9c469db300e1b270a93c6b04c1709ac0f7751136 had started with the conversion
but left out another instance of incorrectly used AM_CFLAGS.
Dave Hart provided the suggestion for this fix.
The main reason for disabling installation is if you're building
libevent as a subpackage for embedding: you want to have your main
package's "make all" build libevent, but you don't want your main
package's "make install" to install libevent.
If the EVHTTP_URI_NONCONFORMANT flag is passed in (which it is when
parsing URIs we get over the wire), then we relax our checks a lot.
Specifically, we do nothing to check for correct characters in the
path, query, and fragment parts of such a URI.
We could do much more here: we could relax our hostname requirements,
deal with spaces differently/better, trap some errors but not others,
etc. But this should solve the worst user-agent compatibility issues
for now; the other issues can wait for a later release.
evhttp needs to be mindful of all hostnames and addresses that clients
use to contact the main server and vhosts to know the difference between
proxy requests and non-proxy requests.
The only changes needed were to handle the fact that the methodname
"epoll (with changelist)" matches the environment variable
EVENT_NOEPOLL rather than the imaginary "EVENT_EPOLL (WITH CHANGELIST)".
Current versions of the Linux kernel don't seem to remove the struct
epitem for a given (file,fd) combo when the fd is closed unless the
file itself is also completely closed. This means that if you do:
fd = dup(fd_orig);
add(fd);
close(fd);
dup2(fd_orig, fd);
add(fd);
you will get an EEXIST when you should have gotten a success. This
could cause warnings and dropped events when using dup and epoll.
The solution is pretty simple: when we get an EEXIST from
EPOLL_CTL_ADD, we retry with EPOLL_CTL_MOD.
Unit test included to demonstrate the bug.
Found due to the patient efforts of Gilad Benjamini; diagnosed with
help from Nicholas Marriott.
