krzysz00 wrote:

Ok, yeah, you're right that doing some sort of stages list to let people 
pipeline better is probably the right move, so I'm withdrawing my broad 
objection to the concept.

I do agree that the `omits`/`ignores` system is kinda backwards when you're 
reading it, but I'm having a hard time thinking of a way to do something 
different that can be easily extended forward-compatibly.

Another phrasing is that "if the `omits` bitfield on an asyncmark is a superset 
of the `ignores` bitfield of an `asyncwait`, then that asyncwait will not 
consider that asyncmark when determining the number of outstanding marks"

https://github.com/llvm/llvm-project/pull/220442
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to