Re: [PATCH 0/1] RFC: poll/select performance on datagram sockets
From: Eric Dumazet <hidden>
Date: 2010-10-30 13:18:11
Also in:
lkml
quoted hunk ↗ jump to hunk
Problem is the peer_wait, that epoll doesnt seem to be plugged into. Bug is in unix_dgram_poll() It calls sock_poll_wait( ... &unix_sk(other)->peer_wait,) only if socket is 'writable'. Its a clear bug Try this patch please ?diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 0ebc777..315716c 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c@@ -2092,7 +2092,7 @@ static unsigned int unix_dgram_poll(struct file *file, struct socket *sock, /* writable? */ writable = unix_writable(sk); - if (writable) { + if (1 /*writable*/) { other = unix_peer_get(sk); if (other) { if (unix_peer(other) != sk) {
Also you'll need to change your program to make the epoll registrations
_after_ sockets connect(), or else you can see that epoll() wont know
about the other peer stuff.
for (i = 0 ; i < NB_CLIENTS ; i++) {
client_fds[i] = socket(AF_UNIX, SOCK_DGRAM, 0);
}
for (i = 0 ; i < NB_CLIENTS ; i++) {
connect(client_fds[i], (struct sockaddr*)&addr_server, sizeof(addr_server));
}
ev.events = EPOLLOUT;
ev.data.fd = client_fds[NB_CLIENTS-1];
if (epoll_ctl(epollfd, EPOLL_CTL_ADD, client_fds[NB_CLIENTS-1], &ev) == -1) {
perror("epoll_ctl: client_fds max");
exit(EXIT_FAILURE);
}
if (trigger == 0) {
ev.events = EPOLLOUT;
ev.data.fd = client_fds[0];
if (epoll_ctl(epollfd, EPOLL_CTL_ADD, client_fds[0], &ev) == -1) {
perror("epoll_ctl: client_fds 0");
exit(EXIT_FAILURE);
}
}