|
@@ -847,7 +847,7 @@ std::array<std::string, 2> mul_str_values = {
|
|
|
"mul_f32", "float"
|
|
"mul_f32", "float"
|
|
|
};
|
|
};
|
|
|
|
|
|
|
|
-std::string& replace(std::string& s, const std::string& from, const std::string& to) {
|
|
|
|
|
|
|
+static std::string& replace(std::string& s, const std::string& from, const std::string& to) {
|
|
|
size_t pos = 0;
|
|
size_t pos = 0;
|
|
|
while ((pos = s.find(from, pos)) != std::string::npos) {
|
|
while ((pos = s.find(from, pos)) != std::string::npos) {
|
|
|
s.replace(pos, from.length(), to);
|
|
s.replace(pos, from.length(), to);
|
|
@@ -856,7 +856,7 @@ std::string& replace(std::string& s, const std::string& from, const std::string&
|
|
|
return s;
|
|
return s;
|
|
|
}
|
|
}
|
|
|
|
|
|
|
|
-std::string generate_kernels() {
|
|
|
|
|
|
|
+static std::string generate_kernels() {
|
|
|
std::stringstream src;
|
|
std::stringstream src;
|
|
|
src << program_source << '\n';
|
|
src << program_source << '\n';
|
|
|
src << k_quants_source << '\n';
|
|
src << k_quants_source << '\n';
|
|
@@ -1788,7 +1788,7 @@ bool ggml_cl_can_mul_mat(const struct ggml_tensor * src0, const struct ggml_tens
|
|
|
return false;
|
|
return false;
|
|
|
}
|
|
}
|
|
|
|
|
|
|
|
-bool ggml_cl_mul_mat_use_f16(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * /* dst */) {
|
|
|
|
|
|
|
+static bool ggml_cl_mul_mat_use_f16(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * /* dst */) {
|
|
|
// If device doesn't support FP16
|
|
// If device doesn't support FP16
|
|
|
if (!fp16_support) {
|
|
if (!fp16_support) {
|
|
|
return false;
|
|
return false;
|