Re: [PATCH] binutils: Run PR binutils/26160 test

["H.J. Lu" <hjl.tools@gmail.com>]

On Mon, Oct 31, 2022 at 6:02 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Mon, Oct 31, 2022 at 5:25 PM Alan Modra <amodra@gmail.com> wrote:
> >
> > On Mon, Oct 31, 2022 at 05:08:04PM -0700, H.J. Lu wrote:
> > > On Mon, Oct 31, 2022 at 4:04 PM Alan Modra <amodra@gmail.com> wrote:
> > > >
> > > > On Mon, Oct 31, 2022 at 09:24:07AM -0700, H.J. Lu wrote:
> > > > > Here is the patch to run the test.  But it failed with
> > > > >
> > > > > readelf: Warning: Cannot fetch indexed address: the .debug_addr
> > > > > section is missing
> > > > >
> > > > > Is this warning valid?
> > > >
> > > > Well, yes, because according to https://gcc.gnu.org/wiki/DebugFission
> > > > .debug_addr is found in the "linked binary".  I think the testcase is
> > > > incomplete without that object file.
> > >
> > > readelf output has been changed.   Here is the patch
> > > to run PR binutils/26160 test.
> >
> > Can we please have the .o file too rather than accepting a warning?
> > That would better test readelf's handling of .dwp files.
>
> The readelf input was generated by dwp which was the part of gold.
> There is no .o file.
>
> > Also, I think it would be nicer if the .dwp file was not compressed,
> > so that the test would run without bzip2 being installed.  It isn't
> > very large.
> >
>
> Since there are other binaries compressed with bzip2, the .dwp
> file doesn't resolve this issue.
>
> --
> H.J.

I am checking it in.


-- 
H.J.