|
|
@@ -2102,6 +2102,8 @@ int main(int argc, char ** argv) {
|
|
|
struct ggml_threadpool_params tpp = ggml_threadpool_params_default(t.n_threads);
|
|
|
if (!parse_cpu_mask(t.cpu_mask, tpp.cpumask)) {
|
|
|
fprintf(stderr, "%s: failed to parse cpu-mask: %s\n", __func__, t.cpu_mask.c_str());
|
|
|
+ llama_free(ctx);
|
|
|
+ llama_model_free(lmodel);
|
|
|
exit(1);
|
|
|
}
|
|
|
tpp.strict_cpu = t.cpu_strict;
|
|
|
@@ -2111,6 +2113,8 @@ int main(int argc, char ** argv) {
|
|
|
struct ggml_threadpool * threadpool = ggml_threadpool_new_fn(&tpp);
|
|
|
if (!threadpool) {
|
|
|
fprintf(stderr, "%s: threadpool create failed : n_threads %d\n", __func__, tpp.n_threads);
|
|
|
+ llama_free(ctx);
|
|
|
+ llama_model_free(lmodel);
|
|
|
exit(1);
|
|
|
}
|
|
|
|
|
|
@@ -2126,6 +2130,8 @@ int main(int argc, char ** argv) {
|
|
|
bool res = test_prompt(ctx, t.n_prompt, t.n_batch, t.n_threads);
|
|
|
if (!res) {
|
|
|
fprintf(stderr, "%s: error: failed to run prompt warmup\n", __func__);
|
|
|
+ llama_free(ctx);
|
|
|
+ llama_model_free(lmodel);
|
|
|
exit(1);
|
|
|
}
|
|
|
}
|
|
|
@@ -2136,6 +2142,8 @@ int main(int argc, char ** argv) {
|
|
|
bool res = test_gen(ctx, 1, t.n_threads);
|
|
|
if (!res) {
|
|
|
fprintf(stderr, "%s: error: failed to run gen warmup\n", __func__);
|
|
|
+ llama_free(ctx);
|
|
|
+ llama_model_free(lmodel);
|
|
|
exit(1);
|
|
|
}
|
|
|
}
|
|
|
@@ -2164,6 +2172,8 @@ int main(int argc, char ** argv) {
|
|
|
bool res = test_prompt(ctx, t.n_depth, t.n_batch, t.n_threads);
|
|
|
if (!res) {
|
|
|
fprintf(stderr, "%s: error: failed to run depth\n", __func__);
|
|
|
+ llama_free(ctx);
|
|
|
+ llama_model_free(lmodel);
|
|
|
exit(1);
|
|
|
}
|
|
|
|
|
|
@@ -2189,6 +2199,8 @@ int main(int argc, char ** argv) {
|
|
|
bool res = test_prompt(ctx, t.n_prompt, t.n_batch, t.n_threads);
|
|
|
if (!res) {
|
|
|
fprintf(stderr, "%s: error: failed to run prompt\n", __func__);
|
|
|
+ llama_free(ctx);
|
|
|
+ llama_model_free(lmodel);
|
|
|
exit(1);
|
|
|
}
|
|
|
}
|
|
|
@@ -2200,6 +2212,8 @@ int main(int argc, char ** argv) {
|
|
|
bool res = test_gen(ctx, t.n_gen, t.n_threads);
|
|
|
if (!res) {
|
|
|
fprintf(stderr, "%s: error: failed to run gen\n", __func__);
|
|
|
+ llama_free(ctx);
|
|
|
+ llama_model_free(lmodel);
|
|
|
exit(1);
|
|
|
}
|
|
|
}
|