This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: PATCH: PR ld/4007: Linker failed to issue an error on bad section in segment
On Thu, Mar 22, 2007 at 08:04:46AM -0700, H. J. Lu wrote:
> On Thu, Mar 22, 2007 at 02:51:20PM +0000, Nick Clifton wrote:
> > Hi H.J.
> >
> > >>>ld/testsuite/
> > >>>2007-02-27 H.J. Lu <hongjiu.lu@intel.com>
> > >>>
> > >>> PR ld/4007
> > >>> * ld-i386/alloc.d: New file.
> > >>> * ld-i386/alloc.s: Likewise.
> > >>> * ld-i386/alloc.t: Likewise.
> > >>> * ld-i386/i386.exp: Run "alloc".
> > >>Is there any reason why this new test should be i386 specific rather
> > >>than, say, ELF specific ?
> > >>
> > >>Approved - but you may wish to change its location.
> > >
> > >The linker script is very specific to i386.
> >
> > Is that a necessary constraint for the test ? ie can this problem occur
> > for toolchains other the i386 based ones, (I think that it can), and if
> > so then is there any reason not to make a more generic test that can
> > check the behaviour of non-i386 based toolchains ?
>
> I don't know. i386 is unique due to address wrap around.
The section addresses in the testcase are OK. You just can't put 2
sections, one with address wrap around and one without, into the
same segment.
H.J.