[TIR] [Analysis] Expose IsOutputBlock to python (#14352)

This patch just exposes an existing analysis API IsOutputBlock to
python. Since many schedule primitives have conditions on output blocks,
this API would be really useful while scheduling

python/tvm/tir/schedule/analysis.py

@@ -140,3 +140,22 @@ def has_block(sch: Schedule, block_name: str) -> bool:
         True if the given block exists in the schedule.
     """
     return _ffi_api.HasBlock(sch, block_name)  # type: ignore
+
+
+def is_output_block(sch: Schedule, block: BlockRV) -> bool:
+    """Check whether the given block is an output block
+
+    Parameters
+    ----------
+    sch : Schedule
+        The schedule object of the block
+    block : BlockRV
+        The blockRV to be checked
+
+    Returns
+    -------
+    yes/no : bool
+        True if the given block is an output block
+
+    """
+    return _ffi_api.IsOutputBlock(sch, block)  # type: ignore

src/tir/schedule/analysis/analysis.cc

@@ -2071,6 +2071,11 @@ TVM_REGISTER_GLOBAL("tir.schedule.GetAutoTensorizeMappingInfo")
     });
 
 TVM_REGISTER_GLOBAL("tir.schedule.HasBlock").set_body_typed(HasBlock);
+TVM_REGISTER_GLOBAL("tir.schedule.IsOutputBlock").set_body_typed([](Schedule sch, BlockRV block) {
+  auto state = sch->state();
+  auto block_sref = sch->GetSRef(block);
+  return IsOutputBlock(state, block_sref, GetScopeRoot(state, block_sref, false));
+});
 
 }  // namespace tir
 }  // namespace tvm

tests/python/unittest/test_tir_schedule_analysis.py

@@ -39,6 +39,7 @@
 from tvm.tir.stmt_functor import pre_order_visit
 from tvm.meta_schedule.testing import te_workload
 from tvm.te import create_prim_func
+from tvm.tir.schedule.analysis import is_output_block
 
 
 def _make_vars(*args: str) -> List[Var]:
@@ -396,5 +397,25 @@ def test_get_auto_tensorize_mapping_info_matmul(n, m, k, expected):
     check_index_map(matmul, "C", WMMA_SYNC_16x16x16_f16f16f32_INTRIN, expected)
 
 
+def test_is_output_block():
+    @T.prim_func
+    def two_elementwise(a: T.handle, c: T.handle) -> None:
+        A = T.match_buffer(a, (128, 128), "float32")
+        B = T.alloc_buffer((128, 128), "float32")
+        C = T.match_buffer(c, (128, 128), "float32")
+        for i, j in T.grid(128, 128):
+            with T.block("B"):
+                vi, vj = T.axis.remap("SS", [i, j])
+                B[vi, vj] = A[vi, vj] * 2.0
+        for i, j in T.grid(128, 128):
+            with T.block("C"):
+                vi, vj = T.axis.remap("SS", [i, j])
+                C[vi, vj] = B[vi, vj] + 1.0
+
+    sch = tvm.tir.Schedule(two_elementwise)
+    block_rv = sch.get_block("C")
+    assert is_output_block(sch, block_rv)
+
+
 if __name__ == "__main__":
     tvm.testing.main()