I was travelling so did not have access to the source. You are correct
and it is technically a bug which apparently neither -ftrapuv nor
gfortran picked up. I have added
ROKMIX = 0.D0 ; ROKSC = 0.D0 ; ROKVL = 0.D0 ; ROKOLD = 0.D0
KVECVL = 0 ; KVECSC = 0 ; KVCOLD =
Laurence Marks píše v Po 14. 10. 2013 v 08:46 -0500:
I was travelling so did not have access to the source. You are correct
and it is technically a bug which apparently neither -ftrapuv nor
gfortran picked up. I have added
ROKMIX = 0.D0 ; ROKSC = 0.D0 ; ROKVL = 0.D0 ; ROKOLD
Thanks. The line
PM1=min(1.D0,PM1)
should be commented out, it is a legacy of when I moved some code
around and does nothing.
N.B., A=0.D0 is the same as A=(0.D0,0.D0) -- there is an implicit
conversion at least for ifort and gfortran (and I believe always),
e.g. test
complex a
It could be that IBMs xlf compiler does speculative execution. (I seem
to remember, that I read that long time ago), causing these problems.
What we often do is something like
allocate (A(1000))
do i=1,N (where N is less than 1000 !!!)
a(i)=b(i)
...
and later we use
do i=1,N
something =
I'm not very familiar with valgrid, but all those reports seem not
relevant to me.
It seems to complain about all allocated arrays, which are not set to
zero globally. But this does NOT mean that one uses uninitialized
variables or assumes that the compiler sets them to zero.
PS: I don't have
Peter Blaha píše v Čt 10. 10. 2013 v 12:22 +0200:
I'm not very familiar with valgrid, but all those reports seem not
relevant to me.
It seems to complain about all allocated arrays, which are not set to
zero globally. But this does NOT mean that one uses uninitialized
variables or assumes
Sorry, but I agree with Peter I am 99.9% certain that this is not a bug
in the mixer. The way to test this is (with ifort) to use -ftrapuv ; the
arrays are not set in mixer.F but elsewhere within some complicated
subroutines. This flag forces a fault if an undefined variable is used.
Laurence Marks píše v Čt 10. 10. 2013 v 09:40 -0500:
Sorry, but I agree with Peter I am 99.9% certain that this is not a
bug in the mixer. The way to test this is (with ifort) to use
-ftrapuv ;
That is not true. According to ifort docs:
-ftrapuv
The option sets any uninitialized local
Dear WIEN2k list,
so after I found out, that ifort by default initializes all variables to
0, I got quite paranoid, since this could hide bugs and easily lead to
changes in behavior with compilers that doesn't do that. So I fired up
valgrind, to see if there are some other occurrences.
This is
9 matches
Mail list logo
