Tao B. Schardl

There's no good way to fix this issue. The misattribution here is happening when the code is compiled with no debug symbols. Compiling with `-g` resolves this issue, but without...

I don't see how modifying the headers or macro definitions affects this specific warning, which is coming from the back-end of the compiler, not the clang front-end. We might be...

That behavior sounds like it may cause problems for our infrastructure for regression-testing Clang, as well as for build systems that try to incorporate OpenCilk. We would need to chase...

Thanks for the bug report. It looks like this issue might be caused by certain versions of libc++. Can you try changing the `offsetof` on line 502 of `/projects/eng/software/src/opencilk/opencilk/cilktools/cilksan/disjointset.h` to...

Is there a particular reason you want to include `cilk/cilk_api.h` in its own namespace? Including headers within namespaces is generally considered a Bad Idea, so I'm curious why you would...

It looks like this PR really contains 1 new commit, but it's currently based on a different branch from `dev/17.x`, which confuses GitHub. Please rebase your change onto `dev/17.x` and...

I tried using this PR to remove the explicit registration and deregistration of reducers in Cilkscale, but it didn't work. Cilkscale might be an odd case, since the relevant structure...

I'm still testing this change with the Cilkscale codebase, and I ran into an issue that I think this change is supposed to handle. I have a modified version of...

Can you add some code-generation tests to this PR, to verify that clang emits the correct LLVM intrinsics? The existing tests mainly check for error messages, and I don't see...

I'm still having some trouble with this PR. I tried using it to convert the `shadow_stack` class member in Cilkscale's implementation from a pointer-to-hyperobject into a hyperobject struct member: https://github.com/neboat/productivity-tools/tree/cilkscale-reducer-registration....