From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: [patch v3 14/36] Hexagon: Add signal functions Date: Fri, 09 Sep 2011 10:12:38 +0200 Message-ID: <3380882.JqAqX4juos@wuerfel> References: <20110909010847.294039464@codeaurora.org> <20110909010916.028280867@codeaurora.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Return-path: In-Reply-To: <20110909010916.028280867@codeaurora.org> Sender: linux-hexagon-owner@vger.kernel.org To: Richard Kuo Cc: linux-arch@vger.kernel.org, linux-hexagon@vger.kernel.org, linux-kernel@vger.kernel.org, Linas Vepstas List-Id: linux-arch.vger.kernel.org On Thursday 08 September 2011 20:09:01 Richard Kuo wrote: > +/* > + * Signal context structure - contains all info to do with the state > + * before the signal handler was invoked. > + */ > +struct sigcontext { > + struct pt_regs sc_regs; > +}; I've recently discussed this with Martin Schwidefsky for an unrelated reason, but I think now that you should define the sigcontext independent of pt_regs. The pt_regs should really be internal > +struct sigframe { > + unsigned long pad[2]; /* Used to be signal tramp */ > + struct sigcontext sc; > + sigset_t mask; > +}; > + > +struct rt_sigframe { > + unsigned long pad[2]; /* Used to be signal tramp */ > + struct siginfo info; > + struct ucontext uc; > +}; I hadn't noticed this earlier, but I think you can remove the struct sigframe definition. There should be absolutely no need to have both sigframe and rt_sigframe. Arnd From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from moutng.kundenserver.de ([212.227.17.9]:56202 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758574Ab1IIIM5 (ORCPT ); Fri, 9 Sep 2011 04:12:57 -0400 From: Arnd Bergmann Subject: Re: [patch v3 14/36] Hexagon: Add signal functions Date: Fri, 09 Sep 2011 10:12:38 +0200 Message-ID: <3380882.JqAqX4juos@wuerfel> In-Reply-To: <20110909010916.028280867@codeaurora.org> References: <20110909010847.294039464@codeaurora.org> <20110909010916.028280867@codeaurora.org> MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="us-ascii" Sender: linux-arch-owner@vger.kernel.org List-ID: To: Richard Kuo Cc: linux-arch@vger.kernel.org, linux-hexagon@vger.kernel.org, linux-kernel@vger.kernel.org, Linas Vepstas Message-ID: <20110909081238.tgTc7a_pcWtj2MF1h7smeVCcuZVZbD2UmRmepK_kszE@z> On Thursday 08 September 2011 20:09:01 Richard Kuo wrote: > +/* > + * Signal context structure - contains all info to do with the state > + * before the signal handler was invoked. > + */ > +struct sigcontext { > + struct pt_regs sc_regs; > +}; I've recently discussed this with Martin Schwidefsky for an unrelated reason, but I think now that you should define the sigcontext independent of pt_regs. The pt_regs should really be internal > +struct sigframe { > + unsigned long pad[2]; /* Used to be signal tramp */ > + struct sigcontext sc; > + sigset_t mask; > +}; > + > +struct rt_sigframe { > + unsigned long pad[2]; /* Used to be signal tramp */ > + struct siginfo info; > + struct ucontext uc; > +}; I hadn't noticed this earlier, but I think you can remove the struct sigframe definition. There should be absolutely no need to have both sigframe and rt_sigframe. Arnd