ilp64 documentation

ilp64 documentation

I don't find any documentation on the ilp64 version, currently available only for IA-64. It has come to my attention that some customers expect such a library not only to take 64-bit integer arguments, but to take 64-bit real arguments for the BLAS s..... functions, and 128-bit real arguments for d.... functions. Others expect it to resemble a library provided on certain HPUX systems, where the BLAS d... functions apparently take 64-bit integer and real arguments.

1 post / 0 new
For more complete information about compiler optimizations, see our Optimization Notice.