reduce log level for io_uring_submit -> -EBUSY message and fix TIME_NS

TIME_NS was never expanded because emper-config.h was not included
in Common.hpp.

The log_level is reduced from Info to Debug because the IoContext reqs_in_uring
has only in debug builds a meaningful value.
And the fact that there where a full CQ during io_uring_submit is captured in the stats
anyway.

Don't record a full CQ before dying in the completer.

emper/Common.hpp

@@ -5,6 +5,8 @@
 #include <functional>
 #include <sstream>	// IWYU pragma: keep
 
+#include "emper-config.h"	 // IWYU pragma: keep
+
 using func_t = std::function<void()>;
 
 #define STRINGIFY(x) #x

emper/io/IoContext.cpp

@@ -99,12 +99,11 @@ auto IoContext::submitPreparedSqes() -> unsigned {
 		// in worker thread -> reapCompletions
 		// in globalCompleter thread -> TODO: deterministically handle global full CQ
 		// for now hope a jam does not happen or will solve itself
-		LOGGER_LOGI("io_submit returned EBUSY trying to submit "
-								<< io_uring_sq_ready(&ring) << " in addition to " << reqs_in_uring);
+		LOGD("io_submit returned EBUSY trying to submit " << io_uring_sq_ready(&ring)
+																											<< " in addition to " << reqs_in_uring);
 
 		if constexpr (callerEnvironment == CallerEnvironment::ANYWHERE) {
 			// we are not busy looping in the globalIo
-			stats.record_io_submit_full_cq();
 			DIE_MSG("Future dropped because global completer SQ is full");
 		}
 

emper/io/Stats.hpp

@@ -203,7 +203,7 @@ class Stats {
 
 	// running mean calculation taken from
 	// https://math.stackexchange.com/questions/106700/incremental-averageing
-	inline void record_io_submit_full_cq(nanoseconds ns = std::chrono::nanoseconds(0)) {
+	inline void record_io_submit_full_cq(nanoseconds ns) {
 		RETURN_IF_NO_STATS();
 		io_submit_full_cq++;