This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: Some benchmarks with Apple's -fast

From: Bradley Lucier <lucier at math dot purdue dot edu>
To: David Edelsohn <dje at watson dot ibm dot com>
Cc: Bradley Lucier <lucier at math dot purdue dot edu>, gcc at gcc dot gnu dot org, nlucier at math dot purdue dot edu, feeley at iro dot umontreal dot ca
Date: Thu, 25 Sep 2003 12:21:50 -0500
Subject: Re: Some benchmarks with Apple's -fast

On Thursday, September 25, 2003, at 12:00 AM, David Edelsohn wrote:

	Are you sure that you need -fno-strict-aliasing, as opposed to the
new Apple -frelax-aliasing option?

No. I looked at Apple's code in gcc/alias.c and don't understand it. And I don't understand Mike Stump's explanation here:

On Tuesday, September 16, 2003, at 12:34 PM, Gabriel Dos Reis wrote:
  -fstrict-aliasing
  -frelax-aliasing
what are the relations between the two, and -fno-strict-aliasing?
Incestuous. strict is at one end of a continuum, no-strict is at the other. And in-between, where bad code and benchmark code lies, is where relaxed fits in. It is the mind reader that knows what they meant, and does the right think anyway, even though the code is wrong.

Also, have you tried profile-feedback?

No.

Brad

References:
- Re: Some benchmarks with Apple's -fast
  - From: David Edelsohn

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]