search

ggerganov / llama.cpp

LLM inference in C/C++
MIT License
64.9k stars 9.31k forks source link

HelloGGML_ASSERT: ggml-metal.m:539: false && "not implemented" #1693

Closed phil-aid closed 1 year ago

phil-aid commented 1 year ago

Prerequisites

Please answer the following questions for yourself before submitting an issue.

Expected Behavior

Please provide a detailed written description of what you were trying to do, and what you expected llama.cpp to do.

Current Behavior

Please provide a detailed written description of what llama.cpp did, instead.

Environment and Context

Please provide detailed information about your computer setup. This is important in case the issue is not reproducible except for under certain specific conditions.

$ lscpu

$ uname -a

$ python3 --version
$ make --version
$ g++ --version

Failure Information (for bugs)

Please help provide information about the failure if this is a bug. If it is not a bug, please remove the rest of this template.

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.

  1. step 1
  2. step 2
  3. step 3
  4. etc.

Failure Logs

Please include any relevant log snippets or files. If it works under one configuration but not under another, please provide logs for both configurations and their corresponding outputs so it is easy to see where behavior changes.

Also, please try to avoid using screenshots if at all possible. Instead, copy/paste the console output and use Github's markdown to cleanly format your logs for easy readability.

Example environment info:

llama.cpp$ git log | head -1
commit 2af23d30434a677c6416812eea52ccc0af65119c

llama.cpp$ lscpu | egrep "AMD|Flags"
Vendor ID:                       AuthenticAMD
Model name:                      AMD Ryzen Threadripper 1950X 16-Core Processor
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 amd_dcm aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb hw_pstate ssbd ibpb vmmcall fsgsbase bmi1 avx2 smep bmi2 rdseed adx smap clflushopt sha_ni xsaveopt xsavec xgetbv1 xsaves clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif overflow_recov succor smca sme sev
Virtualization:                  AMD-V

llama.cpp$ python3 --version
Python 3.10.9

llama.cpp$ pip list | egrep "torch|numpy|sentencepiece"
numpy                         1.24.2
numpydoc                      1.5.0
sentencepiece                 0.1.97
torch                         1.13.1
torchvision                   0.14.1

llama.cpp$ make --version | head -1
GNU Make 4.3

$ md5sum ./models/65B/ggml-model-q4_0.bin
dbdd682cce80e2d6e93cefc7449df487  ./models/65B/ggml-model-q4_0.bin

Example run with the Linux command perf

llama.cpp$ perf stat ./main -m ./models/65B/ggml-model-q4_0.bin -t 16 -n 1024 -p "Please close your issue when it has been answered."
main: seed = 1679149377
llama_model_load: loading model from './models/65B/ggml-model-q4_0.bin' - please wait ...
llama_model_load: n_vocab = 32000
llama_model_load: n_ctx   = 512
llama_model_load: n_embd  = 8192
llama_model_load: n_mult  = 256
llama_model_load: n_head  = 64
llama_model_load: n_layer = 80
llama_model_load: n_rot   = 128
llama_model_load: f16     = 2
llama_model_load: n_ff    = 22016
llama_model_load: n_parts = 8
llama_model_load: ggml ctx size = 41477.73 MB
llama_model_load: memory_size =  2560.00 MB, n_mem = 40960
llama_model_load: loading model part 1/8 from './models/65B/ggml-model-q4_0.bin'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723
llama_model_load: loading model part 2/8 from './models/65B/ggml-model-q4_0.bin.1'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723
llama_model_load: loading model part 3/8 from './models/65B/ggml-model-q4_0.bin.2'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723
llama_model_load: loading model part 4/8 from './models/65B/ggml-model-q4_0.bin.3'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723
llama_model_load: loading model part 5/8 from './models/65B/ggml-model-q4_0.bin.4'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723
llama_model_load: loading model part 6/8 from './models/65B/ggml-model-q4_0.bin.5'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723
llama_model_load: loading model part 7/8 from './models/65B/ggml-model-q4_0.bin.6'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723
llama_model_load: loading model part 8/8 from './models/65B/ggml-model-q4_0.bin.7'
llama_model_load: .......................................................................................... done
llama_model_load: model size =  4869.09 MB / num tensors = 723

system_info: n_threads = 16 / 32 | AVX = 1 | AVX2 = 1 | AVX512 = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | VSX = 0 |

main: prompt: 'Please close your issue when it has been answered.'
main: number of tokens in prompt = 11
     1 -> ''
 12148 -> 'Please'
  3802 -> ' close'
   596 -> ' your'
  2228 -> ' issue'
   746 -> ' when'
   372 -> ' it'
   756 -> ' has'
  1063 -> ' been'
  7699 -> ' answered'
 29889 -> '.'

sampling parameters: temp = 0.800000, top_k = 40, top_p = 0.950000, repeat_last_n = 64, repeat_penalty = 1.300000

Please close your issue when it has been answered.
@duncan-donut: I'm trying to figure out what kind of "support" you need for this script and why, exactly? Is there a question about how the code works that hasn't already been addressed in one or more comments below this ticket, or are we talking something else entirely like some sorta bugfixing job because your server setup is different from mine??
I can understand if your site needs to be running smoothly and you need help with a fix of sorts but there should really be nothing wrong here that the code itself could not handle. And given that I'm getting reports about how it works perfectly well on some other servers, what exactly are we talking? A detailed report will do wonders in helping us get this resolved for ya quickly so please take your time and describe the issue(s) you see as clearly & concisely as possible!!
@duncan-donut: I'm not sure if you have access to cPanel but you could try these instructions. It is worth a shot! Let me know how it goes (or what error message, exactly!) when/if ya give that code a go? [end of text]

main: mem per token = 71159620 bytes
main:     load time = 19309.95 ms
main:   sample time =   168.62 ms
main:  predict time = 223895.61 ms / 888.47 ms per token
main:    total time = 246406.42 ms

 Performance counter stats for './main -m ./models/65B/ggml-model-q4_0.bin -t 16 -n 1024 -p Please close your issue when it has been answered.':

        3636882.89 msec task-clock                #   14.677 CPUs utilized
             13509      context-switches          #    3.714 /sec
              2436      cpu-migrations            #    0.670 /sec
          10476679      page-faults               #    2.881 K/sec
    13133115082869      cycles                    #    3.611 GHz                      (16.77%)
       29314462753      stalled-cycles-frontend   #    0.22% frontend cycles idle     (16.76%)
    10294402631459      stalled-cycles-backend    #   78.39% backend cycles idle      (16.74%)
    23479217109614      instructions              #    1.79  insn per cycle
                                                  #    0.44  stalled cycles per insn  (16.76%)
     2353072268027      branches                  #  647.002 M/sec                    (16.77%)
        1998682780      branch-misses             #    0.08% of all branches          (16.76%)

     247.802177522 seconds time elapsed

    3618.573072000 seconds user
      18.491698000 seconds sys
alienatorZ commented 1 year ago

I am also getting this error.

abc-nix commented 1 year ago

This seems to be a duplicate of #1697. Are you trying to load anything other than a q4_0 quantized model? This is not supported yet for metal (only q4_0 is supported right now).

alienatorZ commented 1 year ago

I am getting nonsense when using Metal. Without it the models perform normally. Is anyone else seeing this? Here is my output:

(base) adam@adams-mbp bin % ./main -m /Users/adam/Documents/Projects/langchainstufff/llama.cpp/models/guanaco-7B.ggmlv3.q4_0.bin -p "I believe the meaning of life is " --ignore-eos -ngl 1 main: build = 613 (827f5ed) main: seed = 1686008097 llama.cpp: loading model from /Users/adam/Documents/Projects/langchainstufff/llama.cpp/models/guanaco-7B.ggmlv3.q4_0.bin llama_model_load_internal: format = ggjt v3 (latest) llama_model_load_internal: n_vocab = 32000 llama_model_load_internal: n_ctx = 512 llama_model_load_internal: n_embd = 4096 llama_model_load_internal: n_mult = 256 llama_model_load_internal: n_head = 32 llama_model_load_internal: n_layer = 32 llama_model_load_internal: n_rot = 128 llama_model_load_internal: ftype = 2 (mostly Q4_0) llama_model_load_internal: n_ff = 11008 llama_model_load_internal: n_parts = 1 llama_model_load_internal: model size = 7B llama_model_load_internal: ggml ctx size = 0.07 MB llama_model_load_internal: mem required = 1932.71 MB (+ 1026.00 MB per state) . llama_init_from_file: kv self size = 256.00 MB ggml_metal_init: allocating ggml_metal_init: using MPS ggml_metal_init: loading '/Users/adam/Documents/Projects/langchainstufff/llama.cpp/build-metal/bin/ggml-metal.metal' ggml_metal_init: loaded kernel_add 0x7fc13bf07080 ggml_metal_init: loaded kernel_mul 0x7fc13bf07860 ggml_metal_init: loaded kernel_mul_row 0x7fc13bf08040 ggml_metal_init: loaded kernel_scale 0x7fc13bf08820 ggml_metal_init: loaded kernel_silu 0x7fc13bf09000 ggml_metal_init: loaded kernel_relu 0x7fc13bf097e0 ggml_metal_init: loaded kernel_soft_max 0x7fc13bf09fc0 ggml_metal_init: loaded kernel_diag_mask_inf 0x7fc13bf0a7a0 ggml_metal_init: loaded kernel_get_rows_q4_0 0x7fc13bf0af80 ggml_metal_init: loaded kernel_rms_norm 0x7fc13bf0b760 ggml_metal_init: loaded kernel_mul_mat_q4_0_f32 0x7fc13bf0bf40 ggml_metal_init: loaded kernel_mul_mat_f16_f32 0x7fc13bf0c890 ggml_metal_init: loaded kernel_rope 0x7fc13bf0d070 ggml_metal_init: loaded kernel_cpy_f32_f16 0x7fc13bf0d850 ggml_metal_init: loaded kernel_cpy_f32_f32 0x7fc13bf0e030 ggml_metal_add_buffer: allocated 'data ' buffer, size = 3616.07 MB ggml_metal_add_buffer: allocated 'eval ' buffer, size = 768.00 MB ggml_metal_add_buffer: allocated 'kv ' buffer, size = 258.00 MB ggml_metal_add_buffer: allocated 'scr0 ' buffer, size = 512.00 MB ggml_metal_add_buffer: allocated 'scr1 ' buffer, size = 512.00 MB

system_info: n_threads = 6 / 12 | AVX = 1 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | VSX = 0 | sampling: repeat_last_n = 64, repeat_penalty = 1.100000, presence_penalty = 0.000000, frequency_penalty = 0.000000, top_k = 40, tfs_z = 1.000000, top_p = 0.950000, typical_p = 1.000000, temp = 0.800000, mirostat = 0, mirostat_lr = 0.100000, mirostat_ent = 5.000000 generate: n_ctx = 512, n_batch = 512, n_predict = -1, n_keep = 0

I believe the meaning of life is 4ceratypenameculgeuenani

alienatorZ commented 1 year ago

I think my issue is I am not on apple silicon? I have an intel cpu MacBook Pro 2018

rafalio commented 1 year ago

Getting same thing, and am on Apple Silicon (m1 max, 32 core). Afaik I compiled everything correctly with LLAMA_METAL=1 and think I quantized the llama models correctly too.

ggerganov commented 1 year ago

Run make clean and retry with latest master. Make a new issue if it still fails

rafalio commented 1 year ago

@ggerganov - I have tried on latest master (590250f) and afaik still seems to be failing for me with same error as here.

I have tried a few commits ago, right after the initial Metal implementation was merged (d1f563a743a83dabc11e125d4a7d64189c16498c), and my same steps are working there correctly, so I think something broke in the subsequent commits.

bakkot commented 1 year ago

@rafalio I suspect you're seeing the same thing as me, which is that the subsequent change in the quantization code broke something; see https://github.com/ggerganov/llama.cpp/issues/1711