On Wed, Oct 16, 2013 at 4:03 PM, Liam Healy <l...@healy.washington.dc.us> wrote: > Indeed, I first approached it this way. But the length needs to be > calculated somewhere, and making it here has the benefit of preserving > the arglist of translate-into-foreign-memory which is parallel to > translate-to-foreign by design. If you think this works, I will commit > it.
Looking at the actual problem we're trying to solve, I think we want to specialise expand-to-foreign-dyn-indirect rather than translate-into-foreign-memory. Does that make sense? -- Luís Oliveira http://kerno.org/~luis/
