[m-rev.] for review: Use uints to represent bit vectors.

[Peter Wang]

On Sun, 15 Apr 2018 08:14:26 -0400 (EDT), Julien Fischer <jfischer at opturion.com> wrote:
> 
> Hi Peter,
> 
> On Sun, 15 Apr 2018, Julien Fischer wrote:
> 
> > On Sun, 15 Apr 2018, Peter Wang wrote:
> >
> >>  library/fat_sparse_bitset.m
> >>  library/sparse_bitset.m:
> >>  library/tree_bitset.m:
> >>     As above.
> >
> > Include the above (i.e. the mail subject) in the log message, othewise
> > the log will look weird.
> >
> >>  ---
> >>  library/fat_sparse_bitset.m | 109
> >>  +++++++++++++++++++++++---------------------
> >>  library/sparse_bitset.m     | 109
> >>  +++++++++++++++++++++++---------------------
> >>  library/tree_bitset.m       |  88 ++++++++++++++++++-----------------
> >>  3 files changed, 157 insertions(+), 149 deletions(-)
> >
> > That's fine otherwise.
> 
> Actually, not so fine in practice as it seems to have broken building
> stage 2; we now get a sequence of errors of the form:
> 

Damn, I had bootchecked in hlc.gc but not asm_fast.gc.
The change to output_uint_expr fixes the problem. Thanks.

Peter