c++/3145

[Franz Sirl]

On Sunday 02 December 2001 15:58, Sam Horwitz wrote:
> This problem appears to be a regression from gcc 2.95.3 which I thought we
> try to fix in gcc 3.0 version.

But the patch is quite invasive and thus considred too dangerous for a stable 
release. I backported the patch to the gcc-3_0-branch though and got 
permission to add it to the contrib/ area as a patch you have to apply 
manually at your own risk.
Maybe, if a lot of people test the patch with gcc-3.0.3 (and no problems 
arise in the mainline due to it),  there is a slight chance it will be 
applied for 3.0.4...

Franz.