@FIR-1044 - lama.cpp: its crashing for new model due to recent bug

Author: Anoop Kapoor

ggml/src/CMakeLists.txt

@@ -318,8 +318,8 @@ if (GGML_CPU_ALL_VARIANTS)
         ggml_add_cpu_backend_variant(x64)
         ggml_add_cpu_backend_variant(sse42        SSE42)
         ggml_add_cpu_backend_variant(sandybridge  SSE42 AVX)
-        ggml_add_cpu_backend_variant(haswell      SSE42 AVX F16C AVX2 BMI2 FMA)
-        ggml_add_cpu_backend_variant(skylakex     SSE42 AVX F16C AVX2 BMI2 FMA AVX512)
+        ggml_add_cpu_backend_variant(haswell      SSE42 F16C BMI2)
+        ggml_add_cpu_backend_variant(skylakex     SSE42 F16C BMI2)
         ggml_add_cpu_backend_variant(icelake      SSE42 AVX F16C AVX2 BMI2 FMA AVX512 AVX512_VBMI AVX512_VNNI)
         ggml_add_cpu_backend_variant(alderlake    SSE42 AVX F16C AVX2 BMI2 FMA AVX_VNNI)
         if (NOT MSVC)

tools/main/main.cpp

@@ -290,7 +290,6 @@ int main(int argc, char ** argv) {
         auto formatted = common_chat_format_single(chat_templates.get(), chat_msgs, new_msg, role == "user", g_params->use_jinja);
         chat_msgs.push_back(new_msg);
         LOG_DBG("formatted: '%s'\n", formatted.c_str());
-        ggml_backend_cleanup();
         return formatted;
     };