This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
Hi, in order to implement type_info::hash_code for C++0x (and then, in a straightforward way, <typeindex>), _Hash_bytes has to be made available to libsupc++, but since we don't want to have libsupc++ depending on libstdc++ facilities, the only option seems moving hash_bytes.cc itself to libsupc++. I did this with the below and everything seems ok (tested x86_64-linux multilib). I'm only a bit nervous about the correctness and performance of hash_code vs the various definitions of type_info::operator==, but I think the below, rather simple, should be fine, in particular should be always true that type_infos comparing equal hash equal. Please review it. I mean to apply (and move to <typeindex>) in one day or so. Thanks, Paolo. /////////////////////
Attachment:
CL_type_info_hash_code
Description: Text document
Attachment:
patch_type_info_hash_code
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |