ipex-llm run DeepSeek-R1-Q4_K_M on hybrid CPU+ARC770 raiss coredump

[xtangxtang]

follow the guide https://github.com/intel/ipex-llm/blob/main/docs/mddocs/Quickstart/llama_cpp_quickstart.md to run DeepSeek-R1-Q4_K_M on hybrid model (INTEL(R) XEON(R) PLATINUM 8592 + ARC770) occurs coredump

The prepare work is

pip install --pre --upgrade ipex-llm[cpp]
mkdir llama-cpp
cd llama-cpp
init-llama-cpp

The env has been set

(ipex-llm) root@emr15195:/home/xtang/llama-cpp# echo $SYCL_CACHE_PERSISTENT
1
(ipex-llm) root@emr15195:/home/xtang/llama-cpp# echo $SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS
1
(ipex-llm) root@emr15195:/home/xtang/llama-cpp# echo $ONEAPI_DEVICE_SELECTOR
level_zero:0

The run command is

numactl -N 0 -m 0 ./llama-cli -m /nvme0/models/DeepSeek-R1-GGUF/DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00001-of-00009.gguf -n 128 --prompt "A conversation between User and Assistant. The user asks a question, and the Assistant solves it. The assistant first thinks about the reasoning process in the mind and then provides the user with the answer. The reasoning process and answer are enclosed within <think> </think> and <answer> </answer> tags, respectively, i.e., <think> reasoning process here </think> <answer> answer here </answer>. User: Question: If \( a > 1 \), then the sum of the real solutions of \( \sqrt{a} - \sqrt{a + x} = x \) is equal to:. Assistant: <think>" -t 48 -e -ngl 2 --color -c 512 --temp 0 --no-context-shift -ot exps=CPU

the output is

ggml_sycl_init: GGML_SYCL_FORCE_MMQ:   no
ggml_sycl_init: SYCL_USE_XMX: yes
ggml_sycl_init: found 1 SYCL devices:
build: 1 (e66308a) with Intel(R) oneAPI DPC++/C++ Compiler 2025.0.4 (2025.0.4.20241205) for x86_64-unknown-linux-gnu
main: llama backend init
main: load the model and apply lora adapter, if any
llama_load_model_from_file: using device SYCL0 (Intel(R) Arc(TM) A770 Graphics) - 15473 MiB free
llama_model_loader: additional 8 GGUFs metadata loaded.
llama_model_loader: loaded meta data with 48 key-value pairs and 1025 tensors from /nvme0/models/DeepSeek-R1-GGUF/DeepSeek-R1-Q4_K_M/DeepSeek-R1-Q4_K_M-00001-of-00009.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              = deepseek2
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 BF16
llama_model_loader: - kv   3:                       general.quantized_by str              = Unsloth
llama_model_loader: - kv   4:                         general.size_label str              = 256x20B
llama_model_loader: - kv   5:                           general.repo_url str              = https://huggingface.co/unsloth
llama_model_loader: - kv   6:                      deepseek2.block_count u32              = 61
llama_model_loader: - kv   7:                   deepseek2.context_length u32              = 163840
llama_model_loader: - kv   8:                 deepseek2.embedding_length u32              = 7168
llama_model_loader: - kv   9:              deepseek2.feed_forward_length u32              = 18432
llama_model_loader: - kv  10:             deepseek2.attention.head_count u32              = 128
llama_model_loader: - kv  11:          deepseek2.attention.head_count_kv u32              = 128
llama_model_loader: - kv  12:                   deepseek2.rope.freq_base f32              = 10000.000000
llama_model_loader: - kv  13: deepseek2.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  14:                deepseek2.expert_used_count u32              = 8
llama_model_loader: - kv  15:        deepseek2.leading_dense_block_count u32              = 3
llama_model_loader: - kv  16:                       deepseek2.vocab_size u32              = 129280
llama_model_loader: - kv  17:            deepseek2.attention.q_lora_rank u32              = 1536
llama_model_loader: - kv  18:           deepseek2.attention.kv_lora_rank u32              = 512
llama_model_loader: - kv  19:             deepseek2.attention.key_length u32              = 192
llama_model_loader: - kv  20:           deepseek2.attention.value_length u32              = 128
llama_model_loader: - kv  21:       deepseek2.expert_feed_forward_length u32              = 2048
llama_model_loader: - kv  22:                     deepseek2.expert_count u32              = 256
llama_model_loader: - kv  23:              deepseek2.expert_shared_count u32              = 1
llama_model_loader: - kv  24:             deepseek2.expert_weights_scale f32              = 2.500000
llama_model_loader: - kv  25:              deepseek2.expert_weights_norm bool             = true
llama_model_loader: - kv  26:               deepseek2.expert_gating_func u32              = 2
llama_model_loader: - kv  27:             deepseek2.rope.dimension_count u32              = 64
llama_model_loader: - kv  28:                deepseek2.rope.scaling.type str              = yarn
llama_model_loader: - kv  29:              deepseek2.rope.scaling.factor f32              = 40.000000
llama_model_loader: - kv  30: deepseek2.rope.scaling.original_context_length u32              = 4096
llama_model_loader: - kv  31: deepseek2.rope.scaling.yarn_log_multiplier f32              = 0.100000
llama_model_loader: - kv  32:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  33:                         tokenizer.ggml.pre str              = deepseek-v3
llama_model_loader: - kv  34:                      tokenizer.ggml.tokens arr[str,129280]  = ["<｜begin▁of▁sentence｜>", "<▒...
llama_model_loader: - kv  35:                  tokenizer.ggml.token_type arr[i32,129280]  = [3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  36:                      tokenizer.ggml.merges arr[str,127741]  = ["Ġ t", "Ġ a", "i n", "Ġ Ġ", "h e...
llama_model_loader: - kv  37:                tokenizer.ggml.bos_token_id u32              = 0
llama_model_loader: - kv  38:                tokenizer.ggml.eos_token_id u32              = 1
llama_model_loader: - kv  39:            tokenizer.ggml.padding_token_id u32              = 128815
llama_model_loader: - kv  40:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  41:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  42:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
llama_model_loader: - kv  43:               general.quantization_version u32              = 2
llama_model_loader: - kv  44:                          general.file_type u32              = 15
llama_model_loader: - kv  45:                                   split.no u16              = 0
llama_model_loader: - kv  46:                        split.tensors.count i32              = 1025
llama_model_loader: - kv  47:                                split.count u16              = 9
llama_model_loader: - type  f32:  361 tensors
llama_model_loader: - type q4_K:  606 tensors
llama_model_loader: - type q6_K:   58 tensors
llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
llm_load_vocab: special tokens cache size = 819
llm_load_vocab: token to piece cache size = 0.8223 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = deepseek2
llm_load_print_meta: vocab type       = BPE
llm_load_print_meta: n_vocab          = 129280
llm_load_print_meta: n_merges         = 127741
llm_load_print_meta: vocab_only       = 0
llm_load_print_meta: n_ctx_train      = 163840
llm_load_print_meta: n_embd           = 7168
llm_load_print_meta: n_layer          = 61
llm_load_print_meta: n_head           = 128
llm_load_print_meta: n_head_kv        = 128
llm_load_print_meta: n_rot            = 64
llm_load_print_meta: n_swa            = 0
llm_load_print_meta: n_embd_head_k    = 192
llm_load_print_meta: n_embd_head_v    = 128
llm_load_print_meta: n_gqa            = 1
llm_load_print_meta: n_embd_k_gqa     = 24576
llm_load_print_meta: n_embd_v_gqa     = 16384
llm_load_print_meta: f_norm_eps       = 0.0e+00
llm_load_print_meta: f_norm_rms_eps   = 1.0e-06
llm_load_print_meta: f_clamp_kqv      = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: f_logit_scale    = 0.0e+00
llm_load_print_meta: n_ff             = 18432
llm_load_print_meta: n_expert         = 256
llm_load_print_meta: n_expert_used    = 8
llm_load_print_meta: causal attn      = 1
llm_load_print_meta: pooling type     = 0
llm_load_print_meta: rope type        = 0
llm_load_print_meta: rope scaling     = yarn
llm_load_print_meta: freq_base_train  = 10000.0
llm_load_print_meta: freq_scale_train = 0.025
llm_load_print_meta: n_ctx_orig_yarn  = 4096
llm_load_print_meta: rope_finetuned   = unknown
llm_load_print_meta: ssm_d_conv       = 0
llm_load_print_meta: ssm_d_inner      = 0
llm_load_print_meta: ssm_d_state      = 0
llm_load_print_meta: ssm_dt_rank      = 0
llm_load_print_meta: ssm_dt_b_c_rms   = 0
llm_load_print_meta: model type       = 671B
llm_load_print_meta: model ftype      = Q4_K - Medium
llm_load_print_meta: model params     = 671.03 B
llm_load_print_meta: model size       = 376.65 GiB (4.82 BPW)
llm_load_print_meta: general.name     = DeepSeek R1 BF16
llm_load_print_meta: BOS token        = 0 '<｜begin▁of▁sentence｜>'
llm_load_print_meta: EOS token        = 1 '<｜end▁of▁sentence｜>'
llm_load_print_meta: EOT token        = 1 '<｜end▁of▁sentence｜>'
llm_load_print_meta: PAD token        = 128815 '<｜PAD▁TOKEN｜>'
llm_load_print_meta: LF token         = 131 'Ä'
llm_load_print_meta: FIM PRE token    = 128801 '<｜fim▁begin｜>'
llm_load_print_meta: FIM SUF token    = 128800 '<｜fim▁hole｜>'
llm_load_print_meta: FIM MID token    = 128802 '<｜fim▁end｜>'
llm_load_print_meta: EOG token        = 1 '<｜end▁of▁sentence｜>'
llm_load_print_meta: max token length = 256
llm_load_print_meta: n_layer_dense_lead   = 3
llm_load_print_meta: n_lora_q             = 1536
llm_load_print_meta: n_lora_kv            = 512
llm_load_print_meta: n_ff_exp             = 2048
llm_load_print_meta: n_expert_shared      = 1
llm_load_print_meta: expert_weights_scale = 2.5
llm_load_print_meta: expert_weights_norm  = 1
llm_load_print_meta: expert_gating_func   = sigmoid
llm_load_print_meta: rope_yarn_log_mul    = 0.1000
llm_load_tensors: offloading 2 repeating layers to GPU
llm_load_tensors: offloaded 2/62 layers to GPU
llm_load_tensors:   CPU_Mapped model buffer size = 46095.40 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 47139.54 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 47232.92 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 46036.25 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 47132.51 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 46036.25 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 47139.54 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 44663.89 MiB
llm_load_tensors:   CPU_Mapped model buffer size = 14105.06 MiB
llm_load_tensors:        SYCL0 model buffer size =   269.34 MiB
...................................................................................................Bus error (core dumped)

[rnwang04]

Hi @xtangxtang , I can't reproduce this error with the same command on our machine (INTEL(R) XEON(R) PLATINUM 8558P + ARC A770).
Here are some tips or questions that I want to further check with you :

• Have you ever tried running it on CPU with -ngl 0 ? It is successful?
• Have you ever successfully run any model on GPU with our llama.cpp ?
• Here is a bus error related issue, not sure if it helps.
• Could you please provide us more detailed info with env-check.sh in https://github.com/intel/ipex-llm/tree/main/python/llm/scripts ?