Boost logo

Boost :

Subject: Re: [boost] Heads up - Release branch closed as of tomorrow
From: Marshall Clow (mclow.lists_at_[hidden])
Date: 2013-06-24 15:29:19


On Jun 24, 2013, at 10:46 AM, John Maddock <john_at_[hidden]> wrote:

>>> OK, I realize this is very late, but I've just fixed
>>> https://svn.boost.org/trac/boost/ticket/8711 which has uncovered a bunch
>>> of related and rather insidious bugs in Multiprecision. The fixes
>>> (patches attached) are mostly a collection of typecasts... it's taken a
>>> while to track down all the locations causing problems which is why this
>>> is so late... sorry! *Iff* enough of the tests cycle in the next couple
>>> of days can I please have permission to commit to release?
>>>
>>> Completely understand if not.
>>
>> Are you asking to merge this changeset:
>> https://svn.boost.org/trac/boost/changeset/84897
>>
>> ... or are there more changes coming? If that's the only change, and the
>> tests cycle and everything looks good, then yes, go ahead.
>
> Yes that's the change - it'll depend how quickly the tests cycle really I guess.

I ran the tests locally for my configurations.

test_arithmetic_cpp_dec_float_2 and test_arithmetic_cpp_dec_float_3 failed on all the C++11 testers.
test_rational_io_cpp_int failed on gcc 4.2.1
test_float_io_cpp_dec_float failed on the released clang in C++11 mode (but passed on the "tip of tree" clang).

All of these also failed in this morning's test run using revision 84896.

So, at least in my tests, no new failures.

-- Marshall

Marshall Clow Idio Software <mailto:mclow.lists_at_[hidden]>

A.D. 1517: Martin Luther nails his 95 Theses to the church door and is promptly moderated down to (-1, Flamebait).
        -- Yu Suzuki


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk