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, > It passes all my tests so far: if Jason can fix the damn serious issue > with class types, I think we should apply it. > > If everything goes well, I also have in my tree: some trivial changes to > the overloads for volatile / const volatile which allow to enable the > code (just use consistently volatile and const volatile also when > passing around tuple<_Args...>); trivial changes int -> size_t in > various places indexing std::tuple (this changed since TR1), then a > check __i >= can also be removed, of course. > > .... and we should add a *ton* of testcases. > So... for now I'm applying this incremental improvement, without attempting anything special for the volatile / const volatile issue (at some point I add an hack which passes the testsuite, but I didn't trust it, treated differently those overloads in an ad-hoc way). Besides what we discussed already, I also noticed inconsistencies in the way temporary tuple<_Args...> were passed around, and decided to uniform everything to use rvalue refs, which seems the right thing to do. If you are aware of testcases that breaks because of that, more than welcome of course... Tested x86_64-linux, committed to mainline. Paolo. /////////////////////
Attachment:
CL_bind
Description: Text document
Attachment:
patch_bind
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |