PATCH RFC: Remove fork from collect2

[Dale Johannesen]

On Mar 19, 2004, at 2:34 PM, Ian Lance Taylor wrote:
> Is there some command line
> option we could use to disable that, to generate code which could be
> linked by either Apple gcc or FSF gcc?  Note in particular that there
> is no floating point code between the vfork and the exec, so there is
> no actual need to save the floating point state.

No.  I thought there was a switch to generate FP save/restore inline,
but there isn't.

> Also note that
> generating calls to functions which are in the user's namespace would
> appear to be a bad thing.

They are not in the user's namespace; external names have a prepended
underscore on Darwin, these names do not.

> 3) Add restFP and saveFP to FSF gcc.  Overall consistency would appear
>    to be a good thing.

It would.  (These routines predate anybody now working at Apple, and
don't "belong" to anybody.)

However, as I said, I don't think that's a general solution to the 
problem.
Indeed, any solution that makes use of the fact that Apple's system
compiler happens to be gcc is not the right approach IMO.

What's wrong with building libiberty on each pass through the bootstrap?
That would solve the general problem, and that library builds very 
quickly.