[Bug fortran/46842] [4.6 Regression] wrong results with MATHMUL(..., TRANSPOSE (func ())) -- 465.tonto test run miscompares

burnus at gcc dot gnu.org gcc-bugzilla@gcc.gnu.org
Fri Dec 10 07:26:00 GMT 2010


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46842

--- Comment #17 from Tobias Burnus <burnus at gcc dot gnu.org> 2010-12-10 07:26:11 UTC ---
With the function call - there is no transpose at all:

    // Temporary descriptor
    atmp.10.dim[0].stride = 1;  // Normal bounds: OK
    atmp.10.dim[0].lbound = 0;
    atmp.10.dim[0].ubound = 1;
    atmp.10.dim[1].stride = 2;
    atmp.10.dim[1].lbound = 0;
    atmp.10.dim[1].ubound = 1;
    atmp.10.data = (void * restrict) &A.11;
    atmp.10.offset = 0;
    func (&atmp.10); // Function call - so far OK.
    D.1585 = &atmp.10; // Before this line a TRANSPOSE is missing
    [...]
    _gfortran_matmul_r4 (&atmp.12, D.1580, D.1585, 0, 0, 0B);



More information about the Gcc-bugs mailing list