There were a couple of places in the code where we manually kept lock
counts to make sure we never accessed resources without holding a
lock, and that we never released a lock we didn't have. The
lock-debugging code already puts counts on _every_ lock when lock
debugging is enabled, so there is no need to keep these counts around
otherwise. This patch rewrites the ASSERT_FOO_LOCKED macros to all
use a common EVLOCK_ASSERT_LOCKED().
We also teach the lock debugging code to keep track of who exactly
holds each lock, so that EVLOCK_ASSERT_LOCKED() means "locked by this
thread."
Original message:
evdns contains a bug related to thread lock.
enable thread lock by evthread_use_pthreads() will cause successive
evdns_base_resolve_ipv4() (and other resolve functions i think) to
hang on EVDNS_LOCK(base) after one or several successful call to
evdns_base_resolve_ipv4().
Previously, our default lock model kind of assumed that every lock was
potentially a read-write lock. This was a poor choice, since
read-write locks are far more expensive than regular locks, and so the
lock API should only use them when we can actually take advantage of
them. Neither our pthreads or win32 lock implementation provided rw
locks.
Now that we have a way (not currently used!) to indicate that we
really want a read-write lock, we shouldn't actually say "lock this
for reading" or "lock this for writing" unless we mean it.
Previously, there was no good way to request different kinds of lock
(say, read/write vs writeonly or recursive vs nonrecursive), or for a
lock function to signal failure (which would be important for a
trylock mode).
This patch revises the lock API to be a bit more useful. The older
lock calls are still supported for now.
We also add a debugging mode to catch common errors in using the
locking APIs.
The entry points are evutil_getaddrinfo and evdns_getaddrinfo respectively.
There are fairly extensive unit tests.
I believe this code conforms to RFC3493 pretty closely, but there are
probably more issues. It should get tested on more platforms.
This code means we can dump the well-intentioned but weirdly-implemented
bufferevent_evdns and evutil_resolve code.
svn:r1537
This function, bufferevent_socket_connect_hostname() can either use
evdns to do the resolve, or use a new function (evutil_resolve) that
uses getaddrinfo or gethostbyname, like http.c does now.
This function is meant to eventually replace the hostname resolution mess in
http.c.
svn:r1496
We need to comb the rest of the code to make sure that we don't blindly wrap
functions in LOCK(x), UNLOCK(x) when those functions might contain a FREE(x)
in the middle.
Rocco Carbone found and reported this bug.
svn:r1384
Generally speaking, it way better to event_assign() an event when you
allocate it than to assign it before every time you event_add it: if
it is already event_add()ed, the assign will mess it up so that it
doesn't _look_ added, and event_add() will insert a second copy.
Later, event_del() will only delete the second copy. Eventually, the
event_base will have a dangling pointer to freed memory. Ouch!
svn:r1307
Others may remain. I wasn't able to get gcc --std=c89 to build libevent
at all, so I don't know what compiler the original reporter is using here.
Note that this change requires us to disable the part of our rpc code
that uses variadic macros when using a non-gcc compiler. This is a
problem if we want our rpc api to be portable.
svn:r1231
The bug could occur when a nameserver was marked as up, but then an
outstanding probe sent to the nameserver failed. Now, evdns_up() cancels
any outstanding probe.
svn:r1140
This patch adds a new set of EVUTIL_IS* functions to replace use of
the ctypes is* functions in all cases where we care about characters'
interpretations in net ascii rather than in the locale. For example,
when we're working with DNS hostnames, we don't want to do the 0x20
hack on non-ascii characters, even if the host thinks they should be
isalpha.
svn:r1114
