BUG/MINOR: fd: Don't clear the update_mask in fd_insert.

Clearing the update_mask bit in fd_insert may lead to duplicate insertion
of fd in fd_updt, that could lead to a write past the end of the array.
Instead, make sure the update_mask bit is cleared by the pollers no matter
what.

This should be backported to 1.8.
[wt: warning: 1.8 doesn't have the lockless fdcache changes and will
 require some careful changes in the pollers]
diff --git a/src/ev_select.c b/src/ev_select.c
index 6e83467..163a458 100644
--- a/src/ev_select.c
+++ b/src/ev_select.c
@@ -57,13 +57,13 @@
 	for (updt_idx = 0; updt_idx < fd_nbupdt; updt_idx++) {
 		fd = fd_updt[updt_idx];
 
+		HA_ATOMIC_AND(&fdtab[fd].update_mask, ~tid_bit);
 		if (!fdtab[fd].owner) {
 			activity[tid].poll_drop++;
 			continue;
 		}
 
 		en = fdtab[fd].state;
-		HA_ATOMIC_AND(&fdtab[fd].update_mask, ~tid_bit);
 
 		/* we have a single state for all threads, which is why we
 		 * don't check the tid_bit. First thread to see the update