Re: [Gm2] cvs regression tests

[Gaius Mulley]

Michael Lambert <address@hidden> writes:

> Hi Gaius,
> 
> This strikes me as odd...  On a Core2 Duo Mac I see
> 
> # of expected passes            12465
> # of unexpected failures        1395
> # of unresolved testcases       1185
> 
> at the end of the output of 'make check-gm2' (using gcc 4.1.2 and the
> latest gm2 CVS).
> 
> I don't understand the discrepancy in the number of tests.
> 
> Michael

Hi Michael,

thanks for pointing this out.  Yes I hadn't checked in my copy of the tcl
library file which the dejagnu used.  I've now checked in my copy and
have just got these results from a LP64 GNU/Linux (Athlon 64) system:

# of expected passes            5982
# of unexpected failures        66
# of unresolved testcases       36

I've downloaded the cvs tarball on a i686 system and rebuilt gm2
and obtained these results:

# of expected passes            5904
# of unexpected failures        78
# of unresolved testcases       36

it still looks as if a few tests have not been checked in (or 'cvs
add'ed) nevertheless at least they are closer than before.. can you
confirm dejagnu tries a similar "# of expected passes"?

regards,
Gaius