Misc. bug: Segmentation fault when importing model to opencl buffer

[zhouzengming]

Name and Version

version: 4737 (5137da7)
built with cc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0 for aarch64-linux-gnu

Operating systems

Linux

Which llama.cpp modules do you know to be affected?

llama-cli

Command line

llama-cli -m /home/aidlux/QualcommLLM/models/Qwen2.5_0.5b.gguf -ngl 1

Problem description & steps to reproduce

The program llama-cli caused a Segmentation falt when initialize model on opencl device. I tried to findout what the problem is with clion debug mode. Here is the call stack and some value:

The program runs correctly when running model on CPU. But when I run with param "-ngl 1" trying to load a layer on GPU, it raised a sagmentation falt.

I compiled llama.cpp on my QCS6490 device whith Adreno643L GPU, setting GGML_OPENCL=ON.

Because of the limit on my device, I tried to reset max buffer size to adapt my device in ggml/src/ggml-opencl/ggml-opencl.cpp: line800-803 as below:

   // Allocate intermediate buffers and images
    size_t max_A_q_d_bytes = 268435456;//311164928;
    size_t max_A_s_d_bytes = 268435456;//38895616;
    size_t max_B_d_bytes = 268435456;//45088768;

below is clinfo about my device:

Number of platforms                               1
  Platform Name                                   QUALCOMM Snapdragon(TM)
  Platform Vendor                                 QUALCOMM
  Platform Version                                OpenCL 3.0 QUALCOMM build: commit unknown
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd
  Platform Host timer resolution                  0ns
  Platform Extensions function suffix             QCOM

  Platform Name                                   QUALCOMM Snapdragon(TM)
Number of devices                                 1
  Device Name                                     QUALCOMM Adreno(TM) 643
  Device Vendor                                   QUALCOMM
  Device Vendor ID                                0x5143
  Device Version                                  OpenCL 2.0 Adreno(TM) 643
  Driver Version                                  OpenCL 3.0 QUALCOMM build: commit unknown Compiler E031.42.02.00
  Device OpenCL C Version                         OpenCL C 2.0 Adreno(TM) 643
  Device Type                                     GPU
  Device Profile                                  FULL_PROFILE
  Device Available                                Yes
  Compiler Available                              Yes
  Linker Available                                Yes
  Max compute units                               2
  Max clock frequency                             1MHz
  Device Partition                                (core)
    Max number of sub-devices                     1
    Supported partition types                     None
    Supported affinity domains                    (n/a)
  Max work item dimensions                        3
  Max work item sizes                             1024x1024x1024
  Max work group size                             1024
  Preferred work group size multiple              128
  Preferred / native vector sizes                 
    char                                                 1 / 1       
    short                                                1 / 1       
    int                                                  1 / 1       
    long                                                 1 / 0       
    half                                                 1 / 1        (cl_khr_fp16)
    float                                                1 / 1       
    double                                               0 / 0        (n/a)
  Half-precision Floating-point support           (cl_khr_fp16)
    Denormals                                     No
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 No
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               No
    Support is emulated in software               No
  Single-precision Floating-point support         (core)
    Denormals                                     No
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 No
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               No
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  No
  Double-precision Floating-point support         (n/a)
  Address bits                                    64, Little-Endian
  Global memory size                              1073741824 (1024MiB)
  Error Correction support                        No
  Max memory allocation                           268435456 (256MiB)
  Unified memory for Host and Device              Yes
  Shared Virtual Memory (SVM) capabilities        (core)
    Coarse-grained buffer sharing                 Yes
    Fine-grained buffer sharing                   Yes
    Fine-grained system sharing                   No
    Atomics                                       Yes
  Minimum alignment for any data type             128 bytes
  Alignment of base address                       1024 bits (128 bytes)
  Page size (QCOM)                                4096 bytes
  External memory padding (QCOM)                  0 bytes
  Preferred alignment for atomics                 
    SVM                                           128 bytes
    Global                                        0 bytes
    Local                                         0 bytes
  Max size for global variable                    65536 (64KiB)
  Preferred total size of global vars             1048576 (1024KiB)
  Global Memory cache type                        Read/Write
  Global Memory cache size                        262144 (256KiB)
  Global Memory cache line size                   64 bytes
  Image support                                   Yes
    Max number of samplers per kernel             16
    Max size for 1D images from buffer            134217728 pixels
    Max 1D or 2D image array size                 2048 images
    Base address alignment for 2D image buffers   64 bytes
    Pitch alignment for 2D image buffers          64 pixels
    Max 2D image size                             16384x16384 pixels
    Max 3D image size                             16384x16384x2048 pixels
    Max number of read image args                 128
    Max number of write image args                64
    Max number of read/write image args           64
  Max number of pipe args                         16
  Max active pipe reservations                    4096
  Max pipe packet size                            1024
  Local memory type                               Local
  Local memory size                               32768 (32KiB)
  Max number of constant args                     8
  Max constant buffer size                        65536 (64KiB)
  Max size of kernel argument                     1024
  Queue properties (on host)                      
    Out-of-order execution                        Yes
    Profiling                                     Yes
  Queue properties (on device)                    
    Out-of-order execution                        Yes
    Profiling                                     Yes
    Preferred size                                655376 (640KiB)
    Max size                                      655376 (640KiB)
  Max queues on device                            1
  Max events on device                            1024
  Prefer user sync for interop                    No
  Profiling timer resolution                      1000ns
  Execution capabilities                          
    Run OpenCL kernels                            Yes
    Run native kernels                            No
  printf() buffer size                            1048576 (1024KiB)
  Built-in kernels                                (n/a)
  Device Extensions                               cl_img_egl_image cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_egl_event cl_khr_egl_image cl_khr_fp16 cl_khr_gl_sharing cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_image2d_from_buffer cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_mipmap_image cl_khr_srgb_image_writes cl_khr_subgroups cl_qcom_accelerated_image_ops cl_qcom_compressed_image cl_qcom_compressed_yuv_image_read cl_qcom_create_buffer_from_image cl_qcom_dot_product8 cl_qcom_ext_host_ptr cl_qcom_ext_host_ptr_iocoherent cl_qcom_extract_image_plane cl_qcom_ion_host_ptr cl_qcom_other_image cl_qcom_perf_hint cl_qcom_priority_hint cl_qcom_protected_context cl_qcom_reqd_sub_group_size cl_qcom_subgroup_shuffle cl_qcom_vector_image_ops 

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  QUALCOMM Snapdragon(TM)
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   Success [QCOM]
  clCreateContext(NULL, ...) [default]            Success [QCOM]
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT)  Success (1)
    Platform Name                                 QUALCOMM Snapdragon(TM)
    Device Name                                   QUALCOMM Adreno(TM) 643
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  Success (1)
    Platform Name                                 QUALCOMM Snapdragon(TM)
    Device Name                                   QUALCOMM Adreno(TM) 643
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  Invalid device type for platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  Success (1)
    Platform Name                                 QUALCOMM Snapdragon(TM)
    Device Name                                   QUALCOMM Adreno(TM) 643

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.2.11
  ICD loader Profile                              OpenCL 2.1
        NOTE:   your OpenCL library only supports OpenCL 2.1,
                but some installed platforms support OpenCL 3.0.
                Programs using 3.0 features may crash
                or behave unexpectedly

First Bad Commit

No response

Relevant log output

Program output with command "llama-cli -m ~/QualcommLLM/models/Qwen2.5_0.5b.gguf -ngl 1":

ggml_opencl: selecting platform: 'QUALCOMM Snapdragon(TM)'
ggml_opencl: selecting device: 'QUALCOMM Adreno(TM) 643'
ggml_opencl: Unsupported Adreno GPU: , using wave size 128, may not work as expected
ggml_opencl: device OpenCL version: OpenCL 2.0 Adreno(TM) 643
ggml_opencl: OpenCL driver: OpenCL 3.0 QUALCOMM build: commit unknown Compiler E031.42.02.00
ggml_opencl: vector subgroup broadcast support: false
ggml_opencl: device FP16 support: true
ggml_opencl: mem base addr align: 1024
ggml_opencl: max mem alloc size: 256 MB
ggml_opencl: SVM coarse grain buffer support: true
ggml_opencl: SVM fine grain buffer support: true
ggml_opencl: SVM fine grain system support: false
ggml_opencl: SVM atomics support: true
ggml_opencl: flattening quantized weights representation as struct of arrays (GGML_OPENCL_SOA_Q)
ggml_opencl: using kernels optimized for Adreno (GGML_OPENCL_USE_ADRENO_KERNELS)
register_backend: registered backend OpenCL (1 devices)
register_device: registered device GPUOpenCL (QUALCOMM Adreno(TM) 643)
register_backend: registered backend CPU (1 devices)
register_device: registered device CPU (CPU)
load_backend: failed to find ggml_backend_init in /home/aidlux/QualcommLLM/dev/llm/llama.cpp/cmake-build-debug/bin/libggml-opencl.so
load_backend: failed to find ggml_backend_init in /home/aidlux/QualcommLLM/dev/llm/llama.cpp/cmake-build-debug/bin/libggml-cpu.so
build: 4737 (5137da7b) with cc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0 for aarch64-linux-gnu (debug)
main: llama backend init
main: load the model and apply lora adapter, if any
llama_model_load_from_file_impl: using device GPUOpenCL (QUALCOMM Adreno(TM) 643) - 0 MiB free
llama_model_loader: loaded meta data with 34 key-value pairs and 290 tensors from /home/aidlux/QualcommLLM/models/Qwen2.5_0.5b.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 0.5B 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              = 0.5B
llama_model_loader: - kv   6:                            general.license str              = apache-2.0
llama_model_loader: - kv   7:                       general.license.link str              = https://huggingface.co/Qwen/Qwen2.5-0...
llama_model_loader: - kv   8:                   general.base_model.count u32              = 1
llama_model_loader: - kv   9:                  general.base_model.0.name str              = Qwen2.5 0.5B
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/Qwen2.5-0.5B
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              = 24
llama_model_loader: - kv  15:                       qwen2.context_length u32              = 32768
llama_model_loader: - kv  16:                     qwen2.embedding_length u32              = 896
llama_model_loader: - kv  17:                  qwen2.feed_forward_length u32              = 4864
llama_model_loader: - kv  18:                 qwen2.attention.head_count u32              = 14
llama_model_loader: - kv  19:              qwen2.attention.head_count_kv u32              = 2
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              = 1
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,151936]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  26:                  tokenizer.ggml.token_type arr[i32,151936]  = [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: - type  f32:  121 tensors
llama_model_loader: - type  f16:  169 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = F16
print_info: file size   = 942.43 MiB (16.00 BPW) 
load: special tokens cache size = 22
load: token to piece cache size = 0.9310 MB
print_info: arch             = qwen2
print_info: vocab_only       = 0
print_info: n_ctx_train      = 32768
print_info: n_embd           = 896
print_info: n_layer          = 24
print_info: n_head           = 14
print_info: n_head_kv        = 2
print_info: n_rot            = 64
print_info: n_swa            = 0
print_info: n_embd_head_k    = 64
print_info: n_embd_head_v    = 64
print_info: n_gqa            = 7
print_info: n_embd_k_gqa     = 128
print_info: n_embd_v_gqa     = 128
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: n_ff             = 4864
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 = 1
print_info: n_ctx_orig_yarn  = 32768
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       = 1B
print_info: model params     = 494.03 M
print_info: general.name     = Qwen2.5 0.5B Instruct
print_info: vocab type       = BPE
print_info: n_vocab          = 151936
print_info: n_merges         = 151387
print_info: BOS token        = 151643 '<|endoftext|>'
print_info: EOS token        = 151645 '<|im_end|>'
print_info: EOT token        = 151645 '<|im_end|>'
print_info: PAD token        = 151643 '<|endoftext|>'
print_info: LF token         = 198 'Ċ'
print_info: FIM PRE token    = 151659 '<|fim_prefix|>'
print_info: FIM SUF token    = 151661 '<|fim_suffix|>'
print_info: FIM MID token    = 151660 '<|fim_middle|>'
print_info: FIM PAD token    = 151662 '<|fim_pad|>'
print_info: FIM REP token    = 151663 '<|repo_name|>'
print_info: FIM SEP token    = 151664 '<|file_sep|>'
print_info: EOG token        = 151643 '<|endoftext|>'
print_info: EOG token        = 151645 '<|im_end|>'
print_info: EOG token        = 151662 '<|fim_pad|>'
print_info: EOG token        = 151663 '<|repo_name|>'
print_info: EOG token        = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors: offloading 1 repeating layers to GPU
load_tensors: offloaded 1/25 layers to GPU
load_tensors:   CPU_Mapped model buffer size =   942.43 MiB
load_tensors:       OpenCL model buffer size =    28.45 MiB
..........................................................................
llama_init_from_model: n_seq_max     = 1
llama_init_from_model: n_ctx         = 4096
llama_init_from_model: n_ctx_per_seq = 4096
llama_init_from_model: n_batch       = 2048
llama_init_from_model: n_ubatch      = 512
llama_init_from_model: flash_attn    = 0
llama_init_from_model: freq_base     = 1000000.0
llama_init_from_model: freq_scale    = 1
llama_init_from_model: n_ctx_per_seq (4096) < n_ctx_train (32768) -- the full capacity of the model will not be utilized
llama_kv_cache_init: kv_size = 4096, offload = 1, type_k = 'f16', type_v = 'f16', n_layer = 24, can_shift = 1
llama_kv_cache_init:        CPU KV buffer size =    46.00 MiB
Segmentation fault