Re: [PATCH 2/3] x86/microcode/intel: Perform a cache flush before ucode update.

[Borislav Petkov <bp@suse.de>]

On Wed, Feb 21, 2018 at 08:49:43AM -0800, Ashok Raj wrote:
> Microcode updates can be safer if the caches are clean.
> Some of the issues around in certain Broadwell parts
> can be addressed by doing a full cache flush.

Take that text...

> 
> Signed-off-by: Ashok Raj <ashok.raj@intel.com>
> Cc: X86 ML <x86@kernel.org>
> Cc: LKML <linux-kernel@vger.kernel.org>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@kernel.org>
> Cc: Tony Luck <tony.luck@intel.com>
> Cc: Andi Kleen <andi.kleen@intel.com>
> Cc: Boris Petkov <bp@suse.de>
> Cc: Tom Lendacky <thomas.lendacky@amd.com>
> Cc: Arjan Van De Ven <arjan.van.de.ven@intel.com>
> ---
>  arch/x86/kernel/cpu/microcode/intel.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/arch/x86/kernel/cpu/microcode/intel.c b/arch/x86/kernel/cpu/microcode/intel.c
> index eff80df..5d32724 100644
> --- a/arch/x86/kernel/cpu/microcode/intel.c
> +++ b/arch/x86/kernel/cpu/microcode/intel.c
> @@ -589,6 +589,7 @@ static int apply_microcode_early(struct ucode_cpu_info *uci, bool early)
>  	if (!mc)
>  		return 0;
>  

... and put it here as a comment.

> +	wbinvd();

This definitely needs to be native_wbinvd(). Early mode don't love pvops.

>  	/* write microcode via MSR 0x79 */
>  	native_wrmsrl(MSR_IA32_UCODE_WRITE, (unsigned long)mc->bits);
>  
> @@ -805,6 +806,7 @@ static enum ucode_state apply_microcode_intel(int cpu)
>  		return UCODE_OK;
>  	}


Ditto.

> +	wbinvd();
>  	/* write microcode via MSR 0x79 */
>  	wrmsrl(MSR_IA32_UCODE_WRITE, (unsigned long)mc->bits);
>  
> -- 
> 2.7.4
> 

-- 
Regards/Gruss,
    Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
--