On Tue, Apr 15, 2014 at 01:03:42PM +0100, Jonathan Wakely wrote:
> On 15 April 2014 12:45, Douglas B Rupp wrote:
> > No I considered that but I think that number will be very small. Will you
> > concede, in hindsight, that it would be better had the name been chosen to
> > be more unique?
>
> No argument from me there, but the same applies to VxWorks, who have
> now chosen the same not-very-distinctive name, but it's even worse in
> their case as that name has been in use by GCC for many years.
>
> This seems rather foolish of VxWorks, but knowing that doesn't solve anything.
>
> Could we install the file as unwind-gcc.h and conditionally install
> another file called unwind.h with the content below if a configure
> test including <unwind.h> fails:
>
> #warning "This header file is deprecated, use unwind-gcc.h instead"
> #include_next "unwind-gcc.h"
Given that unwind.h is a header installed by several compilers, I'd say
even the deprecation warning is highly undesirable.
Jakub
