[Mlir-commits] [mlir] [MLIR][NVVM] Support generating all the ldmatrix intrinsics from NVVM ops (PR #148783)

[Gao Yanfeng]

================
@@ -559,17 +559,47 @@ llvm.func @llvm_nvvm_cp_async_bulk_wait_group() {
 // CHECK-LABEL: @ld_matrix
 llvm.func @ld_matrix(%arg0: !llvm.ptr<3>) {
----------------
Pecco-314 wrote:

That's a good point. But I think it is better to conduct a unified refactoring, splitting this file into multiple files, rather than doing it here.

https://github.com/llvm/llvm-project/pull/148783