This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug math/16731] New: log(1) is -0, should be +0, in FE_DOWNWARD mode


https://sourceware.org/bugzilla/show_bug.cgi?id=16731

            Bug ID: 16731
           Summary: log(1) is -0, should be +0, in FE_DOWNWARD mode
           Product: glibc
           Version: 2.19
            Status: NEW
          Severity: normal
          Priority: P2
         Component: math
          Assignee: unassigned at sourceware dot org
          Reporter: jsm28 at gcc dot gnu.org

According to ISO C, log(1) should be +0 in all rounding modes.  However, for
double on x86_64 at least (probably various other formats and architectures),
in FE_DOWNWARD mode it wrongly returns -0.

(This is likely to be a cause of tests of several other functions producing
wrongly signed zero results when enabled in all rounding modes, although I
haven't listed exactly which functions are affected through calling log.)

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


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]