[m-dev.] --num-reserved-addresses

[Zoltan Somogyi]

On Sun, 4 Feb 2018 15:14:41 +1100 (AEDT), Julien Fischer <jfischer at opturion.com> wrote:
> I have no objections to removing the option.

OK, attached is a diff that does so. (The diff is with -b.)

On my laptop, the compiler is about 0.9 percent faster on speedtest
with this diff than without it, while on oracle1, it is about 1.7% faster.
(Though on both machines there is considerable variation between runs.)

I plan to commit this at the end of monday; anyone who objects
should do so (or ask for more thinking time) before then.

The only part of the capability that this diff deletes that I think
anyone is at all likely to miss is the ability to represent
e.g. the "no" functor of the maybe type with a null pointer
in C# or Java. (This diff does not change anything about
type representations on C.) Then again, I am not sure
whether that capability was ever exercised, and therefore
I am not sure whether it was ever debugged.

Zoltan.




-------------- next part --------------
A non-text attachment was scrubbed...
Name: Log.res
Type: application/octet-stream
Size: 1665 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/developers/attachments/20180204/6895862f/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DIFF.res
Type: application/octet-stream
Size: 111806 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/developers/attachments/20180204/6895862f/attachment-0003.obj>