51 GCC HEAD regressions, 40 new, with your patch on 2005-10-14T21:55:43Z.

[Paolo Carlini]

Andrew Pinski wrote:

> On Oct 15, 2005, at 11:45 AM, GCC regression checker wrote:
>
>> With your recent patch, GCC HEAD has some regression test failures,
>> which used to pass.  There are 40 new failures, and 11
>> failures that existed before and after that patch; 0 failures
>> have been fixed.
>>
>> The new failures are:
>
> This is a latent bug in the tr1 headers:
>         std::size_t result = 14695981039346656037ULL;
>
> Gives now the following correct warning:
> /Users/regress/tbox/native/build/powerpc-apple-darwin7.4.0/libstdc++-
> v3/include/tr1/functional:1171: warning: large integer implicitly 
> truncated to unsigned type

I was aware of this potential problem but had always believed #pragma
GCC system_header was taking care of suppressing it (of course the
warning is not really troublesome because when the involved
specialization is actually used size_t is sufficiently large, 8 bytes).

Will look into it, maybe a #pragma GCC system_header is missing and the
warning was wrongly not emitted because of the bug...

Paolo.