Check failed: kNumAttrs == attrs.size() (2 vs. 1) : ValueError: Incorrect kNumAttrs for instruction: Split

[abacaaaaaa]

When I tried to build the model on my Mac, I got an error:

python3 build.py
Automatically configuring target: metal -keys=metal,gpu -max_function_args=31 -max_num_threads=256 -max_shared_memory_per_block=32768 -max_threads_per_block=1024 -thread_warp_size=32
Load cached module from dist/mod_cache_before_build.pkl and skip tracing. You can use --use-cache=0 to retrace
Traceback (most recent call last):
  File "/Users/b03/Desktop/old/build.py", line 195, in <module>
    build(mod, ARGS)
  File "/Users/b03/Desktop/old/build.py", line 130, in build
    db = ms.database.create(work_dir=args.db_path)
  File "/opt/anaconda3/envs/web2/lib/python3.10/site-packages/tvm/meta_schedule/database/database.py", line 417, in create
    return JSONDatabase(*args, **kwargs)
  File "/opt/anaconda3/envs/web2/lib/python3.10/site-packages/tvm/meta_schedule/database/json_database.py", line 86, in __init__
    self.__init_handle_by_constructor__(
  File "tvm/_ffi/_cython/./object.pxi", line 132, in tvm._ffi._cy3.core.ObjectBase.__init_handle_by_constructor__
  File "tvm/_ffi/_cython/./packed_func.pxi", line 288, in tvm._ffi._cy3.core.ConstructorCall
  File "tvm/_ffi/_cython/./packed_func.pxi", line 277, in tvm._ffi._cy3.core.FuncCall
  File "tvm/_ffi/_cython/./base.pxi", line 182, in tvm._ffi._cy3.core.CHECK_CALL
  File "/opt/anaconda3/envs/web2/lib/python3.10/site-packages/tvm/_ffi/base.py", line 481, in raise_last_ffi_error
    raise py_err
tvm._ffi.base.TVMError: Traceback (most recent call last):
  File "/Users/catalyst/Workspace/mlc-ai-package-self-runner/_work/package/package/tvm/src/support/parallel_for.cc", line 128
RuntimeError: parallel_for_dynamic error with [17:46:32] /Users/catalyst/Workspace/mlc-ai-package-self-runner/_work/package/package/tvm/src/meta_schedule/database/json_database.cc:198: ValueError: Unable to parse TuningRecord, on line 7 of file log_db/database_tuning_record.json. The workload is:
# from tvm.script import ir as I
# from tvm.script import tir as T

@I.ir_module
class Module:
    @T.prim_func
    def main(rxplaceholder: T.Buffer((T.int64(1), T.int64(4), T.int64(64), T.int64(64)), "float32"), T_multiply: T.Buffer((T.int64(1), T.int64(4), T.int64(64), T.int64(64)), "float32")):
        T.func_attr({"op_pattern": 0, "tir.noalias": T.bool(True)})
        # with T.block("root"):
        for ax0, ax1, ax2, ax3 in T.grid(T.int64(1), T.int64(4), T.int64(64), T.int64(64)):
            with T.block("T_multiply"):
                v_ax0, v_ax1, v_ax2, v_ax3 = T.axis.remap("SSSS", [ax0, ax1, ax2, ax3])
                T.reads(rxplaceholder[v_ax0, v_ax1, v_ax2, v_ax3])
                T.writes(T_multiply[v_ax0, v_ax1, v_ax2, v_ax3])
                T_multiply[v_ax0, v_ax1, v_ax2, v_ax3] = T.float32(0.041666667908430099) * rxplaceholder[v_ax0, v_ax1, v_ax2, v_ax3]
The JSONObject of TuningRecord is:
[T.int64(6), [[[["GetBlock", [], ["T_multiply", "main"], ["b0"]], ["GetLoops", ["b0"], [], ["l1", "l2", "l3", "l4"]], ["Fuse", ["l1", "l2", "l3", "l4"], [T.int64(1)], ["l5"]], ["SampleCategorical", [], [[T.int64(32), T.int64(64), T.int64(128), T.int64(256), T.int64(512), T.int64(1024)], [T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666)]], ["v6"]], ["Split", ["l5", "None", "v6"], [T.int64(1)], ["l7", "l8"]], ["Bind", ["l7"], ["blockIdx.x"], []], ["Bind", ["l8"], ["threadIdx.x"], []], ["EnterPostproc", [], [], []]], [[T.int64(3), T.int64(0)]]], [T.float64(2.767325769084694e-05)], {"thread_warp_size": T.int64(32), "host": {"mtriple": "arm64-apple-macos", "tag": "", "kind": "llvm", "mcpu": "apple-latest", "keys": ["arm_cpu", "cpu"]}, "max_threads_per_block": T.int64(1024), "max_function_args": T.int64(31), "max_num_threads": T.int64(256), "tag": "", "max_shared_memory_per_block": T.int64(32768), "kind": "metal", "keys": ["metal", "gpu"]}, [["TENSOR", "float32", [T.int64(1), T.int64(4), T.int64(64), T.int64(64)]], ["TENSOR", "float32", [T.int64(1), T.int64(4), T.int64(64), T.int64(64)]]]]]
The error message is:
[17:46:32] /Users/catalyst/Workspace/mlc-ai-package-self-runner/_work/package/package/tvm/src/meta_schedule/database/database.cc:167: ValueError: Unable to parse the JSON object: [[[["GetBlock", [], ["T_multiply", "main"], ["b0"]], ["GetLoops", ["b0"], [], ["l1", "l2", "l3", "l4"]], ["Fuse", ["l1", "l2", "l3", "l4"], [T.int64(1)], ["l5"]], ["SampleCategorical", [], [[T.int64(32), T.int64(64), T.int64(128), T.int64(256), T.int64(512), T.int64(1024)], [T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666), T.float64(0.16666666666666666)]], ["v6"]], ["Split", ["l5", "None", "v6"], [T.int64(1)], ["l7", "l8"]], ["Bind", ["l7"], ["blockIdx.x"], []], ["Bind", ["l8"], ["threadIdx.x"], []], ["EnterPostproc", [], [], []]], [[T.int64(3), T.int64(0)]]], [T.float64(2.767325769084694e-05)], {"thread_warp_size": T.int64(32), "host": {"mtriple": "arm64-apple-macos", "tag": "", "kind": "llvm", "mcpu": "apple-latest", "keys": ["arm_cpu", "cpu"]}, "max_threads_per_block": T.int64(1024), "max_function_args": T.int64(31), "max_num_threads": T.int64(256), "tag": "", "max_shared_memory_per_block": T.int64(32768), "kind": "metal", "keys": ["metal", "gpu"]}, [["TENSOR", "float32", [T.int64(1), T.int64(4), T.int64(64), T.int64(64)]], ["TENSOR", "float32", [T.int64(1), T.int64(4), T.int64(64), T.int64(64)]]]]
The error is: [17:46:32] /Users/catalyst/Workspace/mlc-ai-package-self-runner/_work/package/package/tvm/src/tir/schedule/primitive/.././instruction_traits.h:387: InternalError: Check failed: kNumAttrs == attrs.size() (2 vs. 1) : ValueError: Incorrect kNumAttrs for instruction: Split

Environments:
Mac M2
Python 3.10
mlc-ai-nightly 0.15.dev315

[senlyu163]

Since the library was developed earlier, you can tune it again without using the tune_db uploaded by the author.