-fdump-translation-unit considered harmful

[Robert Dewar]

Pjotr Kourzanov wrote:
> Richard,
> 
>   One of the objections to exporting tree internals in an open-format
> was that it would make life easier for people that what want to write 
> software that circumvents GPL by reusing internals of the GCC.

Nobody knows if this "circumvents" the GPL. I would rather guess that
any program that "reuses internals of GCC" would go beyond mere
aggregation, and would be considered a deriviative work, but who
knows, none of this has been litigated, and we probably prefer
that it not be.

>   Hence my question: would GPL cover an executable that runs in a 
> separate address space, albeit having some sections /readonly pairwise 
> shared/ with GCC's cc1(plus)?

Separate address spaces are totally irrelevant to coypright law, so
that's for sure a red herring.

>   The http://www.gnu.org/licenses/gpl-faq.html gives a definite answer 
> only when static or dynamic linking is used, or when an RPC-like 
> mechanism is used. Here is a dual borderline case to the one that is 
> explained in the FAQ: insead of dynamically linking and executing just 
> main(), one can start a separate process and have fine-grained 
> interactions (as opposed to RPC) initiated from it. Would the situation 
> I described above be still acceptable, given the GCC's definition of the 
> "arms length" communication?

None of these answers are really definitive, they are all just guesses
in an area where there is little precedential guidance.