That makes sense. If it works and there is no need to verify the tests passes for Windows (with Windows configuration) what we have in master is fine.

Fixed a casing issue with a local var.

Some minor optimizations related to CR.

Removed the changes from threads.ll.

Addressed CR.

Added separate tests for Windows for specific tests.

Fixed a failing Windows test. The issue is that that the order of the operands for add operation has changed. I cant see how these changes could cause the issue, but it is a failing test that blocks push of this Diff.

Merged in master.

Needed merges with the latest master.

Addressed CR feedback.

I suspect there will be conflicts at this time. Posting the source here, so some of the previous commenters could look at it.

Implemented the closure approach in the OMPIRBuilder only.

Just finished significant testing ofg this with our AI codegen. The fix in https://reviews.llvm.org/D92189 whows some issues with number of arguments passed as varargs. The number of varargs that starts showing problems is different for different HW (it seems different for SpirV, Vulcan, MultiCore). I will update this PR.

In D92189#2425639, @jdoerfert wrote:

While only partially related, can you leave a FIXME saying that more than 15 arguments need to be packed in a structure?

This looks fine. Except the failing tests. Once the tests are fixed it should be good to go.

@jdoerfert That makes sense. I looked at the code and that is LLVMIR, so Ill start moving things over.

Addressed some CR feedback on this diff.

Created a small pass to capture omp::ParallelOp parameters.

Addressed some CR feedback.

Rewritten the title and the summary as well.
The tests are in the StandardToLLVM directory (as the rest of the similar conversion tests). This is not part of the LLVM? Or maybe I am wrong?

Resolved CR feedback - updated docs.

Updated the Title and Summary of the diff as requested.

Fixed couple of typos and Diff title/summary issues

Some more formatting and squashing commits.

For some reason a code was removed in a diagnostic message. Added back.

Code formatting changes.

@pifon2a Thanks! That makes sense.

Hello! I am confused about the reasoning why this functionality was removed - we certainly need the it. I did coded up pretty much the same change and someone (thanks Laub, Frank) pointed out this diff. What is the general strategy/approach to lower operations that are not intrinsics in the LLVM.

Resolved a merge conflict in LLVMOps.td due to changes coming from master.

Clean up one more ordinal used for f32 sqrt test.

Addressed CR re: test matching rules and use of a temp f64 const.

Addressed some CR comments re: comments - fixed a type to match an existing type and moved an operation description to a OSD file.