Re: dprintf

[Jeff Johnston <jjohnstn at redhat dot com>]

Shaun Jackman wrote:

> 2005/10/3, Jeff Johnston <jjohnstn@redhat.com>:
>  
>
> > Secondly, it appears you have essentially copied the gnu
> > code you posted earlier which will have a GPL license.
> >    
>
> Sorry, I didn't make myself clear. That was my own implementation of
> the GNU-semantics dprintf, as opposed to the newlib semantics
> __dprintf, which is quite different but unfortunately share very
> similar names.
>
>  
Ok, as long as it was your implementation.

> > First of all, no licensing information.
> >    
>
> Is it acceptable to assign the copyright to the public domain?
>
>  
That's ok.  Just put something in effect that you wrote it and that you 
grant free license to copy modify etc... in all forms.

> > Thirdly, you haven't addressed reentrancy in your patch
> >    
>
> I believe all the called functions (vasprintf, write, free) are
> themselves reentrant. Is this function not reentrant then?
>
>  
I am referring to the fact that there is an _r form of the functions 
that pass the reentrancy struct from function to function.  There is a 
__vasprintf_r and __free_r for example.  These are needed for threading.

> > and you're not using the newlib function define macros
> >    
>
> What is the purpose of the newlib function define macros, and is there
> a newlib hackers reference manual available?
>
>  
They encapsulate the prototype semantics (see _ansi.h).  There is no 
such manual in place.

> > nor have you added any documentation.
> >    
>
> I'll fix this up.
>
>  
Great.  Thanks.

> Cheers,
> Shaun
>