-thread.  For security sensitive non-dumpable
> app, STIBP will be turned on before switching to it for Intel
> processors vulnerable to spectre_v2.

What has this to do with Intel processors?

> -static __always_inline void intel_set_ssb_state(unsigned long tifn)
> +static __always_inline void set_spec_ctrl_state(unsigned long tifn)
>  {
> -	u64 msr = x86_spec_ctrl_base | ssbd_tif_to_spec_ctrl(tifn);
> +	u64 msr = x86_spec_ctrl_base;
> +
> +	if (static_cpu_has(X86_FEATURE_SSBD))
> +		msr |= ssbd_tif_to_spec_ctrl(tifn);
> +
> +	if (cpu_smt_control == CPU_SMT_ENABLED)
> +		msr |= stibp_tif_to_spec_ctrl(tifn);

Oh no. We are not adding yet another conditional into switch to. Either
that's done unconditionally or this wants to have a static key.

>  	wrmsrl(MSR_IA32_SPEC_CTRL, msr);

Thanks,

	tglx