Bug#184048: [m68k] binutils testsuite failures built in a glibc-2.3.1 environment

[Matthias Klose <doko@cs.tu-berlin.de>]

[CC to Andreas]

GOTO Masanori writes:
> At Thu, 7 Aug 2003 08:08:24 +0200,
> Matthias Klose wrote:
> > 
> > [CCing m68k, if the new results look acceptable]
> > 
> > GOTO Masanori writes:
> > > Hi Matthias,
> > > 
> > > At Sun, 9 Mar 2003 08:26:52 +0100,
> > > Matthias Klose wrote:
> > > > Package: libc6-dev
> > > > Version: 2.3.1
> > > > Severity: grave
> > > > 
> > > > Attached is a diff of a binutils built in unstable with gcc-2.95 and
> > > > one built on yesterday's testing (still glibc-2.2.5). Although I
> > > > cannot prove that other build depedencies of binutils are the cause of
> > > > this failures, I start with glibc as the moist obvious one ...
> > > 
> > > I recompiled binutils 2.14.90.0.5-0.2 on m68k with my test built glibc
> > > 2.3.1-1 (2003-07-08 cvs) + gcc 2.95.
> > > 
> > > The result is:
> > > 
> > [...]
> > > 	                === ld Summary ===
> > > 	
> > > 	# of expected passes            172
> > > 	# of unexpected failures        3
> > > 	# of untested testcases         9
> > > 
> > > 
> > > I don't know number of unexpected failures or untested testcases in ld
> > > summary is acceptable or not.  How to act for this bug report?
> > 
> > the current results (included in the binutils m68k package):
> > 
> >                 === ld Summary ===
> > 
> > # of expected passes            173
> > # of unexpected failures        10
> > # of expected failures          1
> > 
> > this looks definitely better than with 2.3.1.
> 
> Exactly.  
> 
> Well, your original report ts-2.13.90.0.18.2-gcc-2.95 has no
> unexpected failures, though...

but fewer testcases as well ...

    FAIL: visibility (hidden_weak) (non PIC, load offset)
    FAIL: visibility (hidden_weak) (PIC main, non PIC so)
    FAIL: visibility (protected_undef_def) (non PIC, load offset)

these three did pass with glibc-2.3.1 and glibc-2.2.3, the second one
did pass with glibc-2.3.1, but not with glibc-2.2.3.

Andreas, do we need to care about these failures?

> > Do you have results with the same configuration, but compiled using
> > current gcc?
> 
> No, I didn't try to build binutils with the current gcc.  I think
> glibc 2.3.2-2 is installed in m68k within a few weeks.