Hi Sean,

This is on our to-do list (the fix is slated for the next release) as we have 
not yet incorporated any of the intrinsics from the Fortran 2008 standard.

Scot


On May 8, 2014, at 3:10 PM, Sean Byland <[email protected]<mailto:[email protected]>> 
wrote:

When building HDF5/1.8.13  with fortran2003 and the Cray compiler I got a build 
failure on tH5P_F03.f90 because of inconsistant implications/usage of SIZEOF. 
To get it to build I changed:

--- tH5P_F03.f90        2014-05-08 14:19:38.000000000 -0500
+++ tH5P_F03.f90.new    2014-04-29 14:16:46.000000000 -0500
@@ -137,7 +137,7 @@

   ! /* Create a compound datatype */

-  CALL h5tcreate_f(H5T_COMPOUND_F, INT(SIZEOF(fill_ctype),size_t), 
comp_type_id, error)
+  CALL h5tcreate_f(H5T_COMPOUND_F, INT((storage_size(fill_ctype)/8),size_t), 
comp_type_id, error)
   CALL check("h5tcreate_f", error, total_error)
   h5off = H5OFFSETOF(C_LOC(fill_ctype), C_LOC(fill_ctype%a))
   CALL h5tinsert_f(comp_type_id, "a", h5off , H5T_NATIVE_REAL, error)

For better compatibility I recommend using Fortran intrinsics like 
"storage_size" and then scaling for bytes/arrays or using C_SIZEOF from the 
ISO_C_BINDING module.


Sean

_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]<mailto:[email protected]>
http://mail.lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org

_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]
http://mail.lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org

Reply via email to