[Mlir-commits] [mlir] [mlir][amdgpu] Shared memory access optimization pass (PR #75627)

[Mehdi Amini]

================
@@ -21,6 +21,10 @@ class ConversionTarget;
 namespace amdgpu {
 
 #define GEN_PASS_DECL_AMDGPUEMULATEATOMICSPASS
+
+/// Create a pass to optimize shared memory reads and writes.
+std::unique_ptr<Pass> createOptimizeSharedMemoryPass();
+
----------------
joker-eph wrote:

This will be auto-generated when you remove the line from TableGen (see below)

https://github.com/llvm/llvm-project/pull/75627