[Bug math/20205] New: [i386/x86_64] nextafterl incorrect incrementing negative subnormals

["jsm28 at gcc dot gnu.org" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=20205

            Bug ID: 20205
           Summary: [i386/x86_64] nextafterl incorrect incrementing
                    negative subnormals
           Product: glibc
           Version: 2.23
            Status: NEW
          Severity: normal
          Priority: P2
         Component: math
          Assignee: unassigned at sourceware dot org
          Reporter: jsm28 at gcc dot gnu.org
  Target Milestone: ---
              Host: i?86-*-* x86_64-*-*

The x86 / x86_64 implementation of nextafterl (also used for nexttowardl)
produces incorrect results when negative subnormals, the low 32 bits of whose
mantissa are zero, are incremented towards zero.  For example, nextafterl
(-0x0.fffffffep-16382L, 0.0L) wrongly produces a NaN.

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