On 09/18/14 16:20, Iain Sandoe wrote:
1. There has been a change made "to make the upper path like the lower path"
(as you said on IRC).
- apparently (from our conversation) you don't expect this to be a general
optimisation improvement.
- but it doesn't seem to be either "obvious" or "a cleanup" to me
- if you are asserting that it is a cleanup, then some explanation is in
order.
Seems reasonable.
2. Apparently you don't think it is necessary to have any testcase to
demonstrate that the new code is working?
- perhaps you are asserting that the code is "correct by inspection"?
A testcase would be nice to add and I'd strongly prefer to have one in
the suite -- even if it's Darwin specific.
My understanding is the problem is Darwin's linker (or dynamic loader?)
can't handle the code we end up generating. While there may be other
systems where one could show the problem due to limitations of the
linker/loader on thos systems, probably the most common will be Darwin.
So we probably need a Darwin specific test.
However, I don't think Kai has a Darwin box to do the necessary testing.
If you, or someone, could build a test and verify it fails without
Kai's patch, then passes with Kai's patch, that'd be quite helpful.
3. You don't seem to think it necessary to amend the comments in the code to
reflect the new functionality?
Kai, can you update the comments, please?
Jeff
