From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id 6BE413858CDB; Mon, 9 Jan 2023 21:36:46 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 6BE413858CDB DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1673300206; bh=zZkw5dP80z6IaL1nYJ8p0A5CxtpTxVv1/VO3MIuIQyE=; h=From:To:Subject:Date:In-Reply-To:References:From; b=oLO7p5gl7ch2Pk3C1vTNdh760/B4do2up/IwYemisCdlCJGPJNnXsxqJqVuuKOF4M qXCwcptBolhA1qciSQGPwGkXW/5ltyUgqv11Hrz+zpIl/HoPAHibmzEVyABbDpDntg f9Y24Mxp4libqF4hOv7qHl/vkl6mz9NUu8ewRWoI= From: "joseph at codesourcery dot com" To: glibc-bugs@sourceware.org Subject: [Bug math/28472] pow(10, i) accuracy Date: Mon, 09 Jan 2023 21:36:46 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: glibc X-Bugzilla-Component: math X-Bugzilla-Version: 2.31 X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: joseph at codesourcery dot com X-Bugzilla-Status: UNCONFIRMED X-Bugzilla-Resolution: X-Bugzilla-Priority: P2 X-Bugzilla-Assigned-To: unassigned at sourceware dot org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://sourceware.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 List-Id: https://sourceware.org/bugzilla/show_bug.cgi?id=3D28472 --- Comment #12 from joseph at codesourcery dot com --- I don't consider special-casing 10 in pow to be reasonable. Integrating=20 better function implementations from CORE-MATH would be reasonable, but=20 it's important there to take account of portability considerations not=20 handled in the CORE-MATH code (glibc supports 32-bit platforms, platforms=20 without all the exception macros defined, platforms where there may be=20 excess precision for intermediate computations, platforms where various=20 __builtin_* functions used in CORE-MATH will be expanded out-of-line and=20 slow, etc., and has various ABI and namespace considerations not covered=20 in the CORE-MATH code as well, so significant work would be needed to turn= =20 such an implementation into something suitable for glibc). Note that for=20 any such new function implementatations, having suitable benchmark inputs=20 in glibc's benchmarks is important first, to demonstrate that a new=20 implementation is at least as fast as well as more accurate. --=20 You are receiving this mail because: You are on the CC list for the bug.=