[MODERATED] Re: [PATCH v7 04/10] TAAv7 4

From: Borislav Petkov <bp@suse.de>
To: speck@linutronix.de
Subject: [MODERATED] Re: [PATCH v7 04/10] TAAv7 4
Date: Wed, 23 Oct 2019 16:39:16 +0200	[thread overview]
Message-ID: <20191023143916.GL12272@zn.tnic> (raw)
In-Reply-To: <20191023132859.GA32452@guptapadev.amr>

On Wed, Oct 23, 2019 at 06:28:59AM -0700, speck for Pawan Gupta wrote:
> cpu_mitigations_off() check after X86_FEATURE_RTM check will fix this.

Ok, the second cmdline switch tsx= makes this a bit trickier than I was
prepared for. I've reverted to your previous order:

static void __init taa_select_mitigation(void)
{
        u64 ia32_cap;

        if (!boot_cpu_has_bug(X86_BUG_TAA)) {
                taa_mitigation = TAA_MITIGATION_OFF;
                return;
        }

        /* TSX previously disabled by tsx=off */
        if (!boot_cpu_has(X86_FEATURE_RTM)) {
                taa_mitigation = TAA_MITIGATION_TSX_DISABLED;
                goto out;
        }

        if (cpu_mitigations_off()) {
                taa_mitigation = TAA_MITIGATION_OFF;
                return;
        }

I've applied the first 9 patches and have incorporated as much of the
feedback as I could. Will try to get a box for which the microcode fits
and test it before I post today or tomorrow and we can continue there.

Thx.

-- 
Regards/Gruss,
    Boris.

SUSE Software Solutions Germany GmbH, GF: Felix Imendörffer, HRB 36809, AG Nürnberg
-- 