From owner-freebsd-toolchain@freebsd.org Thu Mar 14 06:30:16 2019 Return-Path: Delivered-To: freebsd-toolchain@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 1A5F11531767; Thu, 14 Mar 2019 06:30:16 +0000 (UTC) (envelope-from sgk@troutmask.apl.washington.edu) Received: from troutmask.apl.washington.edu (troutmask.apl.washington.edu [128.95.76.21]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) server-signature RSA-PSS (4096 bits) client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "troutmask", Issuer "troutmask" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id 8BF3A8DCEC; Thu, 14 Mar 2019 06:30:14 +0000 (UTC) (envelope-from sgk@troutmask.apl.washington.edu) Received: from troutmask.apl.washington.edu (localhost [127.0.0.1]) by troutmask.apl.washington.edu (8.15.2/8.15.2) with ESMTPS id x2E6U7DR042005 (version=TLSv1.3 cipher=TLS_AES_256_GCM_SHA384 bits=256 verify=NO); Wed, 13 Mar 2019 23:30:07 -0700 (PDT) (envelope-from sgk@troutmask.apl.washington.edu) Received: (from sgk@localhost) by troutmask.apl.washington.edu (8.15.2/8.15.2/Submit) id x2E6U7ke042004; Wed, 13 Mar 2019 23:30:07 -0700 (PDT) (envelope-from sgk) Date: Wed, 13 Mar 2019 23:30:07 -0700 From: Steve Kargl To: John Baldwin Cc: freebsd-toolchain@freebsd.org, freebsd-current@freebsd.org Subject: Re: Optimization bug with floating-point? Message-ID: <20190314063007.GA41995@troutmask.apl.washington.edu> Reply-To: sgk@troutmask.apl.washington.edu References: <20190313024506.GA31746@troutmask.apl.washington.edu> <20190313151635.GA34757@troutmask.apl.washington.edu> <20190313164039.GA35340@troutmask.apl.washington.edu> <20190313212455.GA37717@troutmask.apl.washington.edu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20190313212455.GA37717@troutmask.apl.washington.edu> User-Agent: Mutt/1.11.2 (2019-01-07) X-Rspamd-Queue-Id: 8BF3A8DCEC X-Spamd-Bar: +++ Authentication-Results: mx1.freebsd.org X-Spamd-Result: default: False [3.95 / 15.00]; ARC_NA(0.00)[]; HAS_REPLYTO(0.00)[sgk@troutmask.apl.washington.edu]; FROM_HAS_DN(0.00)[]; RCPT_COUNT_THREE(0.00)[3]; TO_DN_SOME(0.00)[]; TO_MATCH_ENVRCPT_ALL(0.00)[]; MIME_GOOD(-0.10)[text/plain]; MIME_TRACE(0.00)[0:+]; DMARC_NA(0.00)[washington.edu]; AUTH_NA(1.00)[]; REPLYTO_ADDR_EQ_FROM(0.00)[]; RCVD_COUNT_THREE(0.00)[3]; RCVD_TLS_LAST(0.00)[]; NEURAL_SPAM_SHORT(0.82)[0.825,0]; RCVD_IN_DNSWL_MED(-0.20)[21.76.95.128.list.dnswl.org : 127.0.11.2]; MX_GOOD(-0.01)[troutmask.apl.washington.edu]; NEURAL_SPAM_LONG(0.95)[0.950,0]; NEURAL_SPAM_MEDIUM(0.46)[0.462,0]; R_SPF_NA(0.00)[]; FROM_EQ_ENVFROM(0.00)[]; R_DKIM_NA(0.00)[]; SUBJECT_ENDS_QUESTION(1.00)[]; ASN(0.00)[asn:73, ipnet:128.95.0.0/16, country:US]; MID_RHS_MATCH_FROM(0.00)[]; IP_SCORE(0.02)[ip: (0.09), ipnet: 128.95.0.0/16(0.08), asn: 73(0.01), country: US(-0.07)] X-BeenThere: freebsd-toolchain@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Maintenance of FreeBSD's integrated toolchain List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Mar 2019 06:30:16 -0000 On Wed, Mar 13, 2019 at 02:24:55PM -0700, Steve Kargl wrote: > On Wed, Mar 13, 2019 at 10:16:12AM -0700, John Baldwin wrote: > > On 3/13/19 9:40 AM, Steve Kargl wrote: > > > On Wed, Mar 13, 2019 at 09:32:57AM -0700, John Baldwin wrote: > > >> On 3/13/19 8:16 AM, Steve Kargl wrote: > > >>> On Tue, Mar 12, 2019 at 07:45:41PM -0700, Steve Kargl wrote: > > >>>> > > >>>> gcc8 --version > > >>>> gcc8 (FreeBSD Ports Collection) 8.3.0 > > >>>> > > >>>> gcc8 -fno-builtin -o z a.c -lm && ./z > > >>>> gcc8 -O -fno-builtin -o z a.c -lm && ./z > > >>>> gcc8 -O2 -fno-builtin -o z a.c -lm && ./z > > >>>> gcc8 -O3 -fno-builtin -o z a.c -lm && ./z > > >>>> > > >>>> Max ULP: 2.297073 > > >>>> Count: 0 (# of ULP that exceed 21) > > >>> > > >>> clang agrees with gcc8 if one changes ... > > >>> > > >>>> int > > >>>> main(void) > > >>>> { > > >>>> double re, im, u, ur, ui; > > >>>> float complex f; > > >>>> float x, y; > > >>> > > >>> this line to "volatile float x, y". > > >> > > >> So it seems to be a regression in clang 7 vs clang 6? > > > > > > /usr/local/bin/clang60 has the same problem. > > > > > > % /usr/local/bin/clang60 -o z -O2 a.c -lm && ./z > > > Maximum ULP: 23.061242 > > > # of ULP > 21: 39 > > > > > > Adding volatile as in the above "fixes" the problem. > > > > > > AFAICT, this a i386/387 code generation problem. Perhaps, > > > an alignment issue? > > > > Oh, I misread your earlier e-mail to say that clang60 worked. > > > > One issue I'm aware of is that clang does not have any support for the > > special arrangement FreeBSD/i386 uses where it uses different precision > > for registers vs in-memory for some of the floating point types (GCC has > > a special hack that is only used on FreeBSD for this but isn't used on > > any other OS's). I wonder if that could be a factor? Volatile probably > > forces a round trip between memory which might explain why this is the > > case. > > > > I went looking for this special hack. In gcc/gccx/config/i386, > one finds > > /* FreeBSD sets the rounding precision of the FPU to 53 bits. Let the > compiler get the contents of and std::numeric_limits correct. */ > #undef TARGET_96_ROUND_53_LONG_DOUBLE > #define TARGET_96_ROUND_53_LONG_DOUBLE (!TARGET_64BIT) > > So, taking this as a hunch, I added ieeefp.h to my test program > and called 'fpsetprec(FP_PD)' as the first executable statement. > This then results in > > % cc -fno-builtin -m32 -O2 -o z b.o a.c -lm && ./z > Max u: 2.297073 > Count: 0 > > So, is there a way to correctly build clang for i386/387 > to automatically set the precision correctly? > Spent a couple hours wandering in contrib/llvm. Have no idea how to fix clang to actually work on i386/387. Any ideas would be welcomed. AFAICT, all libm float routines need to be modified to conditional include ieeefp.h and call fpsetprec(FP_PD). This will work around issues is FP and libm. FreeBSD needs to issue an erratum about the numerical issues with clang. -- Steve