yizhihenpidehou opened a new issue, #17238: URL: https://github.com/apache/tvm/issues/17238
Thanks for participating in the TVM community! We use https://discuss.tvm.ai for any general usage questions and discussions. The issue tracker is used for actionable items such as feature proposals discussion, roadmaps, and bug tracking. You are always welcomed to post on the forum first :smile_cat: Issues that are inactive for a period of time may get closed. We adopt this policy so that we won't lose track of actionable issues that may fall at the bottom of the pile. Feel free to reopen a new one if you feel there is an additional problem that needs attention when an old one gets closed. ### Expected behavior I want to run the code: ` rt_lib = tvm.build(MyModule, target="llvm") ` ### Actual behavior Traceback (most recent call last): File "/Users/yizhihenpidehou/Desktop/tvm_experiment/lecture2_TensorIR/main.py", line 40, in <module> rt_lib = tvm.build(MyModule, target="llvm") File "/Users/yizhihenpidehou/Desktop/fdu/eg/tvm/python/tvm/driver/build_module.py", line 297, in build rt_mod_host = _driver_ffi.tir_to_runtime(annotated_mods, target_host) File "/Users/yizhihenpidehou/Desktop/fdu/eg/tvm/python/tvm/_ffi/_ctypes/packed_func.py", line 240, in __call__ raise_last_ffi_error() File "/Users/yizhihenpidehou/Desktop/fdu/eg/tvm/python/tvm/_ffi/base.py", line 481, in raise_last_ffi_error raise py_err tvm.error.InternalError: Traceback (most recent call last): File "/Users/yizhihenpidehou/Desktop/fdu/eg/tvm/src/target/codegen.cc", line 72 InternalError: Check failed: (bf != nullptr) is false: target.build.llvm is not enabled ### Environment tvm path: /Users/yizhihenpidehou/Desktop/fdu/eg/tvm/python/tvm/__init__.py Operating System: macOS M2 arm; TVM version:0.18.dev0 <img width="423" alt="image" src="https://github.com/user-attachments/assets/c24a2735-3189-43be-abca-4a3e69d41dbe";> ### Steps to reproduce 1. build tvm from source code; 2. set(USE_LLVM "/opt/homebrew/opt/llvm/bin/llvm-config --link-static") in cmake.config 3. cmake .. & make 4. cd /python subfolder in tvm, and python setup.py install, but I got such error <img width="561" alt="image" src="https://github.com/user-attachments/assets/c301f5ad-3782-4afa-b9df-68a5457f4151";> 5. follow Building with a Conda Environment follow [conda install link](https://tvm.apache.org/docs/install/from_source.html#python-package-installation:~:text=with%20a%20Conda-,Environment,-%C2%B6) 6. run my code ` import sys import numpy as np import tvm from tvm.ir.module import IRModule from tvm.script import tir as T print("tvm path:",tvm.__file__) print("tvm version:",tvm.__version__) dtype = "float32" a_np = np.random.rand(128, 128).astype(dtype) b_np = np.random.rand(128, 128).astype(dtype) c_mm_relu = np.maximum(a_np @ b_np, 0) @tvm.script.ir_module class MyModule: @T.prim_func def mm_relu(A: T.Buffer((128, 128), "float32"), B: T.Buffer((128, 128), "float32"), C: T.Buffer((128, 128), "float32")): T.func_attr({"global_symbol": "mm_relu", "tir.noalias": True}) Y = T.alloc_buffer((128, 128), dtype="float32") for i, j, k in T.grid(128, 128, 128): with T.block("Y"): # [block_axis] = T.axis.[axis_type]([axis_range], [mapped_value]) vi = T.axis.spatial(128, i) vj = T.axis.spatial(128, j) vk = T.axis.reduce(128, k) with T.init(): Y[vi, vj] = T.float32(0) Y[vi, vj] = Y[vi, vj] + A[vi, vk] * B[vk, vj] for i, j in T.grid(128, 128): with T.block("C"): vi = T.axis.spatial(128, i) vj = T.axis.spatial(128, j) C[vi, vj] = T.max(Y[vi, vj], T.float32(0)) rt_lib = tvm.build(MyModule, target="llvm") a_nd = tvm.nd.array(a_np) b_nd = tvm.nd.array(b_np) c_nd = tvm.nd.empty((128, 128), dtype="float32") print(type(c_nd)) func_mm_relu = rt_lib["mm_relu"] func_mm_relu(a_nd, b_nd, c_nd) np.testing.assert_allclose(c_mm_relu, c_nd.numpy(), rtol=1e-5) import tvm from tvm import te import IPython A = te.placeholder((128, 128), "float32", name="A") B = te.placeholder((128, 128), "float32", name="B") k = te.reduce_axis((0, 128), "k") Y = te.compute((128, 128), lambda i, j: te.sum(A[i, k] * B[k, j], axis=k), name="Y") C = te.compute((128, 128), lambda i, j: te.max(Y[i, j], 0), name="C") te_func = te.create_prim_func([A, B, C]).with_attr({"global_symbol": "mm_relu"}) MyModuleFromTE = tvm.IRModule({"mm_relu": te_func}) print(IPython.display.Code(MyModuleFromTE.script(), language="python")) ` ### Triage Please refer to the list of label tags [here](https://github.com/apache/tvm/wiki/Issue-Triage-Labels) to find the relevant tags and add them below in a bullet format (example below). * needs-triage -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
