[MEDIUM] further improve monotonic clock by check forward jumps
The first implementation of the monotonic clock did not verify
forward jumps. The consequence is that a fast changing time may
expire a lot of tasks. While it does seem minor, in fact it is
problematic because most machines which boot with a wrong date
are in the past and suddenly see their time jump by several
years in the future.
The solution is to check if we spent more apparent time in
a poller than allowed (with a margin applied). The margin
is currently set to 1000 ms. It should be large enough for
any poll() to complete.
Tests with randomly jumping clock show that the result is quite
accurate (error less than 1 second at every change of more than
one second).
diff --git a/include/common/time.h b/include/common/time.h
index d615526..82ec402 100644
--- a/include/common/time.h
+++ b/include/common/time.h
@@ -44,6 +44,11 @@
#define TIME_ETERNITY (TV_ETERNITY_MS)
+/* we want to be able to detect time jumps. Fix the maximum wait time to a low
+ * value so that we know the time has changed if we wait longer.
+ */
+#define MAX_DELAY_MS 1000
+
/* returns the lowest delay amongst <old> and <new>, and respects TIME_ETERNITY */
#define MINTIME(old, new) (((new)<0)?(old):(((old)<0||(new)<(old))?(new):(old)))
@@ -84,13 +89,15 @@
return tv;
}
-/* tv_now_mono: sets <date> to the current time (wall clock), <mono> to a value
- * following a monotonic function, and applies any required correction if the
- * time goes backwards. Note that while we could improve it a bit by checking
- * that the new date is not too far in the future, it is not much necessary to
- * do so.
+/* tv_udpate_date: sets <date> to system time, and sets <now> to something as
+ * close as possible to real time, following a monotonic function. The main
+ * principle consists in detecting backwards and forwards time jumps and adjust
+ * an offset to correct them. This function should be called only once after
+ * each poll. The poll's timeout should be passed in <max_wait>, and the return
+ * value in <interrupted> (a non-zero value means that we have not expired the
+ * timeout).
*/
-REGPRM2 struct timeval *tv_now_mono(struct timeval *mono, struct timeval *wall);
+REGPRM2 void tv_update_date(int max_wait, int interrupted);
/*
* sets a struct timeval to its highest value so that it can never happen