[llvm] [AMDGPU] Introduce orderign parameter to atomic intrinsics and introduce new llvm.amdgcn.image.atomic.load intrinsic. (PR #73613)

Nicolai Hähnle via llvm-commits llvm-commits at lists.llvm.org
Tue Dec 12 12:23:14 PST 2023


================
@@ -1154,6 +1154,8 @@ class TargetLoweringBase {
     MaybeAlign align = Align(1);   // alignment
 
     MachineMemOperand::Flags flags = MachineMemOperand::MONone;
+
+    AtomicOrdering ordering = AtomicOrdering::NotAtomic;
----------------
nhaehnle wrote:

This piecemeal addition of fields to `IntrinsicInfo` which ends up not actually capturing the whole breadth of MMO options is problematic.

Could we simply add a `MachineMemOperand *` field in here which, if `nullptr`, supersedes all the other fields?

https://github.com/llvm/llvm-project/pull/73613


More information about the llvm-commits mailing list