This is the mail archive of the
gcc-prs@gcc.gnu.org
mailing list for the GCC project.
bootstrap/6952: gcc 3.1 fails to configure on HP-UX A.09
- From: ralfixx at gmx dot de
- To: gcc-gnats at gcc dot gnu dot org
- Date: 7 Jun 2002 09:45:56 -0000
- Subject: bootstrap/6952: gcc 3.1 fails to configure on HP-UX A.09
- Reply-to: ralfixx at gmx dot de
>Number: 6952
>Category: bootstrap
>Synopsis: gcc 3.1 fails to configure on HP-UX A.09
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: unassigned
>State: open
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Fri Jun 07 02:46:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator: Ralf Fassel
>Release: gcc 3.1
>Organization:
>Environment:
HP-UX venus A.09.05 A 9000/712
gcc 3.1
>Description:
configure fails in gcc-3.1/gcc due to od -c output.
On HP-UX A09, `od -c' lists offset information in decimal by default,
but the configure script expects octal numbers.
This breaks the tests for endianness and floating point format.
I am aware that HP A09 is quite old, but I thought I report it anyway.
>How-To-Repeat:
HP-UX venus A.09.05 A 9000/712
./configure
...
Configuring gcc...
...
checking for CHAR_BIT... yes
checking byte ordering... unknown
configure: error: *** unable to determine endianness
>Fix:
Changing the arguments to od from `od -c' to `od -tc -Ao'.
I don't know whether other systems will break.
*** gcc-3.1/gcc/aclocal.m4~ Tue Jan 22 23:48:45 2002
--- gcc-3.1/gcc/aclocal.m4 Fri Jun 7 11:42:00 2002
***************
*** 935,941 ****
$1
]EOF
if AC_TRY_EVAL(ac_compile); then
! od -c conftest.o |
sed ['s/^[0-7]*[ ]*/ /
s/\*/./g
s/ \\n/*/g
--- 935,941 ----
$1
]EOF
if AC_TRY_EVAL(ac_compile); then
! od -tc -Ao conftest.o |
sed ['s/^[0-7]*[ ]*/ /
s/\*/./g
s/ \\n/*/g
Diff finished at Fri Jun 7 11:42:05
>Release-Note:
>Audit-Trail:
>Unformatted: