llama : absorb llama_grammar in llama_sampling

ggml-ci

Author: ggerganov

Makefile

@@ -886,7 +886,6 @@ OBJ_COMMON = \
 	common/ngram-cache.o \
 	common/sampling.o \
 	common/train.o \
-	common/grammar-parser.o \
 	common/build-info.o \
 	common/json-schema-to-grammar.o
 
@@ -1124,11 +1123,6 @@ common/console.o: \
 	common/console.h
 	$(CXX) $(CXXFLAGS) -c $< -o $@
 
-common/grammar-parser.o: \
-	common/grammar-parser.cpp \
-	common/grammar-parser.h
-	$(CXX) $(CXXFLAGS) -c $< -o $@
-
 common/json-schema-to-grammar.o: \
 	common/json-schema-to-grammar.cpp \
 	common/json-schema-to-grammar.h

common/CMakeLists.txt

@@ -58,8 +58,6 @@ add_library(${TARGET} STATIC
     sampling.cpp
     console.h
     console.cpp
-    grammar-parser.h
-    grammar-parser.cpp
     json.hpp
     json-schema-to-grammar.cpp
     train.h

common/common.cpp

@@ -2127,7 +2127,7 @@ std::tuple<struct llama_model *, struct llama_context *> llama_init_from_gpt_par
         llama_decode(lctx, llama_batch_get_one(tmp.data(), std::min(tmp.size(), (size_t) params.n_batch), 0, 0));
         llama_kv_cache_clear(lctx);
         llama_synchronize(lctx);
-        llama_reset_timings(lctx, nullptr, nullptr);
+        llama_reset_timings(lctx, nullptr);
     }
 
     return std::make_tuple(model, lctx);