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: problem building GCC on PowerPC wiht AltiVec enabled

From: Daniel Egger <degger at fhm dot edu>
To: Franz Sirl <Franz dot Sirl-kernel at lauterbach dot com>
Cc: Alan Modra <amodra at bigpond dot net dot au>,Ilan Finci <ifinci at myrealbox dot com>, gcc at gnu dot org
Date: 23 Jul 2002 14:19:55 +0200
Subject: Re: problem building GCC on PowerPC wiht AltiVec enabled
References: <3D3D1DCC.4010601@myrealbox.com><3D3D1DCC.4010601@myrealbox.com> <5.1.1.5.2.20020723115139.04e5e008@mail.lauterbach.com>

Am Die, 2002-07-23 um 11.59 schrieb Franz Sirl:

> BTW, what I'm still unsure about is the binary compatibility between code 
> compiled with -mno-altivec-abi and -maltivec-abi. Do the different values 
> for STACK_BOUNDARY and ABI_STACK_BOUNDARY create a compatibility problem?

It's only incompatible when using vector varargs as far as I can tell.
Normal (i.e. local vector) usage shows only one difference: The setup
of the vector safe register; this one is useless on Linux anyway since 
all vector registers are safed on a context switch.
 
-- 
Servus,
       Daniel

Attachment: signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil

References:
- problem building GCC on PowerPC wiht AltiVec enabled
  - From: Ilan Finci
- Re: problem building GCC on PowerPC wiht AltiVec enabled
  - From: Franz Sirl

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