News for hackers. Stuff that matters.
Re:C++ Coroutine Theory (2017) by gumby on Tuesday February 06, 2018 @02:00PM (#16318533)
Given the semantics of coroutines, what would you consider a sane call stack? Should each routine have its own stack?

This is not a glib question, I genuinely don't know what the proper answer might be. Perhaps what's needed is a different debugger mechanism?

Re:C++ Coroutine Theory (2017) by spc476 on Tuesday February 06, 2018 @02:25PM (#16318755)
My own implementation of coroutines for C [1] requires each coroutine to have its own stack. It simplifies things quite a bit.

[1] https://github.com/spc476/C-Coroutines [2]

[2] More of a "proof-of-concept" than something really meant for production.