[RHEL7,COMMIT] ms/net: Allow to show socket-specific information in /proc/[pid]/fdinfo/[fd]

Submitted by Konstantin Khorenko on Dec. 26, 2019, 1:58 p.m.


Message ID 201912261358.xBQDwhPw005870@finist-ce7.sw.ru
State New
Series "unix: Show number of scm files in fdinfo"
Headers show

Commit Message

Konstantin Khorenko Dec. 26, 2019, 1:58 p.m.
The commit is pushed to "branch-rh7-3.10.0-1062.7.1.vz7.130.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh7-3.10.0-1062.7.1.vz7.130.5
commit 27e26a1cb241e6404ee4fae6ef98b8ac3e7073a4
Author: Kirill Tkhai <ktkhai@virtuozzo.com>
Date:   Thu Dec 26 16:58:43 2019 +0300

    ms/net: Allow to show socket-specific information in /proc/[pid]/fdinfo/[fd]
    This adds .show_fdinfo to socket_file_ops, so protocols will be able
    to print their specific data in fdinfo.
    Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
    Signed-off-by: David S. Miller <davem@davemloft.net>
    Patchset description:
    This patchset expands the information whether a socket
    contains scmd fds or not:
    $cat /proc/[pid]/fdinfo/[unix_sk_fd] | grep scm_fds
    scm_fds: 1
    This sometimes will be useful to check who owns a counter
    to a resource and prevents it destruction. Say, someone
    opens ploop and places its fd into a unix socket.
    Kirill Tkhai (2):
          net: Allow to show socket-specific information in /proc/[pid]/fdinfo/[fd]
          unix: Show number of pending scm files of receive queue in fdinfo
 include/linux/net.h |  1 +
 net/socket.c        | 13 +++++++++++++
 2 files changed, 14 insertions(+)

Patch hide | download patch | download mbox

diff --git a/include/linux/net.h b/include/linux/net.h
index d1c264888975e..168925acfb65e 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -164,6 +164,7 @@  struct proto_ops {
 	int		(*compat_getsockopt)(struct socket *sock, int level,
 				      int optname, char __user *optval, int __user *optlen);
+	void		(*show_fdinfo)(struct seq_file *m, struct socket *sock);
 	int		(*sendmsg)   (struct kiocb *iocb, struct socket *sock,
 				      struct msghdr *m, size_t total_len);
 	/* Notes for implementing recvmsg:
diff --git a/net/socket.c b/net/socket.c
index 3e1368f686c91..c5f949bf3ac2f 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -135,6 +135,7 @@  static ssize_t sock_sendpage(struct file *file, struct page *page,
 static ssize_t sock_splice_read(struct file *file, loff_t *ppos,
 				struct pipe_inode_info *pipe, size_t len,
 				unsigned int flags);
+static int sock_show_fdinfo(struct seq_file *m, struct file *f);
  *	Socket files have a set of 'special' operations as well as the generic file ones. These don't appear
@@ -157,6 +158,9 @@  static const struct file_operations socket_file_ops = {
 	.sendpage =	sock_sendpage,
 	.splice_write = generic_splice_sendpage,
 	.splice_read =	sock_splice_read,
+	.show_fdinfo =	sock_show_fdinfo,
@@ -1017,6 +1021,15 @@  static ssize_t sock_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	return do_sock_write(&x->async_msg, iocb, iocb->ki_filp, iov, nr_segs);
+static int sock_show_fdinfo(struct seq_file *m, struct file *f)
+	struct socket *sock = f->private_data;
+	if (sock->ops->show_fdinfo)
+		sock->ops->show_fdinfo(m, sock);
+	return 0;
  * Atomic setting of ioctl hooks to avoid race
  * with module unload.