Distribute and run LLMs with a single file.
C++ cross-platformggufllama-cpplocal-ailocal-inferencelocal-llmopen-source-aisingle-file-executablespeech-to-text
diff --git a/ggml/src/ggml-cpu/repack.cpp b/ggml/src/ggml-cpu/repack.cpp
--- a/transcribe.cpp/ggml/src/ggml-cpu/repack.cpp
+++ b/transcribe.cpp/ggml/src/ggml-cpu/repack.cpp
@@ -4723,14 +4723,14 @@ static const ggml::cpu::tensor_traits * ggml_repack_get_optimal_repack_type(cons
return nullptr;
}
-static enum ggml_status ggml_backend_cpu_repack_buffer_init_tensor(ggml_backend_buffer_t buffer, struct ggml_tensor * tensor) {
+static enum ggml_status GGML_CALL ggml_backend_cpu_repack_buffer_init_tensor(ggml_backend_buffer_t buffer, struct ggml_tensor * tensor) {
tensor->extra = (void *) const_cast<ggml::cpu::tensor_traits *>(ggml_repack_get_optimal_repack_type(tensor));
GGML_UNUSED(buffer);
return GGML_STATUS_SUCCESS;
}
-static void ggml_backend_cpu_repack_buffer_set_tensor(ggml_backend_buffer_t buffer, struct ggml_tensor * tensor,
+static void GGML_CALL ggml_backend_cpu_repack_buffer_set_tensor(ggml_backend_buffer_t buffer, struct ggml_tensor * tensor,
const void * data, size_t offset, size_t size) {
GGML_ASSERT(offset == 0);
GGML_ASSERT(size == ggml_nbytes(tensor));
@@ -4742,13 +4742,13 @@ static void ggml_backend_cpu_repack_buffer_set_tensor(ggml_backend_buffer_t buff
GGML_UNUSED(buffer);
}
-static const char * ggml_backend_cpu_repack_buffer_type_get_name(ggml_backend_buffer_type_t buft) {
+static const char * GGML_CALL ggml_backend_cpu_repack_buffer_type_get_name(ggml_backend_buffer_type_t buft) {
return "CPU_REPACK";
GGML_UNUSED(buft);
}
-static ggml_backend_buffer_t ggml_backend_cpu_repack_buffer_type_alloc_buffer(ggml_backend_buffer_type_t buft, size_t size) {
+static ggml_backend_buffer_t GGML_CALL ggml_backend_cpu_repack_buffer_type_alloc_buffer(ggml_backend_buffer_type_t buft, size_t size) {
ggml_backend_buffer_t buffer = ggml_backend_buft_alloc_buffer(ggml_backend_cpu_buffer_type(), size);
if (buffer == nullptr) {
@@ -4763,7 +4763,7 @@ static ggml_backend_buffer_t ggml_backend_cpu_repack_buffer_type_alloc_buffer(gg
return buffer;
}
-static size_t ggml_backend_cpu_repack_buffer_type_get_alignment(ggml_backend_buffer_type_t buft) {
+static size_t GGML_CALL ggml_backend_cpu_repack_buffer_type_get_alignment(ggml_backend_buffer_type_t buft) {
return TENSOR_ALIGNMENT;
GGML_UNUSED(buft);