Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

CUDA illegal memory access | AWS #3163

Closed
nkise-nlab opened this issue Sep 14, 2023 · 9 comments · Fixed by #3220
Closed

CUDA illegal memory access | AWS #3163

nkise-nlab opened this issue Sep 14, 2023 · 9 comments · Fixed by #3220

Comments

@nkise-nlab
Copy link

nkise-nlab commented Sep 14, 2023 •
edited
Loading

Expected Behavior

Running without errors and being able to utilise GPU

Current Behavior

Encountered a CUDA error during runtime:
CUDA error 700 at /llama.cpp/ggml-cuda.cu:6540: an illegal memory access was encountered

Environment and Context

  • Physical (or virtual) hardware you are using, e.g. for Linux:

Hardware: AWS instance ml.g5.12xlarge.
Operating System: x86_64 GNU/Linux
GPUs: 4xT4 16GB

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              48
On-line CPU(s) list: 0-47
Thread(s) per core:  2
Core(s) per socket:  24
Socket(s):           1
NUMA node(s):        1
Vendor ID:           AuthenticAMD
CPU family:          23
Model:               49
Model name:          AMD EPYC 7R32
Stepping:            0
CPU MHz:             2786.228
BogoMIPS:            5600.00
Hypervisor vendor:   KVM
Virtualization type: full
L1d cache:           32K
L1i cache:           32K
L2 cache:            512K
L3 cache:            16384K
NUMA node0 CPU(s):   0-47
Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf tsc_known_freq pni pclmulqdq ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch topoext ssbd ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 clzero xsaveerptr rdpru wbnoinvd arat npt nrip_save rdpid
  • SDK version, e.g. for Linux:
Python 3.11.5
GNU Make 3.82
Built for x86_64-koji-linux-gnu
g++ (GCC) 7.3.1 20180712 (Red Hat 7.3.1-17)

Failure Information

While trying to run llama.cpp, I encountered a CUDA error. This happened with both the models I tried to quantise myself and using TheBlock pre-quantised models. If I remove '-ngl' everything works fine

Models like this are running OK with '-ngl'. The issue seems to be related to GGUF v2.

Steps to Reproduce

Please provide detailed steps for reproducing the issue. We are not sitting in front of your screen, so the more detail the better.

export PATH="/usr/local/cuda-11.8/bin:$PATH"
export LD_LIBRARY_PATH="/usr/local/cuda-11.8/lib64:$LD_LIBRARY_PATH" 

mkdir build
cd build
cmake .. -DLLAMA_CUBLAS=ON
cmake --build . --config Release

./build/bin/main -t 10 -ngl 32 -m ./models/llama2_70b_chat-Q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p

Failure Logs

Example environment info:

llama.cpp$ git log | head -1
commit 71ca2fad7d6c0ef95ef9944fb3a1a843e481f314

Example command

llama.cpp$ ./build/bin/main -t 16 -ngl 16 -m ./models/7B/ggml-model-q4_0.gguf --color --ctx-size 32768 --rope-scale 4 --temp 1.2 --repeat_penalty 1.1 -n -1 -p "$PROMPT"
Log start
main: warning: scaling RoPE frequency by 0.25 (default 1.0)
main: build = 1221 (71ca2fa)
main: seed  = 1694682036
ggml_init_cublas: found 4 CUDA devices:
  Device 0: NVIDIA A10G, compute capability 8.6
  Device 1: NVIDIA A10G, compute capability 8.6
  Device 2: NVIDIA A10G, compute capability 8.6
  Device 3: NVIDIA A10G, compute capability 8.6
llama_model_loader: loaded meta data with 19 key-value pairs and 291 tensors from ./models/7B/ggml-model-q4_0.gguf (version GGUF V2 (latest))
llama_model_loader: - tensor    0:                token_embd.weight q4_0     [  4096, 32000,     1,     1 ]
llama_model_loader: - tensor    1:           blk.0.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor    2:            blk.0.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor    3:            blk.0.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor    4:              blk.0.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor    5:            blk.0.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor    6:              blk.0.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor    7:         blk.0.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor    8:              blk.0.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor    9:              blk.0.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   10:           blk.1.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   11:            blk.1.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   12:            blk.1.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   13:              blk.1.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   14:            blk.1.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   15:              blk.1.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   16:         blk.1.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   17:              blk.1.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   18:              blk.1.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   19:          blk.10.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   20:           blk.10.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   21:           blk.10.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   22:             blk.10.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   23:           blk.10.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   24:             blk.10.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   25:        blk.10.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   26:             blk.10.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   27:             blk.10.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   28:          blk.11.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   29:           blk.11.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   30:           blk.11.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   31:             blk.11.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   32:           blk.11.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   33:             blk.11.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   34:        blk.11.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   35:             blk.11.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   36:             blk.11.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   37:          blk.12.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   38:           blk.12.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   39:           blk.12.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   40:             blk.12.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   41:           blk.12.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   42:             blk.12.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   43:        blk.12.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   44:             blk.12.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   45:             blk.12.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   46:          blk.13.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   47:           blk.13.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   48:           blk.13.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   49:             blk.13.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   50:           blk.13.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   51:             blk.13.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   52:        blk.13.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   53:             blk.13.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   54:             blk.13.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   55:          blk.14.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   56:           blk.14.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   57:           blk.14.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   58:             blk.14.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   59:           blk.14.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   60:             blk.14.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   61:        blk.14.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   62:             blk.14.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   63:             blk.14.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   64:          blk.15.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   65:           blk.15.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   66:           blk.15.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   67:             blk.15.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   68:           blk.15.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   69:             blk.15.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   70:        blk.15.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   71:             blk.15.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   72:             blk.15.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   73:          blk.16.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   74:           blk.16.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   75:           blk.16.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   76:             blk.16.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   77:           blk.16.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   78:             blk.16.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   79:        blk.16.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   80:             blk.16.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   81:             blk.16.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   82:          blk.17.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   83:           blk.17.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   84:           blk.17.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   85:             blk.17.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   86:           blk.17.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   87:             blk.17.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   88:        blk.17.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   89:             blk.17.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   90:             blk.17.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   91:          blk.18.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   92:           blk.18.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor   93:           blk.18.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   94:             blk.18.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor   95:           blk.18.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor   96:             blk.18.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   97:        blk.18.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   98:             blk.18.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor   99:             blk.18.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  100:          blk.19.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  101:           blk.19.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  102:           blk.19.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  103:             blk.19.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  104:           blk.19.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  105:             blk.19.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  106:        blk.19.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  107:             blk.19.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  108:             blk.19.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  109:           blk.2.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  110:            blk.2.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  111:            blk.2.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  112:              blk.2.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  113:            blk.2.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  114:              blk.2.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  115:         blk.2.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  116:              blk.2.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  117:              blk.2.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  118:          blk.20.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  119:           blk.20.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  120:           blk.20.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  121:             blk.20.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  122:           blk.20.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  123:             blk.20.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  124:        blk.20.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  125:             blk.20.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  126:             blk.20.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  127:          blk.21.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  128:           blk.21.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  129:           blk.21.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  130:             blk.21.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  131:           blk.21.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  132:             blk.21.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  133:        blk.21.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  134:             blk.21.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  135:             blk.21.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  136:          blk.22.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  137:           blk.22.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  138:           blk.22.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  139:             blk.22.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  140:           blk.22.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  141:             blk.22.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  142:        blk.22.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  143:             blk.22.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  144:             blk.22.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  145:          blk.23.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  146:           blk.23.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  147:           blk.23.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  148:             blk.23.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  149:           blk.23.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  150:             blk.23.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  151:        blk.23.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  152:             blk.23.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  153:             blk.23.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  154:           blk.3.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  155:            blk.3.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  156:            blk.3.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  157:              blk.3.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  158:            blk.3.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  159:              blk.3.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  160:         blk.3.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  161:              blk.3.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  162:              blk.3.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  163:           blk.4.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  164:            blk.4.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  165:            blk.4.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  166:              blk.4.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  167:            blk.4.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  168:              blk.4.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  169:         blk.4.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  170:              blk.4.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  171:              blk.4.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  172:           blk.5.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  173:            blk.5.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  174:            blk.5.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  175:              blk.5.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  176:            blk.5.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  177:              blk.5.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  178:         blk.5.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  179:              blk.5.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  180:              blk.5.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  181:           blk.6.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  182:            blk.6.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  183:            blk.6.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  184:              blk.6.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  185:            blk.6.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  186:              blk.6.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  187:         blk.6.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  188:              blk.6.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  189:              blk.6.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  190:           blk.7.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  191:            blk.7.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  192:            blk.7.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  193:              blk.7.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  194:            blk.7.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  195:              blk.7.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  196:         blk.7.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  197:              blk.7.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  198:              blk.7.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  199:           blk.8.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  200:            blk.8.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  201:            blk.8.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  202:              blk.8.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  203:            blk.8.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  204:              blk.8.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  205:         blk.8.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  206:              blk.8.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  207:              blk.8.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  208:           blk.9.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  209:            blk.9.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  210:            blk.9.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  211:              blk.9.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  212:            blk.9.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  213:              blk.9.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  214:         blk.9.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  215:              blk.9.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  216:              blk.9.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  217:                    output.weight q6_K     [  4096, 32000,     1,     1 ]
llama_model_loader: - tensor  218:          blk.24.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  219:           blk.24.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  220:           blk.24.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  221:             blk.24.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  222:           blk.24.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  223:             blk.24.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  224:        blk.24.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  225:             blk.24.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  226:             blk.24.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  227:          blk.25.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  228:           blk.25.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  229:           blk.25.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  230:             blk.25.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  231:           blk.25.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  232:             blk.25.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  233:        blk.25.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  234:             blk.25.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  235:             blk.25.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  236:          blk.26.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  237:           blk.26.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  238:           blk.26.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  239:             blk.26.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  240:           blk.26.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  241:             blk.26.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  242:        blk.26.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  243:             blk.26.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  244:             blk.26.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  245:          blk.27.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  246:           blk.27.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  247:           blk.27.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  248:             blk.27.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  249:           blk.27.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  250:             blk.27.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  251:        blk.27.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  252:             blk.27.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  253:             blk.27.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  254:          blk.28.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  255:           blk.28.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  256:           blk.28.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  257:             blk.28.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  258:           blk.28.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  259:             blk.28.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  260:        blk.28.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  261:             blk.28.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  262:             blk.28.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  263:          blk.29.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  264:           blk.29.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  265:           blk.29.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  266:             blk.29.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  267:           blk.29.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  268:             blk.29.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  269:        blk.29.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  270:             blk.29.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  271:             blk.29.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  272:          blk.30.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  273:           blk.30.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  274:           blk.30.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  275:             blk.30.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  276:           blk.30.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  277:             blk.30.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  278:        blk.30.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  279:             blk.30.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  280:             blk.30.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  281:          blk.31.attn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  282:           blk.31.ffn_down.weight q4_0     [ 11008,  4096,     1,     1 ]
llama_model_loader: - tensor  283:           blk.31.ffn_gate.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  284:             blk.31.ffn_up.weight q4_0     [  4096, 11008,     1,     1 ]
llama_model_loader: - tensor  285:           blk.31.ffn_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - tensor  286:             blk.31.attn_k.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  287:        blk.31.attn_output.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  288:             blk.31.attn_q.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  289:             blk.31.attn_v.weight q4_0     [  4096,  4096,     1,     1 ]
llama_model_loader: - tensor  290:               output_norm.weight f32      [  4096,     1,     1,     1 ]
llama_model_loader: - kv   0:                       general.architecture str     
llama_model_loader: - kv   1:                               general.name str     
llama_model_loader: - kv   2:                       llama.context_length u32     
llama_model_loader: - kv   3:                     llama.embedding_length u32     
llama_model_loader: - kv   4:                          llama.block_count u32     
llama_model_loader: - kv   5:                  llama.feed_forward_length u32     
llama_model_loader: - kv   6:                 llama.rope.dimension_count u32     
llama_model_loader: - kv   7:                 llama.attention.head_count u32     
llama_model_loader: - kv   8:              llama.attention.head_count_kv u32     
llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32     
llama_model_loader: - kv  10:                          general.file_type u32     
llama_model_loader: - kv  11:                       tokenizer.ggml.model str     
llama_model_loader: - kv  12:                      tokenizer.ggml.tokens arr     
llama_model_loader: - kv  13:                      tokenizer.ggml.scores arr     
llama_model_loader: - kv  14:                  tokenizer.ggml.token_type arr     
llama_model_loader: - kv  15:                tokenizer.ggml.bos_token_id u32     
llama_model_loader: - kv  16:                tokenizer.ggml.eos_token_id u32     
llama_model_loader: - kv  17:            tokenizer.ggml.unknown_token_id u32     
llama_model_loader: - kv  18:               general.quantization_version u32     
llama_model_loader: - type  f32:   65 tensors
llama_model_loader: - type q4_0:  225 tensors
llama_model_loader: - type q6_K:    1 tensors
llm_load_print_meta: format         = GGUF V2 (latest)
llm_load_print_meta: arch           = llama
llm_load_print_meta: vocab type     = SPM
llm_load_print_meta: n_vocab        = 32000
llm_load_print_meta: n_merges       = 0
llm_load_print_meta: n_ctx_train    = 4096
llm_load_print_meta: n_ctx          = 32768
llm_load_print_meta: n_embd         = 4096
llm_load_print_meta: n_head         = 32
llm_load_print_meta: n_head_kv      = 32
llm_load_print_meta: n_layer        = 32
llm_load_print_meta: n_rot          = 128
llm_load_print_meta: n_gqa          = 1
llm_load_print_meta: f_norm_eps     = 1.0e-05
llm_load_print_meta: f_norm_rms_eps = 1.0e-06
llm_load_print_meta: n_ff           = 11008
llm_load_print_meta: freq_base      = 10000.0
llm_load_print_meta: freq_scale     = 0.25
llm_load_print_meta: model type     = 7B
llm_load_print_meta: model ftype    = mostly Q4_0
llm_load_print_meta: model size     = 6.74 B
llm_load_print_meta: general.name   = LLaMA v2
llm_load_print_meta: BOS token = 1 '<s>'
llm_load_print_meta: EOS token = 2 '</s>'
llm_load_print_meta: UNK token = 0 '<unk>'
llm_load_print_meta: LF token  = 13 '<0x0A>'
llm_load_tensors: ggml ctx size =    0.09 MB
llm_load_tensors: using CUDA for GPU acceleration
ggml_cuda_set_main_device: using device 0 (NVIDIA A10G) as main device
llm_load_tensors: mem required  = 1910.46 MB (+ 16384.00 MB per state)
llm_load_tensors: offloading 16 repeating layers to GPU
llm_load_tensors: offloaded 16/35 layers to GPU
llm_load_tensors: VRAM used: 1738 MB
..................................................................................................
llama_new_context_with_model: kv self size  = 16384.00 MB
llama_new_context_with_model: compute buffer total size = 2073.47 MB
llama_new_context_with_model: VRAM scratch buffer: 2072.00 MB

**CUDA error 700 at /home/ec2-user/SageMaker/Nikita/llama.cpp/ggml-cuda.cu:6540: an illegal memory access was encountered**
@nkise-nlab
Copy link
Author

same problems with 8xA100 instance and using make instead of cmake

@JohannesGaessler
Copy link
Collaborator

Do you still get the error with CUDA_VISIBLE_DEVICES=0?

@CarloNicolini
Copy link

Same here on a AWS g5.12xlarge instance with 4xA10G GPUs, and built with CMake

@ozreact
Copy link

ozreact commented Sep 15, 2023

Same issue attempting to split a model between a 4090 and an A6000.

Do you still get the error with CUDA_VISIBLE_DEVICES=0?

No. If I stick to a single GPU, there's no issue.

@ASH1998
Copy link

ASH1998 commented Sep 16, 2023

Do you still get the error with CUDA_VISIBLE_DEVICES=0?

No, atleast for myself, it works on CUDA_VISIBLE_DEVICES=0, but when I provide multiple GPUs I get the same error. I am using the latest commit on main.

For ex.
This works :

CUDA_VISIBLE_DEVICES=0 ./main -ins -f ./prompts/alpaca.txt -t 24 -ngl 20  -m ~/models/falcon-180b-q5ks.gguf  --color -c 4096  --temp 0.7  --repeat_penalty 1.
1  -s 42  -n -1

This does not work :

CUDA_VISIBLE_DEVICES=0,2,3 ./main -ins -f ./prompts/alpaca.txt -t 24 -ngl 128  -m ~/models/falcon-180b-q5ks.gguf  --color -c 4096  --temp 0.7  --repeat_penalty 1.
1  -s 42  -n -1

I have 4xA100-40G, so I have enough memory to run q5 falcon 180B. This was also discussed in #2160 and #1341 sometime back. I have tested the same commands on same model but ggml quantization on 1f0bccb27929e commit, it worked flawlessly.

Any updates on this is appreciated. Thanks.

@JohannesGaessler
Copy link
Collaborator

I cannot reproduce the issue on my server with 3x P40.

@JohannesGaessler JohannesGaessler mentioned this issue Sep 16, 2023
Merged
@JohannesGaessler
Copy link
Collaborator

JohannesGaessler commented Sep 16, 2023 •
edited
Loading

I think I found the issue, please check whether this fix works: #3163 #3220 .

@slaren
Copy link
Collaborator

slaren commented Sep 17, 2023

I think I found the issue, please check whether this fix works: #3163 .

I think you meant #3220.

@MihaiATK
Copy link

I also get this error[1] after I run convert.py with --outtype q8_0 and then I try to do inference on the output.
The conversion is done based on a merged in qlora on https://huggingface.co/larryvrh/Yi-34B-200K-Llamafied .

This is the command I run:
!CUDA_VISIBLE_DEVICES=0 ./llama.cpp/main -m model-q8_0.gguf -c 8888 -t 4 -ngl 2000000 --temp 0 --file prompt.txt --grammar-file grammar.gbnf

I suspect it has something to do with the 8888 context length param (although the model supports 200k context length) because if I try running it with -c 4095, it works ok.

[1]CUDA error 700 at ggml-cuda.cu:7546: an illegal memory access was encountered current device: 0

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

CUDA: fix scratch buffers being allocated on non-main device JohannesGaessler/llama.cpp
7 participants
@CarloNicolini @slaren @JohannesGaessler @ASH1998 @nkise-nlab @ozreact @MihaiATK