USE_NVTX: OFF USE_GTEST: AUTO SUMMARIZE: OFF USE_IOS_RPC: OFF USE_MSC: OFF USE_ETHOSU: CUDA_VERSION: NOT-FOUND USE_LIBBACKTRACE: AUTO DLPACK_PATH: 3rdparty/dlpack/include USE_TENSORRT_CODEGEN: OFF USE_THRUST: OFF USE_TARGET_ONNX: OFF USE_AOT_EXECUTOR: ON BUILD_DUMMY_LIBTVM: OFF USE_CUDNN: OFF USE_TENSORRT_RUNTIME: OFF USE_ARM_COMPUTE_LIB_GRAPH_EXECUTOR: OFF USE_CCACHE: AUTO USE_ARM_COMPUTE_LIB: OFF USE_CPP_RTVM: USE_OPENCL_GTEST: /path/to/opencl/gtest USE_MKL: OFF USE_PT_TVMDSOOP: OFF MLIR_VERSION: NOT-FOUND USE_CLML: OFF USE_STACKVM_RUNTIME: OFF USE_GRAPH_EXECUTOR_CUDA_GRAPH: OFF ROCM_PATH: /opt/rocm USE_DNNL: OFF USE_VITIS_AI: OFF USE_MLIR: OFF USE_RCCL: OFF USE_LLVM: llvm-config --ignore-libllvm --link-static USE_VERILATOR: OFF USE_TF_TVMDSOOP: OFF USE_THREADS: ON USE_MSVC_MT: OFF BACKTRACE_ON_SEGFAULT: OFF USE_GRAPH_EXECUTOR: ON USE_NCCL: OFF USE_ROCBLAS: OFF GIT_COMMIT_HASH: 79991133c17bb8685185e1f03cc2f688ea37c974 USE_VULKAN: ON USE_RUST_EXT: OFF USE_CUTLASS: OFF USE_CPP_RPC: OFF USE_HEXAGON: OFF USE_CUSTOM_LOGGING: OFF USE_UMA: OFF USE_FALLBACK_STL_MAP: ON USE_SORT: ON USE_RTTI: ON GIT_COMMIT_TIME: 2024-02-21 22:31:30 -0500 USE_HEXAGON_SDK: /path/to/sdk USE_BLAS: none USE_ETHOSN
6条答案
按热度按时间hfsqlsce1#
问题描述
在q4f16_0环境下编译Gemma-2b for Android时,加载模型成功,但聊天出现错误:OpenCL Error Code=-54: CL_INVALID_WORK_GROUP_SIZE。堆栈跟踪显示在"/home/chaoqin/mlcllm/3rdpaty/tvm/scr/runtime/opencl/opencl_module.cc",第90行。
预期行为
在Android平台上使用WebGPU/Vulkan/IOS/Android等平台编译MLC-LLM和TVM-Unity。
环境信息
conda
,来源):condapip
,来源):pipUSE_NVTX: OFF
USE_GTEST: AUTO
SUMMARIZE: OFF
USE_IOS_RPC: OFF
USE_MSC: OFF
USE_ETHOSU:
CUDA_VERSION: NOT-FOUND
USE_LIBBACKTRACE: AUTO
DLPACK_PATH: 3rdparty/dlpack/include
USE_TENSORRT_CODEGEN: OFF
USE_THRUST: OFF
USE_TARGET_ONNX: OFF
USE_AOT_EXECUTOR: ON
BUILD_DUMMY_LIBTVM: OFF
USE_CUDNN: OFF
USE_TENSORRT_RUNTIME: OFF
USE_ARM_COMPUTE_LIB_GRAPH_EXECUTOR: OFF
USE_CCACHE: AUTO
USE_ARM_COMPUTE_LIB: OFF
USE_CPP_RTVM:
USE_OPENCL_GTEST: /path/to/opencl/gtest
USE_MKL: OFF
USE_PT_TVMDSOOP: OFF
MLIR_VERSION: NOT-FOUND
USE_CLML: OFF
USE_STACKVM_RUNTIME: OFF
USE_GRAPH_EXECUTOR_CUDA_GRAPH: OFF
ROCM_PATH: /opt/rocm
USE_DNNL: OFF
USE_VITIS_AI: OFF
USE_MLIR: OFF
USE_RCCL: OFF
USE_LLVM: llvm-config --ignore-libllvm --link-static
USE_VERILATOR: OFF
USE_TF_TVMDSOOP: OFF
USE_THREADS: ON
USE_MSVC_MT: OFF
BACKTRACE_ON_SEGFAULT: OFF
USE_GRAPH_EXECUTOR: ON
USE_NCCL: OFF
USE_ROCBLAS: OFF
GIT_COMMIT_HASH: 79991133c17bb8685185e1f03cc2f688ea37c974
USE_VULKAN: ON
USE_RUST_EXT: OFF
USE_CUTLASS: OFF
USE_CPP_RPC: OFF
USE_HEXAGON: OFF
USE_CUSTOM_LOGGING: OFF
USE_UMA: OFF
USE_FALLBACK_STL_MAP: ON
USE_SORT: ON
USE_RTTI: ON
GIT_COMMIT_TIME: 2024-02-21 22:31:30 -0500
USE_HEXAGON_SDK: /path/to/sdk
USE_BLAS: none
USE_ETHOSN
watbbzwu2#
你好,bulutthecat@qc903113684,对于给您带来的不便深表歉意。您在遇到这个问题时是否检查过#1955是否已包含?或者尝试使用最新版本的软件包?我怀疑这可以通过#1955来解决。谢谢!
gfttwv5a3#
很抱歉,我无法在我这边复现这个问题。在获取最新的更改后,你能重新构建tvm和mlc吗?然后再重新编译模型库?
46qrfjad4#
你好,bulutthecat@qc903113684,对于给您带来的不便深感抱歉。您在遇到这个问题时是否检查过#1955是否已包含?或者尝试使用最新版本的软件包?我怀疑这个问题已经通过#1955修复了。谢谢!
感谢您告知我,如果它能正常工作,我会回复您的。
t0ybt7op5#
这个PR可能解决了问题,但我没有时间进行测试。[#1850]
dtcbnfnu6#
你好,@qc903113684,#1850已被#1822取代,后者已在3周前合并。也就是说,#1822和#1955都可以作为解决此问题潜在的修复方案。