BUG/MEDIUM: time: fix updating of global_now upon clock drift
During commit 7e4a557f6 ("MINOR: time: change the global timeval and the
the global tick at once") the approach made sure that the new now_ms was
always higher than or equal to global_now_ms, but by forgetting the old
value. This can cause the first update to global_now_ms to fail if it's
already out of sync, going back into the loop, and the subsequent call
would then succeed due to commit 4d01f3dcd ("MINOR: time: avoid
overwriting the same values of global_now").
And if it goes out of sync, it will fail to update forever, as observed
by Ashley Penney in github issue #1194, causing incorrect freq counters
calculations everywhere. One possible trigger for this issue is one thread
spinning for a few milliseconds while the other ones continue to work.
The issue really is that old_now_ms ought not to be modified in the loop
as it's used for the CAS. But we don't need to structurally guarantee that
global_now_ms grows monotonically as it's computed from the new global_now
which is already verified for this via the __tv_islt() test. Thus, dropping
any corrections on global_now_ms in the loop is the correct way to proceed
as long as this one is always updated to follow global_now.
No backport is needed, this is only for 2.4-dev.
diff --git a/src/time.c b/src/time.c
index 4ebde37..d6ab185 100644
--- a/src/time.c
+++ b/src/time.c
@@ -236,7 +236,6 @@
do {
tmp_now.tv_sec = (unsigned int)(old_now >> 32);
tmp_now.tv_usec = old_now & 0xFFFFFFFFU;
- old_now_ms = __tv_to_ms(&tmp_now);
if (__tv_islt(&now, &tmp_now))
now = tmp_now;
@@ -246,8 +245,6 @@
*/
new_now = ((ullong)now.tv_sec << 32) + (uint)now.tv_usec;
now_ms = __tv_to_ms(&now);
- if (tick_is_lt(now_ms, old_now_ms))
- now_ms = old_now_ms;
/* let's try to update the global <now> (both in timeval
* and ms forms) or loop again.