search

ggerganov / llama.cpp

LLM inference in C/C++
MIT License
68.35k stars 9.81k forks source link

Bug: [SYCL] SYCL + Docker #10113

Open easyfab opened 3 weeks ago

easyfab commented 3 weeks ago

What happened?

I can't use docker + SYCL when using -ngl >0 With -ngl 0 it's ok

message error : No kernel named _ZTSZZL17rms_norm_f32_syclPKfPfiifPN4sycl3_V15queueEiENKUlRNS3_7handlerEE0_clES7_EUlNS3_7nditemILi3EEEE was found -46 (PI_ERROR_INVALID_KERNEL_NAME)Exception caught at file:/app/ggml/src/ggml-sycl.cpp, line:3528

I tried local build or ghcr.io/ggerganov/llama.cpp:light-intel

For info using ipex-llm intelanalytics/ipex-llm-inference-cpp-xpu docker image is ok with HW -ngl 99

Name and Version

ghcr.io/ggerganov/llama.cpp:light-intel

What operating system are you seeing the problem on?

Linux

Relevant log output

docker run -it --rm -v "$(pwd):/app:Z" --device /dev/dri/renderD128:/dev/dri/renderD128 --device /dev/dri/card0:/dev/dri/card0  ghcr.io/ggerganov/llama.cpp:light-intel -m "/app/Qwen2.5-7B-Instruct-Q5_K_M.gguf" -p "Building a website can be done in 10 simple steps:" -n 400 -e -ngl 33 -c 2048
Unable to find image 'ghcr.io/ggerganov/llama.cpp:light-intel' locally
light-intel: Pulling from ggerganov/llama.cpp
4a023cab5400: Already exists
a0b4ab99a45e: Already exists
29bdda71b58f: Already exists
c3ed24128924: Already exists
b07d42681b62: Already exists
e9ed1c05d4b1: Already exists
64bb6f5a70f1: Already exists
715596b0c2b0: Already exists
fd64f7a4a319: Already exists
55d6a83a1280: Pull complete
Digest: sha256:dfa81744e7d242fb2c02c1bbf235cbf41fa78252a5d0b4b6adaf24083c2f792d
Status: Downloaded newer image for ghcr.io/ggerganov/llama.cpp:light-intel
ggml_sycl_init: GGML_SYCL_FORCE_MMQ:   no
ggml_sycl_init: SYCL_USE_XMX: yes
ggml_sycl_init: found 1 SYCL devices:
build: 3999 (ab3d71f9) with Intel(R) oneAPI DPC++/C++ Compiler 2024.1.2 (2024.1.2.20240508) for x86_64-unknown-linux-gnu
main: llama backend init
main: load the model and apply lora adapter, if any
get_memory_info: [warning] ext_intel_free_memory is not supported (export/set ZES_ENABLE_SYSMAN=1 to support), use total memory as free memory
llama_load_model_from_file: using device SYCL0 (Intel(R) Arc(TM) A770 Graphics) - 15473 MiB free
llama_model_loader: loaded meta data with 38 key-value pairs and 339 tensors from /app/Qwen2.5-7B-Instruct-Q5_K_M.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 7B Instruct
llama_model_loader: - kv   3:                           general.finetune str              = Instruct
llama_model_loader: - kv   4:                           general.basename str              = Qwen2.5
llama_model_loader: - kv   5:                         general.size_label str              = 7B
llama_model_loader: - kv   6:                            general.license str              = apache-2.0
llama_model_loader: - kv   7:                       general.license.link str              = https://huggingface.co/Qwen/Qwen2.5-7...
llama_model_loader: - kv   8:                   general.base_model.count u32              = 1
llama_model_loader: - kv   9:                  general.base_model.0.name str              = Qwen2.5 7B
llama_model_loader: - kv  10:          general.base_model.0.organization str              = Qwen
llama_model_loader: - kv  11:              general.base_model.0.repo_url str              = https://huggingface.co/Qwen/Qwen2.5-7B
llama_model_loader: - kv  12:                               general.tags arr[str,2]       = ["chat", "text-generation"]
llama_model_loader: - kv  13:                          general.languages arr[str,1]       = ["en"]
llama_model_loader: - kv  14:                          qwen2.block_count u32              = 28
llama_model_loader: - kv  15:                       qwen2.context_length u32              = 32768
llama_model_loader: - kv  16:                     qwen2.embedding_length u32              = 3584
llama_model_loader: - kv  17:                  qwen2.feed_forward_length u32              = 18944
llama_model_loader: - kv  18:                 qwen2.attention.head_count u32              = 28
llama_model_loader: - kv  19:              qwen2.attention.head_count_kv u32              = 4
llama_model_loader: - kv  20:                       qwen2.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  21:     qwen2.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  22:                          general.file_type u32              = 17
llama_model_loader: - kv  23:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  24:                         tokenizer.ggml.pre str              = qwen2
llama_model_loader: - kv  25:                      tokenizer.ggml.tokens arr[str,152064]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  26:                  tokenizer.ggml.token_type arr[i32,152064]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  27:                      tokenizer.ggml.merges arr[str,151387]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  28:                tokenizer.ggml.eos_token_id u32              = 151645
llama_model_loader: - kv  29:            tokenizer.ggml.padding_token_id u32              = 151643
llama_model_loader: - kv  30:                tokenizer.ggml.bos_token_id u32              = 151643
llama_model_loader: - kv  31:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  32:                    tokenizer.chat_template str              = {%- if tools %}\n    {{- '<|im_start|>...
llama_model_loader: - kv  33:               general.quantization_version u32              = 2
llama_model_loader: - kv  34:                      quantize.imatrix.file str              = /models_out/Qwen2.5-7B-Instruct-GGUF/...
llama_model_loader: - kv  35:                   quantize.imatrix.dataset str              = /training_dir/calibration_datav3.txt
llama_model_loader: - kv  36:             quantize.imatrix.entries_count i32              = 196
llama_model_loader: - kv  37:              quantize.imatrix.chunks_count i32              = 128
llama_model_loader: - type  f32:  141 tensors
llama_model_loader: - type q5_K:  169 tensors
llama_model_loader: - type q6_K:   29 tensors
llm_load_vocab: special tokens cache size = 22
llm_load_vocab: token to piece cache size = 0.9310 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = qwen2
llm_load_print_meta: vocab type       = BPE
llm_load_print_meta: n_vocab          = 152064
llm_load_print_meta: n_merges         = 151387
llm_load_print_meta: vocab_only       = 0
llm_load_print_meta: n_ctx_train      = 32768
llm_load_print_meta: n_embd           = 3584
llm_load_print_meta: n_layer          = 28
llm_load_print_meta: n_head           = 28
llm_load_print_meta: n_head_kv        = 4
llm_load_print_meta: n_rot            = 128
llm_load_print_meta: n_swa            = 0
llm_load_print_meta: n_embd_head_k    = 128
llm_load_print_meta: n_embd_head_v    = 128
llm_load_print_meta: n_gqa            = 7
llm_load_print_meta: n_embd_k_gqa     = 512
llm_load_print_meta: n_embd_v_gqa     = 512
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             = 18944
llm_load_print_meta: n_expert         = 0
llm_load_print_meta: n_expert_used    = 0
llm_load_print_meta: causal attn      = 1
llm_load_print_meta: pooling type     = 0
llm_load_print_meta: rope type        = 2
llm_load_print_meta: rope scaling     = linear
llm_load_print_meta: freq_base_train  = 1000000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_ctx_orig_yarn  = 32768
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       = ?B
llm_load_print_meta: model ftype      = Q5_K - Medium
llm_load_print_meta: model params     = 7.62 B
llm_load_print_meta: model size       = 5.07 GiB (5.71 BPW)
llm_load_print_meta: general.name     = Qwen2.5 7B Instruct
llm_load_print_meta: BOS token        = 151643 '<|endoftext|>'
llm_load_print_meta: EOS token        = 151645 '<|im_end|>'
llm_load_print_meta: EOT token        = 151645 '<|im_end|>'
llm_load_print_meta: PAD token        = 151643 '<|endoftext|>'
llm_load_print_meta: LF token         = 148848 'ÄĬ'
llm_load_print_meta: FIM PRE token    = 151659 '<|fim_prefix|>'
llm_load_print_meta: FIM SUF token    = 151661 '<|fim_suffix|>'
llm_load_print_meta: FIM MID token    = 151660 '<|fim_middle|>'
llm_load_print_meta: FIM PAD token    = 151662 '<|fim_pad|>'
llm_load_print_meta: FIM REP token    = 151663 '<|repo_name|>'
llm_load_print_meta: FIM SEP token    = 151664 '<|file_sep|>'
llm_load_print_meta: EOG token        = 151643 '<|endoftext|>'
llm_load_print_meta: EOG token        = 151645 '<|im_end|>'
llm_load_print_meta: EOG token        = 151662 '<|fim_pad|>'
llm_load_print_meta: EOG token        = 151663 '<|repo_name|>'
llm_load_print_meta: EOG token        = 151664 '<|file_sep|>'
llm_load_print_meta: max token length = 256
get_memory_info: [warning] ext_intel_free_memory is not supported (export/set ZES_ENABLE_SYSMAN=1 to support), use total memory as free memory
get_memory_info: [warning] ext_intel_free_memory is not supported (export/set ZES_ENABLE_SYSMAN=1 to support), use total memory as free memory
llm_load_tensors: offloading 28 repeating layers to GPU
llm_load_tensors: offloading output layer to GPU
llm_load_tensors: offloaded 29/29 layers to GPU
llm_load_tensors: CPU_Mapped model buffer size =   357.33 MiB
llm_load_tensors:      SYCL0 model buffer size =  4829.59 MiB
.......................................................................................
llama_new_context_with_model: n_ctx      = 2048
llama_new_context_with_model: n_batch    = 2048
llama_new_context_with_model: n_ubatch   = 512
llama_new_context_with_model: flash_attn = 0
llama_new_context_with_model: freq_base  = 1000000.0
llama_new_context_with_model: freq_scale = 1
[SYCL] call ggml_check_sycl
ggml_check_sycl: GGML_SYCL_DEBUG: 0
ggml_check_sycl: GGML_SYCL_F16: no
found 1 SYCL devices:
|  |                   |                                       |       |Max    |        |Max  |Global |                     |
|  |                   |                                       |       |compute|Max work|sub  |mem    |                     |
|ID|        Device Type|                                   Name|Version|units  |group   |group|size   |       Driver version|
|--|-------------------|---------------------------------------|-------|-------|--------|-----|-------|---------------------|
| 0| [level_zero:gpu:0]|                Intel Arc A770 Graphics|    1.3|    512|    1024|   32| 16225M|            1.3.27642|
llama_kv_cache_init:      SYCL0 KV buffer size =   112.00 MiB
llama_new_context_with_model: KV self size  =  112.00 MiB, K (f16):   56.00 MiB, V (f16):   56.00 MiB
llama_new_context_with_model:  SYCL_Host  output buffer size =     0.58 MiB
llama_new_context_with_model:      SYCL0 compute buffer size =   304.00 MiB
llama_new_context_with_model:  SYCL_Host compute buffer size =    11.01 MiB
llama_new_context_with_model: graph nodes  = 986
llama_new_context_with_model: graph splits = 2
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
No kernel named _ZTSZZL17rms_norm_f32_syclPKfPfiifPN4sycl3_V15queueEiENKUlRNS3_7handlerEE0_clES7_EUlNS3_7nd_itemILi3EEEE_ was found -46 (PI_ERROR_INVALID_KERNEL_NAME)Exception caught at file:/app/ggml/src/ggml-sycl.cpp, line:3528
NeoZhangJianyu commented 3 weeks ago

@easyfab Looks like the oneAPI version issue. The binary is built with 2024.1.2.20240508. In the running-time, there should be oneAPI 2024.1.2.20240508 in same time.

easyfab commented 3 weeks ago

So if I understand correctly, we need to modify .devops/llama-server-intel.Dockerfile and .devops/llama-server-intel.Dockerfile ?

ARG ONEAPI_VERSION=2024.1.1-devel-ubuntu22.04 to ~ARG ONEAPI_VERSION=2024.1.2-devel-ubuntu22.04~

Edit : 2024.2.0-1-devel-ubuntu22.04 or 2024.2.1-0-devel-ubuntu22.04 if I look at https://hub.docker.com/r/intel/oneapi/tags

I am a total beginner with git. Could you or someone else submit a PR please.

easyfab commented 3 weeks ago

For info, for me it's the static build that give this error with a shared build it's ok.

I modify the dockerfile like this : I removed :

-DBUILD_SHARED_LIBS=OFF

and add :

COPY --from=build /app/build/ggml/src/libggml.so /libggml.so COPY --from=build /app/build/src/libllama.so /libllama.so

I can close the issue if I'm the only one with this issue.

kkacsh321 commented 3 weeks ago

I have been chasing this same error with a newer A770 while trying to use LocalAI (which uses llama.cpp) I'll have to see if can reproduce