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

Opencl INLINE #5638

Merged
merged 3 commits into from
Dec 25, 2024
Merged

Conversation

magnumripper
Copy link
Member

@magnumripper magnumripper commented Dec 24, 2024

Define NOINLINE as __attribute__((noinline)) which is has been seen working. Replace all uses of the latter with the macro. Drop the questionable "inline" macro and instead define INLINE as static inline, which should be the right thing, for most targets. Use this only for inlines that were replaced by the old inline macro (so not to make an uncontrollable mess at once).

Closes #5618

After cc2f223 we got it too often. This is more like it.
@magnumripper magnumripper force-pushed the opencl-inline branch 2 times, most recently from 93b6e56 to 50fdace Compare December 25, 2024 00:48
This currently ends up with "rotate(x, 32U-(n))" on all platforms anyway
so no net change.
Define NOINLINE as "__attribute__((noinline))" which has been seen
working. Replace all uses of the latter with the macro.

Drop the questionable "inline" macro and instead define INLINE as 'static
inline', which should be the right thing, for anything put POCL and MESA
which we've seen problems with in the past. Then use this macro only for
inlines that were replaced by the old inline macro (for now).

Closes openwall#5618
@magnumripper magnumripper merged commit 55dbabb into openwall:bleeding-jumbo Dec 25, 2024
39 checks passed
@magnumripper magnumripper deleted the opencl-inline branch December 25, 2024 13:06
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 this pull request may close these issues.

OpenCL Add a DECLSPEC or INLINE macro
2 participants