checkasm: Validate the benchmark call configurations even if not benchmarking

This should help catch issues like the one fixed in
185194be, by making sure that we
call the benchmarked function at least once with the given parameters,
even if not benchmarking. Otherwise the benchmark codepath is
essentially dead untested code until somebody works on that piece
of code.

tests/checkasm/checkasm.h

@@ -282,9 +282,9 @@ void checkasm_stack_clobber(uint64_t clobber, ...);
 #ifdef readtime
 #define bench_new(...)\
     do {\
+        func_type *tfunc = func_new;\
+        checkasm_set_signal_handler_state(1);\
         if (checkasm_bench_func()) {\
-            checkasm_set_signal_handler_state(1);\
-            func_type *tfunc = func_new;\
             uint64_t tsum = 0;\
             int tcount = 0;\
             for (int ti = 0; ti < BENCH_RUNS; ti++) {\
@@ -299,9 +299,11 @@ void checkasm_stack_clobber(uint64_t clobber, ...);
                     tcount++;\
                 }\
             }\
-            checkasm_set_signal_handler_state(0);\
             checkasm_update_bench(tcount, tsum);\
+        } else {\
+            tfunc(__VA_ARGS__);\
         }\
+        checkasm_set_signal_handler_state(0);\
     } while (0)
 #else
 #define bench_new(...) do {} while (0)