[Help] Error (core dumped) when running example code from blogpost

[treya-lin]

Hello! I hope it is the right place to ask. I am learning about this repo from this blogpost: https://huggingface.co/blog/introduction-to-ggml

I made examples/demo and tried compiling demo. The multiplication went ok and the result matrix can be printed. but then I added the graph print command as intructed in the blogpost, and it threw segment fault (core dumped).

here is my code

examples/demo/demo.cpp
(if I comment the ggml_graph_print(gf);, it can be compiled and run.
if I uncomment it and recompile, it can be compiled but when I run it it threw core dumped error)

#include "ggml.h"
#include "ggml-cpu.h"

#include <string.h>
#include <stdio.h>

int main(void) {
    // initialize data of matrices
    const int rows_A = 4, cols_A = 2;
    float matrix_A[rows_A * cols_A] = {
        2, 8,
        5, 1,
        4, 2,
        8, 6
    };
    const int rows_B = 3, cols_B = 2;
    float matrix_B[rows_B * cols_B] = {
        10, 5,
        9, 9,
        5, 4
    };

    // 1. Allocate `ggml_context` to store tensor data
    // Calculate the size needed to allocate
    size_t ctx_size = 0;
    ctx_size += rows_A * cols_A * ggml_type_size(GGML_TYPE_F32); // tensor a 
    ctx_size += rows_B * cols_B * ggml_type_size(GGML_TYPE_F32); // tensor b
    ctx_size += rows_A * rows_B * ggml_type_size(GGML_TYPE_F32); // result
    ctx_size += 3 * ggml_tensor_overhead(); // metadata for 3 tensors
    ctx_size += ggml_graph_overhead(); // compute graph
    ctx_size += 1024; //some overhead

    // Allocate `ggml_context` to store tensor data
    struct ggml_init_params params = {
        /*.mem_size = */ ctx_size,
        /*.mem_buffer =*/ NULL,
        /*.no_alloc = */ false,
    };
    struct ggml_context * ctx = ggml_init(params);

    // 2. Create tensors and set data
    struct ggml_tensor * tensor_a = ggml_new_tensor_2d(ctx, GGML_TYPE_F32, cols_A, rows_A);
    struct ggml_tensor * tensor_b = ggml_new_tensor_2d(ctx, GGML_TYPE_F32, cols_B, rows_B);
    memcpy(tensor_a->data, matrix_A, ggml_nbytes(tensor_a));
    memcpy(tensor_b->data, matrix_B, ggml_nbytes(tensor_b));

    // 3. Create a `ggml_cgraph` for mul_mat operation
   struct ggml_cgraph * gf = ggml_new_graph(ctx);


   // result = a*b^T
   // Pay attention: ggml_mul_mat(A,B) ===> B will be transposed internally
   // the result is transposed
   struct ggml_tensor * result = ggml_mul_mat(ctx, tensor_a, tensor_b);

   // Mark the "result" tensor to be computed
   ggml_build_forward_expand(gf, result);

   // Print the graph
   ggml_graph_print(gf);
   // ggml_graph_dump_dot(gf, NULL, "debug.dot");

   // 4. Run the computation
   int n_threads = 1; // Optional: number of threads to perform some operations with multi-threading
   ggml_graph_compute_with_ctx(ctx, gf, n_threads);

   // 5. Retrieve results (output tensors)
   float * result_data = (float *) result->data;
   printf("mul mat (%d x %d)) (transposed result):\n[", (int) result->ne[0], (int) result->ne[1]);
   for (int j = 0; j < result->ne[1]/* rows */; j++) {
       if (j > 0) {
           printf("\n");
       }

       for (int i = 0; i < result->ne[0]/* cols */; i++) {
           printf(" %.2f", result_data[j * result->ne[0] + i]);
       }
   }
   printf(" ]\n");

   // 6. Free memory and exit
   ggml_free(ctx);
   return 0;
}

other relevant codes

1. examples/demo/CMakeLists.txt

set(TEST_TARGET demo)
add_executable(${TEST_TARGET} demo)
target_link_libraries(${TEST_TARGET} PRIVATE ggml)

2. add this at the end of examples/CMakeLists.txt

add_subdirectory(demo)

I dont have much experience with cpp and had trouble debugging. I tried asking some AI tools but it does work very well so I am very puzzled now. Any advice can help!

I am working on a linux server and the env info is:

$ uname -r
6.5.0-35-generic
$ cat /proc/version
Linux version 6.5.0-35-generic (buildd@lcy02-amd64-079) (x86_64-linux-gnu-gcc-12 (Ubuntu 12.3.0-1ubuntu1~22.04) 12.3.0, GNU ld (GNU Binutils for Ubuntu) 2.38) #35~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue May  7 09:00:52 UTC 2

my ggml code version

$ git log
commit 6fcbd60bc72ac3f7ad43f78c87e535f2e6206f58 (HEAD -> master, origin/master, origin/HEAD)
Author: M Refi D.A <[email protected]>
Date:   Thu Nov 21 06:39:37 2024 +0800