On 11/12/2010 11:24 AM, Daniel James wrote:
On 12 November 2010 15:46, Eric Niebler <eric@boostpro.com> wrote:
On 11/12/2010 3:45 AM, Daniel James wrote:
On 12 November 2010 04:21, Marshall Clow <mclow.lists@gmail.com> wrote:
Interesting that that test is failing (on Windows) for mingw-gcc 4.4.5, but passing for mingw-gcc 4.4.3 and mingw-gcc 4.5.1
That might be because the mingw-gcc 4.4.5 tester has a config issue.
... which John Maddock closed as invalid. I don't know what to make of this.
He closed it because it looked like the issue was with the tester's configuration, not with Boost.Config. Looking at the latest test results it seems to have been fixed (at least the failures in my libraries have been), so if the relevant test is still failing it's probably something else.
Ah. We do NOT have a repro of this bug on the release branch. I can't justify holding 1.45 for this unless someone can confirm the problem happens on release. This bug is no longer a showstopper. Jim, if you can verify that the problem happens on the release branch with this platform/compiler combination, feel free to change the severity back. -- Eric Niebler BoostPro Computing http://www.boostpro.com