From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vineet Gupta Subject: Re: pt_regs leak into userspace (was Re: [PATCH v3 20/71] ARC: Signal handling) Date: Fri, 15 Feb 2013 13:05:00 +0530 Message-ID: <511DE524.3070901@synopsys.com> References: <1359024639-21915-1-git-send-email-vgupta@synopsys.com> <1359024639-21915-9-git-send-email-vgupta@synopsys.com> <51189D2F.4030000@synopsys.com> <5118C458.2070205@synopsys.com> <20130211140704.GI4503@ZenIV.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: Received: from us01smtp2.synopsys.com ([198.182.44.80]:60453 "EHLO kiruna.synopsys.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932857Ab3BOHgQ (ORCPT ); Fri, 15 Feb 2013 02:36:16 -0500 In-Reply-To: Sender: linux-arch-owner@vger.kernel.org List-ID: To: Jonas Bonn Cc: Al Viro , Arnd Bergmann , "linux-arch@vger.kernel.org" , "linux-kernel@vger.kernel.org" On Friday 15 February 2013 12:53 PM, Jonas Bonn wrote: > On 11 February 2013 15:07, Al Viro wrote: > >> I'd suggest asking itanic folks; they do *not* put callee-saved stuff into >> sigcontext. AFAICS, they don't have setcontext() implemented as a syscall >> at all - it's done as sigprocmask() + doing to callee-saved registers what >> longjmp() does. > > Just to round off this discussion, after giving it some more thought I > agree that the case where you would need callee-saved registers > restored is probably rather pathological. Any sane use of > get/set/swapcontext is manageable without this. > > So, Vineet, I'm now convinced your approach is sound. I will probably > amend the OpenRISC arch to behave similarly. Consider your entire > patch Acked now. I'll add your ack to the the signal handling patch as the contention was primarily on sigcontext bits. Thx, -Vineet