From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 5433 invoked by alias); 1 Aug 2003 11:37:23 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 5426 invoked from network); 1 Aug 2003 11:37:23 -0000 Received: from unknown (HELO air.cs.bath.ac.uk) (138.38.108.3) by sources.redhat.com with SMTP; 1 Aug 2003 11:37:23 -0000 Received: from fire.cs.bath.ac.uk ([138.38.108.1] helo=ralph.cs.bath.ac.uk) by air.cs.bath.ac.uk with esmtp (Exim 3.16 #3) id 19iYDq-00082g-00 for gcc@gcc.gnu.org; Fri, 01 Aug 2003 12:37:22 +0100 Subject: Re: GCC From: Martin To: gcc@gcc.gnu.org In-Reply-To: <3F2A4C90.7020609@cyclicode.net> References: <1059727535.32270.507.camel@ralph.cs.bath.ac.uk> <3F2A4C90.7020609@cyclicode.net> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Fri, 01 Aug 2003 14:07:00 -0000 Message-Id: <1059737842.32270.628.camel@ralph.cs.bath.ac.uk> Mime-Version: 1.0 X-SW-Source: 2003-08/txt/msg00022.txt.bz2 > > It strikes me this is part of a larger problem. It seems there is (in > > many companies / institutions there is a large divide between the people > > who design processors and the people who write compilers. > Well it's going a little off-topic, [apologies to list (admins) if this thread is considered OT] > but this doesn't always happen. Oh no. I was just saying that it seems to me that it happens too often - there are many noteable examples where it hasn't and they've all been technically very strong architectures (and easy ones to write compilers for). It seems there is a real advantage to doing things like this - I was just pondering why it doesn't happen more often. > There will always be situations where for very good reasons specific > instructions will be added to an ISA but that may be hard to utilize > well within gcc. There may well be very good reasons why these > instructions make sense though and arguing that a more pure-RISC > strategy (one that makes it easy to target with compilers) will solve > all of the world's problems isn't the right answer either ;-) Far from it - I'm not saying that people should only design processors to make compiling for them as simple as possible. It just strikes me that with a little more communication between the developer communities better results could be achieved. Just my opinion and perhaps this kind of thing happens a lot more than I though and I am horribly mistaken, > FWIW though I'd note that I've generally found that the most bizarre > instructions do not originate with the processor designers but from > operating system and application developers who demand magic > instructions to save cycles in critical places. These are usually also > the same people who complain when the compiler can't use their magic > instructions very well :-) :-D I can't say I'm surprised by this - another developer community who should probabily work more with compiler and processor design teams. Cheers, - Martin