Re: [Gcl-devel] Re: Control flow bug involving tagbody, return-from, unwind protect

[Camm Maguire]

Greetings!

I have a new iteration now checked in if you would like to check it
out.  Some heavy work on array type propagation and inlining, a minor
redo of the autodeclaration code, and some general type work to
support sequence indices and long long fixnums.  Passed all misc tests
and 96x2k random tests of size 1k.  

If you go over the comments I've attached, should time permit, you can
get an idea of what I'm thinking regarding future development in this
regard.  Feedback is always most appreciated.

How do you typically use your random-type-prop 'harness' for
accelerated bug detection?

Take care,

"Paul F. Dietz" <address@hidden> writes:

> Camm Maguire wrote:
> 
> > As to your other misc test, it passes in my not-yet-committed tree,
> > but I'll save extracting the patch snippet until I can consolidate
> > everything.  How are we doing on these random-tester issues, btw, vis
> > a vis the 'competition/?
> 
> 
> clisp passes everything.
> 
> sbcl/cmucl have outstanding unfixed bugs.
> 
> acl 7.0 still has a few random tester bugs, and one random type prop
> bug (that weird one involving isqrt and member types.)
> 
> ecl doesn't do well at all.
> 
> I don't know about the current state of lispworks, or any of the
> lisps for PPC.
> 
> abcl was passing all the misc tests when I last checked.
> 
>       Paul
> 
> 
> 
> _______________________________________________
> Gcl-devel mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/gcl-devel
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah