Skip to contentSkip to conversation

Add c/r support for listening INET sockets

RFC PATCH In review Dan Smith 3 participants 6 messages Sep 28, 2009 – Sep 30, 2009 3 messages not in the archive Networking

lore
Share & tools
lore

Dan Smith

author
#0 lore raw

RFC PATCH 1/1 Add c/r support for listening INET sockets

This is an incremental step towards supporting checkpoint/restart on AF_INET sockets. In this scenario, any sockets that were in TCP_LISTEN state are restored as they were. Any that were connected are forced to TCP_CLOSE. This should cover a range of use cases that involve applications that are tolerant of such an interruption.

Cc Oren Laadan <[email protected]>
Cc Andrew Morton <akpm-de/[email protected]>
---
 include/linux/checkpoint_hdr.h |   11 ++
 include/net/inet_common.h      |    9 ++
 net/checkpoint.c               |    9 ++
 net/ipv4/Makefile              |    1 +
 net/ipv4/af_inet.c             |    6 +
 net/ipv4/checkpoint.c          |  204 ++++++++++++++++++++++++++++++++++++++++
 6 files changed, 240 insertions(+), 0 deletions(-)
 create mode 100644 net/ipv4/checkpoint.c
include/linux/checkpoint_hdr.h +11 -0
index 2ed523f..b5ce115 100644
@@ -15,9 +15,11 @@
 #ifdef __KERNEL__
 #include <linux/socket.h>
 #include <linux/un.h>
+#include <linux/in.h>
 #else
 #include <sys/socket.h>
 #include <sys/un.h>
+#include <sys/in.h>
 #endif
 
 /*
@@ -106,6 +108,7 @@ enum {
 	CKPT_HDR_SOCKET_QUEUE,
 	CKPT_HDR_SOCKET_BUFFER,
 	CKPT_HDR_SOCKET_UNIX,
+	CKPT_HDR_SOCKET_INET,
 
 	CKPT_HDR_TAIL = 9001,
 
@@ -470,6 +473,14 @@ struct ckpt_hdr_socket_unix {
 	struct sockaddr_un raddr;
 } __attribute__ ((aligned(8)));
 
+struct ckpt_hdr_socket_inet {
+	struct ckpt_hdr h;
+	__u32 laddr_len;
+	__u32 raddr_len;
+	struct sockaddr_in laddr;
+	struct sockaddr_in raddr;
+} __attribute__((aligned(8)));
+
 struct ckpt_hdr_file_socket {
 	struct ckpt_hdr_file common;
 	__s32 sock_objref;
include/net/inet_common.h +9 -0
index 18c7732..7ade732 100644
@@ -45,6 +45,15 @@ extern int			inet_ctl_sock_create(struct sock **sk,
 						     unsigned char protocol,
 						     struct net *net);
 
+#ifdef CONFIG_CHECKPOINT
+struct ckpt_ctx;
+struct ckpt_hdr_socket;
+extern int inet_checkpoint(struct ckpt_ctx *ctx, struct socket *sock);
+extern int inet_collect(struct ckpt_ctx *ctx, struct socket *sock);
+extern int inet_restore(struct ckpt_ctx *cftx, struct socket *sock,
+			struct ckpt_hdr_socket *h);
+#endif /* CONFIG_CHECKPOINT */
+
 static inline void inet_ctl_sock_destroy(struct sock *sk)
 {
 	sk_release_kernel(sk);
net/checkpoint.c +9 -0
index a11ec7a..6960389 100644
@@ -711,6 +711,15 @@ struct sock *do_sock_restore(struct ckpt_ctx *ctx)
 	if (ret < 0)
 		goto err;
 
+	if ((h->sock_common.family == AF_INET) &&
+	    (h->sock.state != TCP_LISTEN)) {
+		/* Temporary hack to enable restore of TCP_LISTEN sockets
+		 * while forcing anything else to a closed state
+		 */
+		sock->sk->sk_state = TCP_CLOSE;
+		sock->state = SS_UNCONNECTED;
+	}
+
 	ckpt_hdr_put(ctx, h);
 	return sock->sk;
  err:
net/ipv4/Makefile +1 -0
index 80ff87c..c00d8ce 100644
@@ -49,6 +49,7 @@ obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
 obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
 obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
 obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
+obj-$(CONFIG_CHECKPOINT) += checkpoint.o
 
 obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
 		      xfrm4_output.o
net/ipv4/af_inet.c +6 -0
index 566ea6c..7828885 100644
@@ -857,6 +857,9 @@ const struct proto_ops inet_stream_ops = {
 	.mmap		   = sock_no_mmap,
 	.sendpage	   = tcp_sendpage,
 	.splice_read	   = tcp_splice_read,
+	.checkpoint        = inet_checkpoint,
+	.restore           = inet_restore,
+	.collect           = inet_collect,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_sock_common_setsockopt,
 	.compat_getsockopt = compat_sock_common_getsockopt,
@@ -882,6 +885,9 @@ const struct proto_ops inet_dgram_ops = {
 	.recvmsg	   = sock_common_recvmsg,
 	.mmap		   = sock_no_mmap,
 	.sendpage	   = inet_sendpage,
+	.checkpoint        = inet_checkpoint,
+	.restore           = inet_restore,
+	.collect           = inet_collect,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_sock_common_setsockopt,
 	.compat_getsockopt = compat_sock_common_getsockopt,
net/ipv4/checkpoint.c new file +204 -0

204 lines not on this page · raw message

Signature · 1 line
1.6.2.5

A message that is not in the archive (sent off-list or not yet received): 1254164330-23072-1-git-send-email-danms-r/[email protected]

#2 lore raw
Quoted 83 lines · Dan Smith

This is an incremental step towards supporting checkpoint/restart on AF_INET sockets. In this scenario, any sockets that were in TCP_LISTEN state are restored as they were. Any that were connected are forced to TCP_CLOSE. This should cover a range of use cases that involve applications that are tolerant of such an interruption.

Cc Oren Laadan <[email protected]>
Cc Andrew Morton <akpm-de/[email protected]>
---
 include/linux/checkpoint_hdr.h |   11 ++
 include/net/inet_common.h      |    9 ++
 net/checkpoint.c               |    9 ++
 net/ipv4/Makefile              |    1 +
 net/ipv4/af_inet.c             |    6 +
 net/ipv4/checkpoint.c          |  204 ++++++++++++++++++++++++++++++++++++++++
 6 files changed, 240 insertions(+), 0 deletions(-)
 create mode 100644 net/ipv4/checkpoint.c
include/linux/checkpoint_hdr.h +11 -0
index 2ed523f..b5ce115 100644
@@ -15,9 +15,11 @@
 #ifdef __KERNEL__
 #include <linux/socket.h>
 #include <linux/un.h>
+#include <linux/in.h>
 #else
 #include <sys/socket.h>
 #include <sys/un.h>
+#include <sys/in.h>
 #endif
 
 /*
@@ -106,6 +108,7 @@ enum {
 	CKPT_HDR_SOCKET_QUEUE,
 	CKPT_HDR_SOCKET_BUFFER,
 	CKPT_HDR_SOCKET_UNIX,
+	CKPT_HDR_SOCKET_INET,
 
 	CKPT_HDR_TAIL = 9001,
 
@@ -470,6 +473,14 @@ struct ckpt_hdr_socket_unix {
 	struct sockaddr_un raddr;
 } __attribute__ ((aligned(8)));
 
+struct ckpt_hdr_socket_inet {
+	struct ckpt_hdr h;
+	__u32 laddr_len;
+	__u32 raddr_len;
+	struct sockaddr_in laddr;
+	struct sockaddr_in raddr;
+} __attribute__((aligned(8)));
+
 struct ckpt_hdr_file_socket {
 	struct ckpt_hdr_file common;
 	__s32 sock_objref;
include/net/inet_common.h +9 -0
index 18c7732..7ade732 100644
@@ -45,6 +45,15 @@ extern int			inet_ctl_sock_create(struct sock **sk,
 						     unsigned char protocol,
 						     struct net *net);
 
+#ifdef CONFIG_CHECKPOINT
+struct ckpt_ctx;
+struct ckpt_hdr_socket;
+extern int inet_checkpoint(struct ckpt_ctx *ctx, struct socket *sock);
+extern int inet_collect(struct ckpt_ctx *ctx, struct socket *sock);
+extern int inet_restore(struct ckpt_ctx *cftx, struct socket *sock,
+			struct ckpt_hdr_socket *h);
+#endif /* CONFIG_CHECKPOINT */
+
 static inline void inet_ctl_sock_destroy(struct sock *sk)
 {
 	sk_release_kernel(sk);
net/checkpoint.c +0 -0
index a11ec7a..6960389 100644

net/checkpoint.c struct sock *do_sock_restore(struct ckpt_ctx *ctx) · line 719

Quoted 9 lines
net/checkpoint.c +6 -0
@@ -711,6 +711,15 @@ struct sock *do_sock_restore(struct ckpt_ctx *ctx)
 	if (ret < 0)
 		goto err;
 
+	if ((h->sock_common.family == AF_INET) &&
+	    (h->sock.state != TCP_LISTEN)) {
+		/* Temporary hack to enable restore of TCP_LISTEN sockets
+		 * while forcing anything else to a closed state
+		 */
+		sock->sk->sk_state = TCP_CLOSE;

I'll admit by this point I'm a bit confused about the order in which these things are restored. Does do_sock_restore() run before inet_restore() below (curious whether its check for TCP_CLOSE will catch these particular sockets).

Quoted 76 lines
net/checkpoint.c +3 -0
+		sock->state = SS_UNCONNECTED;
+	}
+
 	ckpt_hdr_put(ctx, h);
 	return sock->sk;
  err:
net/ipv4/Makefile +1 -0
index 80ff87c..c00d8ce 100644
@@ -49,6 +49,7 @@ obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
 obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
 obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
 obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
+obj-$(CONFIG_CHECKPOINT) += checkpoint.o
 
 obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
 		      xfrm4_output.o
net/ipv4/af_inet.c +6 -0
index 566ea6c..7828885 100644
@@ -857,6 +857,9 @@ const struct proto_ops inet_stream_ops = {
 	.mmap		   = sock_no_mmap,
 	.sendpage	   = tcp_sendpage,
 	.splice_read	   = tcp_splice_read,
+	.checkpoint        = inet_checkpoint,
+	.restore           = inet_restore,
+	.collect           = inet_collect,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_sock_common_setsockopt,
 	.compat_getsockopt = compat_sock_common_getsockopt,
@@ -882,6 +885,9 @@ const struct proto_ops inet_dgram_ops = {
 	.recvmsg	   = sock_common_recvmsg,
 	.mmap		   = sock_no_mmap,
 	.sendpage	   = inet_sendpage,
+	.checkpoint        = inet_checkpoint,
+	.restore           = inet_restore,
+	.collect           = inet_collect,
 #ifdef CONFIG_COMPAT
 	.compat_setsockopt = compat_sock_common_setsockopt,
 	.compat_getsockopt = compat_sock_common_getsockopt,
net/ipv4/checkpoint.c new file +28 -0
new file mode 100644
index 0000000..881e723
@@ -0,0 +1,204 @@
+/*
+ *  Copyright 2009 IBM Corporation
+ *
+ *  Author(s): Dan Smith <danms-r/[email protected]>
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License as
+ *  published by the Free Software Foundation, version 2 of the
+ *  License.
+ */
+
+#include <linux/namei.h>
+#include <linux/checkpoint.h>
+#include <linux/checkpoint_hdr.h>
+#include <linux/tcp.h>
+#include <linux/in.h>
+#include <linux/deferqueue.h>
+#include <net/tcp_states.h>
+#include <net/tcp.h>
+
+struct dq_sock {
+	struct ckpt_ctx *ctx;
+	struct sock *sk;
+};
+
+struct dq_buffers {
+	struct ckpt_ctx *ctx;
+	struct sock *sk;

net/ipv4/checkpoint.c line 38

Quoted 10 lines
net/ipv4/checkpoint.c new file +10 -0
+};
+
+int inet_checkpoint(struct ckpt_ctx *ctx, struct socket *sock)
+{
+	struct ckpt_hdr_socket_inet *in;
+	int ret = -EINVAL;
+
+       in = ckpt_hdr_get_type(ctx, sizeof(*in), CKPT_HDR_SOCKET_INET);
+       if (!in)
+               goto out;

Better to just return... I don't think this'll cause a crash (it only makes ckpt_hdr_put() pass ((struct ckpt_hdr *)NULL)->len to _ckpt_hdr_put which then ignores it and does a safe kfree(NULL)), but it won't be safe if the implementation of the ckpt_hdr_*() calls end up changing.

Quoted 173 lines
net/ipv4/checkpoint.c new file +166 -0
+
+       ret = ckpt_sock_getnames(ctx, sock,
+				(struct sockaddr *)&in->laddr, &in->laddr_len,
+				(struct sockaddr *)&in->raddr, &in->raddr_len);
+       if (ret)
+	       goto out;
+
+       ret = ckpt_write_obj(ctx, (struct ckpt_hdr *) in);
+ out:
+       ckpt_hdr_put(ctx, in);
+
+       return ret;
+}
+
+int inet_collect(struct ckpt_ctx *ctx, struct socket *sock)
+{
+       return ckpt_obj_collect(ctx, sock->sk, CKPT_OBJ_SOCK);
+}
+
+static int inet_read_buffer(struct ckpt_ctx *ctx, struct sk_buff_head *queue)
+{
+	struct ckpt_hdr_socket_buffer *h;
+	int len;
+	int ret;
+	struct sk_buff *skb;
+
+	h = ckpt_read_obj_type(ctx, sizeof(*h), CKPT_HDR_SOCKET_BUFFER);
+	if (IS_ERR(h))
+		return PTR_ERR(h);
+
+	len = _ckpt_read_obj_type(ctx, NULL, 0, CKPT_HDR_BUFFER);
+	if (len < 0) {
+		ret = len;
+		goto out;
+	} else if (len > SKB_MAX_ALLOC) {
+		ckpt_debug("Socket buffer too big (%i > %lu)",
+			   len, SKB_MAX_ALLOC);
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	skb = alloc_skb(len, GFP_KERNEL);
+	if (!skb) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = ckpt_kread(ctx, skb_put(skb, len), len);
+	if (ret < 0)
+		goto out;
+
+	spin_lock(&queue->lock);
+	skb_queue_tail(queue, skb);
+	spin_unlock(&queue->lock);
+ out:
+	ckpt_hdr_put(ctx, h);
+
+	if (ret < 0)
+		kfree_skb(skb);
+
+	return ret;
+}
+
+static int inet_read_buffers(struct ckpt_ctx *ctx, struct sk_buff_head *queue)
+{
+	struct ckpt_hdr_socket_queue *h;
+	int ret = 0;
+	int i;
+
+	h = ckpt_read_obj_type(ctx, sizeof(*h), CKPT_HDR_SOCKET_QUEUE);
+	if (IS_ERR(h))
+		return PTR_ERR(h);
+
+	for (i = 0; i < h->skb_count; i++) {
+		ret = inet_read_buffer(ctx, queue);
+		ckpt_debug("read inet buffer %i: %i", i, ret);
+		if (ret < 0)
+			goto out;
+
+		if (ret > h->total_bytes) {
+			ckpt_debug("Buffers exceeded claim");
+			ret = -EINVAL;
+			goto out;
+		}
+
+		h->total_bytes -= ret;
+		ret = 0;
+	}
+
+	ret = h->skb_count;
+ out:
+	ckpt_hdr_put(ctx, h);
+
+	return ret;
+}
+
+static int inet_deferred_restore_buffers(void *data)
+{
+	struct dq_buffers *dq = (struct dq_buffers *)data;
+	struct ckpt_ctx *ctx = dq->ctx;
+	struct sock *sk = dq->sk;
+	int ret;
+
+	ret = inet_read_buffers(ctx, &sk->sk_receive_queue);
+	ckpt_debug("(R) inet_read_buffers: %i\n", ret);
+	if (ret < 0)
+		return ret;
+
+	ret = inet_read_buffers(ctx, &sk->sk_write_queue);
+	ckpt_debug("(W) inet_read_buffers: %i\n", ret);
+
+	return ret;
+}
+
+static int inet_defer_restore_buffers(struct ckpt_ctx *ctx, struct sock *sk)
+{
+	struct dq_buffers dq;
+
+	dq.ctx = ctx;
+	dq.sk = sk;
+
+	return deferqueue_add(ctx->files_deferq, &dq, sizeof(dq),
+			      inet_deferred_restore_buffers, NULL);
+}
+
+int inet_restore(struct ckpt_ctx *ctx,
+		 struct socket *sock,
+		 struct ckpt_hdr_socket *h)
+{
+	struct ckpt_hdr_socket_inet *in;
+	int ret = 0;
+
+	in = ckpt_read_obj_type(ctx, sizeof(*in), CKPT_HDR_SOCKET_INET);
+	if (IS_ERR(in))
+		return PTR_ERR(in);
+
+	/* Listening sockets and those that are closed but have a local
+	 * address need to call bind()
+	 */
+	if ((h->sock.state == TCP_LISTEN) ||
+	    ((h->sock.state == TCP_CLOSE) && (in->laddr_len > 0))) {
+		sock->sk->sk_reuse = 2;
+		inet_sk(sock->sk)->freebind = 1;
+		ret = sock->ops->bind(sock,
+				      (struct sockaddr *)&in->laddr,
+				      in->laddr_len);
+		ckpt_debug("inet bind: %i\n", ret);
+		if (ret < 0)
+			goto out;
+
+		if (h->sock.state == TCP_LISTEN) {
+			ret = sock->ops->listen(sock, h->sock.backlog);
+			ckpt_debug("inet listen: %i\n", ret);
+			if (ret < 0)
+				goto out;
+		}
+	} else {
+		if (!sock_flag(sock->sk, SOCK_DEAD))
+			ret = inet_defer_restore_buffers(ctx, sock->sk);
+	}
+ out:
+	ckpt_hdr_put(ctx, in);
+
+	return ret;
+ }
+
Signature · 6 lines
1.6.2.5

_______________________________________________
Containers mailing list
Containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org
https://lists.linux-foundation.org/mailman/listinfo/containers

A message that is not in the archive (sent off-list or not yet received): 20090929195025.GA14956-r/[email protected]

author
#4 lore raw
Quoted 6 lines

net/checkpoint.c struct sock *do_sock_restore(struct ckpt_ctx *ctx) · line 719

Quoted 6 lines
net/checkpoint.c +6 -0
+	if ((h->sock_common.family == AF_INET) &&
+	    (h->sock.state != TCP_LISTEN)) {
+		/* Temporary hack to enable restore of TCP_LISTEN sockets
+		 * while forcing anything else to a closed state
+		 */
+		sock->sk->sk_state = TCP_CLOSE;

SH> I'll admit by this point I'm a bit confused about the order in SH> which these things are restored. Does do_sock_restore() run SH> before inet_restore() below (curious whether its check for SH> TCP_CLOSE will catch these particular sockets).

It runs after inet_restore(), which was called above via sock->ops->restore(). Thus, inet_restore() can largely ignore the connected sockets and then they're forced closed here (since inet_restore() runs before the common restore routine, otherwise it could set them closed itself).

Quoted 8 lines

net/ipv4/checkpoint.c line 38

Quoted 8 lines
net/ipv4/checkpoint.c new file +8 -0
+int inet_checkpoint(struct ckpt_ctx *ctx, struct socket *sock)
+{
+	struct ckpt_hdr_socket_inet *in;
+	int ret = -EINVAL;
+
+       in = ckpt_hdr_get_type(ctx, sizeof(*in), CKPT_HDR_SOCKET_INET);
+       if (!in)
+               goto out;

SH> Better to just return... I don't think this'll cause a crash (it SH> only makes ckpt_hdr_put() pass ((struct ckpt_hdr *)NULL)->len to SH> _ckpt_hdr_put which then ignores it and does a safe kfree(NULL)), SH> but it won't be safe if the implementation of the ckpt_hdr_*() SH> calls end up changing.

Sure. Also note my broken whitespace that came as a result of me cutting patches into pieces. I'll fix that too.

Signature · 3 lines
Dan Smith
IBM Linux Technology Center
email: danms-r/[email protected]
#5 lore raw
In reply to a message not in the archive
Quoted 6 lines · Dan Smith

This is an incremental step towards supporting checkpoint/restart on AF_INET sockets. In this scenario, any sockets that were in TCP_LISTEN state are restored as they were. Any that were connected are forced to TCP_CLOSE. This should cover a range of use cases that involve applications that are tolerant of such an interruption.

It's important to mention that pending connections - those that were received and are awaiting to be accepted on the listening socket are not handled either (does checkpoint fail or ignore them ?)

Oren.

A message that is not in the archive (sent off-list or not yet received): [email protected]

author
#7 lore raw

OL> It's important to mention that pending connections - those that OL> were received and are awaiting to be accepted on the listening OL> socket are not handled either (does checkpoint fail or ignore them OL> ?)

It ignores them and presumably they'll get an RST after the restart happens and claims no knowledge of the pending connection.

Signature · 3 lines
Dan Smith
IBM Linux Technology Center
email: danms-r/[email protected]
#8 lore raw
In reply to a message not in the archive
Quoted 10 lines · Dan Smith

This is an incremental step towards supporting checkpoint/restart on AF_INET sockets. In this scenario, any sockets that were in TCP_LISTEN state are restored as they were. Any that were connected are forced to TCP_CLOSE. This should cover a range of use cases that involve applications that are tolerant of such an interruption.

Cc Oren Laadan <[email protected]>
Cc Andrew Morton <akpm-de/[email protected]>

---

Good stuff... a few comments below.

[...]

Quoted 5 lines
net/ipv4/checkpoint.c new file +5 -0
+
+static int inet_read_buffer(struct ckpt_ctx *ctx, struct sk_buff_head *queue)
+{
+	struct ckpt_hdr_socket_buffer *h;
+	int len;

net/ipv4/checkpoint.c line 71

Quoted 10 lines
net/ipv4/checkpoint.c new file +10 -0
+	int ret;
+	struct sk_buff *skb;
+
+	h = ckpt_read_obj_type(ctx, sizeof(*h), CKPT_HDR_SOCKET_BUFFER);
+	if (IS_ERR(h))
+		return PTR_ERR(h);
+
+	len = _ckpt_read_obj_type(ctx, NULL, 0, CKPT_HDR_BUFFER);
+	if (len < 0) {
+		ret = len;

Note: skb wasn't initialized yet, ret will be negative (more below)...

Quoted 16 lines
net/ipv4/checkpoint.c new file +16 -0
+		goto out;
+	} else if (len > SKB_MAX_ALLOC) {
+		ckpt_debug("Socket buffer too big (%i > %lu)",
+			   len, SKB_MAX_ALLOC);
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	skb = alloc_skb(len, GFP_KERNEL);
+	if (!skb) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	ret = ckpt_kread(ctx, skb_put(skb, len), len);
+	if (ret < 0)

net/ipv4/checkpoint.c line 97

Quoted 10 lines
net/ipv4/checkpoint.c new file +10 -0
+		goto out;
+
+	spin_lock(&queue->lock);
+	skb_queue_tail(queue, skb);
+	spin_unlock(&queue->lock);
+ out:
+	ckpt_hdr_put(ctx, h);
+
+	if (ret < 0)
+		kfree_skb(skb);

skb may be uninitialized at this point (do I sound like gcc ?)

Quoted 18 lines
net/ipv4/checkpoint.c new file +18 -0
+
+	return ret;
+}
+
+static int inet_read_buffers(struct ckpt_ctx *ctx, struct sk_buff_head *queue)
+{
+	struct ckpt_hdr_socket_queue *h;
+	int ret = 0;
+	int i;
+
+	h = ckpt_read_obj_type(ctx, sizeof(*h), CKPT_HDR_SOCKET_QUEUE);
+	if (IS_ERR(h))
+		return PTR_ERR(h);
+
+	for (i = 0; i < h->skb_count; i++) {
+		ret = inet_read_buffer(ctx, queue);
+		ckpt_debug("read inet buffer %i: %i", i, ret);
+		if (ret < 0)

net/ipv4/checkpoint.c line 125

Quoted 10 lines
net/ipv4/checkpoint.c new file +10 -0
+			goto out;
+
+		if (ret > h->total_bytes) {
+			ckpt_debug("Buffers exceeded claim");
+			ret = -EINVAL;
+			goto out;
+		}
+
+		h->total_bytes -= ret;
+		ret = 0;
		^^^^^^^^

Unnecessary ?

Quoted 38 lines
net/ipv4/checkpoint.c new file +38 -0
+	}
+
+	ret = h->skb_count;
+ out:
+	ckpt_hdr_put(ctx, h);
+
+	return ret;
+}
+
+static int inet_deferred_restore_buffers(void *data)
+{
+	struct dq_buffers *dq = (struct dq_buffers *)data;
+	struct ckpt_ctx *ctx = dq->ctx;
+	struct sock *sk = dq->sk;
+	int ret;
+
+	ret = inet_read_buffers(ctx, &sk->sk_receive_queue);
+	ckpt_debug("(R) inet_read_buffers: %i\n", ret);
+	if (ret < 0)
+		return ret;
+
+	ret = inet_read_buffers(ctx, &sk->sk_write_queue);
+	ckpt_debug("(W) inet_read_buffers: %i\n", ret);
+
+	return ret;
+}
+
+static int inet_defer_restore_buffers(struct ckpt_ctx *ctx, struct sock *sk)
+{
+	struct dq_buffers dq;
+
+	dq.ctx = ctx;
+	dq.sk = sk;
+
+	return deferqueue_add(ctx->files_deferq, &dq, sizeof(dq),
+			      inet_deferred_restore_buffers, NULL);
+}
+

net/ipv4/checkpoint.c line 173

Quoted 10 lines
net/ipv4/checkpoint.c new file +10 -0
+int inet_restore(struct ckpt_ctx *ctx,
+		 struct socket *sock,
+		 struct ckpt_hdr_socket *h)
+{
+	struct ckpt_hdr_socket_inet *in;
+	int ret = 0;
+
+	in = ckpt_read_obj_type(ctx, sizeof(*in), CKPT_HDR_SOCKET_INET);
+	if (IS_ERR(in))
+		return PTR_ERR(in);

I think you need to at least sanitize the in->laddr_len field - the ops->bind() below assumes its valid.

Quoted 21 lines
net/ipv4/checkpoint.c new file +21 -0
+
+	/* Listening sockets and those that are closed but have a local
+	 * address need to call bind()
+	 */
+	if ((h->sock.state == TCP_LISTEN) ||
+	    ((h->sock.state == TCP_CLOSE) && (in->laddr_len > 0))) {
+		sock->sk->sk_reuse = 2;
+		inet_sk(sock->sk)->freebind = 1;
+		ret = sock->ops->bind(sock,
+				      (struct sockaddr *)&in->laddr,
+				      in->laddr_len);
+		ckpt_debug("inet bind: %i\n", ret);
+		if (ret < 0)
+			goto out;
+
+		if (h->sock.state == TCP_LISTEN) {
+			ret = sock->ops->listen(sock, h->sock.backlog);
+			ckpt_debug("inet listen: %i\n", ret);
+			if (ret < 0)
+				goto out;
+		}

net/ipv4/checkpoint.c line 204

Quoted 10 lines
net/ipv4/checkpoint.c new file +10 -0
+	} else {
+		if (!sock_flag(sock->sk, SOCK_DEAD))
+			ret = inet_defer_restore_buffers(ctx, sock->sk);
+	}
+ out:
+	ckpt_hdr_put(ctx, in);
+
+	return ret;
+ }
+

Oren.