Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 30 Jan 2009 12:13:52 +1100
From:      Andrew Reilly <andrew-freebsd@areilly.bpc-users.org>
To:        Garance A Drosihn <drosih@rpi.edu>
Cc:        Christoph Mallon <christoph.mallon@gmx.de>, freebsd-current@freebsd.org
Subject:   Re: Alternatives to gcc (was Re: gcc 4.3: when will it becomestandard compiler?)
Message-ID:  <20090130011352.GA85741@duncan.reilly.home>
In-Reply-To: <p06240800c5a7c06bb469@[128.113.24.47]>
References:  <20090128155340.GA75143@lpthe.jussieu.fr> <7508A5B5-C6D2-498A-AEA1-D84E85F1D743@mac.com> <200901291243.00378.doconnor@gsoft.com.au> <49811242.7030106@delphij.net> <498116AB.2060105@gmail.com> <49816854.9060909@gmx.de> <p06240800c5a7c06bb469@[128.113.24.47]>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Jan 29, 2009 at 03:26:29PM -0500, Garance A Drosihn wrote:
> At 9:27 AM +0100 1/29/09, Christoph Mallon wrote:
> >Eitan Adler schrieb:
> >>To quote from the FAQ:
> >>Who does this change affect?
> >>    Nobody who is currently using GCC should be affected by this change.
> >
> >Let me repeat that:
> >"Nobody who is currently using GCC *should* be affected by this change."
> >
> >Emphasize mine.
> 
> There have been a couple of different issues brought up in this thread,
> and I've lost track of which one we're on.  But one of the issues is
> the use of LLVM+gcc, and my guess is that the above statement would not
> apply to that.  That is just my guess, of course.

It seems to me that LLVM+gcc is precicely the condition that the
GCC maintainers are attempting to prevent with the GPLv3 version
of the libgcc exception.  It almost certainly doesn't constitute
an "Eligible Compilation Process".  They've historically refused
to break out a useful intermediate (parse tree or RTL) form
explicitly so that gcc couldn't be used as a convenient front
end for a proprietary code generator.  But, on the other hand,
they didn't seem to mind adding the pico-java back-end, which
is much the same idea, so perhaps this speculation is totally
wrong.  In that case I have no idea what use case they're so
keen to prevent.

Cheers,

Andrew




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090130011352.GA85741>