public inbox for glibc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug math/14796] New: fma use of Dekker multiplication in directed rounding modes unsafe
@ 2012-11-01 23:43 jsm28 at gcc dot gnu.org
2012-11-03 19:51 ` [Bug math/14796] " jsm28 at gcc dot gnu.org
2014-06-14 11:16 ` fweimer at redhat dot com
0 siblings, 2 replies; 3+ messages in thread
From: jsm28 at gcc dot gnu.org @ 2012-11-01 23:43 UTC (permalink / raw)
To: glibc-bugs
http://sourceware.org/bugzilla/show_bug.cgi?id=14796
Bug #: 14796
Summary: fma use of Dekker multiplication in directed rounding
modes unsafe
Product: glibc
Version: 2.16
Status: NEW
Severity: normal
Priority: P2
Component: math
AssignedTo: unassigned@sourceware.org
ReportedBy: jsm28@gcc.gnu.org
Classification: Unclassified
fma uses Dekker multiplication, but in directed rounding modes this is unsafe
(at least for odd mantissa lengths); the rounding mode should be set to
FE_TONEAREST before using Dekker multiplication / Knuth addition. Example,
when using dbl-64/s_fma.c: fma (0x0.fffffffffffff8p0, 0x0.fffffffffffff8p0,
-0x0.fffffffffffffp0) returns 0x1p-1074 in FE_DOWNWARD mode; it should return
0x1p-106 in all rounding modes.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug math/14796] fma use of Dekker multiplication in directed rounding modes unsafe
2012-11-01 23:43 [Bug math/14796] New: fma use of Dekker multiplication in directed rounding modes unsafe jsm28 at gcc dot gnu.org
@ 2012-11-03 19:51 ` jsm28 at gcc dot gnu.org
2014-06-14 11:16 ` fweimer at redhat dot com
1 sibling, 0 replies; 3+ messages in thread
From: jsm28 at gcc dot gnu.org @ 2012-11-03 19:51 UTC (permalink / raw)
To: glibc-bugs
http://sourceware.org/bugzilla/show_bug.cgi?id=14796
Joseph Myers <jsm28 at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |FIXED
--- Comment #1 from Joseph Myers <jsm28 at gcc dot gnu.org> 2012-11-03 19:50:54 UTC ---
Fixed for 2.17 by:
commit 5b5b04d6282df0364424c6f2c0462e5c1a4394b0
Author: Joseph Myers <joseph@codesourcery.com>
Date: Sat Nov 3 19:48:53 2012 +0000
Make fma use of Dekker and Knuth algorithms use round-to-nearest (bug
14796).
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug math/14796] fma use of Dekker multiplication in directed rounding modes unsafe
2012-11-01 23:43 [Bug math/14796] New: fma use of Dekker multiplication in directed rounding modes unsafe jsm28 at gcc dot gnu.org
2012-11-03 19:51 ` [Bug math/14796] " jsm28 at gcc dot gnu.org
@ 2014-06-14 11:16 ` fweimer at redhat dot com
1 sibling, 0 replies; 3+ messages in thread
From: fweimer at redhat dot com @ 2014-06-14 11:16 UTC (permalink / raw)
To: glibc-bugs
https://sourceware.org/bugzilla/show_bug.cgi?id=14796
Florian Weimer <fweimer at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Flags| |security-
--
You are receiving this mail because:
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2014-06-14 11:16 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-11-01 23:43 [Bug math/14796] New: fma use of Dekker multiplication in directed rounding modes unsafe jsm28 at gcc dot gnu.org
2012-11-03 19:51 ` [Bug math/14796] " jsm28 at gcc dot gnu.org
2014-06-14 11:16 ` fweimer at redhat dot com
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).