🚀 go-pugleaf

In article <OZ.93Jan6152628@ursa.sis.yorku.ca> oz@ursa.sis.yorku.ca (Ozan Yigit) writes:
>Michael R Castle writes:
>
>   Writing a compiler is a daunting project (ask the FSF!).  Writing a portable
>   compiler, even more so.  
>
>There is LCC from princeton. The backends [that optimize as well
>or better as GCC] are not available except by special educational
>arrangement, but one can [for example] adapt arizona PO as a
>backend...

The question was about public domain compilers; as your posting makes
clear, LCC isn't even freely distributable.  As for your claim that
LCC optimizes as well as or better than GCC, this may be true, but
it would be interesting to see actual benchmarks (e.g. SPECmark
values) to back up such claims.

And take a look at this:
>    Production code generators for the VAX, MIPS, and Motorola 68020 are
>    available for research use to universities willing to execute a license
>    agreement. Contact Chris Fraser (cwf@research.att.com) for details.
>    These code generators are automatically generated from compact
>    specifications, but neither the specifications nor their processor is
>    available, only the generated code generators.

Even if a university executes a license, the most interesting part of
the compiler for those who would port to new platforms is not available.



--
Joe Buck	jbuck@ohm.berkeley.edu