[PATCH/RFC] New module refcounting for net_proto_family

From: Max Krasnyansky (maxk@qualcomm.com)
Date: Wed Dec 18 2002 - 10:25:55 EST

Hi Folks,

It seems that new module code is going to stay with us at least for a
while :).
So it's probably time to start fixing old interfaces to use new
refcounting scheme.

Here is a patch for sock_create() and stuff that uses net_proto_family
interface. Tested with modified af_bluetooth.c and seems to work fine.
Other families are unaffected for now because their owner field == NULL.

If people are ok with this aproach I will also fix af_unix and other
families and push all fixes into my BK tree.

(URL in case if my mailer messed up spaces
http://bluez.sourceforge.net/patches/npf_refcnt_patch-2.5.52.gz)

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or
higher.
# This patch includes the following deltas:
# ChangeSet 1.888 -> 1.889
# net/socket.c 1.39 -> 1.40
# include/linux/net.h 1.7 -> 1.8
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 02/12/17 maxk@qualcomm.com 1.889
# Convert generic socket code to new module refcounting.
# --------------------------------------------
#
diff -Nru a/include/linux/net.h b/include/linux/net.h
--- a/include/linux/net.h Tue Dec 17 20:02:04 2002
+++ b/include/linux/net.h Tue Dec 17 20:02:04 2002
@@ -76,6 +76,8 @@
 
         short type;
         unsigned char passcred;
+
+ struct module *owner;
 };
 
 struct scm_cookie;
@@ -124,6 +126,8 @@
         short authentication;
         short encryption;
         short encrypt_net;
+
+ struct module *owner;
 };
 
 struct net_proto
diff -Nru a/net/socket.c b/net/socket.c
--- a/net/socket.c Tue Dec 17 20:02:04 2002
+++ b/net/socket.c Tue Dec 17 20:02:04 2002
@@ -470,6 +470,8 @@
 
         sock = SOCKET_I(inode);
 
+ sock->owner = NULL;
+
         inode->i_mode = S_IFSOCK|S_IRWXUGO;
         inode->i_sock = 1;
         inode->i_uid = current->fsuid;
@@ -517,6 +519,8 @@
                 return;
         }
         sock->file=NULL;
+
+ module_put(sock->owner);
 }
 
 static int __sock_sendmsg(struct kiocb *iocb, struct socket *sock,
struct msghdr *msg, int size)
@@ -964,8 +968,9 @@
 
 int sock_create(int family, int type, int protocol, struct socket
**res)
 {
- int i;
+ struct net_proto_family *npf;
         struct socket *sock;
+ int err;
 
         /*
          * Check protocol is in range
@@ -990,14 +995,8 @@
         }
                 
 #if defined(CONFIG_KMOD) && defined(CONFIG_NET)
- /* Attempt to load a protocol module if the find failed.
- *
- * 12/09/1996 Marcin: But! this makes REALLY only sense, if the user
- * requested real, full-featured networking support upon
configuration.
- * Otherwise module support will break!
- */
- if (net_families[family]==NULL)
- {
+ /* Attempt to load a protocol module if the find failed. */
+ if (net_families[family]==NULL) {
                 char module_name[30];
                 sprintf(module_name,"net-pf-%d",family);
                 request_module(module_name);
@@ -1005,29 +1004,31 @@
 #endif
 
         net_family_read_lock();
- if (net_families[family] == NULL) {
- i = -EAFNOSUPPORT;
+
+ npf = net_families[family];
+ if (!npf || !try_module_get(npf->owner)) {
+ err = -EAFNOSUPPORT;
                 goto out;
         }
+
+ /*
+ * Allocate the socket and allow the family to set things up. if
+ * the protocol is 0, the family is instructed to select an
appropriate
+ * default.
+ */
 
-/*
- * Allocate the socket and allow the family to set things up. if
- * the protocol is 0, the family is instructed to select an appropriate
- * default.
- */
-
- if (!(sock = sock_alloc()))
- {
+ sock = sock_alloc();
+ if (!sock) {
                 printk(KERN_WARNING "socket: no more sockets\n");
- i = -ENFILE; /* Not exactly a match, but its the
+ err = -ENFILE; /* Not exactly a match, but its the
                                            closest posix thing */
                 goto out;
         }
 
         sock->type = type;
+ sock->owner = npf->owner;
 
- if ((i = net_families[family]->create(sock, protocol)) < 0)
- {
+ if ((err = npf->create(sock, protocol)) < 0) {
                 sock_release(sock);
                 goto out;
         }
@@ -1036,7 +1037,7 @@
 
 out:
         net_family_read_unlock();
- return i;
+ return err;
 }
 
 asmlinkage long sys_socket(int family, int type, int protocol)
@@ -1201,6 +1202,9 @@
         newsock->type = sock->type;
         newsock->ops = sock->ops;
 
+ try_module_get(sock->owner);
+ newsock->owner = sock->owner;
+
         err = sock->ops->accept(sock, newsock, sock->file->f_flags);
         if (err < 0)
                 goto out_release; 

-- 
Max Krasnyansky <maxk@qualcomm.com>
Qualcomm

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

This archive was generated by hypermail 2b29 : Mon Dec 23 2002 - 22:00:20 EST