This is the mail archive of the
binutils@sourceware.cygnus.com
mailing list for the binutils project.
Re: Experience building binutils-2.9.5 (CVS) under VAX Ultrix 4.3
- To: ian at zembu dot com (Ian Lance Taylor)
- Subject: Re: Experience building binutils-2.9.5 (CVS) under VAX Ultrix 4.3
- From: "John David Anglin" <dave at hiauly1 dot hia dot nrc dot ca>
- Date: Fri, 31 Mar 2000 00:28:39 -0500 (EST)
- Cc: binutils at sourceware dot cygnus dot com
> Date: Thu, 30 Mar 2000 17:44:28 -0500 (EST)
> From: "John David Anglin" <dave@hiauly1.hia.nrc.ca>
>
> 2) There are a couple of check problems with ld:
>
> Running /usr/local/src/gnu/binutils-2.9.5/ld/testsuite/ld-cdtest/cdtest.exp ...
> FAIL: cdtest
> FAIL: cdtest with -Ur
> ...
>
> --- cdtest.dat Mon May 3 03:29:08 1999
> +++ ../../tmpdir/cdtest.out Wed Mar 29 15:38:22 2000
> @@ -4,6 +4,8 @@
> Constructing Foo(4) "default-foo"
> Initializing Foo(5) "default-foo" with Foo(4)
> Destructing Foo(4) "default-foo" (remaining foos: 4)
> +Initializing Foo(5) "default-foo" with Foo(5)
> +Destructing Foo(5) "default-foo" (remaining foos: 4)
> Constructing Foo(5) "other_foo1"
> Constructing Foo(6) "other_foo2"
> Copying Foo(5) "other_foo1" to Foo(6)
>
> This appears to be a result of using gcc-2.8.1 to build the package.
>
> That's kind of weird, actually. That would seem to be a bug
> somewhere. I can't remember seeing this before. It might indicate a
> real problem in the compiler or the linker. The compiler is quite
> old, of course, but it might still be worth investigating.
The compiler is actually not that old. It was gcc-2.8.1 with CVS patches
through to 990109 (roughly, the end of separate gcc/egcs development).
Dave
--
J. David Anglin dave.anglin@nrc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6605)