Re: Test failure

[Robert T. Short]

On 01/29/2012 03:41 AM, Michael Goffioul wrote:
> On Sat, Jan 28, 2012 at 5:25 PM, Ben Abbott<address@hidden>  wrote:
> > On Jan 27, 2012, at 5:51 PM, Robert T. Short wrote:
> >
> > > I just compiled octave (stable branch) from the mercurial sources.  I just 
> > > installed debian testing, so this is a completely fresh compile and test.
> > >
> > > I get the following from the make check step.
> > >
> > >   scripts/general/postpad.m .............................. PASS    9/9
> > >   scripts/general/prepad.m ............................... PASS    9/9
> > >   scripts/general/profile.m ..............................profiler.cc:400: void 
> > > profile_data_accumulator::add_current_time(): Assertion `last_time>= 0.0&&  
> > > last_time<= t' failed.
> > > panic: Aborted -- stopping myself...
> > > make[1]: *** [check] Aborted
> > > make[1]: Leaving directory 
> > > `/home/rtshort/Mirrored/octave/Sources/octave-full/test'
> > > make: *** [check] Error 2
> > >
> > >
> > > My tip is
> > > changeset:   14266:37ca58f9a887
> > > tag:         tip
> > > parent:      14264:284656167c25
> > > user:        Ben Abbott<address@hidden>
> > > date:        Tue Jan 24 20:52:12 2012 -0500
> > > summary:     When root_figure units = normalized, screensize = [0 0 1 1].
> > >
> > > I have attached the configure output.  I assume this is something really 
> > > trivial, but have no idea what.
> > >
> > > Help?
> > >
> > > Bob
> > > <log.configure-full>
> > I also see this on Ubuntu 10.04 (same tip).
> >
> > Ben
> I also see this sometimes (under Linux and Windows). It must be
> related to some timer resolution. Usually, trying to run it a second
> time works fine.
>
> Michael.
It is consistent.  Fails every time.  I tried the development branch 
too, same thing.  It sounds like it isn't something as simple as a 
missing library, so I will file a bug report.  Not urgent for me, i 
stripped out the tests and all the other tests pass.

Bob