From mboxrd@z Thu Jan  1 00:00:00 1970
From: Dave Hansen <dave.hansen@linux.intel.com>
Subject: Re: [PATCH 03/11] x86/fpu: make __raw_xsave_addr() use feature number
 instead of mask
Date: Fri, 12 Oct 2018 08:52:03 -0700
Message-ID: <225ca4ed-a0c8-9566-4a40-00d7876b623b@linux.intel.com>
References: <20181004140547.13014-1-bigeasy@linutronix.de>
 <20181004140547.13014-4-bigeasy@linutronix.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
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>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        linux-kernel@vger.kernel.org
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <20181004140547.13014-4-bigeasy@linutronix.de>
Content-Language: en-US
Sender: linux-kernel-owner@vger.kernel.org
List-Id: kvm.vger.kernel.org

On 10/04/2018 07:05 AM, Sebastian Andrzej Siewior wrote:
> Most users of __raw_xsave_addr() use a feature number, shift it to a
> mask and then __raw_xsave_addr() shifts it back to the feature number.
> 
> Make __raw_xsave_addr() use the feature number as argument.

This generally looks like a nice cleanup.  Thanks for taking a look at it!

> diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c
> index 87a57b7642d36..38d0b5ea40425 100644
> --- a/arch/x86/kernel/fpu/xstate.c
> +++ b/arch/x86/kernel/fpu/xstate.c
> @@ -811,10 +811,8 @@ void fpu__resume_cpu(void)
>   *
>   * Note: does not work for compacted buffers.
>   */
> -void *__raw_xsave_addr(struct xregs_state *xsave, int xstate_feature_mask)
> +void *__raw_xsave_addr(struct xregs_state *xsave, int feature_nr)
>  {

Could we call this 'xfeature_nr' consistently?

> -	int feature_nr = fls64(xstate_feature_mask) - 1;
> -
>  	if (!xfeature_enabled(feature_nr)) {
>  		WARN_ON_FPU(1);
>  		return NULL;
> @@ -842,6 +840,7 @@ void *__raw_xsave_addr(struct xregs_state *xsave, int xstate_feature_mask)
>   */
>  void *get_xsave_addr(struct xregs_state *xsave, int xstate_feature)
>  {
> +	int feature_nr;
>  	/*
>  	 * Do we even *have* xsave state?
>  	 */
> @@ -869,7 +868,8 @@ void *get_xsave_addr(struct xregs_state *xsave, int xstate_feature)
>  	if (!(xsave->header.xfeatures & xstate_feature))
>  		return NULL;
>  
> -	return __raw_xsave_addr(xsave, xstate_feature);
> +	feature_nr = fls64(xstate_feature) - 1;
> +	return __raw_xsave_addr(xsave, feature_nr);
>  }

Should we also be using a feature number for get_xsave_addr()?  In other
words, could you take a look and see how widely we should be doing this
kind of conversion and not just limit it to __raw_xsave_addr()?