Experimental Physics and
Industrial Control System

Ralph Lange <[email protected]> · Fri, 04 Dec 2009 09:15:37 -0500

As procServ uses fork() to spawn its child and a bunch of other

*nix-type system calls, I was assuming it would need the cygwin runtime

library anyway.

(I don't know how native Windows implements these, and - honestly - I

don't really want to know, either.)

So the idea was to build procServ on a cygwin system, distribute the

EXE, and run it on any system that has the cygwin DLL.

As far as I understand, CMake would not solve the fork() et al. issue

and require CMake to be installed for the build instead of cygwin. As I

assume cygwin to be more common on Windows systems than CMake, I don't

see the immediate advantage of a build system switch.

Ralph

On Thu 03 Dec 2009 18:44:07 J. Lewis Muir wrote:

On 12/3/09 4:51 PM, Benjamin Franksen wrote:

On Donnerstag, 3. Dezember 2009, Ralph Lange wrote:

The major milestone at the horizon is getting it to run on Windows
machines, to finally get reasonable remote access and logging for all
these embedded Windows soft IOCs running in Scopes, LabViews systems and
such. So switching to the autotools framework was really just a first

step ...

Hmmm. Autotools need a unix-like (bourne-) shell. On windows this means

you'll have to have some unix emulation installed (cygwin, whatever).

Cheers
Ben

Or maybe the procServ build system could be switched to CMake:

  http://www.cmake.org/

Lewis

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System