Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757713AbcCUTnj (ORCPT ); Mon, 21 Mar 2016 15:43:39 -0400 Received: from one.firstfloor.org ([193.170.194.197]:37997 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756082AbcCUTni (ORCPT ); Mon, 21 Mar 2016 15:43:38 -0400 Date: Mon, 21 Mar 2016 20:43:36 +0100 From: Andi Kleen To: Brian Gerst Cc: Andi Kleen , the arch/x86 maintainers , Andy Lutomirski , Linux Kernel Mailing List , Andi Kleen Subject: Re: [PATCH 4/9] x86: Enumerate kernel FSGS capability in AT_HWCAP2 Message-ID: <20160321194336.GC5083@two.firstfloor.org> References: <1458576969-13309-1-git-send-email-andi@firstfloor.org> <1458576969-13309-5-git-send-email-andi@firstfloor.org> <20160321185429.GY5083@two.firstfloor.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 687 Lines: 17 > It would be less complicated actually, as normal userspace would just > continue to call arch_prctl() as it does today. Glibc would implement We already have that through the system call, no advantage of putting it into a vsyscall. Also the experience with getcpu and similar so far is that vsyscalls are too slow for the users who want really fast paths. So they're prefer to use the direct instructions anyways. > arch_prctl() just like it does with gettimeofday() -- with an ifunc > selector [1] that calls the VDSO function if it is available, or the > syscall if not. No custom assembly needed. vdso always needs custom assembler, please see how glibc implements it. -Andi