Paolo Carlini wrote: > Hum... > > In fact, I have just tried both the testcases included in c++/6678 and > could *not* reproduce the problem on my PII, linux2.4.16, glibc2.2.5 > system... ... on the other hand, I *could* reproduce it with gcc3.0.4.... I'm going to annotate this in the audit trail... Paolo.