This is the mail archive of the
glibc-bugs@sourceware.org
mailing list for the glibc project.
[Bug math/14328] New: ctan, ctanh inaccuracy for subnormals in round-upwards mode
- From: "jsm28 at gcc dot gnu.org" <sourceware-bugzilla at sourceware dot org>
- To: glibc-bugs at sources dot redhat dot com
- Date: Tue, 03 Jul 2012 20:10:21 +0000
- Subject: [Bug math/14328] New: ctan, ctanh inaccuracy for subnormals in round-upwards mode
- Auto-submitted: auto-generated
http://sourceware.org/bugzilla/show_bug.cgi?id=14328
Bug #: 14328
Summary: ctan, ctanh inaccuracy for subnormals in round-upwards
mode
Product: glibc
Version: 2.15
Status: NEW
Severity: normal
Priority: P2
Component: math
AssignedTo: unassigned@sourceware.org
ReportedBy: jsm28@gcc.gnu.org
CC: liubov.dmitrieva@gmail.com
Classification: Unclassified
(This bug report is for an issue identified in test logs sent off-list by
Liubov Dmitrieva.)
ctan and ctanh functions (for all floating-point types) give inaccurate results
(with spurious underflow exceptions), in round-upwards mode only, in some cases
when the imaginary part of the argument (for ctan) or the real part (for ctanh)
is subnormal. For example, ctan (0x1.921fb54442d18p+0 + 0x1p-1074i) gives (on
x86_64) a result with imaginary part 2.635438829887015752e-291, but it should
be about half that; likewise, ctanh (0x1p-1074 + 0x1.921fb54442d18p+0i) gives
real part 2.635438829887015752e-291, but it should be about half that.
I'm working on a patch for this issue.
--
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.