This is the mail archive of the gcc-patches@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: Patch for __float80 varargs passing on ia64-hpux

From: Mark Mitchell <mark at codesourcery dot com>
To: "Joseph S. Myers" <joseph at codesourcery dot com>
Cc: gcc-patches at gcc dot gnu dot org, wilson at specifixinc dot com, Richard Henderson <rth at redhat dot com>
Date: Fri, 17 Jun 2005 20:49:13 -0700
Subject: Re: Patch for __float80 varargs passing on ia64-hpux
References: <Pine.LNX.4.61.0506132031070.14594@digraph.polyomino.org.uk>

Joseph S. Myers wrote:

This patch fixes a bug in passing __float80 values in variable
arguments on ia64-hpux: they were not being split into general
registers in the correct layout and order.

Bootstrapped with no regressions on ia64-hp-hpux11.23, and I've tested
that the ABI for passing __float80 values now seems to be internally
consistent and consistent with the HP compiler.  OK to commit?

I was hoping one of the IA64 maintainers would review this patch.

My concern is that on IA64 GNU/Linux, something bad would happen. However, since your patch is conditioned on BYTES_BIG_ENDIAN, that concern probably does not apply; you can build little-endian code on GNU/Linux, but I don't think anybody does. I also looked at the Itanium Software Conventions document, and it seems to support your change.

Therefore, your patch is OK. Of course, if the IA64 maintainers object, please honor their objections.

--
Mark Mitchell
CodeSourcery, LLC
mark@codesourcery.com
(916) 791-8304

Follow-Ups:
- Re: Patch for __float80 varargs passing on ia64-hpux
  - From: James E Wilson

References:
- Patch for __float80 varargs passing on ia64-hpux
  - From: Joseph S. Myers

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