From nobody Mon Mar 18 09:53:48 2024 X-Original-To: dev-commits-src-all@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 4Tyqtj13CBz5D7d4; Mon, 18 Mar 2024 09:53:49 +0000 (UTC) (envelope-from git@FreeBSD.org) Received: from mxrelay.nyi.freebsd.org (mxrelay.nyi.freebsd.org [IPv6:2610:1c1:1:606c::19:3]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (4096 bits) client-digest SHA256) (Client CN "mxrelay.nyi.freebsd.org", Issuer "R3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4Tyqtj06Bcz4nPg; Mon, 18 Mar 2024 09:53:49 +0000 (UTC) (envelope-from git@FreeBSD.org) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=freebsd.org; s=dkim; t=1710755629; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=uBZl3hKCK6WpKS4IgxZibgTS2tHF2h2Vc//lvQfYX+M=; b=pSl8PdND0Y+mhXaBEqyslbJFcBGo5eFCLeutTgrpoi0YcTBCmJ4s2Cm2jlQebarhVvrFNy 2HTWRPGWyNiJzaHcBGaquCin4l+xZ1aqfcJOFJaaoetlRiFCNVzqQ4gI0gF9d6sY1fxlJQ sq9TgtWNKsHDXL/Drp/jv3CzIi10EjLwuvyb+31c/WhnYPtwPuRtsK+QkCGXHA7CYYyFRE BvuyeODTm/+SuwKC5YqhUUB9L9zSH+0xfrkii8lUZmQ6Gq00bp/md4FJCpsKEVMdz/f0KT tHdWolMA2UBO9yJCEse2zd8wV/zcnVr1j0e8VyG8935wpVWeXHYPFQY6HybeWA== ARC-Seal: i=1; s=dkim; d=freebsd.org; t=1710755629; a=rsa-sha256; cv=none; b=JV+YXhKaWMeia5SpHhjaqnW/HFyq9u3rbFkJr5ANoAARcnhdVmWJxEEfvlDA3JpFSdgvJO z6afknQSqJsnJRcxIfJ+a5A9mc9pPMmaM0jIvGqVW0Om4omOMjg4v320tlXypINRW/NZro E+NjyMtxqdFIvS7tp5hPZv7tzG9gF25llgU0VzLNZUbn5CKu4/n1Hp99IiGhVQDjP7UeI8 XJg0Guh602uNVKyXAepfAgQIO135+6asw6ckNxsj3VeVsykutP8ZEilVPxj0ReZ31l8gPv M5GP/0NlHq8lTw8TkFskd/7PoexKih+/ziQ8k3dE1aN/NqFoh81qIXiIryPuOw== ARC-Authentication-Results: i=1; mx1.freebsd.org; none ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=freebsd.org; s=dkim; t=1710755629; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=uBZl3hKCK6WpKS4IgxZibgTS2tHF2h2Vc//lvQfYX+M=; b=D5fzlY6Xgzo07jgt6BjMKpe5mK/22U8ErbLn70APB6Co5ByKygN0gA6xEoN0nLPWmZWLmo HblwTJBPfecSaaQELc66Ynb6q8k+28s5EsSP6nU9DAHW1rMcOjXeHFNJJII6Nrr/vju/G6 75MkRW8uGahhItlvPbW6zE0G5OCTNxtaGy8xNiV5BltGygmriEZ0ixMMPy80c/SF5q+Wdu l8VzYU3RcpwWRijMcqQic7KzCXCmcwaD7eHCsJmC80sPe2ulv3cDUcPBuAGMPzYiQ+wpAW +l4hrtf4zru9r25DBjhmZ1vM696VPfcj0qwEpadI/iDpWCH0N1fe+jLlWvOxeQ== Received: from gitrepo.freebsd.org (gitrepo.freebsd.org [IPv6:2610:1c1:1:6068::e6a:5]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (Client did not present a certificate) by mxrelay.nyi.freebsd.org (Postfix) with ESMTPS id 4Tyqth6q19zJ8H; Mon, 18 Mar 2024 09:53:48 +0000 (UTC) (envelope-from git@FreeBSD.org) Received: from gitrepo.freebsd.org ([127.0.1.44]) by gitrepo.freebsd.org (8.17.1/8.17.1) with ESMTP id 42I9rmFj090543; Mon, 18 Mar 2024 09:53:48 GMT (envelope-from git@gitrepo.freebsd.org) Received: (from git@localhost) by gitrepo.freebsd.org (8.17.1/8.17.1/Submit) id 42I9rmbZ090540; Mon, 18 Mar 2024 09:53:48 GMT (envelope-from git) Date: Mon, 18 Mar 2024 09:53:48 GMT Message-Id: <202403180953.42I9rmbZ090540@gitrepo.freebsd.org> To: src-committers@FreeBSD.org, dev-commits-src-all@FreeBSD.org, dev-commits-src-main@FreeBSD.org From: Mark Murray Subject: git: e38f2308273c - main - lib/msun: Fix tgammal(3) on IEEE 128-bit platforms List-Id: Commit messages for all branches of the src repository List-Archive: https://lists.freebsd.org/archives/dev-commits-src-all List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-dev-commits-src-all@freebsd.org X-BeenThere: dev-commits-src-all@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-Git-Committer: markm X-Git-Repository: src X-Git-Refname: refs/heads/main X-Git-Reftype: branch X-Git-Commit: e38f2308273c8a51ec45f013d22c963590917cca Auto-Submitted: auto-generated The branch main has been updated by markm: URL: https://cgit.FreeBSD.org/src/commit/?id=e38f2308273c8a51ec45f013d22c963590917cca commit e38f2308273c8a51ec45f013d22c963590917cca Author: Mark Murray AuthorDate: 2024-03-01 15:53:58 +0000 Commit: Mark Murray CommitDate: 2024-03-18 09:48:43 +0000 lib/msun: Fix tgammal(3) on IEEE 128-bit platforms Undo the 80-bit "stub" implementation of the 128-bit long double tgammal(3) function. The latest (as of Feb 2024) version of the src/contrib/arm-optimised-routines library includes a standalone, full 128-bit replacement. This needs a small bit of wrapping to fit it in, but is otherwise a drop-in replacement. Testing this is hard, as most maths packages blow up as soon as their 80-bit floating-point capability is exceeded. With 128-bit tgammal(), this is easy to do, and this is the range that needs to be checked the most carefully. Using my copy of Maple, I was able to check that the output was within a few ULP of the correct answer, right up to the point of 128-bit over- and underflow. Additionally, the results are no worse, and indeed better than the 80-bit version. Steve Kargl sent me his libm testing code, which I used to verify that the excpetions for certain key values were correct. Tested in this case were +-Inf, +-NaN, +-1 and +-0. Differential Revision: https://reviews.freebsd.org/D44168 Reviewed by: theraven, andrew, imp --- lib/msun/Makefile | 2 +- lib/msun/ld128/b_tgammal.c | 53 +++++----------------------------------------- lib/msun/man/lgamma.3 | 19 ++++++++--------- 3 files changed, 15 insertions(+), 59 deletions(-) diff --git a/lib/msun/Makefile b/lib/msun/Makefile index 6c8c6c3e8009..24989749a502 100644 --- a/lib/msun/Makefile +++ b/lib/msun/Makefile @@ -26,7 +26,7 @@ CFLAGS+= -I${.CURDIR}/x86 CFLAGS+= -I${.CURDIR}/ld80 .elif ${LDBL_PREC} == 113 .PATH: ${.CURDIR}/ld128 -CFLAGS+= -I${.CURDIR}/ld128 +CFLAGS+= -I${.CURDIR}/ld128 -I${SRCTOP}/contrib/arm-optimized-routines/math .endif CFLAGS+= -I${.CURDIR}/${ARCH_SUBDIR} diff --git a/lib/msun/ld128/b_tgammal.c b/lib/msun/ld128/b_tgammal.c index 4bae4f3aded6..6df7264a4c9e 100644 --- a/lib/msun/ld128/b_tgammal.c +++ b/lib/msun/ld128/b_tgammal.c @@ -1,55 +1,12 @@ /*- * SPDX-License-Identifier: BSD-2-Clause * - * Copyright (c) 2013 David Chisnall - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * - * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND - * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. + * Copyright (c) 2024 The FreeBSD Foundation */ -#include -#include - /* - * If long double is not the same size as double, then these will lose - * precision and we should emit a warning whenever something links against - * them. + * This is a pure C function generously donated by ARM. + * See src/contrib/arm-optimized-routines/math/tgamma128.[ch]. */ -#if (LDBL_MANT_DIG > 53) -#define WARN_IMPRECISE(x) \ - __warn_references(x, # x " has lower than advertised precision"); -#else -#define WARN_IMPRECISE(x) -#endif -/* - * Declare the functions as weak variants so that other libraries providing - * real versions can override them. - */ -#define DECLARE_WEAK(x)\ - __weak_reference(imprecise_## x, x);\ - WARN_IMPRECISE(x) - -#define DECLARE_IMPRECISE(f) \ - long double imprecise_ ## f ## l(long double v) { return f(v); }\ - DECLARE_WEAK(f ## l) - -DECLARE_IMPRECISE(tgamma); +#define tgamma128 tgammal +#include "tgamma128.c" diff --git a/lib/msun/man/lgamma.3 b/lib/msun/man/lgamma.3 index 8c0298ec8299..a77d524373fb 100644 --- a/lib/msun/man/lgamma.3 +++ b/lib/msun/man/lgamma.3 @@ -25,7 +25,7 @@ .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd December 8, 2017 +.Dd March 1, 2024 .Dt LGAMMA 3 .Os .Sh NAME @@ -167,15 +167,6 @@ non-positive integers. For large non-integer negative values, .Fn tgamma will underflow. -.Sh BUGS -To conform with newer C/C++ standards, a stub implementation for -.Nm tgammal -was committed to the math library, where -.Nm tgammal -is mapped to -.Nm tgamma . -Thus, the numerical accuracy is at most that of the 53-bit double -precision implementation. .Sh SEE ALSO .Xr math 3 .Sh STANDARDS @@ -212,3 +203,11 @@ The .Fn tgamma function appeared in .Fx 5.0 . +The 128-bit +.Ft long double +version of +.Fn tgammal +replaced the 80-bit stub version in +version in +.Fx 16 , +thanks to an appropriate implementation from Arm.