Wrong hastable key_eq implementation

[François Dumont]

Attached patch applied.

2012-01-05  François Dumont <fdumont@gcc.gnu.org>

         * include/bits/hashtable_policy.h (_Hashtable_base<>::_M_eq()):
         protected rather than private, use it...
         * include/bits/hashtable.h (_Hashtable<>::key_eq()): ... here.
         * testsuite/23_containers/unordered_set/observers.cc: New.
         * testsuite/23_containers/unordered_multiset/observers.cc: New.
         * testsuite/23_containers/unordered_map/observers.cc: New.
         * testsuite/23_containers/unordered_multimap/observers.cc: New.

ChangeLog has been roll.

François

On 01/04/2012 10:56 PM, Paolo Carlini wrote:
> Hi,
>>
>> I was quite surprise to discover this regression introduced by recent 
>> hashtable modifications.
>>
>> Tested under linux  x86_64, ok to commit ?
> Ok, thanks.
>> If so and not done at that time, can I do the ChangeLog roll ?
>>
>> I plan to do:
>> svn mv ChangeLog ChangeLog-2011
>> svn add ChangeLog
>>
>> and move all 2012 entries to newly created ChangeLog.
>>
>> Ok ?
> Sure.
>
> Thanks!
> Paolo.
>
> PS: to the list too: over the next week or so I will be 
> traveling/vacationing, thus don't be surprised if I will not reply 
> immediately to messages posted to the GCC mailing lists...
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: key_eq.patch
Type: text/x-patch
Size: 5865 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/libstdc++/attachments/20120105/609ae223/attachment.bin>