Intel® Fortran Compiler

ifort 14.0 dot_product vectorizing differently with different array subscripts

Greetings,

I was looking at the vectorization report for a code and I found different vectorization behavior for dot_product() based on the array subscripts. The following statements are identical except for the lower bound of the array indexes (in the first it is the variable "n1", in the second it is the constant "1"):

 (75)   disap = disap + dot_product(ym(n1:nmax2), pMergeBulk(n1:nmax2,n))
    
 (77)   disap = disap + dot_product(ym(1:nmax2), pMergeBulk(1:nmax2,n))

after compiling with vec-report6 I got the following results:

Intel MPI compiler wrapper wrongly strips off quotation marks

This bug pertains to the Intel MPI compiler wrapper (mpiifort, perhaps
also mpiicc) incorrectly stripping off already-quoted quote
characters.

Background:
I recently found that Intel MPI compiler wrapper (compiler suite
version 17.4.196) fails to compile nwchem 5.1.1 properly due to a bug
in the "mpiifort" wrapper. This also affects "mpiicc" for an obvious
reason.

Here is a test case that simulates that problem--no need to get nwchem
version 5.1.1 (file: helloworld.F90)

.text can not be used when mking a shared object

I am working on the project of porting numerous codes from f77 TruUnix to Linux f90 64bit.   Really struggling with some of this.   The contempt28 program uses the shared library and pgplot to create plotting files.   The contempt28 is integrated with using the pgplot with the shared libraries.

I am trying to make this work have stalled getting the shared library built.The build order is pgplot (done), the shlib(shared library) and link to pgplot.  Then build contempt28 and link to shlib and pgplot.   

Any way to TRAP integer overflows ?

This has probably been covered - maybe there is an article somewhere ?

I have a program that makes use of INTEGER arithmetic in quite a few places.

 

I would like to KNOW when an integer operation causes a bad result.

I was wondering if there is a way to automatically find out when this occurs,

other than putting in dozens of IF tests.

For example, if I have m=N*7, if there is an overflow, m and N will have different signs

Problem netCDF library: undefined symbol nf_inq_libvers_

hi, everyone

I´m trying to compile a netCDF library in linux, but it is not working. I don´t know what´s happening, I think is a compilation problem.

Someone can help me to solve this error "./CCTM_V5g_Linux3_x86_64ifort: symbol lookup error: ./CCTM_V5g_Linux3_x86_64ifor            t: undefined symbol: nf_inq_libvers_"

 

Thanks.

 

 

 

Subscribe to Intel® Fortran Compiler