1
1
1
框架组件
尝试无修改编译新版llama.cpp,能启动,不能对话,过程分享
2025-10-29 09:47:59
282
opendas/llama.cpp太老,想编一个新版。 希望官方能出一个如何自己移植的教程。 我只是自己尝试而已。 结果能正常识别K100AI,能加载,不能对话。有兴趣可以看看 ----- 1. clone git hub仓库 2. 按照官方编译rocm的方法编译 ```bash HIPCXX="$(hipconfig -l)/clang" HIP_PATH="$(hipconfig -R)" \ HIP_DEVICE_LIB_PATH=/opt/dtk-25.04/amdgcn/bitcode \ HSA_OVERRIDE_GFX_VERSION=9.2.8 \ cmake -S . -B build -DGGML_HIP=ON -DGPU_TARGETS=gfx928 -DCMAKE_BUILD_TYPE=Release \ && cmake --build build --config Release --target llama-server -- -j 128 ``` 链接时 报错说缺少 "clang_rt.builtins-x86_64" ```bash [ 0%] Built target build_info [ 6%] Built target ggml-base [ 6%] Linking HIP shared library ../../../bin/libggml-hip.so [ 16%] Built target ggml-cpu clang-15: warning: argument unused during compilation: '--gpu-max-threads-per-block=1024' [-Wunused-command-line-argument] /usr/bin/ld: cannot find -lclang_rt.builtins-x86_64: No such file or directory clang-15: error: linker command failed with exit code 1 (use -v to see invocation) gmake[3]: *** [ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/build.make:1572: bin/libggml-hip.so] Error 1 gmake[2]: *** [CMakeFiles/Makefile2:1817: ggml/src/ggml-hip/CMakeFiles/ggml-hip.dir/all] Error 2 gmake[1]: *** [CMakeFiles/Makefile2:3997: tools/server/CMakeFiles/llama-server.dir/rule] Error 2 gmake: *** [Makefile:1479: llama-server] Error 2 ``` 3. 自己编译llvm 15.0.0,准备补上rt库 ```bash git clone --branch llvmorg-15.0.0 --depth 1 https://github.com/llvm/llvm-project.git sudo apt update sudo apt install -y cmake ninja-build python3 git build-essential \ libffi-dev libedit-dev zlib1g-dev libncurses5-dev libxml2-dev cd llvm-project mkdir build && cd build cmake -G Ninja ../llvm \ -DLLVM_ENABLE_PROJECTS="clang;clang-tools-extra;lld" \ -DLLVM_ENABLE_RUNTIMES="compiler-rt" \ -DCMAKE_BUILD_TYPE=Release \ -DCMAKE_INSTALL_PREFIX=/opt/clang-15.0.0 ``` 4. 复制到lib `cp ./lib/clang/15.0.0/lib/x86_64-unknown-linux-gnu/libclang_rt.builtins.a /opt/dtk/llvm/lib/libclang_rt.builtins-x86_64.a` 6. 重新编译 llama.cpp ```bash export LIBRARY_PATH=/opt/dtk/llvm/lib:$LIBRARY_PATH export LD_LIBRARY_PATH=/opt/dtk/llvm/lib:$LD_LIBRARY_PATH cmake --build buildx --config Release --target llama-server -- -j 128 ``` 7. 居然编译通过了 ``` ... [ 96%] Building CXX object common/CMakeFiles/common.dir/json-schema-to-grammar.cpp.o [ 96%] Building CXX object common/CMakeFiles/common.dir/regex-partial.cpp.o [ 96%] Building CXX object tools/mtmd/CMakeFiles/mtmd.dir/clip.cpp.o [ 96%] Building CXX object tools/mtmd/CMakeFiles/mtmd.dir/mtmd-audio.cpp.o [ 96%] Building CXX object common/CMakeFiles/common.dir/log.cpp.o [ 96%] Building CXX object tools/mtmd/CMakeFiles/mtmd.dir/mtmd-helper.cpp.o [ 98%] Building CXX object common/CMakeFiles/common.dir/speculative.cpp.o [ 98%] Building CXX object common/CMakeFiles/common.dir/sampling.cpp.o [ 98%] Linking CXX shared library ../../bin/libmtmd.so [ 98%] Built target mtmd [ 98%] Linking CXX static library libcommon.a [ 98%] Built target common [ 98%] Generating index.html.gz.hpp [ 98%] Generating loading.html.hpp [100%] Building CXX object tools/server/CMakeFiles/llama-server.dir/server.cpp.o [100%] Linking CXX executable ../../bin/llama-server [100%] Built target llama-server --- ``` 8. 尝试运行,成功加载Qwen2.5-Coder-14B-GGUF ``` ./buildx/bin/llama-server -m /srv/disk2/workspace/Qwen2.5-Coder-14B-GGUF/qwen2.5-coder-14b-q8_0.gguf --alias qwen2.5 -ngl 99 --port 8810 ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no ggml_cuda_init: found 8 ROCm devices: Device 0: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 Device 1: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 Device 2: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 Device 3: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 Device 4: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 Device 5: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 Device 6: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 Device 7: K100_AI, gfx928:sramecc+:xnack- (0x928), VMM: no, Wave Size: 64 build: 6276 (f7207b04) with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu system info: n_threads = 128, n_threads_batch = 128, total_threads = 255 system_info: n_threads = 128 (n_threads_batch = 128) / 255 | ROCm : NO_VMM = 1 | PEER_MAX_BATCH_SIZE = 128 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | REPACK = 1 | main: binding port with default address family main: HTTP server is listening, hostname: 127.0.0.1, port: 8810, http threads: 254 main: loading model srv load_model: loading model '/srv/disk2/workspace/Qwen2.5-Coder-14B-GGUF/qwen2.5-coder-14b-q8_0.gguf' llama_model_load_from_file_impl: using device ROCm0 (K100_AI) - 7894 MiB free llama_model_load_from_file_impl: using device ROCm1 (K100_AI) - 2193 MiB free llama_model_load_from_file_impl: using device ROCm2 (K100_AI) - 5380 MiB free llama_model_load_from_file_impl: using device ROCm3 (K100_AI) - 1827 MiB free llama_model_load_from_file_impl: using device ROCm4 (K100_AI) - 1825 MiB free llama_model_load_from_file_impl: using device ROCm5 (K100_AI) - 3094 MiB free llama_model_load_from_file_impl: using device ROCm6 (K100_AI) - 3429 MiB free llama_model_load_from_file_impl: using device ROCm7 (K100_AI) - 7945 MiB free llama_model_loader: loaded meta data with 33 key-value pairs and 579 tensors from /srv/disk2/workspace/Qwen2.5-Coder-14B-GGUF/qwen2.5-coder-14b-q8_0.gguf (version GGUF V3 (latest)) llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. llama_model_loader: - kv 0: general.architecture str = qwen2 llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.name str = Qwen2.5 Coder 14B llama_model_loader: - kv 3: general.basename str = Qwen2.5-Coder llama_model_loader: - kv 4: general.size_label str = 14B llama_model_loader: - kv 5: general.license str = apache-2.0 llama_model_loader: - kv 6: general.license.link str = https://huggingface.co/Qwen/Qwen2.5-C... llama_model_loader: - kv 7: general.base_model.count u32 = 1 llama_model_loader: - kv 8: general.base_model.0.name str = Qwen2.5 14B llama_model_loader: - kv 9: general.base_model.0.organization str = Qwen llama_model_loader: - kv 10: general.base_model.0.repo_url str = https://huggingface.co/Qwen/Qwen2.5-14B llama_model_loader: - kv 11: general.tags arr[str,5] = ["code", "qwen", "qwen-coder", "codeq... llama_model_loader: - kv 12: general.languages arr[str,1] = ["en"] llama_model_loader: - kv 13: qwen2.block_count u32 = 48 llama_model_loader: - kv 14: qwen2.context_length u32 = 32768 llama_model_loader: - kv 15: qwen2.embedding_length u32 = 5120 llama_model_loader: - kv 16: qwen2.feed_forward_length u32 = 13824 llama_model_loader: - kv 17: qwen2.attention.head_count u32 = 40 llama_model_loader: - kv 18: qwen2.attention.head_count_kv u32 = 8 llama_model_loader: - kv 19: qwen2.rope.freq_base f32 = 1000000.000000 llama_model_loader: - kv 20: qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 21: general.file_type u32 = 7 llama_model_loader: - kv 22: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 23: tokenizer.ggml.pre str = qwen2 llama_model_loader: - kv 24: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 25: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 26: tokenizer.ggml.merges arr[str,151387] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 27: tokenizer.ggml.eos_token_id u32 = 151645 llama_model_loader: - kv 28: tokenizer.ggml.padding_token_id u32 = 151643 llama_model_loader: - kv 29: tokenizer.ggml.bos_token_id u32 = 151643 llama_model_loader: - kv 30: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 31: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>... llama_model_loader: - kv 32: general.quantization_version u32 = 2 llama_model_loader: - type f32: 241 tensors llama_model_loader: - type q8_0: 338 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q8_0 print_info: file size = 14.62 GiB (8.50 BPW) load: printing all EOG tokens: load: - 151643 ('<|endoftext|>') load: - 151645 ('<|im_end|>') load: - 151662 ('<|fim_pad|>') load: - 151663 ('<|repo_name|>') load: - 151664 ('<|file_sep|>') load: special tokens cache size = 22 load: token to piece cache size = 0.9310 MB print_info: arch = qwen2 print_info: vocab_only = 0 print_info: n_ctx_train = 32768 print_info: n_embd = 5120 print_info: n_layer = 48 print_info: n_head = 40 print_info: n_head_kv = 8 print_info: n_rot = 128 print_info: n_swa = 0 print_info: is_swa_any = 0 print_info: n_embd_head_k = 128 print_info: n_embd_head_v = 128 print_info: n_gqa = 5 print_info: n_embd_k_gqa = 1024 print_info: n_embd_v_gqa = 1024 print_info: f_norm_eps = 0.0e+00 print_info: f_norm_rms_eps = 1.0e-05 print_info: f_clamp_kqv = 0.0e+00 print_info: f_max_alibi_bias = 0.0e+00 print_info: f_logit_scale = 0.0e+00 print_info: f_attn_scale = 0.0e+00 print_info: n_ff = 13824 print_info: n_expert = 0 print_info: n_expert_used = 0 print_info: causal attn = 1 print_info: pooling type = -1 print_info: rope type = 2 print_info: rope scaling = linear print_info: freq_base_train = 1000000.0 print_info: freq_scale_train = 1 print_info: n_ctx_orig_yarn = 32768 print_info: rope_finetuned = unknown print_info: model type = 14B print_info: model params = 14.77 B print_info: general.name = Qwen2.5 Coder 14B print_info: vocab type = BPE print_info: n_vocab = 152064 print_info: n_merges = 151387 print_info: BOS token = 151643 '<|endoftext|>' print_info: EOS token = 151645 '<|im_end|>' print_info: EOT token = 151645 '<|im_end|>' print_info: PAD token = 151643 '<|endoftext|>' print_info: LF token = 198 'Ċ' print_info: FIM PRE token = 151659 '<|fim_prefix|>' print_info: FIM SUF token = 151661 '<|fim_suffix|>' print_info: FIM MID token = 151660 '<|fim_middle|>' print_info: FIM PAD token = 151662 '<|fim_pad|>' print_info: FIM REP token = 151663 '<|repo_name|>' print_info: FIM SEP token = 151664 '<|file_sep|>' print_info: EOG token = 151643 '<|endoftext|>' print_info: EOG token = 151645 '<|im_end|>' print_info: EOG token = 151662 '<|fim_pad|>' print_info: EOG token = 151663 '<|repo_name|>' print_info: EOG token = 151664 '<|file_sep|>' print_info: max token length = 256 load_tensors: loading model tensors, this can take a while... (mmap = true) load_tensors: offloading 48 repeating layers to GPU load_tensors: offloading output layer to GPU load_tensors: offloaded 49/49 layers to GPU load_tensors: ROCm0 model buffer size = 3347.67 MiB load_tensors: ROCm1 model buffer size = 836.92 MiB load_tensors: ROCm2 model buffer size = 2231.78 MiB load_tensors: ROCm3 model buffer size = 836.92 MiB load_tensors: ROCm4 model buffer size = 557.95 MiB load_tensors: ROCm5 model buffer size = 1394.86 MiB load_tensors: ROCm6 model buffer size = 1394.86 MiB load_tensors: ROCm7 model buffer size = 3578.65 MiB load_tensors: CPU_Mapped model buffer size = 788.91 MiB .......................................................................................... llama_context: constructing llama_context llama_context: n_seq_max = 1 llama_context: n_ctx = 4096 llama_context: n_ctx_per_seq = 4096 llama_context: n_batch = 2048 llama_context: n_ubatch = 512 llama_context: causal_attn = 1 llama_context: flash_attn = 0 llama_context: kv_unified = false llama_context: freq_base = 1000000.0 llama_context: freq_scale = 1 llama_context: n_ctx_per_seq (4096) < n_ctx_train (32768) -- the full capacity of the model will not be utilized llama_context: ROCm_Host output buffer size = 0.58 MiB llama_kv_cache: ROCm0 KV buffer size = 192.00 MiB llama_kv_cache: ROCm1 KV buffer size = 48.00 MiB llama_kv_cache: ROCm2 KV buffer size = 128.00 MiB llama_kv_cache: ROCm3 KV buffer size = 48.00 MiB llama_kv_cache: ROCm4 KV buffer size = 32.00 MiB llama_kv_cache: ROCm5 KV buffer size = 80.00 MiB llama_kv_cache: ROCm6 KV buffer size = 80.00 MiB llama_kv_cache: ROCm7 KV buffer size = 160.00 MiB llama_kv_cache: size = 768.00 MiB ( 4096 cells, 48 layers, 1/1 seqs), K (f16): 384.00 MiB, V (f16): 384.00 MiB llama_context: pipeline parallelism enabled (n_copies=4) llama_context: ROCm0 compute buffer size = 448.02 MiB llama_context: ROCm1 compute buffer size = 448.02 MiB llama_context: ROCm2 compute buffer size = 448.02 MiB llama_context: ROCm3 compute buffer size = 448.02 MiB llama_context: ROCm4 compute buffer size = 448.02 MiB llama_context: ROCm5 compute buffer size = 448.02 MiB llama_context: ROCm6 compute buffer size = 448.02 MiB llama_context: ROCm7 compute buffer size = 448.03 MiB llama_context: ROCm_Host compute buffer size = 58.04 MiB llama_context: graph nodes = 1830 llama_context: graph splits = 9 common_init_from_params: added <|endoftext|> logit bias = -inf common_init_from_params: added <|im_end|> logit bias = -inf common_init_from_params: added <|fim_pad|> logit bias = -inf common_init_from_params: added <|repo_name|> logit bias = -inf common_init_from_params: added <|file_sep|> logit bias = -inf common_init_from_params: setting dry_penalty_last_n to ctx_size = 4096 common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable) srv init: initializing slots, n_slots = 1 slot init: id 0 | task -1 | new slot n_ctx_slot = 4096 main: model loaded main: chat template, chat_template: {%- if tools %} {{- '<|im_start|>system\n' }} {%- if messages[0]['role'] == 'system' %} {{- messages[0]['content'] }} {%- else %} {{- 'You are a helpful assistant.' }} {%- endif %} {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within XML tags:\n" }} {%- for tool in tools %} {{- "\n" }} {{- tool | tojson }} {%- endfor %} {{- "\n\n\nFor each function call, return a json object with function name and arguments within XML tags:\n\n{\"name\": , \"arguments\": }\n<|im_end|>\n" }} {%- else %} {%- if messages[0]['role'] == 'system' %} {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }} {%- else %} {{- '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }} {%- endif %} {%- endif %} {%- for message in messages %} {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %} {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }} {%- elif message.role == "assistant" %} {{- '<|im_start|>' + message.role }} {%- if message.content %} {{- '\n' + message.content }} {%- endif %} {%- for tool_call in message.tool_calls %} {%- if tool_call.function is defined %} {%- set tool_call = tool_call.function %} {%- endif %} {{- '\n\n{"name": "' }} {{- tool_call.name }} {{- '", "arguments": ' }} {{- tool_call.arguments | tojson }} {{- '}\n' }} {%- endfor %} {{- '<|im_end|>\n' }} {%- elif message.role == "tool" %} {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %} {{- '<|im_start|>user' }} {%- endif %} {{- '\n\n' }} {{- message.content }} {{- '\n' }} {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} {{- '<|im_end|>\n' }} {%- endif %} {%- endif %} {%- endfor %} {%- if add_generation_prompt %} {{- '<|im_start|>assistant\n' }} {%- endif %} , example_format: '<|im_start|>system You are a helpful assistant<|im_end|> <|im_start|>user Hello<|im_end|> <|im_start|>assistant Hi there<|im_end|> <|im_start|>user How are you?<|im_end|> <|im_start|>assistant ' main: server is listening on http://127.0.0.1:8810 - starting the main loop srv update_slots: all slots are idle ``` 9. 尝试推理报错 客户端 ``` curl http://127.0.0.1:8810/v1/chat/completions \ -H "Content-Type: application/json" \ -d '{ "model": "llama-2-7b", "messages": [ {"role": "user", "content": "你好,请简单介绍一下你自己。"} ], "temperature": 0.7 }' curl: (52) Empty reply from server ``` 服务端 ``` srv params_from_: Chat format: Content-only slot launch_slot_: id 0 | task 0 | processing task slot update_slots: id 0 | task 0 | new prompt, n_ctx_slot = 4096, n_keep = 0, n_prompt_tokens = 14 slot update_slots: id 0 | task 0 | kv cache rm [0, end) slot update_slots: id 0 | task 0 | prompt processing progress, n_past = 14, n_tokens = 14, progress = 1.000000 slot update_slots: id 0 | task 0 | prompt done, n_past = 14, n_tokens = 14 Invalid address access: 0x7ef726456000, Error code: 1. >>>>>>>> KERNEL VMFault !!!! <<<<<< >>>>>>>> PID: 3086204, SIGNAL: 0 !!!! <<<<<< =========================================================================================== =========> HOSTQUEUE <0x55b21576c520>: VMFault HSA QUEUE ANALYSIS <========= HOSTQUEUE <0x55b21576c520>: get hsa queue W/R ptr: write index: 869, read index: 860 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 860 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0xb02 HOSTQUEUE <0x55b21576c520>: setup: 3 HOSTQUEUE <0x55b21576c520>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b21576c520>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b21576c520>: group_segment_size: 43072 HOSTQUEUE <0x55b21576c520>: private_segment_size: 240 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x0 HOSTQUEUE <0x55b21576c520>: kernel_object: 0x7efb6471a380 HOSTQUEUE <0x55b21576c520>: device id: 0 HOSTQUEUE <0x55b21576c520>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b21576c520>: kernel name: _ZL9mul_mat_qIL9ggml_type8ELi16ELb0EEvPKcPKiS4_S4_PfS5_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS: size: 116 <<<<<<<<< 00 40 bf 21 f7 7e 00 00 00 b0 c0 8f f9 7e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 f0 61 fa 7e 00 00 00 00 e0 5d f9 7e 00 00 00 14 00 00 00 36 00 00 0e 00 00 00 a0 00 00 00 0e 00 00 00 00 36 00 00 01 00 00 00 01 00 00 00 00 c0 21 00 c0 4e 00 00 00 f4 02 00 01 00 00 00 01 00 00 00 00 c0 21 00 c0 4e 00 00 00 f4 02 00 0e 00 00 00 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b21576c520>: ptr arg index: 0, ptr: 0x7ef721bf4000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef655000000, size byte: 3510288384 HOSTQUEUE <0x55b21576c520>: ptr arg index: 1, ptr: 0x7ef98fc0b000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef98fc0b000, size byte: 104192 HOSTQUEUE <0x55b21576c520>: ptr arg index: 4, ptr: 0x7efa61f06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 5, ptr: 0x7ef95de00000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef95de00000, size byte: 1032192 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 861 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0xb02 HOSTQUEUE <0x55b21576c520>: setup: 3 HOSTQUEUE <0x55b21576c520>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b21576c520>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b21576c520>: group_segment_size: 64 HOSTQUEUE <0x55b21576c520>: private_segment_size: 0 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x0 HOSTQUEUE <0x55b21576c520>: kernel_object: 0x7efb6471a400 HOSTQUEUE <0x55b21576c520>: device id: 0 HOSTQUEUE <0x55b21576c520>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b21576c520>: kernel name: _ZL24mul_mat_q_stream_k_fixupIL9ggml_type8ELi16ELb0EEvPKiS2_PfPKfiiiiiiiii HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS: size: 68 <<<<<<<<< 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 f0 61 fa 7e 00 00 00 00 e0 5d f9 7e 00 00 00 14 00 00 00 36 00 00 0e 00 00 00 00 36 00 00 01 00 00 00 00 f4 02 00 01 00 00 00 00 f4 02 00 0e 00 00 00 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b21576c520>: ptr arg index: 2, ptr: 0x7efa61f06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 3, ptr: 0x7ef95de00000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef95de00000, size byte: 1032192 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 862 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0xb02 HOSTQUEUE <0x55b21576c520>: setup: 3 HOSTQUEUE <0x55b21576c520>: workgroup: x:256, y:1, z:1 HOSTQUEUE <0x55b21576c520>: grid: x:193536, y:1, z:1 HOSTQUEUE <0x55b21576c520>: group_segment_size: 0 HOSTQUEUE <0x55b21576c520>: private_segment_size: 0 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x0 HOSTQUEUE <0x55b21576c520>: kernel_object: 0x7f026d86bdc0 HOSTQUEUE <0x55b21576c520>: device id: 0 HOSTQUEUE <0x55b21576c520>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b21576c520>: kernel name: _ZL21unary_gated_op_kernelIXadL_ZL7op_silufEEfEvPKT0_S2_PS0_llll HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS: size: 56 <<<<<<<<< 00 60 40 60 fa 7e 00 00 00 60 f0 61 fa 7e 00 00 00 60 a0 63 fa 7e 00 00 00 f4 02 00 00 00 00 00 00 36 00 00 00 00 00 00 00 36 00 00 00 00 00 00 00 36 00 00 00 00 00 00 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b21576c520>: ptr arg index: 0, ptr: 0x7efa60406000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 1, ptr: 0x7efa61f06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 2, ptr: 0x7efa63a06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 863 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0xb02 HOSTQUEUE <0x55b21576c520>: setup: 3 HOSTQUEUE <0x55b21576c520>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b21576c520>: grid: x:1792, y:27, z:1 HOSTQUEUE <0x55b21576c520>: group_segment_size: 0 HOSTQUEUE <0x55b21576c520>: private_segment_size: 0 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x0 HOSTQUEUE <0x55b21576c520>: kernel_object: 0x7f026dea13c0 HOSTQUEUE <0x55b21576c520>: device id: 0 HOSTQUEUE <0x55b21576c520>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b21576c520>: kernel name: _ZL17quantize_mmq_q8_1IL18mmq_q8_1_ds_layout0EEvPKfPKiPvlllllii HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS: size: 72 <<<<<<<<< 00 60 a0 63 fa 7e 00 00 00 00 00 00 00 00 00 00 00 00 e0 5d f9 7e 00 00 00 36 00 00 00 00 00 00 00 36 00 00 00 00 00 00 00 f4 02 00 00 00 00 00 00 f4 02 00 00 00 00 00 00 36 00 00 00 00 00 00 0e 00 00 00 01 00 00 00 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b21576c520>: ptr arg index: 0, ptr: 0x7efa63a06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 2, ptr: 0x7ef95de00000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef95de00000, size byte: 1032192 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 864 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0xb02 HOSTQUEUE <0x55b21576c520>: setup: 3 HOSTQUEUE <0x55b21576c520>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b21576c520>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b21576c520>: group_segment_size: 43072 HOSTQUEUE <0x55b21576c520>: private_segment_size: 240 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x0 HOSTQUEUE <0x55b21576c520>: kernel_object: 0x7efb6471a380 HOSTQUEUE <0x55b21576c520>: device id: 0 HOSTQUEUE <0x55b21576c520>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b21576c520>: kernel name: _ZL9mul_mat_qIL9ggml_type8ELi16ELb0EEvPKcPKiS4_S4_PfS5_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS: size: 116 <<<<<<<<< 00 c0 43 1d f7 7e 00 00 00 00 e0 5d f9 7e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 a0 5f fa 7e 00 00 00 50 c2 8f f9 7e 00 00 00 36 00 00 00 14 00 00 0e 00 00 00 b0 01 00 00 0e 00 00 00 00 14 00 00 01 00 00 00 01 00 00 00 00 c0 21 00 a0 d4 00 00 00 18 01 00 01 00 00 00 01 00 00 00 00 c0 21 00 a0 d4 00 00 00 18 01 00 0e 00 00 00 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b21576c520>: ptr arg index: 0, ptr: 0x7ef71d43c000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef655000000, size byte: 3510288384 HOSTQUEUE <0x55b21576c520>: ptr arg index: 1, ptr: 0x7ef95de00000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef95de00000, size byte: 1032192 HOSTQUEUE <0x55b21576c520>: ptr arg index: 4, ptr: 0x7efa5fa06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 5, ptr: 0x7ef98fc25000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef98fc25000, size byte: 1032192 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 865 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0xb02 HOSTQUEUE <0x55b21576c520>: setup: 3 HOSTQUEUE <0x55b21576c520>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b21576c520>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b21576c520>: group_segment_size: 64 HOSTQUEUE <0x55b21576c520>: private_segment_size: 0 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x0 HOSTQUEUE <0x55b21576c520>: kernel_object: 0x7efb6471a400 HOSTQUEUE <0x55b21576c520>: device id: 0 HOSTQUEUE <0x55b21576c520>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b21576c520>: kernel name: _ZL24mul_mat_q_stream_k_fixupIL9ggml_type8ELi16ELb0EEvPKiS2_PfPKfiiiiiiiii HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS: size: 68 <<<<<<<<< 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 a0 5f fa 7e 00 00 00 50 c2 8f f9 7e 00 00 00 36 00 00 00 14 00 00 0e 00 00 00 00 14 00 00 01 00 00 00 00 18 01 00 01 00 00 00 00 18 01 00 0e 00 00 00 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b21576c520>: ptr arg index: 2, ptr: 0x7efa5fa06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 3, ptr: 0x7ef98fc25000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7ef98fc25000, size byte: 1032192 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 866 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0xb02 HOSTQUEUE <0x55b21576c520>: setup: 3 HOSTQUEUE <0x55b21576c520>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b21576c520>: grid: x:35840, y:1, z:1 HOSTQUEUE <0x55b21576c520>: group_segment_size: 0 HOSTQUEUE <0x55b21576c520>: private_segment_size: 0 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x0 HOSTQUEUE <0x55b21576c520>: kernel_object: 0x7f026de6fb80 HOSTQUEUE <0x55b21576c520>: device id: 0 HOSTQUEUE <0x55b21576c520>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b21576c520>: kernel name: _ZL11k_bin_bcastIXadL_ZL6op_addffEEfffEvPKT0_PKT1_PT2_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS: size: 92 <<<<<<<<< 00 60 a0 5f fa 7e 00 00 00 60 00 5f fa 7e 00 00 00 60 a0 5f fa 7e 00 00 00 18 01 00 01 00 00 00 01 00 00 00 01 00 00 00 00 18 01 00 01 00 00 00 01 00 00 00 01 00 00 00 00 18 01 00 00 18 01 00 00 18 01 00 00 18 01 00 00 18 01 00 00 18 01 00 00 18 01 00 00 18 01 00 00 18 01 00 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b21576c520>: ptr arg index: 0, ptr: 0x7efa5fa06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 1, ptr: 0x7efa5f006000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: ptr arg index: 2, ptr: 0x7efa5fa06000 HOSTQUEUE <0x55b21576c520>: origin ptr: 0x7efa59800000, size byte: 469786624 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 867 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH BARRIER AND COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0x1503 HOSTQUEUE <0x55b21576c520>: dep_signal[0]: 0x0 HOSTQUEUE <0x55b21576c520>: dep_signal[1]: 0x0 HOSTQUEUE <0x55b21576c520>: dep_signal[2]: 0x0 HOSTQUEUE <0x55b21576c520>: dep_signal[3]: 0x0 HOSTQUEUE <0x55b21576c520>: dep_signal[4]: 0x0 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x7f036f3ea480 HOSTQUEUE <0x55b21576c520>: >>>>>>>> DUMP AQL 0x7f026df9a000 PACKET WITH INDEX 868 <<<<<<<<< HOSTQUEUE <0x55b21576c520>: >>>>>>>> MATCH BARRIER AND COMMAND <<<<<<<<< HOSTQUEUE <0x55b21576c520>: header: 0x1503 HOSTQUEUE <0x55b21576c520>: dep_signal[0]: 0x7f036f3ea400 HOSTQUEUE <0x55b21576c520>: dep_signal[1]: 0x0 HOSTQUEUE <0x55b21576c520>: dep_signal[2]: 0x0 HOSTQUEUE <0x55b21576c520>: dep_signal[3]: 0x0 HOSTQUEUE <0x55b21576c520>: dep_signal[4]: 0x0 HOSTQUEUE <0x55b21576c520>: completion_signal: 0x7f036f3ea380 =========================================================================================== =========> HOSTQUEUE <0x55b21732b9e0>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2177337f0>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b215770ef0>: VMFault HSA QUEUE ANALYSIS <========= HOSTQUEUE <0x55b215770ef0>: get hsa queue W/R ptr: write index: 111, read index: 93 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 93 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH BARRIER AND COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0x103 HOSTQUEUE <0x55b215770ef0>: dep_signal[0]: 0x7f036f3ea380 HOSTQUEUE <0x55b215770ef0>: dep_signal[1]: 0x0 HOSTQUEUE <0x55b215770ef0>: dep_signal[2]: 0x0 HOSTQUEUE <0x55b215770ef0>: dep_signal[3]: 0x0 HOSTQUEUE <0x55b215770ef0>: dep_signal[4]: 0x0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x7f036f3e9100 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 94 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:1024, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:14336, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 128 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026de8b140 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL12rms_norm_f32ILi1024ELb1EEvPKfPfilllfS1_llliiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 104 <<<<<<<<< 00 00 e0 3d fa 7e 00 00 00 60 c0 42 fa 7e 00 00 00 14 00 00 00 00 00 00 00 14 00 00 00 00 00 00 00 18 01 00 00 00 00 00 00 18 01 00 00 00 00 00 ac c5 27 37 00 00 00 00 00 00 40 17 f6 7e 00 00 00 14 00 00 00 00 00 00 00 14 00 00 00 00 00 00 00 14 00 00 00 00 00 00 00 14 00 00 01 00 00 00 01 00 00 00 01 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa3de00000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7efa42c06000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 7, ptr: 0x7ef617400000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef617400000, size byte: 877572096 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 95 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:1792, y:10, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026de493c0 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL17quantize_mmq_q8_1IL18mmq_q8_1_ds_layout0EEvPKfPKiPvlllllii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 72 <<<<<<<<< 00 60 c0 42 fa 7e 00 00 00 00 00 00 00 00 00 00 00 b0 20 8f f9 7e 00 00 00 14 00 00 00 00 00 00 00 14 00 00 00 00 00 00 00 18 01 00 00 00 00 00 00 18 01 00 00 00 00 00 00 14 00 00 00 00 00 00 0e 00 00 00 01 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa42c06000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7ef98f20b000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f20b000, size byte: 104192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 96 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 43072 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 240 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7efb5ff1a380 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL9mul_mat_qIL9ggml_type8ELi16ELb0EEvPKcPKiS4_S4_PfS5_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 116 <<<<<<<<< 00 50 40 17 f6 7e 00 00 00 b0 20 8f f9 7e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 60 43 fa 7e 00 00 00 50 22 8f f9 7e 00 00 00 14 00 00 00 14 00 00 0e 00 00 00 a0 00 00 00 0e 00 00 00 00 14 00 00 01 00 00 00 01 00 00 00 00 80 0c 00 c0 4e 00 00 00 18 01 00 01 00 00 00 01 00 00 00 00 80 0c 00 c0 4e 00 00 00 18 01 00 0e 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7ef617405000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef617400000, size byte: 877572096 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7ef98f20b000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f20b000, size byte: 104192 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 4, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 5, ptr: 0x7ef98f225000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f225000, size byte: 1032192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 97 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 64 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7efb5ff1a400 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL24mul_mat_q_stream_k_fixupIL9ggml_type8ELi16ELb0EEvPKiS2_PfPKfiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 68 <<<<<<<<< 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 60 43 fa 7e 00 00 00 50 22 8f f9 7e 00 00 00 14 00 00 00 14 00 00 0e 00 00 00 00 14 00 00 01 00 00 00 00 18 01 00 01 00 00 00 00 18 01 00 0e 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 3, ptr: 0x7ef98f225000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f225000, size byte: 1032192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 98 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:2560, y:14, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026d2cfb80 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL11k_bin_bcastIXadL_ZL6op_addffEEfffEvPKT0_PKT1_PT2_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 92 <<<<<<<<< 00 60 60 43 fa 7e 00 00 00 50 3c 1b f6 7e 00 00 00 60 60 43 fa 7e 00 00 00 14 00 00 0e 00 00 00 01 00 00 00 01 00 00 00 00 14 00 00 01 00 00 00 01 00 00 00 01 00 00 00 00 14 00 00 00 18 01 00 00 18 01 00 00 14 00 00 00 18 01 00 00 18 01 00 00 14 00 00 00 14 00 00 00 14 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7ef61b3c5000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef617400000, size byte: 877572096 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 99 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:1, y:256, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:560, y:256, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026d24ae00 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL9rope_neoxILb1ELb0EfEvPKT1_PS0_iiiiiPKifff14rope_corr_dimsfPKf HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 80 <<<<<<<<< 00 60 60 43 fa 7e 00 00 00 60 00 44 fa 7e 00 00 80 00 00 00 28 00 00 00 80 00 00 00 00 14 00 00 80 00 00 00 00 00 00 00 00 08 c0 3f fa 7e 00 00 00 00 80 3f 00 00 00 00 00 00 80 3f 00 00 b8 41 00 00 20 42 ac 4b 4e 3f 00 00 00 00 00 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7efa44006000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 5, ptr: 0x7efa3fc00800 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 100 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:1792, y:10, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026de493c0 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL17quantize_mmq_q8_1IL18mmq_q8_1_ds_layout0EEvPKfPKiPvlllllii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 72 <<<<<<<<< 00 60 c0 42 fa 7e 00 00 00 00 00 00 00 00 00 00 00 b0 20 8f f9 7e 00 00 00 14 00 00 00 00 00 00 00 14 00 00 00 00 00 00 00 18 01 00 00 00 00 00 00 18 01 00 00 00 00 00 00 14 00 00 00 00 00 00 0e 00 00 00 01 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa42c06000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7ef98f20b000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f20b000, size byte: 104192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 101 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 43072 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 240 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7efb5ff1a380 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL9mul_mat_qIL9ggml_type8ELi16ELb0EEvPKcPKiS4_S4_PfS5_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 116 <<<<<<<<< 00 50 e9 18 f6 7e 00 00 00 b0 20 8f f9 7e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 60 43 fa 7e 00 00 00 50 22 8f f9 7e 00 00 00 14 00 00 00 04 00 00 0e 00 00 00 a0 00 00 00 0e 00 00 00 00 04 00 00 01 00 00 00 01 00 00 00 00 80 02 00 c0 4e 00 00 00 38 00 00 01 00 00 00 01 00 00 00 00 80 02 00 c0 4e 00 00 00 38 00 00 0e 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7ef618e95000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef617400000, size byte: 877572096 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7ef98f20b000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f20b000, size byte: 104192 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 4, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 5, ptr: 0x7ef98f225000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f225000, size byte: 1032192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 102 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 64 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7efb5ff1a400 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL24mul_mat_q_stream_k_fixupIL9ggml_type8ELi16ELb0EEvPKiS2_PfPKfiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 68 <<<<<<<<< 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 60 43 fa 7e 00 00 00 50 22 8f f9 7e 00 00 00 14 00 00 00 04 00 00 0e 00 00 00 00 04 00 00 01 00 00 00 00 38 00 00 01 00 00 00 00 38 00 00 0e 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 3, ptr: 0x7ef98f225000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f225000, size byte: 1032192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 103 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:512, y:14, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026d2cfb80 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL11k_bin_bcastIXadL_ZL6op_addffEEfffEvPKT0_PKT1_PT2_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 92 <<<<<<<<< 00 60 60 43 fa 7e 00 00 00 a0 3c 1b f6 7e 00 00 00 60 60 43 fa 7e 00 00 00 04 00 00 0e 00 00 00 01 00 00 00 01 00 00 00 00 04 00 00 01 00 00 00 01 00 00 00 01 00 00 00 00 04 00 00 00 38 00 00 00 38 00 00 00 04 00 00 00 38 00 00 00 38 00 00 00 04 00 00 00 04 00 00 00 04 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7ef61b3ca000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef617400000, size byte: 877572096 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 104 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:1, y:256, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:112, y:256, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026d24ae00 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL9rope_neoxILb1ELb0EfEvPKT1_PS0_iiiiiPKifff14rope_corr_dimsfPKf HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 80 <<<<<<<<< 00 60 60 43 fa 7e 00 00 00 60 80 43 fa 7e 00 00 80 00 00 00 08 00 00 00 80 00 00 00 00 04 00 00 80 00 00 00 00 00 00 00 00 08 c0 3f fa 7e 00 00 00 00 80 3f 00 00 00 00 00 00 80 3f 00 00 b8 41 00 00 20 42 ac 4b 4e 3f 00 00 00 00 00 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7efa43806000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 5, ptr: 0x7efa3fc00800 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 105 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:1792, y:10, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026de493c0 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL17quantize_mmq_q8_1IL18mmq_q8_1_ds_layout0EEvPKfPKiPvlllllii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 72 <<<<<<<<< 00 60 c0 42 fa 7e 00 00 00 00 00 00 00 00 00 00 00 b0 20 8f f9 7e 00 00 00 14 00 00 00 00 00 00 00 14 00 00 00 00 00 00 00 18 01 00 00 00 00 00 00 18 01 00 00 00 00 00 00 14 00 00 00 00 00 00 0e 00 00 00 01 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa42c06000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7ef98f20b000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f20b000, size byte: 104192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 106 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 43072 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 240 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7efb5ff1a380 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL9mul_mat_qIL9ggml_type8ELi16ELb0EEvPKcPKiS4_S4_PfS5_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 116 <<<<<<<<< 00 50 3e 19 f6 7e 00 00 00 b0 20 8f f9 7e 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 60 43 fa 7e 00 00 00 50 22 8f f9 7e 00 00 00 14 00 00 00 04 00 00 0e 00 00 00 a0 00 00 00 0e 00 00 00 00 04 00 00 01 00 00 00 01 00 00 00 00 80 02 00 c0 4e 00 00 00 38 00 00 01 00 00 00 01 00 00 00 00 80 02 00 c0 4e 00 00 00 38 00 00 0e 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7ef6193e5000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef617400000, size byte: 877572096 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7ef98f20b000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f20b000, size byte: 104192 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 4, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 5, ptr: 0x7ef98f225000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f225000, size byte: 1032192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 107 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:64, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:7680, y:8, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 64 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7efb5ff1a400 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL24mul_mat_q_stream_k_fixupIL9ggml_type8ELi16ELb0EEvPKiS2_PfPKfiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 68 <<<<<<<<< 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 60 60 43 fa 7e 00 00 00 50 22 8f f9 7e 00 00 00 14 00 00 00 04 00 00 0e 00 00 00 00 04 00 00 01 00 00 00 00 38 00 00 01 00 00 00 00 38 00 00 0e 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 3, ptr: 0x7ef98f225000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef98f225000, size byte: 1032192 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 108 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:128, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:512, y:14, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026d2cfb80 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL11k_bin_bcastIXadL_ZL6op_addffEEfffEvPKT0_PKT1_PT2_iiiiiiiiiiiiiiiii HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 92 <<<<<<<<< 00 60 60 43 fa 7e 00 00 00 b0 3c 1b f6 7e 00 00 00 60 60 43 fa 7e 00 00 00 04 00 00 0e 00 00 00 01 00 00 00 01 00 00 00 00 04 00 00 01 00 00 00 01 00 00 00 01 00 00 00 00 04 00 00 00 38 00 00 00 38 00 00 00 04 00 00 00 38 00 00 00 38 00 00 00 04 00 00 00 04 00 00 00 04 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7ef61b3cb000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7ef617400000, size byte: 877572096 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 109 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:256, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:14336, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026d823c40 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL10k_set_rowsIf6__halfEvPKT_PKlPT0_lllllllllllllllll HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 160 <<<<<<<<< 00 60 80 43 fa 7e 00 00 00 30 c0 3f fa 7e 00 00 00 00 00 bf fa 7e 00 00 00 04 00 00 00 00 00 00 0e 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 0e 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 00 04 00 00 00 00 00 00 00 38 00 00 00 00 00 00 00 38 00 00 00 00 00 00 01 00 00 00 00 00 00 00 0e 00 00 00 00 00 00 00 0e 00 00 00 00 00 00 00 00 04 00 00 00 00 00 00 00 00 40 00 00 00 00 00 00 00 40 00 00 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa43806000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7efa3fc03000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efabf000000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efabf000000, size byte: 50331648 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP AQL 0x7f026defc000 PACKET WITH INDEX 110 <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: >>>>>>>> MATCH KERNEL COMMAND <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: header: 0xb02 HOSTQUEUE <0x55b215770ef0>: setup: 3 HOSTQUEUE <0x55b215770ef0>: workgroup: x:256, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: grid: x:14336, y:1, z:1 HOSTQUEUE <0x55b215770ef0>: group_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: private_segment_size: 0 HOSTQUEUE <0x55b215770ef0>: completion_signal: 0x0 HOSTQUEUE <0x55b215770ef0>: kernel_object: 0x7f026d823c40 HOSTQUEUE <0x55b215770ef0>: device id: 1 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> FIND MATCH KERNEL DETAILS <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: kernel name: _ZL10k_set_rowsIf6__halfEvPKT_PKlPT0_lllllllllllllllll HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS: size: 160 <<<<<<<<< 00 60 60 43 fa 7e 00 00 00 60 00 40 fa 7e 00 00 00 00 80 bf fa 7e 00 00 01 00 00 00 00 00 00 00 00 38 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 00 38 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 00 38 00 00 00 00 00 00 00 38 00 00 00 00 00 00 01 00 00 00 00 00 00 00 00 38 00 00 00 00 00 00 00 38 00 00 00 00 00 00 01 00 00 00 00 00 00 00 00 00 40 00 00 00 00 00 00 00 40 00 00 00 00 00 HOSTQUEUE <0x55b215770ef0>: >>>>>>>> DUMP KERNEL ARGS PTR INFO <<<<<<<<< HOSTQUEUE <0x55b215770ef0>: ptr arg index: 0, ptr: 0x7efa43606000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 1, ptr: 0x7efa40006000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efa3d400000, size byte: 469786624 HOSTQUEUE <0x55b215770ef0>: ptr arg index: 2, ptr: 0x7efabf800000 HOSTQUEUE <0x55b215770ef0>: origin ptr: 0x7efabf000000, size byte: 50331648 =========================================================================================== =========> HOSTQUEUE <0x55b217520b40>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217842df0>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b21756b210>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b215776020>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217948080>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2156c8a40>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2175b5e80>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217a4d510>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2156cdb50>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217600ae0>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217b52980>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2156d2c80>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217c57e60>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b21764b890>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217d5d520>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2156d7dd0>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217696680>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2156dff10>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b217e62aa0>: VMFault HSA QUEUE ANALYSIS <========= =========================================================================================== =========> HOSTQUEUE <0x55b2176e1430>: VMFault HSA QUEUE ANALYSIS <========= >>>>>>>> KERNEL VMFault Analysis END !!!! <<<<<< >>>>>>>> KERNEL VMFault Analysis END !!!! <<<<<< Aborted (core dumped) ```
全部评论(3)
只看楼主
Z
zc
1楼
你好,我使用dtk25.04.2的镜像(docker pull image.sourcefind.cn:5000/dcu/admin/base/pytorch:2.5.1-ubuntu22.04-dtk25.04.2-py3.10),按照opendas上面的操作可以正常编译运行,请求能够正常返回,其中编译的时候操作为:export LIBRARY_PATH=/opt/dtk/llvm/lib/clang/17.0.0/lib/linux/:$LIBRARY_PATH,您可以再试一下
2025/11/20
_
_mg32p2jatogh
回复 @zc
这个版本的llama.cpp不支持qwen3moe模型,llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen3moe',有没有官方移植的更新版本的llama.cpp提供一下
1/12
_
_mg32p2jatogh
回复 @zc
你好,我在使用您回答中的镜像时启动时报错llama_new_context_with_model: graph splits = 3
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
ggml_cuda_compute_forward: RMS_NORM failed
ROCm error: invalid device function
  current device: 0, in function ggml_cuda_compute_forward at /data/llama.cpp/ggml/src/ggml-cuda/ggml-cuda.cu:2174
  err
/data/llama.cpp/ggml/src/ggml-cuda/ggml-cuda.cu:70: ROCm error
[New LWP 7076]
[New LWP 7983]
[New LWP 7984]
[New LWP 7985]
[New LWP 7986]
能否解答一下:weiqu:emoji
1/12
没有更多内容了
--
楼主
Level.4
0 帖子
楼主热帖
暂无数据