[v3] libstdc++ vs. icc 8.1

Nathan Myers ncm-nospam@cantrip.org
Wed Apr 7 04:35:00 GMT 2004


On Tue, Apr 06, 2004 at 10:52:48PM -0500, Benjamin Kosnik wrote:
> 
> >I have a question why should we put in workarounds in libstdc++ for ICC 
> >at all.
> 
> Because libstdc++ was wrong. It's fixed now. Sometimes, having more than
> one compiler run through the same sources is a good thing.

I'll go farther: having someone willing to run our code against another 
compiler and report oddities is worth quite a lot of hassle and clutter.
Ideally we would take this information and use it to make regression 
tests so that Gcc would (eventually) be equipped to report the same 
problems their compiler did.

Nathan Myers
ncm-nospam@cantrip.org



More information about the Libstdc++ mailing list