This is the mail archive of the gcc-bugs@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]

Re: Very nasty C bug in gcc 2.95.2

To: ken at gordian dot com
Subject: Re: Very nasty C bug in gcc 2.95.2
From: Geoff Keating <geoffk at geoffk dot org>
Date: Thu, 15 Mar 2001 00:25:39 -0800
CC: gcc-bugs at gcc dot gnu dot org
References: <3AB014C0.D4074553@gordian.com> <jmsnkfvhxn.fsf@geoffk.org> <3AB06D7F.6DACBEC3@gordian.com>
Reply-to: Geoff Keating <geoffk at redhat dot com>

> Date: Wed, 14 Mar 2001 23:21:35 -0800
> From: Ken Settle <ken@gordian.com>

> Apparently, if the offset is >= 1 && <= the stack alignment boundry, it
> uses ori, and it uses addi otherwise, on all 3 cpus mentioned (ppc, arm,
> mips).  I'm confused about why anyone would prefer an ori in this
> circumstance, rather than an addi.

It's because logical operations are easier to simplify than arithmetic
ones, so GCC tries to convert arithmetic operations to logical ones.

-- 
- Geoffrey Keating <geoffk@geoffk.org>

References:
- Very nasty C bug in gcc 2.95.2
  - From: Ken Settle
- Re: Very nasty C bug in gcc 2.95.2
  - From: Geoff Keating
- Re: Very nasty C bug in gcc 2.95.2
  - From: Ken Settle

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