Date: Wed, 8 Jan 1997 08:27:54 +0800 (GMT)
From: Orlando Andico <orly AT gibson DOT eee DOT upd DOT edu DOT ph>
To: Benjamin D Chambers <chambersb AT juno DOT com>
cc: djgpp AT delorie DOT com
Subject: Re: Intel ASM to AT&T ASM question
In-Reply-To: <19970106.200214.4975.2.chambersb@juno.com>
Message-ID: <Pine.SGI.3.93.970108082252.19523D-100000@gibson.eee.upd.edu.ph>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Mon, 6 Jan 1997, Benjamin D Chambers wrote:

[..]
> Of course it is.  I have yet to see a compiler that really stacks up to
> gcc - this doesn't need pointing out :)
[..]

somewhat offtopic, but..  the IRIX 5.3 IDO C compiler (32-bit) running on
an R5000 (32-bit CPU with some 64-bit instructions) is about 10% faster
than GCC; the IRIX 6.3 IDO C compiler (64-bit) running on an R10000 (pure
64-bit machine; 64-bit OS) is about 30% faster than GCC;

i did a dumb test to check the above -- compile the time-dishonored
Dhrystone benchmark. for GCC I used

  gcc -O10 -static -mcpu=r4000 -mips3 ...

(but there's no difference between -O2, -O3, and -O10, gcc-2.7.2) for IRIX
I used

  cc -O2 -mips3 ...

note that the IRIX IDO has a faster mode (-O3) but it's kind of
complicated (you have to deal with these things called ucode object files)
and I don't know how to invoke its magic..
 
.-----------------------------------------------------------------.
| Orlando Andico                email: orly AT gibson DOT eee DOT upd DOT edu DOT ph |
| IRC Lab/EE Dept/UP Diliman   http://gibson.eee.upd.edu.ph/~orly |
|  "through adventure we are not adventuresome" -- 10000 Maniacs  |
`-----------------------------------------------------------------'