Now there are appropriate "for each event", "for each fd", and "for
each signal" helpers that they can use; this makes the code a bit
simpler, and far less duplicated.
This lets me turn back on the functions I disabled when removing
eventlist.
Additionally, check more lists for circularity in
event_base_assert_ok().
Add typedefs for the callback types.
Name fewer things "ctx".
Adds an implementation of Floyd's tortoise-and-hare algorithm to check
for circularity in TAILQs and LISTs, to avoid the abuse of flags that
event_base_assert_ok() was doing before.
Suggested by Dave Hart.
Linux provides some features that allow avoiding extra calls to
fcntl when creating new nonblocking/close-on-exec sockets, so
we can add wrapper functions to emulate those when they are not
available.
Additionally, even when we are emulating those functions, we can
take a fast path that cuts our fcntl calls in half: we don't need to
look up the previous value of a file's flags when we have just
created it.
Dongsheng Song reports that when building on windows with NDEBUG, you
run into an attempt to do EVUTIL_ASSERT(x) where x is a bitfield,
which turns into _EVUTIL_NIL_CONDITION(x), which takes sizeof(x),
which is illegal. This patch fixes _EVUTIL_NIL_CONDITION to work on
bitfields too.
This patch splits the formerly windows-only case of evutil_socketpair()
into an (internal-use-only) function named evutil_ersatz_socketpair(), and
makes it build and work right on non-Windows hosts.
We need this for convenience to test sendfile on solaris, where socketpair
can't give you an AF_INET pair, and sendfile() won't work on AF_UNIX.
It turns out that the happy fun Linux kernel is deprecating sysctl,
and using sysctl to fetch entropy will spew messages in the kernel
logs. Let's not do that. Instead, let's call sysctl for our
entropy only when all other means fail.
Additionally, let's add another means, and try
/proc/sys/kernel/random/uuid if /dev/urandom fails.
The old logging code was littered with places where we stored messages in
static char[] fields. This is fine in a single-threaded program, but if you
ever tried to log evdns messages from two threads at once, you'd hit a race.
This patch also refactors evdns's debug_ntop function into a more useful
evutil_sockaddr_port_format() function, with unit tests.
If the user sets a bind address to use for nameservers, and a
nameserver happens to be on 127.0.0.1, the nameserver will generally
fail. This patch alters this behavior so that the bind address is
only applied when the nameserver is on a non-loopback address.
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
