2011/8/29 Marc-Andre Lafortune <ruby-core-mailing-list / marc-andre.ca>:
> On Sat, Aug 27, 2011 at 4:32 AM, NARUSE, Yui <naruse / airemix.jp> wrote:
>> (2011/08/27 4:40), Marc-Andre Lafortune wrote:
>>> On Thu, Aug 25, 2011 at 9:46 PM, Yui NARUSE <naruse / airemix.jp> wrote:
>>>> r33061 tried to fix this but the result isn't changed.
>>>
>>> Sorry, I do not understand. r33061 fixes Integer#round, which had a
>>> completely different issue from Float#round (see redmine #5228)
>>
>> ./ruby -ve'p 42.0.round(308)'
>> ruby 1.9.4dev (2011-08-26 trunk 33073) [x86_64-freebsd8.2]
>> Infinity
>
> How is this in any way related to r33061?

It happens after r33061.
You can know after r33066, RubySpec F is increased.
20110825T020101Z ruby 1.9.4dev (2011-08-25 trunk 33066)
[x86_64-freebsd8.2] (fbsd) 1851W rubyspec:3F0E
(diff:src,btest,make,install-nodoc,test-all,rubyspec)
http://59.106.172.211/~chkbuild/ruby-trunk/summary.html

It suspects a commit between r33047 and r33066 breaks it.
...but in this case, it is because you add a test to RubySpec, sorry.

>> Adding tests to RubySpec is not enough.
>> You should also add tests to CRuby's test-all.
>
> I respectfully disagree.

RubySpec is not CRuby's implementation test, it is a different thing.
You must add a test to test-all when you change CRuby's behavior.

>>> Can someone either give me access or update it?
>>
>> Added.
>
> Thanks, but somehow I still can't edit the page
> http://redmine.ruby-lang.org/projects/redmine/wiki/VersionControlSystem

I added you a edit bit; did you unlocked the page?
(I don't know if you unlock it then can edit but...)

--=20
NARUSE, Yui =A0<naruse / airemix.jp>