From mboxrd@z Thu Jan 1 00:00:00 1970 From: Oleg Nesterov Subject: Re: [PATCH v4 2/5] x86,entry: Only call user_exit if TIF_NOHZ Date: Tue, 29 Jul 2014 21:32:32 +0200 Message-ID: <20140729193232.GA8153@redhat.com> References: <7123b2489cc5d1d5abb7bcf1364ca729cab3e6ca.1406604806.git.luto@amacapital.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <7123b2489cc5d1d5abb7bcf1364ca729cab3e6ca.1406604806.git.luto@amacapital.net> Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org List-help: List-unsubscribe: List-software: Ecartis version 1.0.0 List-subscribe: List-owner: List-post: List-archive: To: Andy Lutomirski Cc: linux-kernel@vger.kernel.org, Kees Cook , Will Drewry , x86@kernel.org, linux-arm-kernel@lists.infradead.org, linux-mips@linux-mips.org, linux-arch@vger.kernel.org, linux-security-module@vger.kernel.org, Alexei Starovoitov , hpa@zytor.com, Frederic Weisbecker List-Id: linux-arch.vger.kernel.org On 07/28, Andy Lutomirski wrote: > > @@ -1449,7 +1449,12 @@ long syscall_trace_enter(struct pt_regs *regs) > { > long ret = 0; > > - user_exit(); > + /* > + * If TIF_NOHZ is set, we are required to call user_exit() before > + * doing anything that could touch RCU. > + */ > + if (test_thread_flag(TIF_NOHZ)) > + user_exit(); Personally I still think this change just adds more confusion, but I leave this to you and Frederic. It is not that "If TIF_NOHZ is set, we are required to call user_exit()", we need to call user_exit() just because we enter the kernel. TIF_NOHZ is just the implementation detail which triggers this slow path. At least it should be correct, unless I am confused even more than I think. Oleg. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:60466 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751397AbaG2Teq (ORCPT ); Tue, 29 Jul 2014 15:34:46 -0400 Date: Tue, 29 Jul 2014 21:32:32 +0200 From: Oleg Nesterov Subject: Re: [PATCH v4 2/5] x86,entry: Only call user_exit if TIF_NOHZ Message-ID: <20140729193232.GA8153@redhat.com> References: <7123b2489cc5d1d5abb7bcf1364ca729cab3e6ca.1406604806.git.luto@amacapital.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <7123b2489cc5d1d5abb7bcf1364ca729cab3e6ca.1406604806.git.luto@amacapital.net> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Andy Lutomirski Cc: linux-kernel@vger.kernel.org, Kees Cook , Will Drewry , x86@kernel.org, linux-arm-kernel@lists.infradead.org, linux-mips@linux-mips.org, linux-arch@vger.kernel.org, linux-security-module@vger.kernel.org, Alexei Starovoitov , hpa@zytor.com, Frederic Weisbecker Message-ID: <20140729193232.bIcM0RHotCghppSToMecWwcfqjaJexRKPrpzDz9lWfE@z> On 07/28, Andy Lutomirski wrote: > > @@ -1449,7 +1449,12 @@ long syscall_trace_enter(struct pt_regs *regs) > { > long ret = 0; > > - user_exit(); > + /* > + * If TIF_NOHZ is set, we are required to call user_exit() before > + * doing anything that could touch RCU. > + */ > + if (test_thread_flag(TIF_NOHZ)) > + user_exit(); Personally I still think this change just adds more confusion, but I leave this to you and Frederic. It is not that "If TIF_NOHZ is set, we are required to call user_exit()", we need to call user_exit() just because we enter the kernel. TIF_NOHZ is just the implementation detail which triggers this slow path. At least it should be correct, unless I am confused even more than I think. Oleg.