[Bug math/19595] New: [ldbl-128ibm] fmodl incorrect for results in subnormal double range

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

https://sourceware.org/bugzilla/show_bug.cgi?id=19595

            Bug ID: 19595
           Summary: [ldbl-128ibm] fmodl incorrect for results in subnormal
                    double range
           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: ---

The ldbl-128ibm implementation of fmodl produces completely incorrect results
when the correct result is in the range of subnormal double values.  I have a
patch for 2.24.  E.g.:

Failure: Test: fmod (0x1p1023, 0x3p-1074)
Result:
 is:          3.19334449525555311678003001794392e+293  
0x1.000000000000200000000000000p+975
 should be:   9.88131291682493088353137585736443e-324  
0x0.000000000000200000000000000p-1022
 difference:  3.19334449525555311678003001794392e+293  
0x1.000000000000200000000000000p+975
 ulp       :  inf
 max.ulp   :  1.0000

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