From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <sourceware-bugzilla@sourceware.org>
Received: by sourceware.org (Postfix, from userid 48)
	id 94F2D3858C74; Wed, 28 Feb 2024 11:32:09 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 94F2D3858C74
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org;
	s=default; t=1709119929;
	bh=Hyg5C6FNh0PDT0kWc/rhufD02CD3DcLMnkA9JM7q3gM=;
	h=From:To:Subject:Date:From;
	b=SS1+XGMwDzflORcuWqodtZCqXVX88bwMeoLd3kvXQqMTJL1JMlVoG4urVsw0/rIPY
	 WUDQrOipqPah960b6fGj+4edHXvRU19aCln53PZsDjIxNZXZUSl23isageSlciMpzo
	 oEBVSpt5iCV4YdNPz1MT78X0B2LhRZFxsmLP7DDI=
From: "newbie-02 at gmx dot de" <sourceware-bugzilla@sourceware.org>
To: glibc-bugs@sourceware.org
Subject: [Bug math/31426] New: log10( 10^n ) poor accuracy in bin80 datatype,
 log10l( 1E+07l ) -> 7.000000000000000000434...
Date: Wed, 28 Feb 2024 11:32:09 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: glibc
X-Bugzilla-Component: math
X-Bugzilla-Version: unspecified
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: newbie-02 at gmx dot de
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: bug_id short_desc product version bug_status
 bug_severity priority component assigned_to reporter target_milestone
 attachments.created
Message-ID: <bug-31426-131@http.sourceware.org/bugzilla/>
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: <glibc-bugs.sourceware.org>

https://sourceware.org/bugzilla/show_bug.cgi?id=3D31426

            Bug ID: 31426
           Summary: log10( 10^n ) poor accuracy in bin80 datatype, log10l(
                    1E+07l ) -> 7.000000000000000000434...
           Product: glibc
           Version: unspecified
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: math
          Assignee: unassigned at sourceware dot org
          Reporter: newbie-02 at gmx dot de
  Target Milestone: ---

Created attachment 15381
  --> https://sourceware.org/bugzilla/attachment.cgi?id=3D15381&action=3Ded=
it
Snippet demonstrating log10l imprecision.

hello @all,=20=20

quite sure you'll tell me it's not a bug but the 'feature of imprecision' in
bin-FP-math :-(=20

log10l( 1E+07l ) produces 7.000000000000000000434..., while=20
logl( 1E+07l ) / logl( 10.0l ) would hold with 7.0,=20

Same as with pow10s, having them exact at integral powers of ten=20
would help to form a reliable grid of stable points, and=20
find better ways for correct rounding, which=20
would help for better math.=20

Thus if possible to improve without messing other points I'd=20
appreciate, if not I'd like some explanation why and which=20
calculation path is taken / preferred, and a code pointer to=20
have a look myself.=20

'Me bad' is always an option, tried to evaluate and present=20
with best intention and skills available.=20

If 'notabug' pls. leave open or unconfirmed as hook for others=20
who at some time in the future might have a good idea.

--=20
You are receiving this mail because:
You are on the CC list for the bug.=