The definition of the EV_CHANGE_xxx macros partially relies on the
definition of a different set of macros (EV_SIGNAL, EV_PERSIST, etc.).
There are two issues I see with that.
First, both EV_CHANGE_ADD and EV_TIMEOUT are defined as 1, which might lead
to situations where a bit can be interpreted as two different things., e.g.
following this code in event_changelist_add
if (events & (EV_READ|EV_SIGNAL)) {
change->read_change = EV_CHANGE_ADD |
(events & (EV_ET|EV_PERSIST|EV_SIGNAL));
}
if (events & EV_WRITE) {
change->write_change = EV_CHANGE_ADD |
(events & (EV_ET|EV_PERSIST|EV_SIGNAL));
}
I haven't actually found any buggy scenario, but it is a bit fishy.
Second, the definitions combine constant values (e.g. 0x02) with macro usage
(e.g. EV_SIGNAL). This style easily allows human errors where two of these
macros will end up with the same value.
I don't know the code well enough to suggest a patch so I'll have to settle
for raising the issue.
Gilad
***********************************************************************
To unsubscribe, send an e-mail to majord...@freehaven.net with
unsubscribe libevent-users in the body.
