linux/net/tipc/msg.c

857 lines
22 KiB
C
Raw Normal View History

/*
* net/tipc/msg.c: TIPC message header routines
*
* Copyright (c) 2000-2006, 2014-2015, Ericsson AB
* Copyright (c) 2005, 2010-2011, Wind River Systems
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the names of the copyright holders nor the names of its
* contributors may be used to endorse or promote products derived from
* this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include <net/sock.h>
#include "core.h"
#include "msg.h"
#include "addr.h"
#include "name_table.h"
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
#include "crypto.h"
#define MAX_FORWARD_SIZE 1024
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
#ifdef CONFIG_TIPC_CRYPTO
#define BUF_HEADROOM ALIGN(((LL_MAX_HEADER + 48) + EHDR_MAX_SIZE), 16)
#define BUF_TAILROOM (TIPC_AES_GCM_TAG_SIZE)
#else
#define BUF_HEADROOM (LL_MAX_HEADER + 48)
#define BUF_TAILROOM 16
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
#endif
static unsigned int align(unsigned int i)
{
return (i + 3) & ~3u;
}
/**
* tipc_buf_acquire - creates a TIPC message buffer
* @size: message size (including TIPC header)
*
* Returns a new buffer with data pointers set to the specified size.
*
* NOTE: Headroom is reserved to allow prepending of a data link header.
* There may also be unrequested tailroom present at the buffer's end.
*/
struct sk_buff *tipc_buf_acquire(u32 size, gfp_t gfp)
{
struct sk_buff *skb;
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
#ifdef CONFIG_TIPC_CRYPTO
unsigned int buf_size = (BUF_HEADROOM + size + BUF_TAILROOM + 3) & ~3u;
#else
unsigned int buf_size = (BUF_HEADROOM + size + 3) & ~3u;
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
#endif
skb = alloc_skb_fclone(buf_size, gfp);
if (skb) {
skb_reserve(skb, BUF_HEADROOM);
skb_put(skb, size);
skb->next = NULL;
}
return skb;
}
2015-02-05 21:36:36 +08:00
void tipc_msg_init(u32 own_node, struct tipc_msg *m, u32 user, u32 type,
u32 hsize, u32 dnode)
{
memset(m, 0, hsize);
msg_set_version(m);
msg_set_user(m, user);
msg_set_hdr_sz(m, hsize);
msg_set_size(m, hsize);
2015-02-05 21:36:36 +08:00
msg_set_prevnode(m, own_node);
msg_set_type(m, type);
if (hsize > SHORT_H_SIZE) {
2015-02-05 21:36:36 +08:00
msg_set_orignode(m, own_node);
msg_set_destnode(m, dnode);
}
}
2015-02-05 21:36:36 +08:00
struct sk_buff *tipc_msg_create(uint user, uint type,
uint hdr_sz, uint data_sz, u32 dnode,
u32 onode, u32 dport, u32 oport, int errcode)
{
struct tipc_msg *msg;
struct sk_buff *buf;
buf = tipc_buf_acquire(hdr_sz + data_sz, GFP_ATOMIC);
if (unlikely(!buf))
return NULL;
msg = buf_msg(buf);
2015-02-05 21:36:36 +08:00
tipc_msg_init(onode, msg, user, type, hdr_sz, dnode);
msg_set_size(msg, hdr_sz + data_sz);
msg_set_origport(msg, oport);
msg_set_destport(msg, dport);
msg_set_errcode(msg, errcode);
if (hdr_sz > SHORT_H_SIZE) {
msg_set_orignode(msg, onode);
msg_set_destnode(msg, dnode);
}
return buf;
}
/* tipc_buf_append(): Append a buffer to the fragment list of another buffer
* @*headbuf: in: NULL for first frag, otherwise value returned from prev call
* out: set when successful non-complete reassembly, otherwise NULL
* @*buf: in: the buffer to append. Always defined
* out: head buf after successful complete reassembly, otherwise NULL
* Returns 1 when reassembly complete, otherwise 0
*/
int tipc_buf_append(struct sk_buff **headbuf, struct sk_buff **buf)
{
struct sk_buff *head = *headbuf;
struct sk_buff *frag = *buf;
struct sk_buff *tail = NULL;
struct tipc_msg *msg;
u32 fragid;
int delta;
bool headstolen;
if (!frag)
goto err;
msg = buf_msg(frag);
fragid = msg_type(msg);
frag->next = NULL;
skb_pull(frag, msg_hdr_sz(msg));
if (fragid == FIRST_FRAGMENT) {
if (unlikely(head))
goto err;
if (unlikely(skb_unclone(frag, GFP_ATOMIC)))
goto err;
head = *headbuf = frag;
*buf = NULL;
TIPC_SKB_CB(head)->tail = NULL;
if (skb_is_nonlinear(head)) {
skb_walk_frags(head, tail) {
TIPC_SKB_CB(head)->tail = tail;
}
} else {
skb_frag_list_init(head);
}
return 0;
}
if (!head)
goto err;
if (skb_try_coalesce(head, frag, &headstolen, &delta)) {
kfree_skb_partial(frag, headstolen);
} else {
tail = TIPC_SKB_CB(head)->tail;
if (!skb_has_frag_list(head))
skb_shinfo(head)->frag_list = frag;
else
tail->next = frag;
head->truesize += frag->truesize;
head->data_len += frag->len;
head->len += frag->len;
TIPC_SKB_CB(head)->tail = frag;
}
if (fragid == LAST_FRAGMENT) {
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
TIPC_SKB_CB(head)->validated = 0;
if (unlikely(!tipc_msg_validate(&head)))
goto err;
*buf = head;
TIPC_SKB_CB(head)->tail = NULL;
*headbuf = NULL;
return 1;
}
*buf = NULL;
return 0;
err:
kfree_skb(*buf);
kfree_skb(*headbuf);
*buf = *headbuf = NULL;
return 0;
}
tipc: add smart nagle feature We introduce a feature that works like a combination of TCP_NAGLE and TCP_CORK, but without some of the weaknesses of those. In particular, we will not observe long delivery delays because of delayed acks, since the algorithm itself decides if and when acks are to be sent from the receiving peer. - The nagle property as such is determined by manipulating a new 'maxnagle' field in struct tipc_sock. If certain conditions are met, 'maxnagle' will define max size of the messages which can be bundled. If it is set to zero no messages are ever bundled, implying that the nagle property is disabled. - A socket with the nagle property enabled enters nagle mode when more than 4 messages have been sent out without receiving any data message from the peer. - A socket leaves nagle mode whenever it receives a data message from the peer. In nagle mode, messages smaller than 'maxnagle' are accumulated in the socket write queue. The last buffer in the queue is marked with a new 'ack_required' bit, which forces the receiving peer to send a CONN_ACK message back to the sender upon reception. The accumulated contents of the write queue is transmitted when one of the following events or conditions occur. - A CONN_ACK message is received from the peer. - A data message is received from the peer. - A SOCK_WAKEUP pseudo message is received from the link level. - The write queue contains more than 64 1k blocks of data. - The connection is being shut down. - There is no CONN_ACK message to expect. I.e., there is currently no outstanding message where the 'ack_required' bit was set. As a consequence, the first message added after we enter nagle mode is always sent directly with this bit set. This new feature gives a 50-100% improvement of throughput for small (i.e., less than MTU size) messages, while it might add up to one RTT to latency time when the socket is in nagle mode. Acked-by: Ying Xue <ying.xue@windreiver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-30 21:00:41 +08:00
/**
* tipc_msg_append(): Append data to tail of an existing buffer queue
* @hdr: header to be used
* @m: the data to be appended
* @mss: max allowable size of buffer
* @dlen: size of data to be appended
* @txq: queue to appand to
* Returns the number og 1k blocks appended or errno value
*/
int tipc_msg_append(struct tipc_msg *_hdr, struct msghdr *m, int dlen,
int mss, struct sk_buff_head *txq)
{
struct sk_buff *skb;
tipc: add smart nagle feature We introduce a feature that works like a combination of TCP_NAGLE and TCP_CORK, but without some of the weaknesses of those. In particular, we will not observe long delivery delays because of delayed acks, since the algorithm itself decides if and when acks are to be sent from the receiving peer. - The nagle property as such is determined by manipulating a new 'maxnagle' field in struct tipc_sock. If certain conditions are met, 'maxnagle' will define max size of the messages which can be bundled. If it is set to zero no messages are ever bundled, implying that the nagle property is disabled. - A socket with the nagle property enabled enters nagle mode when more than 4 messages have been sent out without receiving any data message from the peer. - A socket leaves nagle mode whenever it receives a data message from the peer. In nagle mode, messages smaller than 'maxnagle' are accumulated in the socket write queue. The last buffer in the queue is marked with a new 'ack_required' bit, which forces the receiving peer to send a CONN_ACK message back to the sender upon reception. The accumulated contents of the write queue is transmitted when one of the following events or conditions occur. - A CONN_ACK message is received from the peer. - A data message is received from the peer. - A SOCK_WAKEUP pseudo message is received from the link level. - The write queue contains more than 64 1k blocks of data. - The connection is being shut down. - There is no CONN_ACK message to expect. I.e., there is currently no outstanding message where the 'ack_required' bit was set. As a consequence, the first message added after we enter nagle mode is always sent directly with this bit set. This new feature gives a 50-100% improvement of throughput for small (i.e., less than MTU size) messages, while it might add up to one RTT to latency time when the socket is in nagle mode. Acked-by: Ying Xue <ying.xue@windreiver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-30 21:00:41 +08:00
int accounted, total, curr;
int mlen, cpy, rem = dlen;
struct tipc_msg *hdr;
skb = skb_peek_tail(txq);
accounted = skb ? msg_blocks(buf_msg(skb)) : 0;
total = accounted;
tipc: fix NULL pointer dereference in streaming syzbot found the following crash: general protection fault, probably for non-canonical address 0xdffffc0000000019: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x00000000000000c8-0x00000000000000cf] CPU: 1 PID: 7060 Comm: syz-executor394 Not tainted 5.7.0-rc6-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__tipc_sendstream+0xbde/0x11f0 net/tipc/socket.c:1591 Code: 00 00 00 00 48 39 5c 24 28 48 0f 44 d8 e8 fa 3e db f9 48 b8 00 00 00 00 00 fc ff df 48 8d bb c8 00 00 00 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 e2 04 00 00 48 8b 9b c8 00 00 00 48 b8 00 00 00 RSP: 0018:ffffc90003ef7818 EFLAGS: 00010202 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff8797fd9d RDX: 0000000000000019 RSI: ffffffff8797fde6 RDI: 00000000000000c8 RBP: ffff888099848040 R08: ffff88809a5f6440 R09: fffffbfff1860b4c R10: ffffffff8c305a5f R11: fffffbfff1860b4b R12: ffff88809984857e R13: 0000000000000000 R14: ffff888086aa4000 R15: 0000000000000000 FS: 00000000009b4880(0000) GS:ffff8880ae700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020000140 CR3: 00000000a7fdf000 CR4: 00000000001406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: tipc_sendstream+0x4c/0x70 net/tipc/socket.c:1533 sock_sendmsg_nosec net/socket.c:652 [inline] sock_sendmsg+0xcf/0x120 net/socket.c:672 ____sys_sendmsg+0x32f/0x810 net/socket.c:2352 ___sys_sendmsg+0x100/0x170 net/socket.c:2406 __sys_sendmmsg+0x195/0x480 net/socket.c:2496 __do_sys_sendmmsg net/socket.c:2525 [inline] __se_sys_sendmmsg net/socket.c:2522 [inline] __x64_sys_sendmmsg+0x99/0x100 net/socket.c:2522 do_syscall_64+0xf6/0x7d0 arch/x86/entry/common.c:295 entry_SYSCALL_64_after_hwframe+0x49/0xb3 RIP: 0033:0x440199 ... This bug was bisected to commit 0a3e060f340d ("tipc: add test for Nagle algorithm effectiveness"). However, it is not the case, the trouble was from the base in the case of zero data length message sending, we would unexpectedly make an empty 'txq' queue after the 'tipc_msg_append()' in Nagle mode. A similar crash can be generated even without the bisected patch but at the link layer when it accesses the empty queue. We solve the issues by building at least one buffer to go with socket's header and an optional data section that may be empty like what we had with the 'tipc_msg_build()'. Note: the previous commit 4c21daae3dbc ("tipc: Fix NULL pointer dereference in __tipc_sendstream()") is obsoleted by this one since the 'txq' will be never empty and the check of 'skb != NULL' is unnecessary but it is safe anyway. Reported-by: syzbot+8eac6d030e7807c21d32@syzkaller.appspotmail.com Fixes: c0bceb97db9e ("tipc: add smart nagle feature") Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-03 13:06:01 +08:00
do {
tipc: add smart nagle feature We introduce a feature that works like a combination of TCP_NAGLE and TCP_CORK, but without some of the weaknesses of those. In particular, we will not observe long delivery delays because of delayed acks, since the algorithm itself decides if and when acks are to be sent from the receiving peer. - The nagle property as such is determined by manipulating a new 'maxnagle' field in struct tipc_sock. If certain conditions are met, 'maxnagle' will define max size of the messages which can be bundled. If it is set to zero no messages are ever bundled, implying that the nagle property is disabled. - A socket with the nagle property enabled enters nagle mode when more than 4 messages have been sent out without receiving any data message from the peer. - A socket leaves nagle mode whenever it receives a data message from the peer. In nagle mode, messages smaller than 'maxnagle' are accumulated in the socket write queue. The last buffer in the queue is marked with a new 'ack_required' bit, which forces the receiving peer to send a CONN_ACK message back to the sender upon reception. The accumulated contents of the write queue is transmitted when one of the following events or conditions occur. - A CONN_ACK message is received from the peer. - A data message is received from the peer. - A SOCK_WAKEUP pseudo message is received from the link level. - The write queue contains more than 64 1k blocks of data. - The connection is being shut down. - There is no CONN_ACK message to expect. I.e., there is currently no outstanding message where the 'ack_required' bit was set. As a consequence, the first message added after we enter nagle mode is always sent directly with this bit set. This new feature gives a 50-100% improvement of throughput for small (i.e., less than MTU size) messages, while it might add up to one RTT to latency time when the socket is in nagle mode. Acked-by: Ying Xue <ying.xue@windreiver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-30 21:00:41 +08:00
if (!skb || skb->len >= mss) {
skb = tipc_buf_acquire(mss, GFP_KERNEL);
if (unlikely(!skb))
return -ENOMEM;
skb_orphan(skb);
skb_trim(skb, MIN_H_SIZE);
hdr = buf_msg(skb);
skb_copy_to_linear_data(skb, _hdr, MIN_H_SIZE);
msg_set_hdr_sz(hdr, MIN_H_SIZE);
msg_set_size(hdr, MIN_H_SIZE);
__skb_queue_tail(txq, skb);
total += 1;
}
hdr = buf_msg(skb);
curr = msg_blocks(hdr);
mlen = msg_size(hdr);
tipc: fix kernel WARNING in tipc_msg_append() syzbot found the following issue: WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 check_copy_size include/linux/thread_info.h:150 [inline] WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 copy_from_iter include/linux/uio.h:144 [inline] WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 tipc_msg_append+0x49a/0x5e0 net/tipc/msg.c:242 Kernel panic - not syncing: panic_on_warn set ... This happens after commit 5e9eeccc58f3 ("tipc: fix NULL pointer dereference in streaming") that tried to build at least one buffer even when the message data length is zero... However, it now exposes another bug that the 'mss' can be zero and the 'cpy' will be negative, thus the above kernel WARNING will appear! The zero value of 'mss' is never expected because it means Nagle is not enabled for the socket (actually the socket type was 'SOCK_SEQPACKET'), so the function 'tipc_msg_append()' must not be called at all. But that was in this particular case since the message data length was zero, and the 'send <= maxnagle' check became true. We resolve the issue by explicitly checking if Nagle is enabled for the socket, i.e. 'maxnagle != 0' before calling the 'tipc_msg_append()'. We also reinforce the function to against such a negative values if any. Reported-by: syzbot+75139a7d2605236b0b7f@syzkaller.appspotmail.com Fixes: c0bceb97db9e ("tipc: add smart nagle feature") Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-11 18:07:35 +08:00
cpy = min_t(size_t, rem, mss - mlen);
tipc: add smart nagle feature We introduce a feature that works like a combination of TCP_NAGLE and TCP_CORK, but without some of the weaknesses of those. In particular, we will not observe long delivery delays because of delayed acks, since the algorithm itself decides if and when acks are to be sent from the receiving peer. - The nagle property as such is determined by manipulating a new 'maxnagle' field in struct tipc_sock. If certain conditions are met, 'maxnagle' will define max size of the messages which can be bundled. If it is set to zero no messages are ever bundled, implying that the nagle property is disabled. - A socket with the nagle property enabled enters nagle mode when more than 4 messages have been sent out without receiving any data message from the peer. - A socket leaves nagle mode whenever it receives a data message from the peer. In nagle mode, messages smaller than 'maxnagle' are accumulated in the socket write queue. The last buffer in the queue is marked with a new 'ack_required' bit, which forces the receiving peer to send a CONN_ACK message back to the sender upon reception. The accumulated contents of the write queue is transmitted when one of the following events or conditions occur. - A CONN_ACK message is received from the peer. - A data message is received from the peer. - A SOCK_WAKEUP pseudo message is received from the link level. - The write queue contains more than 64 1k blocks of data. - The connection is being shut down. - There is no CONN_ACK message to expect. I.e., there is currently no outstanding message where the 'ack_required' bit was set. As a consequence, the first message added after we enter nagle mode is always sent directly with this bit set. This new feature gives a 50-100% improvement of throughput for small (i.e., less than MTU size) messages, while it might add up to one RTT to latency time when the socket is in nagle mode. Acked-by: Ying Xue <ying.xue@windreiver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-30 21:00:41 +08:00
if (cpy != copy_from_iter(skb->data + mlen, cpy, &m->msg_iter))
return -EFAULT;
msg_set_size(hdr, mlen + cpy);
skb_put(skb, cpy);
rem -= cpy;
total += msg_blocks(hdr) - curr;
tipc: fix kernel WARNING in tipc_msg_append() syzbot found the following issue: WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 check_copy_size include/linux/thread_info.h:150 [inline] WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 copy_from_iter include/linux/uio.h:144 [inline] WARNING: CPU: 0 PID: 6808 at include/linux/thread_info.h:150 tipc_msg_append+0x49a/0x5e0 net/tipc/msg.c:242 Kernel panic - not syncing: panic_on_warn set ... This happens after commit 5e9eeccc58f3 ("tipc: fix NULL pointer dereference in streaming") that tried to build at least one buffer even when the message data length is zero... However, it now exposes another bug that the 'mss' can be zero and the 'cpy' will be negative, thus the above kernel WARNING will appear! The zero value of 'mss' is never expected because it means Nagle is not enabled for the socket (actually the socket type was 'SOCK_SEQPACKET'), so the function 'tipc_msg_append()' must not be called at all. But that was in this particular case since the message data length was zero, and the 'send <= maxnagle' check became true. We resolve the issue by explicitly checking if Nagle is enabled for the socket, i.e. 'maxnagle != 0' before calling the 'tipc_msg_append()'. We also reinforce the function to against such a negative values if any. Reported-by: syzbot+75139a7d2605236b0b7f@syzkaller.appspotmail.com Fixes: c0bceb97db9e ("tipc: add smart nagle feature") Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-11 18:07:35 +08:00
} while (rem > 0);
tipc: add smart nagle feature We introduce a feature that works like a combination of TCP_NAGLE and TCP_CORK, but without some of the weaknesses of those. In particular, we will not observe long delivery delays because of delayed acks, since the algorithm itself decides if and when acks are to be sent from the receiving peer. - The nagle property as such is determined by manipulating a new 'maxnagle' field in struct tipc_sock. If certain conditions are met, 'maxnagle' will define max size of the messages which can be bundled. If it is set to zero no messages are ever bundled, implying that the nagle property is disabled. - A socket with the nagle property enabled enters nagle mode when more than 4 messages have been sent out without receiving any data message from the peer. - A socket leaves nagle mode whenever it receives a data message from the peer. In nagle mode, messages smaller than 'maxnagle' are accumulated in the socket write queue. The last buffer in the queue is marked with a new 'ack_required' bit, which forces the receiving peer to send a CONN_ACK message back to the sender upon reception. The accumulated contents of the write queue is transmitted when one of the following events or conditions occur. - A CONN_ACK message is received from the peer. - A data message is received from the peer. - A SOCK_WAKEUP pseudo message is received from the link level. - The write queue contains more than 64 1k blocks of data. - The connection is being shut down. - There is no CONN_ACK message to expect. I.e., there is currently no outstanding message where the 'ack_required' bit was set. As a consequence, the first message added after we enter nagle mode is always sent directly with this bit set. This new feature gives a 50-100% improvement of throughput for small (i.e., less than MTU size) messages, while it might add up to one RTT to latency time when the socket is in nagle mode. Acked-by: Ying Xue <ying.xue@windreiver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-30 21:00:41 +08:00
return total - accounted;
}
/* tipc_msg_validate - validate basic format of received message
*
* This routine ensures a TIPC message has an acceptable header, and at least
* as much data as the header indicates it should. The routine also ensures
* that the entire message header is stored in the main fragment of the message
* buffer, to simplify future access to message header fields.
*
* Note: Having extra info present in the message header or data areas is OK.
* TIPC will ignore the excess, under the assumption that it is optional info
* introduced by a later release of the protocol.
*/
bool tipc_msg_validate(struct sk_buff **_skb)
{
struct sk_buff *skb = *_skb;
struct tipc_msg *hdr;
int msz, hsz;
/* Ensure that flow control ratio condition is satisfied */
if (unlikely(skb->truesize / buf_roundup_len(skb) >= 4)) {
skb = skb_copy_expand(skb, BUF_HEADROOM, 0, GFP_ATOMIC);
if (!skb)
return false;
kfree_skb(*_skb);
*_skb = skb;
}
if (unlikely(TIPC_SKB_CB(skb)->validated))
return true;
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
if (unlikely(!pskb_may_pull(skb, MIN_H_SIZE)))
return false;
hsz = msg_hdr_sz(buf_msg(skb));
if (unlikely(hsz < MIN_H_SIZE) || (hsz > MAX_H_SIZE))
return false;
if (unlikely(!pskb_may_pull(skb, hsz)))
return false;
hdr = buf_msg(skb);
if (unlikely(msg_version(hdr) != TIPC_VERSION))
return false;
msz = msg_size(hdr);
if (unlikely(msz < hsz))
return false;
if (unlikely((msz - hsz) > TIPC_MAX_USER_MSG_SIZE))
return false;
if (unlikely(skb->len < msz))
return false;
tipc: introduce TIPC encryption & authentication This commit offers an option to encrypt and authenticate all messaging, including the neighbor discovery messages. The currently most advanced algorithm supported is the AEAD AES-GCM (like IPSec or TLS). All encryption/decryption is done at the bearer layer, just before leaving or after entering TIPC. Supported features: - Encryption & authentication of all TIPC messages (header + data); - Two symmetric-key modes: Cluster and Per-node; - Automatic key switching; - Key-expired revoking (sequence number wrapped); - Lock-free encryption/decryption (RCU); - Asynchronous crypto, Intel AES-NI supported; - Multiple cipher transforms; - Logs & statistics; Two key modes: - Cluster key mode: One single key is used for both TX & RX in all nodes in the cluster. - Per-node key mode: Each nodes in the cluster has one specific TX key. For RX, a node requires its peers' TX key to be able to decrypt the messages from those peers. Key setting from user-space is performed via netlink by a user program (e.g. the iproute2 'tipc' tool). Internal key state machine: Attach Align(RX) +-+ +-+ | V | V +---------+ Attach +---------+ | IDLE |---------------->| PENDING |(user = 0) +---------+ +---------+ A A Switch| A | | | | | | Free(switch/revoked) | | (Free)| +----------------------+ | |Timeout | (TX) | | |(RX) | | | | | | v | +---------+ Switch +---------+ | PASSIVE |<----------------| ACTIVE | +---------+ (RX) +---------+ (user = 1) (user >= 1) The number of TFMs is 10 by default and can be changed via the procfs 'net/tipc/max_tfms'. At this moment, as for simplicity, this file is also used to print the crypto statistics at runtime: echo 0xfff1 > /proc/sys/net/tipc/max_tfms The patch defines a new TIPC version (v7) for the encryption message (- backward compatibility as well). The message is basically encapsulated as follows: +----------------------------------------------------------+ | TIPCv7 encryption | Original TIPCv2 | Authentication | | header | packet (encrypted) | Tag | +----------------------------------------------------------+ The throughput is about ~40% for small messages (compared with non- encryption) and ~9% for large messages. With the support from hardware crypto i.e. the Intel AES-NI CPU instructions, the throughput increases upto ~85% for small messages and ~55% for large messages. By default, the new feature is inactive (i.e. no encryption) until user sets a key for TIPC. There is however also a new option - "TIPC_CRYPTO" in the kernel configuration to enable/disable the new code when needed. MAINTAINERS | add two new files 'crypto.h' & 'crypto.c' in tipc Acked-by: Ying Xue <ying.xue@windreiver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-11-08 13:05:11 +08:00
TIPC_SKB_CB(skb)->validated = 1;
return true;
}
tipc: fix changeover issues due to large packet In conjunction with changing the interfaces' MTU (e.g. especially in the case of a bonding) where the TIPC links are brought up and down in a short time, a couple of issues were detected with the current link changeover mechanism: 1) When one link is up but immediately forced down again, the failover procedure will be carried out in order to failover all the messages in the link's transmq queue onto the other working link. The link and node state is also set to FAILINGOVER as part of the process. The message will be transmited in form of a FAILOVER_MSG, so its size is plus of 40 bytes (= the message header size). There is no problem if the original message size is not larger than the link's MTU - 40, and indeed this is the max size of a normal payload messages. However, in the situation above, because the link has just been up, the messages in the link's transmq are almost SYNCH_MSGs which had been generated by the link synching procedure, then their size might reach the max value already! When the FAILOVER_MSG is built on the top of such a SYNCH_MSG, its size will exceed the link's MTU. As a result, the messages are dropped silently and the failover procedure will never end up, the link will not be able to exit the FAILINGOVER state, so cannot be re-established. 2) The same scenario above can happen more easily in case the MTU of the links is set differently or when changing. In that case, as long as a large message in the failure link's transmq queue was built and fragmented with its link's MTU > the other link's one, the issue will happen (there is no need of a link synching in advance). 3) The link synching procedure also faces with the same issue but since the link synching is only started upon receipt of a SYNCH_MSG, dropping the message will not result in a state deadlock, but it is not expected as design. The 1) & 3) issues are resolved by the last commit that only a dummy SYNCH_MSG (i.e. without data) is generated at the link synching, so the size of a FAILOVER_MSG if any then will never exceed the link's MTU. For the 2) issue, the only solution is trying to fragment the messages in the failure link's transmq queue according to the working link's MTU so they can be failovered then. A new function is made to accomplish this, it will still be a TUNNEL PROTOCOL/FAILOVER MSG but if the original message size is too large, it will be fragmented & reassembled at the receiving side. Acked-by: Ying Xue <ying.xue@windriver.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-07-24 09:56:12 +08:00
/**
* tipc_msg_fragment - build a fragment skb list for TIPC message
*
* @skb: TIPC message skb
* @hdr: internal msg header to be put on the top of the fragments
* @pktmax: max size of a fragment incl. the header
* @frags: returned fragment skb list
*
* Returns 0 if the fragmentation is successful, otherwise: -EINVAL
* or -ENOMEM
*/
int tipc_msg_fragment(struct sk_buff *skb, const struct tipc_msg *hdr,
int pktmax, struct sk_buff_head *frags)
{
int pktno, nof_fragms, dsz, dmax, eat;
struct tipc_msg *_hdr;
struct sk_buff *_skb;
u8 *data;
/* Non-linear buffer? */
if (skb_linearize(skb))
return -ENOMEM;
data = (u8 *)skb->data;
dsz = msg_size(buf_msg(skb));
dmax = pktmax - INT_H_SIZE;
if (dsz <= dmax || !dmax)
return -EINVAL;
nof_fragms = dsz / dmax + 1;
for (pktno = 1; pktno <= nof_fragms; pktno++) {
if (pktno < nof_fragms)
eat = dmax;
else
eat = dsz % dmax;
/* Allocate a new fragment */
_skb = tipc_buf_acquire(INT_H_SIZE + eat, GFP_ATOMIC);
if (!_skb)
goto error;
skb_orphan(_skb);
__skb_queue_tail(frags, _skb);
/* Copy header & data to the fragment */
skb_copy_to_linear_data(_skb, hdr, INT_H_SIZE);
skb_copy_to_linear_data_offset(_skb, INT_H_SIZE, data, eat);
data += eat;
/* Update the fragment's header */
_hdr = buf_msg(_skb);
msg_set_fragm_no(_hdr, pktno);
msg_set_nof_fragms(_hdr, nof_fragms);
msg_set_size(_hdr, INT_H_SIZE + eat);
}
return 0;
error:
__skb_queue_purge(frags);
__skb_queue_head_init(frags);
return -ENOMEM;
}
/**
* tipc_msg_build - create buffer chain containing specified header and data
* @mhdr: Message header, to be prepended to data
* @m: User message
* @dsz: Total length of user data
* @pktmax: Max packet size that can be used
* @list: Buffer or chain of buffers to be returned to caller
*
* Note that the recursive call we are making here is safe, since it can
* logically go only one further level down.
*
* Returns message data size or errno: -ENOMEM, -EFAULT
*/
int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset,
int dsz, int pktmax, struct sk_buff_head *list)
{
int mhsz = msg_hdr_sz(mhdr);
struct tipc_msg pkthdr;
int msz = mhsz + dsz;
int pktrem = pktmax;
struct sk_buff *skb;
int drem = dsz;
int pktno = 1;
char *pktpos;
int pktsz;
int rc;
msg_set_size(mhdr, msz);
/* No fragmentation needed? */
if (likely(msz <= pktmax)) {
skb = tipc_buf_acquire(msz, GFP_KERNEL);
/* Fall back to smaller MTU if node local message */
if (unlikely(!skb)) {
if (pktmax != MAX_MSG_SIZE)
return -ENOMEM;
rc = tipc_msg_build(mhdr, m, offset, dsz, FB_MTU, list);
if (rc != dsz)
return rc;
if (tipc_msg_assemble(list))
return dsz;
return -ENOMEM;
}
skb_orphan(skb);
__skb_queue_tail(list, skb);
skb_copy_to_linear_data(skb, mhdr, mhsz);
pktpos = skb->data + mhsz;
if (copy_from_iter_full(pktpos, dsz, &m->msg_iter))
return dsz;
rc = -EFAULT;
goto error;
}
/* Prepare reusable fragment header */
2015-02-05 21:36:36 +08:00
tipc_msg_init(msg_prevnode(mhdr), &pkthdr, MSG_FRAGMENTER,
FIRST_FRAGMENT, INT_H_SIZE, msg_destnode(mhdr));
msg_set_size(&pkthdr, pktmax);
msg_set_fragm_no(&pkthdr, pktno);
tipc: clean up handling of message priorities Messages transferred by TIPC are assigned an "importance priority", -an integer value indicating how to treat the message when there is link or destination socket congestion. There is no separate header field for this value. Instead, the message user values have been chosen in ascending order according to perceived importance, so that the message user field can be used for this. This is not a good solution. First, we have many more users than the needed priority levels, so we end up with treating more priority levels than necessary. Second, the user field cannot always accurately reflect the priority of the message. E.g., a message fragment packet should really have the priority of the enveloped user data message, and not the priority of the MSG_FRAGMENTER user. Until now, we have been working around this problem in different ways, but it is now time to implement a consistent way of handling such priorities, although still within the constraint that we cannot allocate any more bits in the regular data message header for this. In this commit, we define a new priority level, TIPC_SYSTEM_IMPORTANCE, that will be the only one used apart from the four (lower) user data levels. All non-data messages map down to this priority. Furthermore, we take some free bits from the MSG_FRAGMENTER header and allocate them to store the priority of the enveloped message. We then adjust the functions msg_importance()/msg_set_importance() so that they read/set the correct header fields depending on user type. This small protocol change is fully compatible, because the code at the receiving end of a link currently reads the importance level only from user data messages, where there is no change. Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-14 04:08:11 +08:00
msg_set_importance(&pkthdr, msg_importance(mhdr));
/* Prepare first fragment */
skb = tipc_buf_acquire(pktmax, GFP_KERNEL);
if (!skb)
return -ENOMEM;
skb_orphan(skb);
__skb_queue_tail(list, skb);
pktpos = skb->data;
skb_copy_to_linear_data(skb, &pkthdr, INT_H_SIZE);
pktpos += INT_H_SIZE;
pktrem -= INT_H_SIZE;
skb_copy_to_linear_data_offset(skb, INT_H_SIZE, mhdr, mhsz);
pktpos += mhsz;
pktrem -= mhsz;
do {
if (drem < pktrem)
pktrem = drem;
if (!copy_from_iter_full(pktpos, pktrem, &m->msg_iter)) {
rc = -EFAULT;
goto error;
}
drem -= pktrem;
if (!drem)
break;
/* Prepare new fragment: */
if (drem < (pktmax - INT_H_SIZE))
pktsz = drem + INT_H_SIZE;
else
pktsz = pktmax;
skb = tipc_buf_acquire(pktsz, GFP_KERNEL);
if (!skb) {
rc = -ENOMEM;
goto error;
}
skb_orphan(skb);
__skb_queue_tail(list, skb);
msg_set_type(&pkthdr, FRAGMENT);
msg_set_size(&pkthdr, pktsz);
msg_set_fragm_no(&pkthdr, ++pktno);
skb_copy_to_linear_data(skb, &pkthdr, INT_H_SIZE);
pktpos = skb->data + INT_H_SIZE;
pktrem = pktsz - INT_H_SIZE;
} while (1);
msg_set_type(buf_msg(skb), LAST_FRAGMENT);
return dsz;
error:
__skb_queue_purge(list);
__skb_queue_head_init(list);
return rc;
}
/**
* tipc_msg_bundle - Append contents of a buffer to tail of an existing one
* @bskb: the bundle buffer to append to
* @msg: message to be appended
* @max: max allowable size for the bundle buffer
*
* Returns "true" if bundling has been performed, otherwise "false"
*/
static bool tipc_msg_bundle(struct sk_buff *bskb, struct tipc_msg *msg,
u32 max)
{
struct tipc_msg *bmsg = buf_msg(bskb);
u32 msz, bsz, offset, pad;
msz = msg_size(msg);
bsz = msg_size(bmsg);
offset = align(bsz);
pad = offset - bsz;
if (unlikely(skb_tailroom(bskb) < (pad + msz)))
return false;
if (unlikely(max < (offset + msz)))
return false;
skb_put(bskb, pad + msz);
skb_copy_to_linear_data_offset(bskb, offset, msg, msz);
msg_set_size(bmsg, offset + msz);
msg_set_msgcnt(bmsg, msg_msgcnt(bmsg) + 1);
return true;
}
/**
* tipc_msg_try_bundle - Try to bundle a new message to the last one
* @tskb: the last/target message to which the new one will be appended
* @skb: the new message skb pointer
* @mss: max message size (header inclusive)
* @dnode: destination node for the message
* @new_bundle: if this call made a new bundle or not
*
* Return: "true" if the new message skb is potential for bundling this time or
* later, in the case a bundling has been done this time, the skb is consumed
* (the skb pointer = NULL).
* Otherwise, "false" if the skb cannot be bundled at all.
*/
bool tipc_msg_try_bundle(struct sk_buff *tskb, struct sk_buff **skb, u32 mss,
u32 dnode, bool *new_bundle)
{
struct tipc_msg *msg, *inner, *outer;
u32 tsz;
/* First, check if the new buffer is suitable for bundling */
msg = buf_msg(*skb);
if (msg_user(msg) == MSG_FRAGMENTER)
return false;
if (msg_user(msg) == TUNNEL_PROTOCOL)
return false;
if (msg_user(msg) == BCAST_PROTOCOL)
return false;
if (mss <= INT_H_SIZE + msg_size(msg))
tipc: improve link congestion algorithm The link congestion algorithm used until now implies two problems. - It is too generous towards lower-level messages in situations of high load by giving "absolute" bandwidth guarantees to the different priority levels. LOW traffic is guaranteed 10%, MEDIUM is guaranted 20%, HIGH is guaranteed 30%, and CRITICAL is guaranteed 40% of the available bandwidth. But, in the absence of higher level traffic, the ratio between two distinct levels becomes unreasonable. E.g. if there is only LOW and MEDIUM traffic on a system, the former is guaranteed 1/3 of the bandwidth, and the latter 2/3. This again means that if there is e.g. one LOW user and 10 MEDIUM users, the former will have 33.3% of the bandwidth, and the others will have to compete for the remainder, i.e. each will end up with 6.7% of the capacity. - Packets of type MSG_BUNDLER are created at SYSTEM importance level, but only after the packets bundled into it have passed the congestion test for their own respective levels. Since bundled packets don't result in incrementing the level counter for their own importance, only occasionally for the SYSTEM level counter, they do in practice obtain SYSTEM level importance. Hence, the current implementation provides a gap in the congestion algorithm that in the worst case may lead to a link reset. We now refine the congestion algorithm as follows: - A message is accepted to the link backlog only if its own level counter, and all superior level counters, permit it. - The importance of a created bundle packet is set according to its contents. A bundle packet created from messges at levels LOW to CRITICAL is given importance level CRITICAL, while a bundle created from a SYSTEM level message is given importance SYSTEM. In the latter case only subsequent SYSTEM level messages are allowed to be bundled into it. This solves the first problem described above, by making the bandwidth guarantee relative to the total number of users at all levels; only the upper limit for each level remains absolute. In the example described above, the single LOW user would use 1/11th of the bandwidth, the same as each of the ten MEDIUM users, but he still has the same guarantee against starvation as the latter ones. The fix also solves the second problem. If the CRITICAL level is filled up by bundle packets of that level, no lower level packets will be accepted any more. Suggested-by: Gergely Kiss <gergely.kiss@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-14 22:46:17 +08:00
return false;
/* Ok, but the last/target buffer can be empty? */
if (unlikely(!tskb))
return true;
/* Is it a bundle already? Try to bundle the new message to it */
if (msg_user(buf_msg(tskb)) == MSG_BUNDLER) {
*new_bundle = false;
goto bundle;
}
/* Make a new bundle of the two messages if possible */
tsz = msg_size(buf_msg(tskb));
if (unlikely(mss < align(INT_H_SIZE + tsz) + msg_size(msg)))
return true;
if (unlikely(pskb_expand_head(tskb, INT_H_SIZE, mss - tsz - INT_H_SIZE,
GFP_ATOMIC)))
return true;
inner = buf_msg(tskb);
skb_push(tskb, INT_H_SIZE);
outer = buf_msg(tskb);
tipc_msg_init(msg_prevnode(inner), outer, MSG_BUNDLER, 0, INT_H_SIZE,
dnode);
msg_set_importance(outer, msg_importance(inner));
msg_set_size(outer, INT_H_SIZE + tsz);
msg_set_msgcnt(outer, 1);
*new_bundle = true;
bundle:
if (likely(tipc_msg_bundle(tskb, msg, mss))) {
consume_skb(*skb);
*skb = NULL;
}
return true;
}
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
/**
* tipc_msg_extract(): extract bundled inner packet from buffer
* @skb: buffer to be extracted from.
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
* @iskb: extracted inner buffer, to be returned
* @pos: position in outer message of msg to be extracted.
* Returns position of next msg
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
* Consumes outer buffer when last packet extracted
* Returns true when when there is an extracted buffer, otherwise false
*/
bool tipc_msg_extract(struct sk_buff *skb, struct sk_buff **iskb, int *pos)
{
struct tipc_msg *hdr, *ihdr;
int imsz;
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
*iskb = NULL;
if (unlikely(skb_linearize(skb)))
goto none;
hdr = buf_msg(skb);
if (unlikely(*pos > (msg_data_sz(hdr) - MIN_H_SIZE)))
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
goto none;
ihdr = (struct tipc_msg *)(msg_data(hdr) + *pos);
imsz = msg_size(ihdr);
if ((*pos + imsz) > msg_data_sz(hdr))
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
goto none;
*iskb = tipc_buf_acquire(imsz, GFP_ATOMIC);
if (!*iskb)
goto none;
skb_copy_to_linear_data(*iskb, ihdr, imsz);
if (unlikely(!tipc_msg_validate(iskb)))
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
goto none;
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
*pos += align(imsz);
return true;
none:
kfree_skb(skb);
kfree_skb(*iskb);
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 21:36:41 +08:00
*iskb = NULL;
return false;
}
/**
* tipc_msg_reverse(): swap source and destination addresses and add error code
* @own_node: originating node id for reversed message
* @skb: buffer containing message to be reversed; will be consumed
* @err: error code to be set in message, if any
* Replaces consumed buffer with new one when successful
* Returns true if success, otherwise false
*/
bool tipc_msg_reverse(u32 own_node, struct sk_buff **skb, int err)
{
struct sk_buff *_skb = *skb;
struct tipc_msg *_hdr, *hdr;
int hlen, dlen;
if (skb_linearize(_skb))
goto exit;
_hdr = buf_msg(_skb);
dlen = min_t(uint, msg_data_sz(_hdr), MAX_FORWARD_SIZE);
hlen = msg_hdr_sz(_hdr);
if (msg_dest_droppable(_hdr))
goto exit;
if (msg_errcode(_hdr))
goto exit;
/* Never return SHORT header */
if (hlen == SHORT_H_SIZE)
hlen = BASIC_H_SIZE;
/* Don't return data along with SYN+, - sender has a clone */
if (msg_is_syn(_hdr) && err == TIPC_ERR_OVERLOAD)
dlen = 0;
/* Allocate new buffer to return */
*skb = tipc_buf_acquire(hlen + dlen, GFP_ATOMIC);
if (!*skb)
goto exit;
memcpy((*skb)->data, _skb->data, msg_hdr_sz(_hdr));
memcpy((*skb)->data + hlen, msg_data(_hdr), dlen);
/* Build reverse header in new buffer */
hdr = buf_msg(*skb);
msg_set_hdr_sz(hdr, hlen);
msg_set_errcode(hdr, err);
msg_set_non_seq(hdr, 0);
msg_set_origport(hdr, msg_destport(_hdr));
msg_set_destport(hdr, msg_origport(_hdr));
msg_set_destnode(hdr, msg_prevnode(_hdr));
msg_set_prevnode(hdr, own_node);
msg_set_orignode(hdr, own_node);
msg_set_size(hdr, hlen + dlen);
skb_orphan(_skb);
kfree_skb(_skb);
return true;
exit:
kfree_skb(_skb);
*skb = NULL;
return false;
}
bool tipc_msg_skb_clone(struct sk_buff_head *msg, struct sk_buff_head *cpy)
{
struct sk_buff *skb, *_skb;
skb_queue_walk(msg, skb) {
_skb = skb_clone(skb, GFP_ATOMIC);
if (!_skb) {
__skb_queue_purge(cpy);
pr_err_ratelimited("Failed to clone buffer chain\n");
return false;
}
__skb_queue_tail(cpy, _skb);
}
return true;
}
/**
* tipc_msg_lookup_dest(): try to find new destination for named message
* @skb: the buffer containing the message.
* @err: error code to be used by caller if lookup fails
* Does not consume buffer
* Returns true if a destination is found, false otherwise
*/
bool tipc_msg_lookup_dest(struct net *net, struct sk_buff *skb, int *err)
{
struct tipc_msg *msg = buf_msg(skb);
u32 dport, dnode;
u32 onode = tipc_own_addr(net);
if (!msg_isdata(msg))
return false;
if (!msg_named(msg))
return false;
if (msg_errcode(msg))
return false;
*err = TIPC_ERR_NO_NAME;
if (skb_linearize(skb))
return false;
msg = buf_msg(skb);
if (msg_reroute_cnt(msg))
return false;
dnode = tipc_scope2node(net, msg_lookup_scope(msg));
dport = tipc_nametbl_translate(net, msg_nametype(msg),
msg_nameinst(msg), &dnode);
if (!dport)
return false;
msg_incr_reroute_cnt(msg);
if (dnode != onode)
msg_set_prevnode(msg, onode);
msg_set_destnode(msg, dnode);
msg_set_destport(msg, dport);
*err = TIPC_OK;
return true;
}
/* tipc_msg_assemble() - assemble chain of fragments into one message
*/
bool tipc_msg_assemble(struct sk_buff_head *list)
{
struct sk_buff *skb, *tmp = NULL;
if (skb_queue_len(list) == 1)
return true;
while ((skb = __skb_dequeue(list))) {
skb->next = NULL;
if (tipc_buf_append(&tmp, &skb)) {
__skb_queue_tail(list, skb);
return true;
}
if (!tmp)
break;
}
__skb_queue_purge(list);
__skb_queue_head_init(list);
pr_warn("Failed do assemble buffer\n");
return false;
}
/* tipc_msg_reassemble() - clone a buffer chain of fragments and
* reassemble the clones into one message
*/
bool tipc_msg_reassemble(struct sk_buff_head *list, struct sk_buff_head *rcvq)
{
struct sk_buff *skb, *_skb;
struct sk_buff *frag = NULL;
struct sk_buff *head = NULL;
int hdr_len;
/* Copy header if single buffer */
if (skb_queue_len(list) == 1) {
skb = skb_peek(list);
hdr_len = skb_headroom(skb) + msg_hdr_sz(buf_msg(skb));
_skb = __pskb_copy(skb, hdr_len, GFP_ATOMIC);
if (!_skb)
return false;
__skb_queue_tail(rcvq, _skb);
return true;
}
/* Clone all fragments and reassemble */
skb_queue_walk(list, skb) {
frag = skb_clone(skb, GFP_ATOMIC);
if (!frag)
goto error;
frag->next = NULL;
if (tipc_buf_append(&head, &frag))
break;
if (!head)
goto error;
}
__skb_queue_tail(rcvq, frag);
return true;
error:
pr_warn("Failed do clone local mcast rcv buffer\n");
kfree_skb(head);
return false;
}
bool tipc_msg_pskb_copy(u32 dst, struct sk_buff_head *msg,
struct sk_buff_head *cpy)
{
struct sk_buff *skb, *_skb;
skb_queue_walk(msg, skb) {
_skb = pskb_copy(skb, GFP_ATOMIC);
if (!_skb) {
__skb_queue_purge(cpy);
return false;
}
msg_set_destnode(buf_msg(_skb), dst);
__skb_queue_tail(cpy, _skb);
}
return true;
}
/* tipc_skb_queue_sorted(); sort pkt into list according to sequence number
* @list: list to be appended to
* @seqno: sequence number of buffer to add
* @skb: buffer to add
*/
bool __tipc_skb_queue_sorted(struct sk_buff_head *list, u16 seqno,
struct sk_buff *skb)
{
struct sk_buff *_skb, *tmp;
if (skb_queue_empty(list) || less(seqno, buf_seqno(skb_peek(list)))) {
__skb_queue_head(list, skb);
return true;
}
if (more(seqno, buf_seqno(skb_peek_tail(list)))) {
__skb_queue_tail(list, skb);
return true;
}
skb_queue_walk_safe(list, _skb, tmp) {
if (more(seqno, buf_seqno(_skb)))
continue;
if (seqno == buf_seqno(_skb))
break;
__skb_queue_before(list, _skb, skb);
return true;
}
kfree_skb(skb);
return false;
}
void tipc_skb_reject(struct net *net, int err, struct sk_buff *skb,
struct sk_buff_head *xmitq)
{
if (tipc_msg_reverse(tipc_own_addr(net), &skb, err))
__skb_queue_tail(xmitq, skb);
}