Re: [boost] Boost regression notification (2005-10-06 [RC_1_33_0])

From: Martin Wille (mw8329_at_[hidden])
Date: 2005-10-06 14:30:14

John Maddock wrote:

> We really need to get 1.33.1 out the door, but this list of regressions 
> seems to be getting worse not better.  Is there anything that can be done to 
> improve things?  Any library authors stuck for ideas and need assistance?

(This message basically is a repost of a problem I already had reported 
on the testing list.)

We have an additional problem that does not become visible in the 
reports Doug sends to this list. For gcc 4.0.2, there are a lot of 
additional regressions. They look like a compiler problem to me, but I 
didn't seriously investigate the problem.

If someone would be able to track the problems down, be to a bug in 
Boost or in the compiler, then this would definitely help. (The gcc 
4.0.1 toolset is relevant for the release. I think we should not ship
Boost 1.33.1 without being able to say what exactly goes wrong with gcc 
4.0.2.)

The complete list of test results (including the toolsets not considered 
  relevant for the release) can be found at http://tinyurl.com/9s72x [1].

Mainly affected by the new problems are: IOstreams, Iterator and Spirit.
(Jonathan already tried to fix IOstreams, but without luck).

Regards,
m

[1] 
http://engineering.meta-comm.com/boost-regression/CVS-RC_1_33_0/developer/summary.html
Send instant messages to your online friends http://au.messenger.yahoo.com