RE: [Axiom-developer] RE: algebra Makefiles with explicitdependencies, bootstrap, fixed-points etc.

[Bill Page]

Tim, Steve,

On Sunday, January 09, 2005 9:41 PM I wrote:
>
> On Sunday, January 09, 2005 5:09 PM Tim wrote:
> > 
> > I'd suggest that the "fixed-point" bootstrap be merged into 
> > axiom--algebra--1 so we can test it first.
> > By test it I mean that we have to compare, line by line, the
> > output of the current input files to see if anything breaks.
> > There are also a set of known broken input files
> > (see src/input/Makefile.pamphlet) and we need to see if this
> > fixes any known problems there.
> 
> That should be pretty easy since we already save the output as
> .output files. I think we just need to cache the first version
> of these files somewhere and then delete them and re-run 'make'
> to create a new set. Then run a diff. I will do this later
> tonight with my current axiom--windows--1 build and let you
> know the result.
> 

I created a new directory called `int/output' and I copied all
the files named `int/input/*.output' to int/output and deleted
them from int/input. These files were created when I did the
initial build of Axiom using the bootstrap files.

Then I ran 'make' again to create a new set of `int/input/*.output'
files. These output files are generated using the last *.o algebra
files that were built during the fixedPoint iteration.

Comparing the two sets of files using the command

  $ diff -au  -I 'Version' -I 'Time' -x '*\.input' int/output int/input

there are no differences.

The extra options -I on diff say to ignore any lines containing
the word 'Version' (since the version of Axiom that generated these
files are always different) and also to ignore any lines containing
'Time' since the timing changed slightly between the two different
versions (the fixedPoint version is apparently very slightly faster),
and finally the option -x says to not compare the *.input files
since I didn't copy them to int/output.

So it seems that the fixedPoint iteration did not break anything
that we previously knew about and did not fix anything that we
knew was already broken. Of course this is not such a strong
test because the current set of input files do not really cover
that much of what Axiom can do. But I think that it shows that
we would not be making a big mistake if we decide to update
the bootstrap lisp chunks in the src/algebra/*.spad.pamphlet
files.

Steve, could you also run this sort of test on your linux
version of the fixedPoint solution?

Regards,
Bill Page.