[m-rev.] For review: --inline-linear-tail-rec-sccs

[Zoltan Somogyi]

For review by anyone.

This is something we discussed earlier this year.

I am particularly interested in whether we should allow
a procedure in the SCC to have *more* than one tail recursive
call (to the same callee) despite the code size implications.
(This would obviously require a new option name, and updated
documentation.)

Zoltan.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Log.linear_rec
Type: application/octet-stream
Size: 4318 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20170316/29f5c7ea/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DIFF.linear_rec
Type: application/octet-stream
Size: 142860 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20170316/29f5c7ea/attachment-0003.obj>