llama.cpp
llama.cpp copied to clipboard
Published
1 month ago
•
ggml-org
ggml-org
Potential memory allocation leak
Name and Version
docker exec -it llama-server ./llama-server --version
ggml_vulkan: Found 1 Vulkan devices: ggml_vulkan: 0 = AMD Radeon RX 6600 XT (RADV NAVI23) (radv) | uma: 0 | fp16: 1 | warp size: 32 | shared memory: 65536 | matrix cores: none version: 4943 (18b663d8) built with cc (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0 for x86_64-linux-gnu
Operating systems
Linux
Which llama.cpp modules do you know to be affected?
llama-server
Command line
--mlock or --flash-attn did not help with or without :(
services:
llama-server:
#image: ghcr.io/ggml-org/llama.cpp:server-vulkan
image: llama-server
container_name: llama-server
devices:
- /dev/dri/renderD128:/dev/dri/renderD128
- /dev/dri/card1:/dev/dri/card1
ports:
- 8080:8080
volumes:
- ./Models:/Models
entrypoint: /app/llama-server -m /Models/google_gemma-3-12b-it-Q4_K_M.gguf --port 8080 --ctx-size 16384 --gpu-layers 15 --temp 1.0 --top-k 64 --top-p 0.95 --mlock # --flash-attn
restart: unless-stopped
Problem description & steps to reproduce
Hello guy, after few long prompt in different conversation, Gemma start to going VERY slow like 0.1 token/s and the M.2 is used at 100% in read even if my ram is not fully used and my vram too.
Potential memory allocation leak ? (i am very good to describe the problem, sorry 😄)
As you can see in my RAM screen, memory "Commitée" is maybe the cause of the error because every time model reload (like i stop the container docker stop llama-server or kill the container docker rm -f llama-server) this memory "commitée" continue to be more and more huge
Here is the model load after few prompt
RAM with only model load (not in generation) :
VRAM with only model load (not in generation) :
SSD M.2 (not in generation) :
During generation with the problem of ram after few prompt :
RAM with only model load (in generation) :
VRAM with only model load (in generation) :
M.2 with only model load (in generation) :
After closing and reopening my computer session (ubuntu 24.10) before first model loading:
RAM with no model load :
VRAM with no model load :
M.2 with no model load :
After closing and reopening my computer session (ubuntu 24.10) with only first model (during generation) :
Speed :
RAM after reboot (in generation) :
VRAM after reboot (in generation) :
M.2 after reboot (in generation) :
First Bad Commit
No response
Relevant log output
After reboot without any problem generation :
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = AMD Radeon RX 6600 XT (RADV NAVI23) (radv) | uma: 0 | fp16: 1 | warp size: 32 | shared memory: 65536 | matrix cores: none
build: 4943 (18b663d8) with cc (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0 for x86_64-linux-gnu
system info: n_threads = 6, n_threads_batch = 6, total_threads = 12
system_info: n_threads = 6 (n_threads_batch = 6) / 12 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | AARCH64_REPACK = 1 |
main: HTTP server is listening, hostname: 0.0.0.0, port: 8080, http threads: 11
main: loading model
srv load_model: loading model '/Models/google_gemma-3-12b-it-Q4_K_M.gguf'
llama_model_load_from_file_impl: using device Vulkan0 (AMD Radeon RX 6600 XT (RADV NAVI23)) - 8176 MiB free
llama_model_loader: loaded meta data with 44 key-value pairs and 626 tensors from /Models/google_gemma-3-12b-it-Q4_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 = gemma3
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = Gemma 3 12b It
llama_model_loader: - kv 3: general.finetune str = it
llama_model_loader: - kv 4: general.basename str = gemma-3
llama_model_loader: - kv 5: general.size_label str = 12B
llama_model_loader: - kv 6: general.license str = gemma
llama_model_loader: - kv 7: general.base_model.count u32 = 1
llama_model_loader: - kv 8: general.base_model.0.name str = Gemma 3 12b Pt
llama_model_loader: - kv 9: general.base_model.0.organization str = Google
llama_model_loader: - kv 10: general.base_model.0.repo_url str = https://huggingface.co/google/gemma-3...
llama_model_loader: - kv 11: general.tags arr[str,1] = ["image-text-to-text"]
llama_model_loader: - kv 12: gemma3.context_length u32 = 131072
llama_model_loader: - kv 13: gemma3.embedding_length u32 = 3840
llama_model_loader: - kv 14: gemma3.block_count u32 = 48
llama_model_loader: - kv 15: gemma3.feed_forward_length u32 = 15360
llama_model_loader: - kv 16: gemma3.attention.head_count u32 = 16
llama_model_loader: - kv 17: gemma3.attention.layer_norm_rms_epsilon f32 = 0.000001
llama_model_loader: - kv 18: gemma3.attention.key_length u32 = 256
llama_model_loader: - kv 19: gemma3.attention.value_length u32 = 256
llama_model_loader: - kv 20: gemma3.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 21: gemma3.attention.sliding_window u32 = 1024
llama_model_loader: - kv 22: gemma3.attention.head_count_kv u32 = 8
llama_model_loader: - kv 23: gemma3.rope.scaling.type str = linear
llama_model_loader: - kv 24: gemma3.rope.scaling.factor f32 = 8.000000
llama_model_loader: - kv 25: tokenizer.ggml.model str = llama
llama_model_loader: - kv 26: tokenizer.ggml.pre str = default
llama_model_loader: - kv 27: tokenizer.ggml.tokens arr[str,262144] = ["<pad>", "<eos>", "<bos>", "<unk>", ...
llama_model_loader: - kv 28: tokenizer.ggml.scores arr[f32,262144] = [-1000.000000, -1000.000000, -1000.00...
llama_model_loader: - kv 29: tokenizer.ggml.token_type arr[i32,262144] = [3, 3, 3, 3, 3, 4, 3, 3, 3, 3, 3, 3, ...
llama_model_loader: - kv 30: tokenizer.ggml.bos_token_id u32 = 2
llama_model_loader: - kv 31: tokenizer.ggml.eos_token_id u32 = 1
llama_model_loader: - kv 32: tokenizer.ggml.unknown_token_id u32 = 3
llama_model_loader: - kv 33: tokenizer.ggml.padding_token_id u32 = 0
llama_model_loader: - kv 34: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 35: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 36: tokenizer.chat_template str = {{ bos_token }}\n{%- if messages[0]['r...
llama_model_loader: - kv 37: tokenizer.ggml.add_space_prefix bool = false
llama_model_loader: - kv 38: general.quantization_version u32 = 2
llama_model_loader: - kv 39: general.file_type u32 = 15
llama_model_loader: - kv 40: quantize.imatrix.file str = /models_out/gemma-3-12b-it-GGUF/googl...
llama_model_loader: - kv 41: quantize.imatrix.dataset str = /training_dir/calibration_datav3.txt
llama_model_loader: - kv 42: quantize.imatrix.entries_count i32 = 336
llama_model_loader: - kv 43: quantize.imatrix.chunks_count i32 = 129
llama_model_loader: - type f32: 289 tensors
llama_model_loader: - type q4_K: 288 tensors
llama_model_loader: - type q6_K: 49 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Medium
print_info: file size = 6.79 GiB (4.96 BPW)
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: special tokens cache size = 6414
load: token to piece cache size = 1.9446 MB
print_info: arch = gemma3
print_info: vocab_only = 0
print_info: n_ctx_train = 131072
print_info: n_embd = 3840
print_info: n_layer = 48
print_info: n_head = 16
print_info: n_head_kv = 8
print_info: n_rot = 256
print_info: n_swa = 1024
print_info: n_swa_pattern = 6
print_info: n_embd_head_k = 256
print_info: n_embd_head_v = 256
print_info: n_gqa = 2
print_info: n_embd_k_gqa = 2048
print_info: n_embd_v_gqa = 2048
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-06
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 = 6.2e-02
print_info: n_ff = 15360
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: causal attn = 1
print_info: pooling type = 0
print_info: rope type = 2
print_info: rope scaling = linear
print_info: freq_base_train = 1000000.0
print_info: freq_scale_train = 0.125
print_info: n_ctx_orig_yarn = 131072
print_info: rope_finetuned = unknown
print_info: ssm_d_conv = 0
print_info: ssm_d_inner = 0
print_info: ssm_d_state = 0
print_info: ssm_dt_rank = 0
print_info: ssm_dt_b_c_rms = 0
print_info: model type = 12B
print_info: model params = 11.77 B
print_info: general.name = Gemma 3 12b It
print_info: vocab type = SPM
print_info: n_vocab = 262144
print_info: n_merges = 0
print_info: BOS token = 2 '<bos>'
print_info: EOS token = 1 '<eos>'
print_info: EOT token = 106 '<end_of_turn>'
print_info: UNK token = 3 '<unk>'
print_info: PAD token = 0 '<pad>'
print_info: LF token = 248 '<0x0A>'
print_info: EOG token = 1 '<eos>'
print_info: EOG token = 106 '<end_of_turn>'
print_info: max token length = 48
load_tensors: loading model tensors, this can take a while... (mmap = true)
make_cpu_buft_list: disabling extra buffer types (i.e. repacking) since a GPU device is available
load_tensors: offloading 15 repeating layers to GPU
load_tensors: offloaded 15/49 layers to GPU
load_tensors: Vulkan0 model buffer size = 1952.34 MiB
load_tensors: CPU_Mapped model buffer size = 5003.78 MiB
............................warning: failed to mlock 832323584-byte buffer (after previously locking 0 bytes): Cannot allocate memory
Try increasing RLIMIT_MEMLOCK ('ulimit -l' as root).
..............................................................
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 16384
llama_context: n_ctx_per_seq = 16384
llama_context: n_batch = 2048
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = 0
llama_context: freq_base = 1000000.0
llama_context: freq_scale = 0.125
llama_context: n_ctx_per_seq (16384) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 1.00 MiB
init: kv_size = 16384, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 48, can_shift = 1
init: Vulkan0 KV buffer size = 1920.00 MiB
init: CPU KV buffer size = 4224.00 MiB
llama_context: KV self size = 6144.00 MiB, K (f16): 3072.00 MiB, V (f16): 3072.00 MiB
llama_context: Vulkan0 compute buffer size = 1314.53 MiB
llama_context: Vulkan_Host compute buffer size = 72.01 MiB
llama_context: graph nodes = 2023
llama_context: graph splits = 565 (with bs=512), 4 (with bs=1)
common_init_from_params: setting dry_penalty_last_n to ctx_size = 16384
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 = 16384
main: model loaded
main: chat template, chat_template: {{ bos_token }}
{%- if messages[0]['role'] == 'system' -%}
{%- if messages[0]['content'] is string -%}
{%- set first_user_prefix = messages[0]['content'] + '
' -%}
{%- else -%}
{%- set first_user_prefix = messages[0]['content'][0]['text'] + '
' -%}
{%- endif -%}
{%- set loop_messages = messages[1:] -%}
{%- else -%}
{%- set first_user_prefix = "" -%}
{%- set loop_messages = messages -%}
{%- endif -%}
{%- for message in loop_messages -%}
{%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
{{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
{%- endif -%}
{%- if (message['role'] == 'assistant') -%}
{%- set role = "model" -%}
{%- else -%}
{%- set role = message['role'] -%}
{%- endif -%}
{{ '<start_of_turn>' + role + '
' + (first_user_prefix if loop.first else "") }}
{%- if message['content'] is string -%}
{{ message['content'] | trim }}
{%- elif message['content'] is iterable -%}
{%- for item in message['content'] -%}
{%- if item['type'] == 'image' -%}
{{ '<start_of_image>' }}
{%- elif item['type'] == 'text' -%}
{{ item['text'] | trim }}
{%- endif -%}
{%- endfor -%}
{%- else -%}
{{ raise_exception("Invalid content type") }}
{%- endif -%}
{{ '<end_of_turn>
' }}
{%- endfor -%}
{%- if add_generation_prompt -%}
{{'<start_of_turn>model
'}}
{%- endif -%}
, example_format: '<start_of_turn>user
You are a helpful assistant
Hello<end_of_turn>
<start_of_turn>model
Hi there<end_of_turn>
<start_of_turn>user
How are you?<end_of_turn>
<start_of_turn>model
'
main: server is listening on http://0.0.0.0:8080 - starting the main loop
srv update_slots: all slots are idle
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
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 = 16384, n_keep = 0, n_prompt_tokens = 438
slot update_slots: id 0 | task 0 | kv cache rm [0, end)
slot update_slots: id 0 | task 0 | prompt processing progress, n_past = 438, n_tokens = 438, progress = 1.000000
slot update_slots: id 0 | task 0 | prompt done, n_past = 438, n_tokens = 438
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
slot release: id 0 | task 0 | stop processing: n_past = 1246, truncated = 0
slot print_timing: id 0 | task 0 |
prompt eval time = 4265.06 ms / 438 tokens ( 9.74 ms per token, 102.70 tokens per second)
eval time = 147795.38 ms / 809 tokens ( 182.69 ms per token, 5.47 tokens per second)
total time = 152060.44 ms / 1247 tokens
srv log_server_r: request: POST /v1/chat/completions 172.18.0.1 200
srv update_slots: all slots are idle
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
After few prompt with problem generation (log start by my previous and last request that didn't use M.2 to generate answer) :
srv log_server_r: request: POST /v1/chat/completions 172.18.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 16901 | processing task
slot update_slots: id 0 | task 16901 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 2014
slot update_slots: id 0 | task 16901 | kv cache rm [11, end)
slot update_slots: id 0 | task 16901 | prompt processing progress, n_past = 2014, n_tokens = 2003, progress = 0.994538
slot update_slots: id 0 | task 16901 | prompt done, n_past = 2014, n_tokens = 2003
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv log_server_r: request: GET /health 127.0.0.1 200
srv cancel_tasks: cancel task, id_task = 16901
srv log_server_r: request: POST /v1/chat/completions 172.18.0.1 200
