Date: Sat, 15 Sep 2012 10:03:22 -0600 From: Warner Losh <imp@bsdimp.com> To: Roman Divacky <rdivacky@freebsd.org> Cc: Ian Lepore <freebsd@damnhippie.dyndns.org>, Tijl Coosemans <tijl@coosemans.org>, current@freebsd.org, Dimitry Andric <dim@freebsd.org>, toolchain@freebsd.org, Steve Kargl <sgk@troutmask.apl.washington.edu> Subject: Re: Clang as default compiler November 4th Message-ID: <FEC16D7B-E10F-47CF-B0D6-F89F4EFE3021@bsdimp.com> In-Reply-To: <20120915091455.GA93876@freebsd.org> References: <504F4645.4070900@FreeBSD.org> <504F4A6B.4010001@coosemans.org> <504F5101.8090906@FreeBSD.org> <505101C3.70203@freebsd.org> <20120913020833.GA8255@troutmask.apl.washington.edu> <1347550332.1110.108.camel@revolution.hippie.lan> <20120913161024.GA13846@troutmask.apl.washington.edu> <20120914202319.GB5244@lor.one-eyed-alien.net> <20120915001808.GA70215@troutmask.apl.washington.edu> <20120915010600.GA70426@troutmask.apl.washington.edu> <20120915091455.GA93876@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sep 15, 2012, at 3:14 AM, Roman Divacky wrote: > LLVM by default turns these: >=20 > case LibFunc::copysign: case LibFunc::copysignf: case = LibFunc::copysignl: > case LibFunc::fabs: case LibFunc::fabsf: case = LibFunc::fabsl: > case LibFunc::sin: case LibFunc::sinf: case = LibFunc::sinl: > case LibFunc::cos: case LibFunc::cosf: case = LibFunc::cosl: > case LibFunc::sqrt: case LibFunc::sqrtf: case = LibFunc::sqrtl: > case LibFunc::floor: case LibFunc::floorf: case = LibFunc::floorl: > case LibFunc::nearbyint: case LibFunc::nearbyintf: case = LibFunc::nearbyintl: > case LibFunc::ceil: case LibFunc::ceilf: case = LibFunc::ceill: > case LibFunc::rint: case LibFunc::rintf: case = LibFunc::rintl: > case LibFunc::trunc: case LibFunc::truncf: case = LibFunc::truncl: > case LibFunc::log2: case LibFunc::log2f: case = LibFunc::log2l: > case LibFunc::exp2: case LibFunc::exp2f: case = LibFunc::exp2l: >=20 > from lib calls to direct code (ie. instruction or sequence of). This = is not a "bug" > but feature ;) >=20 > I am not sure what the rules for the transformation should be. Allow = it only with > -ffast-math ? Disallow it on i386? Really, no idea. In the past, gcc has only done this with -ffast-math, but my 'in the = past' is a long time ago... > Steve, you tested on i386? Can you test on amd64? >=20 > Do you guys have any opinion what to do here? >=20 > On Fri, Sep 14, 2012 at 06:06:00PM -0700, Steve Kargl wrote: >> On Fri, Sep 14, 2012 at 05:18:08PM -0700, Steve Kargl wrote: >>>=20 >>> A third class of failure appears to be that clang emits >>> i387 fpu instructions for at least sinf and cosf instead=20 >>> of calls to the library routines. AFAIK, the library >>> routines are faster and more accurate. >>>=20 >>=20 >> Yep. Clang has problems with at least sinf on i386 FreeBSD. >>=20 >> % pwd >> /usr/home/kargl/trunk/math/sine >>=20 >> % make clean && make CC=3Dcc testf >> cc -o testf -O2 -pipe -static -I/usr/local/include -I../mp testf.c \ >> -L/usr/local/lib -L../mp -lsgk -lmpfr -lgmp -lm >>=20 >> % ./testf -m 0 -M 1e20 -r >> ULP Range | >> -----------+------------------------- >> [0.0:0.6] | 1006424 (100.00%) >> (0.6:0.7] | 0 ( 0.00%) >> (0.7:0.8] | 0 ( 0.00%) >> (0.8:0.9] | 0 ( 0.00%) >> (0.9:1.0] | 0 ( 0.00%) >> (1.0:2.0] | 0 ( 0.00%) >> (2.0:3.0] | 0 ( 0.00%) >> 3.0 < ULP | 0 ( 0.00%) >> -----------+------------------------- >> Count | 1006424 >> Max ULP | 0.50084 >> Max ULP x | 53462490661259313152.000000 0x1.72f876p+65 >>=20 >> % make clean && make CC=3Dclang testf >> clang -o testf -O2 -pipe -static -I/usr/local/include -I../mp testf.c = \ >> -L/usr/local/lib -L../mp -lsgk -lmpfr -lgmp -lm >>=20 >> % ./testf -m 0 -M 1e20 -r >> ULP Range | >> -----------+------------------------- >> [0.0:0.6] | 1 ( 0.00%) >> (0.6:0.7] | 0 ( 0.00%) >> (0.7:0.8] | 0 ( 0.00%) >> (0.8:0.9] | 0 ( 0.00%) >> (0.9:1.0] | 0 ( 0.00%) >> (1.0:2.0] | 0 ( 0.00%) >> (2.0:3.0] | 0 ( 0.00%) >> 3.0 < ULP | 999998 (100.00%) >> -----------+------------------------- >> Count | 999999 >> Max ULP | 1328505256679420125050194353979392.00000 >> Max ULP x | 75516780764213542912.000000 0x1.06006p+66 >>=20 >> --=20 >> Steve >> _______________________________________________ >> freebsd-current@freebsd.org mailing list >> http://lists.freebsd.org/mailman/listinfo/freebsd-current >> To unsubscribe, send any mail to = "freebsd-current-unsubscribe@freebsd.org" > _______________________________________________ > freebsd-toolchain@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-toolchain > To unsubscribe, send any mail to = "freebsd-toolchain-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?FEC16D7B-E10F-47CF-B0D6-F89F4EFE3021>