I propose to replace Section 10.5.1 with the following text. I'll
bring some copies to the meeting.
Rajeev
\subsection{Datatypes in Heterogeneous Environments}
In heterogeneous environments, where the file data representation is
internal, external32, or user-defined, care must be taken in
constructing etypes and filetypes. Any of the datatype constructor
functions may be used; however, for those functions that accept
displacements in bytes, the displacements must be specified in terms
of their values in the file for the file data representation being
used. The function \func{MPI\_FILE\_GET\_TYPE\_EXTENT} can be used to
calculate the extents of datatypes in the file.
Datatypes passed as arguments to read/write routines specify the
data layout in memory; therefore, they must always be constructed using
displacements corresponding to displacements in memory.
\begin{users}
File data representations in internal, external32, or user-defined modes
may be different from corresponding data representations in
memory. Therefore, for these file data representations, it is
important not to use hardwired byte offsets for file positioning,
including the initial displacement that specifies the view.
\end{users}