From owner-freebsd-current@FreeBSD.ORG Wed Mar 26 17:55:15 2014 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id A5CD4D99; Wed, 26 Mar 2014 17:55:15 +0000 (UTC) Received: from bigwig.baldwin.cx (bigwig.baldwin.cx [IPv6:2001:470:1f11:75::1]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 795CC32A; Wed, 26 Mar 2014 17:55:15 +0000 (UTC) Received: from jhbbsd.localnet (unknown [209.249.190.124]) by bigwig.baldwin.cx (Postfix) with ESMTPSA id 4CC16B93B; Wed, 26 Mar 2014 13:55:14 -0400 (EDT) From: John Baldwin To: Adrian Chadd Subject: Re: signal 8 (floating point exception) upon resume Date: Wed, 26 Mar 2014 13:43:22 -0400 User-Agent: KMail/1.13.5 (FreeBSD/8.4-CBSD-20130906; KDE/4.5.5; amd64; ; ) References: <201403251546.51476.jhb@freebsd.org> In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201403261343.22370.jhb@freebsd.org> X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.7 (bigwig.baldwin.cx); Wed, 26 Mar 2014 13:55:14 -0400 (EDT) Cc: "freebsd-mobile@freebsd.org" , freebsd-current , Konstantin Belousov , Bruce Evans , Bruce Evans , Jung-uk Kim X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Mar 2014 17:55:15 -0000 On Tuesday, March 25, 2014 5:38:50 pm Adrian Chadd wrote: > On 25 March 2014 12:46, John Baldwin wrote: > > On Sunday, March 23, 2014 4:41:24 pm Adrian Chadd wrote: > >> [snip] > >> > >> Hi, > >> > >> As part of this thread, a whole lot of stuff was thrown around to try > >> and fix / improve the correctness of this. > >> > >> But it still happens to me in -HEAD i386. I updated to r263418 and > >> it's now doing it around 30-50% of the time I resume. > > > > Yes, nothing has changed in HEAD. > > > >> So, since I really am trying to avoid getting neck deep in learning > >> (by myself) a new thing right now, would someone be willing to help me > >> through the process of (a) learning how this is all supposed to work > >> (which thanks to jhb and bde, I think I've learnt from the posts in > >> this thread) and (b) some things to try out? I'll be able to report > >> the results of this pretty quickly. > > > > You can try www.freebsd.org/~jhb/patches/i386_fpu_suspend2.patch. You > > could have tried the first patch I posted here earlier when I first > > posted it as well. :) > > There was a lot of chatter, I thought it was prudent to let it all > settle before jumping in. > > Anyway: You do understand C well enough to fix simple typos? I don't have any i386 machines around, but I'll work on cross-building. > --- npx.o --- > /usr/home/adrian/work/freebsd/head/src/sys/i386/isa/npx.c:769:18: > error: declaration of 'union safefpu' will not be visible outside of > this function [-Werror,-Wvisibility] > npxsuspend(union safefpu *addr) > ^ s/safe/save/ > /usr/home/adrian/work/freebsd/head/src/sys/i386/isa/npx.c:779:8: > error: implicit declaration of function 'rcr' is invalid in C99 > [-Werror,-Wimplicit-function-declaration] > cr0 = rcr(0); > ^ Probably just need #include -- John Baldwin