forums.silverfrost.com

Paul,

I think you need to look at this one.

!
! With either the 32 bit or 64 bit complier, with /checkmate the following
! error stops compilation at the line indicated:
!
! error 967 - ...

John,

The problem with using AI is you have to learn to ask the correct questions!

Here's my final attempt at this problem (for now anyway!)

In this case AI was very helpful! Although it's a ...

John,

Thanks for sharing this summary of your observations on this which are very useful.

This is clearly a thought provoking thread.

I have been experimenting with a random pivot. I don't s ...

I think the error reported by the compiler in this case is not correct.

I don't understand how an implied do loop / array constructor can reference an intrinsic function, but not a function defin ...

Lester, These answer both of your questions.

program q1
implicit none
integer :: n
real*8 :: fact
integer :: i
do i = 1, 25
n = i
fact = gamma( mi ...

The following code generates a run time access violation when compiled with the 64 bit compiler.

The 32 bit compiler returns the correct value of 1.d0

program bug_in_product
implicit none
real ...

The code I posted earlier is not portable to other compilers, because of a programmer error!

Code below is, which explicitly deals with negative and zero values of n.

For n .ge. 1 the implied d ...

Lester,

Using the 64 bit compiler, you could do this calculation in one line of fortran.

Use an implied do loop within an array constructor, and then apply the product intrinsic to the array.
...

Please consider adding shiftr and shiftl to the wish list.

I had to resort to adding a ftn95 specific module in order to compile and run a much larger program with ftn95

Thanks

Ken

winap ...

John,

You might want to look at this implementation, which is the fastest I have seen for large data sets. It sorts the input array and also returns index values.

https://github.com/gher-uliege ...

Dan,

Try this. Everything in a single function.

winapp
Integer function LibraryVersion64Win()
use iso_fortran_env , only : compiler_version, compiler_options
implicit none
C_EX ...

So my guess, was way out - again.

Adding (kind=3) to all integer declarations removes the warning.

program test
implicit none
integer(kind=3) iarray(7)
integer(kind=3) ...

Dan,

use iso_fortran_env , only : compiler_version

*should* hide everything else in the module.

program p
use iso_fortran_env , only : compiler_version
implicit none

write(*, ...

Dan,

iso_fortran_env is a "standard" module introduced in Fortran 2008.

COMPILER_VERSION and COMPILER_OPTIONS were added to the module as part of Fortran 2018.

Steve,

I am guessing here, but if QUICKSORTDP_index and QSRT_PARTITIONDP_index are not contained within a module, since you have assumed size arrays (*), then the recursive QUICKSORTDP_index requir ...