Hi folks,
> Paul G. wrote:
> > Hi folks,
> >
> > Here is something I came across when I began my port of Python for
> > Msys.
> >
> >
> >
> >>Cygwin: With recent (relative to the time of writing, 2001-12-19)
> >> Cygwin installations, there are problems with the
> >> interaction of dynamic linking and fork().
> >
> >
> > ... there is much more, but most of it is in terms of suggested
> > "work arounds" and is, for the most part, not necessarily applicable
> > here...
> >
> > Is it possible that the problem may yet exist under Msys? Has
> > anyone ran in to problems with the interaction of dynamic linking
> > and fork()?
> >
> > For my part, I've been using a lot of .dlls for the various ports I
> > am working on, and have not seen anything that might indicate that
> > the same problems exist under Msys.
> >
>
> Assuming you're not using msysDVLPR then it's not a problem because
> there is no fork().
Well, I am using (have installed) the MsysDTK. (/usr/bin or simply /bin) is being searched
for binutis. /usr/Mingw /bin is lower on the search list (path) than is Msys/1.0/bin (/usr/bin).
That would indicate to me that I am in fact using the MsysDTK binutils.
In fact, if an Msys binutil does not exist(/usr/bin), then my path assures me that I will use
the binutil (if it exists) from the /usr/mingw/bin instead. Otherwise, the search for a given
binutil will fail.
MsysDvlpr is installed. I typically do not invoke the msysdvlpr shell script until later in the
porting process, however, as my pattern is to first test/port using Mingw runtime:
MINGW32_NT-4.0 TALIESIN 1.0.8(0.46/3/2) 2002-12-04 18:14 i686 unknown
(if no problems...then...)
test/port for Msys runtime.
As to the reply, Earnie, thanks. I had kind of thought that was the case, but am still
wanting to get some verificatioin, and you have provided that, at least I think you have...
Thanks,
Paul G.
|
