linux-kernel - Re: [PATCH] x86/srso: Correct the mitigation status when SMT is disabled

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230815212751.xhsyn7iwj2gwpuk5@treble>
Date:   Tue, 15 Aug 2023 14:27:51 -0700
From:   Josh Poimboeuf <jpoimboe@...nel.org>
To:     Borislav Petkov <bp@...en8.de>
Cc:     Nikolay Borisov <nik.borisov@...e.com>, X86 ML <x86@...nel.org>,
        Josh Poimboeuf <jpoimboe@...hat.com>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] x86/srso: Correct the mitigation status when SMT is
 disabled

On Tue, Aug 15, 2023 at 10:17:53PM +0200, Borislav Petkov wrote:
> On Tue, Aug 15, 2023 at 12:58:31PM -0700, Josh Poimboeuf wrote:
> > AFAICT, nowhere in the spec does it say the SRSO_NO bit won't get set by
> > future (fixed) HW.  In fact I'd expect it will, similar to other *_NO
> > flags.
> 
> I'm pretty sure it won't.
> 
> SRSO_NO is synthesized by the hypervisor *software*. Nothing else.

Citation needed.

> It is there so that you don't check microcode version in the guest which
> is nearly impossible anyway.
> 
> > Regardless, here SRSO_NO seems to mean two different things: "reported
> > safe by host (or HW)" and "not reported safe on Zen1/2 with SMT not
> > possible".
> 
> Huh?

Can you clarify what doesn't make sense?

> > Also, in this code, the SRSO_NO+SMT combo doesn't seem logically
> > possible, as srso_show_state() only gets called if X86_BUG_SRSO is set,
> > which only happens if SRSO_NO is not set by the HW/host in the first
> > place.  So here, if boot_cpu_has(X86_FEATURE_SRSO_NO), it means SRSO_NO
> > was manually set by srso_select_mitigation(), and SMT can't possibly be
> > enabled.
> 
> Have you considered the case where Linux would set SRSO_NO when booting
> on future hardware, which is fixed?
> 
> There SRSO_NO and SMT will very much be possible.

How is that relevant to my comment?  The bug bit still wouldn't get set
and srso_show_state() still wouldn't be called.

-- 
Josh