[prev in list] [next in list] [prev in thread] [next in thread] 

List:       hurd-bug
Subject:    Re: Confusing definitions and declarations of mig_dealloc_reply_port()
From:       Svante Signell <svante.signell () gmail ! com>
Date:       2015-11-06 22:21:00
Message-ID: 1446848460.2910.6.camel () gmail ! com
[Download RAW message or body]

On Fri, 2015-11-06 at 20:56 +0100, Samuel Thibault wrote:
> Svante Signell, on Thu 05 Nov 2015 10:30:27 +0100, wrote:
> > -    __hurd_local_reply_port = __mach_reply_port ();
> > +    port = __mach_reply_port ();
> 
> That's not the same.

See the next message:
port = __hurd_local_reply_port = __mach_reply_port ();

> > How many users are there of GNU/Hurd, especially running old
> > releases?
> 
> All the current users.  If you introduce this change, then all
> existing
> code will stop building with the newer mig, and conversely, the
> modified
> glibc will not build with the old mig.
> 
> Really, this change doesn't deserve such a pain.

So you say it is impossible to do such a change?
I don't believe so, but am not proposing that change since you object
to it so much.

What about: mig first, glibc after that, anything else? And of course
all 8? users would have to upgrade to the latest releases.

[prev in list] [next in list] [prev in thread] [next in thread]