linux-kernel - Re: [PATCH] x86/tsx: Use MSR_TSX

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: <20220215181931.wxfsn2a3npg7xmi2@guptapa-mobl1.amr.corp.intel.com>
Date:   Tue, 15 Feb 2022 10:19:31 -0800
From:   Pawan Gupta <pawan.kumar.gupta@...ux.intel.com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     Andrew Cooper <andrew.cooper3@...rix.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>, x86@...nel.org,
        "H. Peter Anvin" <hpa@...or.com>, Andi Kleen <ak@...ux.intel.com>,
        Tony Luck <tony.luck@...el.com>, linux-kernel@...r.kernel.org,
        antonio.gomez.iglesias@...ux.intel.com, neelima.krishnan@...el.com,
        stable@...r.kernel.org
Subject: Re: [PATCH] x86/tsx: Use MSR_TSX_CTRL to clear CPUID bits

On 15.02.2022 17:31, Borislav Petkov wrote:
>On Tue, Feb 15, 2022 at 04:11:03AM -0800, Pawan Gupta wrote:
>> That is exactly what this patch is fixing. Please let me know if you
>> have any questions.
>
>Just one: does the explanation I've written for this mess, sound about
>right?

I admit it has gotten complicated with so many bits associated with TSX.
Your explanation is accurate. I just have a small suggestion below.

>+/*
>+ * Disabling TSX is not a trivial business.
>+ *
>+ * First of all, there's a CPUID bit: X86_FEATURE_RTM_ALWAYS_ABORT
>+ * which says that TSX is practically disabled (all transactions are
>+ * aborted by default). When that bit is set, the kernel unconditionally
>+ * disables TSX.
>+ *
>+ * In order to do that, however, it needs to dance a bit:
>+ *
>+ * 1. The first method to disable it is through MSR_TSX_FORCE_ABORT and
>+ * the MSR is present only when *two* CPUID bits are set:

s/MSR/MSR bit MSR_TFA_TSX_CPUID_CLEAR/

Thanks,
Pawan