RE: [RFC 9/9] Unify windows specifics into common/windows-hdep files

["Pierre Muller" <pierre.muller@ics-cnrs.unistra.fr>]

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Eli Zaretskii
> Envoyé : mercredi 30 mars 2011 21:47
> À : Pierre Muller
> Cc : gdb-patches@sourceware.org
> Objet : Re: [RFC 9/9] Unify windows specifics into common/windows-hdep
files
> 
> > From: "Pierre Muller" <pierre.muller@ics-cnrs.unistra.fr>
> > Date: Wed, 30 Mar 2011 14:05:17 +0200
> >
> >    if (strcasecmp (buf, "ntdll.dll") == 0)
> >      {
> > -      GetSystemDirectory (buf, sizeof (buf));
> > +      GetSystemDirectoryA (buf, sizeof (buf));
> 
> This will be a problem if we ever support a _UNICODE build.

  No because buf always a char array in the code and
thus require the use of GetSystemDirectoryA function
even if Unicode functions are used elsewhere.

> > +      if (windows_conv_path (WINDOWS_POSIX_TO_NATIVE, exec_file,
> real_path,
> > +			     __PMAX * sizeof (win_buf_t)) < 0)
> >  	error (_("Error starting executable: %d"), errno);
> 
> But windows_conv_path never returns a negative value, it returns zero
> if it fails.

  At least for Cygwin, windows_conv_path simply returns cygwinconv_path
result, his is why I didn't change the code arounf it:
I agree that I need to improve the mingw part of that function
according to your comments for RFC 1/9.
 
> Thanks.

Thank you for your comments

Pierre