From mboxrd@z Thu Jan  1 00:00:00 1970
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Subject: Re: [PATCH 18/24] x86/fpu: Prepare copy_fpstate_to_sigframe() for
 TIF_NEED_FPU_LOAD
Date: Mon, 1 Apr 2019 10:24:24 +0200
Message-ID: <20190401082424.uljjh5cejj33gb3r@linutronix.de>
References: <20190321202632.16810-1-bigeasy@linutronix.de>
 <20190321202632.16810-19-bigeasy@linutronix.de>
 <alpine.DEB.2.21.1903312013390.2476@nanos.tec.linutronix.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Cc: linux-kernel@vger.kernel.org, x86@kernel.org,
        Andy Lutomirski <luto@kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Radim =?utf-8?B?S3LEjW3DocWZ?= <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: Thomas Gleixner <tglx@linutronix.de>
Return-path: <linux-kernel-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <alpine.DEB.2.21.1903312013390.2476@nanos.tec.linutronix.de>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: kvm.vger.kernel.org

On 2019-03-31 20:20:25 [+0200], Thomas Gleixner wrote:
> 
> I think this should do the following:
> 
> 	fpregs_lock();
> 	if (!test_thread_flag(TIF_NEED_FPU_LOAD)) {
> 		pagefault_disable();
> 		ret = copy_fpu_to_user(...);
> 		pagefault_enable();
> 		if (!res)
> 			return 0;
> 		copy_fpregs_to_fpstate(fpu); 
> 	}
> 	fpregs_unlock();
> 
> The point is that in most cases the direct store from the FPU registers to
> user space will succeed simply because the stack is accessible and you only
> do the store in kernel memory and copy when that fails.

let me try that.

> Thanks,
> 
> 	tglx
> 

Sebastian