> But I guess all of this can be encapsulated so it doesn't matter.
> windows already have a native spawn function if I remember right.
Yes. AFAIK that's actually the preferred way of starting a process on
MsWin. Fork+exec in one go.
> also, the above could also be achieved using posix_spawn() on platforms
> having it (or vfork if we want to mess with the gory details, I prefer
> not).
>
> The async aspect is mostly unrelated to implementation I would say.
>
> so +0.5 actually...
Unfortunately it's implementation-dependent. For all we know
posix_spawn may be a library convenience wrapper around fork/exec..
(vfork if we're lucky).
-- /kinkieReceived on Mon Feb 22 2010 - 09:20:49 MST
This archive was generated by hypermail 2.2.0 : Tue Feb 23 2010 - 12:00:09 MST