[Mlir-commits] [mlir] [MLIR] Implement emulation of static indexing subbyte type vector stores (PR #115922)

[llvmlistbot at llvm.org]

================
@@ -309,6 +314,99 @@ emulatedVectorLoad(OpBuilder &rewriter, Location loc, Value base,
       newLoad);
 }
 
+/// Emits `memref.generic_atomic_rmw` op to store a subbyte-sized value to a
+/// byte in memory, with a mask. The `valueToStore` is a vector of subbyte-sized
+/// elements, with size of 8 bits, and the mask is used to select which elements
+/// to store.
+///
+/// Inputs:
+///   linearizedMemref = |a|b|c|d| : <4xi2> (<1xi8>)
+///   linearizedIndex = 2
+///   valueToStore = |e|f|g|h| : vector<4xi2>
+///   mask = |0|0|1|1| : vector<4xi1>
+///
+/// Result:
+///   linearizedMemref = |a|b|g|h| : <4xi2> (<1xi8>)
+static void atomicStore(OpBuilder &builder, Location loc,
+                        MemRefValue linearizedMemref, Value linearizedIndex,
+                        VectorValue valueToStore, Value mask,
+                        int64_t numSrcElemsPerDest) {
----------------
lialan wrote:

Thanks! Actually this was not needed! I removed it.

https://github.com/llvm/llvm-project/pull/115922