[Error] tvm.build error in the course 2 example “Constructing Tensor Program”

[lindylin1817]

I installed TVM from its latest github main version in these days and tried the "constructing tensor program" of the 2nd course. My code is

import tvm
from tvm.ir.module import IRModule
from tvm.script import tir as T
import numpy as np

@tvm.script.ir_module
class MyModule:
    @T.prim_func
    def func(A: T.Buffer[128, "float32"],
             B: T.Buffer[128, "float32"],
             C: T.Buffer[128, "float32"]):
        # extra annotations for the function
        T.func_attr({"global_symbol": "main", "tir.noalias": True})
        for i in range(128):
            with T.block("C"):
                # declare a data parallel iterator on spatial domain
                vi = T.axis.spatial(128, i)
                C[vi] = A[vi] + B[vi]

print(type(MyModule))
print(MyModule.script())

rt_mod = tvm.build(MyModule, target="llvm")  # The module for CPU backends.
print(type(rt_mod))
~

But when I run it, there was the following error. Thanks for any help!

Traceback (most recent call last):
  File "test1.py", line 25, in <module>
    rt_mod = tvm.build(MyModule, target="llvm")  # The module for CPU backends.
  File "/home/xwv/xwv/tvm-main/python/tvm/driver/build_module.py", line 282, in build
    rt_mod_host = _driver_ffi.tir_to_runtime(annotated_mods, target_host)
  File "/home/xwv/xwv/tvm-main/python/tvm/_ffi/_ctypes/packed_func.py", line 237, in __call__
    raise get_last_ffi_error()
tvm._ffi.base.TVMError: Traceback (most recent call last):
  9: TVMFuncCall
  8: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::runtime::Module (tvm::runtime::Map<tvm::Target, tvm::IRModule, void, void> const&, tvm::Target)>::AssignTypedLambda<tvm::{lambda(tvm::runtime::Map<tvm::Target, tvm::IRModule, void, void> const&, tvm::Target)#6}>(tvm::{lambda(tvm::runtime::Map<tvm::Target, tvm::IRModule, void, void> const&, tvm::Target)#6}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
  7: tvm::TIRToRuntime(tvm::runtime::Map<tvm::Target, tvm::IRModule, void, void> const&, tvm::Target const&)
  6: tvm::codegen::Build(tvm::IRModule, tvm::Target)
  5: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::runtime::Module (tvm::IRModule, tvm::Target)>::AssignTypedLambda<tvm::codegen::{lambda(tvm::IRModule, tvm::Target)#1}>(tvm::codegen::{lambda(tvm::IRModule, tvm::Target)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
  4: tvm::codegen::LLVMModuleNode::Init(tvm::IRModule const&, tvm::Target const&)
  3: void tvm::codegen::CodeGenLLVM::AddFunctionsOrdered<__gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > >, tvm::codegen::CodeGenLLVM::AddFunctionsOrdered<__gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > > >(__gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > >, __gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > >)::{lambda(auto:1)#1}>(__gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > >, __gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > >, tvm::codegen::CodeGenLLVM::AddFunctionsOrdered<__gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > > >(__gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > >, __gnu_cxx::__normal_iterator<tvm::tir::PrimFunc*, std::vector<tvm::tir::PrimFunc, std::allocator<tvm::tir::PrimFunc> > >)::{lambda(auto:1)#1})
  2: tvm::codegen::CodeGenCPU::AddFunction(tvm::tir::PrimFunc const&)
  1: tvm::codegen::CodeGenCPU::AddDebugInformation(tvm::tir::PrimFunc, llvm::Function*)
  0: tvm::codegen::CodeGenCPU::GetDebugType(tvm::Type const&, llvm::Type*)
  File "/home/xwv/xwv/tvm-main/src/target/llvm/codegen_cpu.cc", line 281
TVMError:
---------------------------------------------------------------
An error occurred during the execution of TVM.
For more information, please see: https://tvm.apache.org/docs/errors.html
---------------------------------------------------------------
  Check failed: (ptr_type != nullptr) is false: Got LLVM pointer type from non-pointer IR type: handle

[Hzfengsy]

Did you turn on the LLVM config during CMake? see here

BTW, we use a different version of TVM for our course, so please install it by

pip install mlc-ai-nightly -f https://mlc.ai/wheels

[lindylin1817]

@Hzfengsy Sorry for replying late. Thanks for your suggestion! I checked in my previous case, didn't set LLVM config ON when compiling TVM. I turned it ON in config file and rebuilt TVM this time. But I got the exactly same error as previous case. I don't know why...
BTW, I can't "pip install" your "https://mlc.ai/wheels" here, always network error. We might need a mirror site here ...

[Hzfengsy]

If you are in China mainland, a proxy may help :)