Mangling extension for C++20 Modules

[urnathan]

Both GCC and Clang added mangling for c++20 module purview symbols. While we tried to do the same thing, we failed :( It also turns out the scheme we designed is undemangleable. Plus the modules-ts evolved.

Here is a newer scheme: https://drive.google.com/file/d/1qQjqptzOFT_lfXH8L6-iD9nCRi34wjft/view?usp=sharing

1. is demangleable
2. changes to the strong-ownership model
3. Addresses issue P1874: Dynamic Initialization Order of Non-Local Variables in Modules #99

[tchaikov]

the referenced google document

Both GCC and Clang added mangling for c++20 module purview symbols. While we tried to do the same thing, we failed :( It also turns out the scheme we designed is undemangleable. Plus the modules-ts evolved.

Here is a newer scheme: https://drive.google.com/file/d/1qQjqptzOFT_lfXH8L6-iD9nCRi34wjft/view?usp=sharing

hello,

the referenced google document is not reachable anymore. was it moved / removed ?

1. is demangleable
2. changes to the strong-ownership model
3. Addresses issue P1874: Dynamic Initialization Order of Non-Local Variables in Modules #99

[ChuanqiXu9]

@urnathan hi, would you like to send the documentation again? This is somewhat embrassing that the implementation were already merged into GCC and Clang but the document doesn't merge into the Itanium C++ ABI specification... so they is the status quo standard now. I think it may be better to summarize them into the ABI specification formally.

CC @rjmccall I found this when I try to summarize the ABI related things with modules

[urnathan]

On 1/11/24 01:58, Chuanqi Xu wrote: @urnathan <https://github.com/urnathan> hi, would you like to send the documentation again? This is somewhat embrassing that the implementation were already merged into GCC and Clang but the document doesn't merge into the Itanium C++ ABI specification... so they is the status quo standard now. I think it may be better to summarize them into the ABI specification formally. — Reply to this email directly, view it on GitHub <#134 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADD67GPV3RJGQOL2D4Y5VRLYN6EQNAVCNFSM5M5ZNSN2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOBYGY2DCMJWHE2A>. You are receiving this because you were mentioned.Message ID: ***@***.***>

#144

…

-- Nathan Sidwell