This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
target/2725: Trouble using -mipsX without -mabi on IRIX
- To: gcc-gnats at gcc dot gnu dot org
- Subject: target/2725: Trouble using -mipsX without -mabi on IRIX
- From: reichelt at igpm dot rwth-aachen dot de
- Date: 3 May 2001 10:01:24 -0000
- Reply-To: reichelt at igpm dot rwth-aachen dot de
>Number: 2725
>Category: target
>Synopsis: Trouble using -mipsX without -mabi on IRIX
>Confidential: no
>Severity: non-critical
>Priority: medium
>Responsible: unassigned
>State: open
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Thu May 03 03:06:00 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator: Volker Reichelt
>Release: gcc 2.95.3, gcc 3.0 20010430 (prerelease)
>Organization:
>Environment:
IRIX 6.5
>Description:
The file consisting of a single line
#include <new>
won't compile under IRIX 6.5, if I try
to compile it using "-mips3" or "-mips4"
without explicitly specifying "-mabi=...".
Instead I get the error message
.../include/g++-v3/new:55: `operator new'
takes type `size_t' (`long unsigned int') as first parameter
.../include/g++-v3/new:56: `operator new'
takes type `size_t' (`long unsigned int') as first parameter
etc. According to the documentation mabi should
default to 64 when using "-mips3" or "-mips4".
This seems to be the case for the compiler.
The preprocessor however gets the symbols of the mabi n32.
>How-To-Repeat:
g++ -mips4 -c MabiBug.cpp (for the error message)
g++ -mips4 -dM -E -c MabiBug.cpp (for the wrong symbols)
g++ -mips4 -mabi=64 -dM -E -c MabiBug.cpp (for the correct symbols))
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted: