Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Move to C11 and adopt _Atomic, static_asserts, and thread_local #485

Open
madolson opened this issue May 10, 2024 · 3 comments · May be fixed by #490
Open

Move to C11 and adopt _Atomic, static_asserts, and thread_local #485

madolson opened this issue May 10, 2024 · 3 comments · May be fixed by #490
Assignees

Comments

@madolson
Copy link
Member

As per 445a4b6, we implemented custom logic to compile with C11 when atomics were available, and C99 when they were not available. However, with the EoL of CentOS 7, we believe that all default compilers should be able to compile with C99. This gives us access to three pieces of functionality we use:

  1. Static asserts, we have some custom code defined to support it but now we can standardize on the C11 functionality.
  2. native thread_local support. We can avoid _thread.
  3. Native atomics support, so we can remove our custom logic supporting atomics.

This change should target Valkey 8. We will continue to use the old gcc version on Valkey 7.2.

@adetunjii
Copy link

adetunjii commented May 11, 2024

@madolson Are we sticking to the default memory ordering for _Atomic operations (ex: incr, decr, ....) for C11 which is sequentially consistent memory ordering memory_order_seq_cst, or are we going with the one used for the custom logic which is relaxed (memory_order_relaxed).

See: https://en.cppreference.com/w/c/atomic/memory_order

@adetunjii adetunjii linked a pull request May 12, 2024 that will close this issue
@madolson
Copy link
Member Author

Historically we have predominantly used atomics for counters in other threads, which are fine with memory_order_relaxed and see better performance.

@adetunjii
Copy link

@madolson please take a look at the PR to see if it's good, so I can finish up the remaining parts

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants