[libcxx-commits] [libcxx] [libc++] Protect the libc++ implementation from CUDA SDK's `__noinline__` macro (PR #73838)

[Mark de Wever via libcxx-commits]

mordante wrote:

> > Our CI resources are limited and they slow down development, at least for me it does regularly.
> 
> Thanks for the feedback on the CI. Could you please file a bug about it where a meaningful conversation can take place. We're working on improving the CI experience and feedback from developers is super valuable.

The main issue is the CI takes a long time and when a lot of people have patches the feedback cycle can take quite some time limiting to one iteration per day. This is mainly an issue when the patch affects system specific behaviour. For example, locales. I don't think there is a good solution for that. Typically I prefer to work on these things in the weekend, then the CI is less busy.

The main point I want to make is that these work-arounds have a cost for us. It was not the intention to bash our CI.

https://github.com/llvm/llvm-project/pull/73838