module {
  tt.func public @triton__0d1d2de(%arg0: !tt.ptr<f32, 1> {tt.divisibility = 16 : i32}, %arg1: !tt.ptr<bf16, 1> {tt.divisibility = 16 : i32}, %arg2: i32 {tt.divisibility = 16 : i32, tt.max_divisibility = 16 : i32}) attributes {noinline = false} {
    %cst = arith.constant dense<12865792> : tensor<1024xi32>
    %c1024_i32 = arith.constant 1024 : i32
    %0 = tt.get_program_id x : i32
    %1 = arith.muli %0, %c1024_i32 : i32
    %2 = tt.make_range {end = 1024 : i32, start = 0 : i32} : tensor<1024xi32>
    %3 = tt.splat %1 : (i32) -> tensor<1024xi32>
    %4 = arith.addi %3, %2 : tensor<1024xi32>
    %5 = arith.cmpi slt, %4, %cst : tensor<1024xi32>
    %6 = tt.splat %arg0 : (!tt.ptr<f32, 1>) -> tensor<1024x!tt.ptr<f32, 1>>
    %7 = tt.addptr %6, %4 : tensor<1024x!tt.ptr<f32, 1>>, tensor<1024xi32>
    %8 = tt.load %7, %5 {cache = 1 : i32, evict = 1 : i32, isVolatile = false} : tensor<1024xf32>
    %9 = tt.splat %arg1 : (!tt.ptr<bf16, 1>) -> tensor<1024x!tt.ptr<bf16, 1>>
    %10 = tt.addptr %9, %4 : tensor<1024x!tt.ptr<bf16, 1>>, tensor<1024xi32>
    %11 = arith.truncf %8 : tensor<1024xf32> to tensor<1024xbf16>
    tt.store %10, %11, %5 {cache = 1 : i32, evict = 1 : i32} : tensor<1024xbf16>
    tt.return
  }
}