This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: Results for 3.2 20020710 (experimental) testsuite on mipseb-unknown-netbsd
- From: Jason R Thorpe <thorpej at wasabisystems dot com>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Wed, 17 Jul 2002 16:52:52 -0700
- Subject: Re: Results for 3.2 20020710 (experimental) testsuite on mipseb-unknown-netbsd
- Organization: Wasabi Systems, Inc.
- References: <20020712001920.AE89B7DA4@yeah-baby.shagadelic.org>
On Thu, Jul 11, 2002 at 05:19:20PM -0700, Jason R Thorpe wrote:
> LAST_UPDATED: Wed Jul 10 19:33:28 UTC 2002
>
> Target is mipseb-unknown-netbsd
> Host is i386-unknown-netbsdelf
>
> === g++ tests ===
>
>
> Running target nbsd-r4400
> FAIL: g++.dg/bprob/bprob-1.C compilation, -g -fprofile-arcs
> UNRESOLVED: g++.dg/bprob/bprob-1.C execution, -g -fprofile-arcs
> UNRESOLVED: g++.dg/bprob/bprob-1.C compilation, -g -fbranch-probabilities
> UNRESOLVED: g++.dg/bprob/bprob-1.C execution, -g -fbranch-probabilities
.
.
.
So, these failures are a result of -Dinhibit_libc when libgcc is built.
inhibit_libc is suppressed if --with-headers is specified. However,
--with-headers has some annoying semantics (at least, I find them to
be annoying :-)
--with-headers wants to copy headers from the specified source directory
to $prefix/$target/sys-include. However, in my case, sys-include is already
populated by virtue of being a symbolic link into the NFS root area of a
diskless client system (the same is done with the target system's libs in
$prefix/$target/lib).
Is there a way to suppress inhibit_libc if the sys-include and lib
directories are already populated? Or do I have to bite the bullet
and needlessly waste disk space?
--
-- Jason R. Thorpe <thorpej@wasabisystems.com>