Stas Sergeev

Yeah, still bound to wrapper size by `__ASYM()` macro and `glob_asmdefs.h`. So tried patch 8c8a69eb5 to use `new`. This boots a bit but crashes. The problem is here: https://github.com/stsp/fdpp/blob/no_pnew/fdpp/farptr.hpp#L418 Obviously...

Got it to work with 1db08dd1, but it still doesn't fully boot, as with `new` the dtor is never called that way. Non-functional idea, it seems.

Turns out gcc allows to modify rvalue members with `-fpermissive`, but clang doesn't. :(

Added 2 tickets to clang, asking for gcc-compatible extensions: https://bugs.llvm.org/show_bug.cgi?id=40694 https://bugs.llvm.org/show_bug.cgi?id=40670 Of course it is very unlikely that they will be implemented. It seems clang doesn't like language extensions.

http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2018/p0476r2.html This paper may provide an alternative to the currently used placement new.

> This paper may provide an alternative to the Nope, it gives us nothing. But this does something very interesting: http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2018/p0593r2.html

In C++23 there will be `std::start_lifetime_as()`: http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p0593r5.html So this UB should wait 4 years or more. I would be surprised though if DSE can actually break our code because of...

No, will not run from HMA. I made it so that fdpp is fully contained in UMBs. HMA is no longer needed. So we need to implement this: http://www.ctyme.com/intr/rb-4818.htm and...

Maybe there are some differences, but please see the HMA-specific MCB structure layout here: http://www.ctyme.com/intr/rb-4818.htm

So if you add some tests, I can add an impl. Currently things to test are 4a01 and 4a02. 4a0[34] should follow.