leejet / stable-diffusion.cpp

Stable Diffusion in pure C/C++
MIT License
2.9k stars 232 forks source link

Error when compile with latest GGML: ggml_quantize_chunk #265

Closed paulocoutinhox closed 1 month ago

paulocoutinhox commented 1 month ago

Hi,

Error when compile with latest GGML.

/Users/paulo/Developer/workspaces/cpp/ai-kit/vendor/stable-diffusion/model.cpp:608: error: no matching function for call to 'ggml_quantize_chunk'
/Users/paulo/Developer/workspaces/cpp/ai-kit/vendor/stable-diffusion/model.cpp:608:13: error: no matching function for call to 'ggml_quantize_chunk'
            ggml_quantize_chunk(dst_type, (float*)src_data_f32, dst, 0, nrows, n_per_row, hist, im);
            ^~~~~~~~~~~~~~~~~~~
/Users/paulo/Developer/workspaces/cpp/ai-kit/vendor/stable-diffusion/../ggml/include/ggml/ggml.h:2234:21: note: candidate function not viable: requires 7 arguments, but 8 were provided
    GGML_API size_t ggml_quantize_chunk(
                    ^

Thanks.

paulocoutinhox commented 1 month ago

I tried remove the "hist" param to follow quantize chunk method params, but still with errors:

ggml_metal_init: allocating
ggml_metal_init: picking default device: Apple A15 GPU
ggml_metal_init: loading '/private/var/containers/Bundle/Application/BF17FE80-AC05-47AF-8D5E-280CA37F4AA3/AiKit.app/default.metallib'
ggml_metal_init: GPU name:   Apple A15 GPU
ggml_metal_init: GPU family: MTLGPUFamilyApple8  (1008)
ggml_metal_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_init: GPU family: MTLGPUFamilyMetal3  (5001)
ggml_metal_init: simdgroup reduction support   = true
ggml_metal_init: simdgroup matrix mul. support = true
ggml_metal_init: hasUnifiedMemory              = true
ggml_metal_init: recommendedMaxWorkingSetSize  =  2863.32 MB
Assertion failed: (false), function ggml_quantize_chunk, file ggml.c, line 21621.
paulocoutinhox commented 1 month ago

Fixed. Thanks.