Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Thomas Gleixner <tglx@linutronix.de>
To:     Adamos Ttofari <attofari@amazon.de>
Cc:     attofari@amazon.de, Ingo Molnar <mingo@redhat.com>,
        Borislav Petkov <bp@alien8.de>,
        Dave Hansen <dave.hansen@linux.intel.com>, x86@kernel.org,
        "H. Peter Anvin" <hpa@zytor.com>, Kyle Huey <me@kylehuey.com>,
        "Chang S. Bae" <chang.seok.bae@intel.com>,
        Andrew Cooper <andrew.cooper3@citrix.com>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] x86: fpu: Keep xfd_state always in sync with
 MSR_IA32_XFD
In-Reply-To: <20230512113900.56393-1-attofari@amazon.de>
References: <20230511152818.13839-1-attofari@amazon.de>
 <20230512113900.56393-1-attofari@amazon.de>
Date:   Fri, 12 May 2023 14:52:30 +0200
Message-ID: <87y1ltbtkh.ffs@tglx>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On Fri, May 12 2023 at 11:38, Adamos Ttofari wrote:
>  	if (cpu_feature_enabled(X86_FEATURE_XFD))
> -		wrmsrl(MSR_IA32_XFD, init_fpstate.xfd);
> +		xfd_set_state(init_fpstate.xfd);

> @@ -914,8 +915,7 @@ void fpu__resume_cpu(void)
>  				     xfeatures_mask_independent());
>  	}
>  
> -	if (fpu_state_size_dynamic())
> -		wrmsrl(MSR_IA32_XFD, current->thread.fpu.fpstate->xfd);
> +	xfd_update_state(current->thread.fpu.fpstate);

How is that supposed to work?

    this_cpu(xfd_state) == current->thread.fpu.fpstate->xfd

So the MSR write won't happen.

Thanks,

        tglx