[m-rev.] diff: disable mfilterjavac by default

[Paul Bone]

On Tue, Oct 15, 2013 at 04:41:03PM +1100, Julien Fischer wrote:
> On Tue, Oct 15, 2013 at 4:39 PM, Paul Bone <paul at bone.id.au> wrote:
> 
> > On Tue, Apr 23, 2013 at 05:41:43PM +1000, Julien Fischer wrote:
> > > Disable mfilterjavac for now.
> > >
> > > mfilterjavac is currently causing problems when installing the java grade
> > > from
> > > the stage 2 dir.  It can be re-enabled when the following sequence of
> > > actions
> > > works correctly:
> > >
> > > (1) Build stage 1 using rotd-2013-04-18
> > > (2) Bootcheck (or otherwise build the stage 2 compiler).
> > > (3) cd stage2 && mmake install (with the java grade as one of the grades
> > >     to install).
> >
> > Hi Julien.
> >
> > I just tried these steps, except with ROTD 2013-10-01 and it works for me.
> > Perhaps because the ROTD understands the --mfilterjavac option.  Is there
> > something that I'm missing?
> >
> 
> In terms of my original point, yes, rotd-2013-10-01 has mfilterjavac
> disabled
> by default -- you need an rotd that would attempt to use it by default.
> 
> Also my original step should be:
> 
>    cd stage2 && make install
> 
> (i.e. not mmake).  Note that for none of this do I have a Mercury
> installation
> in my PATH.
> 
> Since I wrote the above, I wonder whether the above was intended to work --
> at the moment it does work, with the exception of for the java grade with
> mfilterjavac
> enabled.  In particular, the compiler has no trouble finding mfiltercc at
> this point, which
> leads me to suspect your original change that introduced mfilterjavac left
> something
> out of a M(m)akefile somewhere.
> 
> I'm using a workspace based on the current master with mfilterjavac
> > re-enabled.
> 
> 
> I think you have misunderstood; it is rote-2013-04-18 I am bootchecking
> above,
> not a separate workspace.

Yes, I misunderstood.  However rotd-2013-04-18 does not contain the
mfilterjavac change.  So i don't know why it would be broken.  The next
ROTD, 2013-04-29, contains your change disabling mfilterjavac, so it
probably isn't broken.  Is there an ROTD that's already built that I can
test or do I have to generate one?


-- 
Paul Bone
http://www.bone.id.au