Re: [cdt-dev] is Win98 a supported platform ?

["Alain Magloire" <alain@xxxxxxx>]

> 
> At 11:52 AM 4/21/2003 -0400, Alain Magloire wrote:
> 
> >         Our Spawner JNI code does not work on Win98 ?
> 
> I don't see 98 as an important platform. You can't get anything Windows 
> except XP from Dell and we work on XP just fine. This is a problem that 
> time will solve because the number of active use Win 98 boxes is going to 
> go down at a nice clip now.
> 
> >This make the CDT non-functionnal on Win98 using the Spawner class.
> >A few solutions:
> >
> >(1) Fall back to java.lang.Process
> 
> 
> My personal opinion is that this is unworkable with the current version of 
> java.lang.Process. Process is just missing too many important features that 
> Spawner provides.
> 
> I still have not packaged them up and submitted the patch, but we have a 
> version of the spawner that allows you to set the priority on process 
> invocation. Given Win32's bad time slicing for greedy PRI_NORMAL tasks, 
> we've found it's absolutely essential to getting reasonable UI 
> responsiveness whenever you are spawning anything of consequence in the 
> background.
> 

Cool, it sounds like a very usefull addition back to the CDT trunk.

On another topic, related to the "can not suspend" PR's and thread.
GDB is simply unreliable to depend on it when trying to drop a SIGINT
to the application, i.e. the behaviour changes depending on the platform,
if option --tty is pass or not etc ..

So we are looking at the solution to enhance Spawner

public native static raise(int pid, int sig);
or
public native static raise(Object pid, int sig);

To interrupt the program, will that be harmfull in your enviroment?
instead of the dropping the SIGINT to gdb?