[LLVMdev] Compiler Driver Decisions

[Reid Spencer]

On Mon, 2004-08-02 at 17:31, Chris Lattner wrote:
> On Mon, 2 Aug 2004, Misha Brukman wrote:
> 
> > On Mon, Aug 02, 2004 at 07:23:06PM -0500, Chris Lattner wrote:
> > > > > > 1. Name = llvmcc
> > > > >
> > > > > Why not 'llvmc' "llvm compiler"?  What does the extra C mean?
> > > >
> > > > I dunno. Perhaps cause Misha liked it. But, you do have a point there.
> >
> > LLVMCC = LLVM Compiler Collection, a la GCC
> > After all, it's going to be the "driver", like GCC, and unify
> > front-ends, so I should be able to do:
> >
> > % llvmcc a.java -o a.o
> > % llvmcc b.cpp -o b.o
> >
> > Right?
> 
> Absolutely.  The problem is that "C compiler" is what people think of when
> they see CC.  This we certainly are not.  If we are really a compiler of
> code, why not just call it llvmc?  Also, just because GCC set a precedent
> here does not mean that it needs to be followed.  Their renaming to
> compiler collection is largely due to historical reasons.
> 
I actually like Misha's point here. Most people that have used GCC
recently realize that the CC means "Compiler Collection" and not "C
Compiler" which is appropriate given what it does. Since we intend to be
front end language agnostic and the driver tool will support multiple
front end languages, "Compiler Collection" is appropriate for LLVM too.

I agree that llvm is overloaded and should be avoided. So its either
llvmc or llvmcc. My vote is for the latter.

Reid.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20040802/4424f4bc/attachment.sig>