From mboxrd@z Thu Jan  1 00:00:00 1970
From: Dave Hansen <dave.hansen@intel.com>
Subject: Re: [PATCH 12/22] x86/fpu: Only write PKRU if it is different from
 current
Date: Mon, 25 Feb 2019 10:08:10 -0800
Message-ID: <198b2412-e81d-6c55-c74f-c2cab60fc5f2@intel.com>
References: <20190221115020.12385-1-bigeasy@linutronix.de>
 <20190221115020.12385-13-bigeasy@linutronix.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Cc: x86@kernel.org, Andy Lutomirski <luto@kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        kvm@vger.kernel.org, "Jason A. Donenfeld" <Jason@zx2c4.com>,
        Rik van Riel <riel@surriel.com>,
        Dave Hansen <dave.hansen@linux.intel.com>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        linux-kernel@vger.kernel.org
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <20190221115020.12385-13-bigeasy@linutronix.de>
Content-Language: en-US
Sender: linux-kernel-owner@vger.kernel.org
List-Id: kvm.vger.kernel.org

On 2/21/19 3:50 AM, Sebastian Andrzej Siewior wrote:
> @@ -111,6 +111,12 @@ static inline void __write_pkru(u32 pkru)
>  {
>  	u32 ecx = 0, edx = 0;
>  
> +	/*
> +	 * WRPKRU is relatively expensive compared to RDPKRU.
> +	 * Avoid WRPKRU when it would not change the value.
> +	 */
> +	if (pkru == __read_pkru())
> +		return;
>  	/*
>  	 * "wrpkru" instruction.  Loads contents in EAX to PKRU,
>  	 * requires that ecx = edx = 0.

Could we do this in write_pkru() instead?  I tried to keep the __
versions of read and write as simple and symmetric as possible.