2005-08-10 03:14:34 +00:00
|
|
|
/*
|
|
|
|
* net/dccp/proto.c
|
|
|
|
*
|
|
|
|
* An implementation of the DCCP protocol
|
|
|
|
* Arnaldo Carvalho de Melo <acme@conectiva.com.br>
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify it
|
|
|
|
* under the terms of the GNU General Public License version 2 as
|
|
|
|
* published by the Free Software Foundation.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/dccp.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/skbuff.h>
|
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/in.h>
|
|
|
|
#include <linux/if_arp.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/random.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
|
|
|
#include <linux/slab.h>
|
2005-08-10 03:14:34 +00:00
|
|
|
#include <net/checksum.h>
|
|
|
|
|
2005-12-27 04:43:12 +00:00
|
|
|
#include <net/inet_sock.h>
|
2005-08-10 03:14:34 +00:00
|
|
|
#include <net/sock.h>
|
|
|
|
#include <net/xfrm.h>
|
|
|
|
|
2007-10-24 03:23:30 +00:00
|
|
|
#include <asm/ioctls.h>
|
2005-08-10 03:14:34 +00:00
|
|
|
#include <linux/spinlock.h>
|
|
|
|
#include <linux/timer.h>
|
|
|
|
#include <linux/delay.h>
|
|
|
|
#include <linux/poll.h>
|
|
|
|
|
|
|
|
#include "ccid.h"
|
|
|
|
#include "dccp.h"
|
[DCCP]: Initial feature negotiation implementation
Still needs more work, but boots and doesn't crashes, even
does some negotiation!
18:38:52.174934 127.0.0.1.43458 > 127.0.0.1.5001: request <change_l ack_ratio 2, change_r ccid 2, change_l ccid 2>
18:38:52.218526 127.0.0.1.5001 > 127.0.0.1.43458: response <nop, nop, change_l ack_ratio 2, confirm_r ccid 2 2, confirm_l ccid 2 2, confirm_r ack_ratio 2>
18:38:52.185398 127.0.0.1.43458 > 127.0.0.1.5001: <nop, confirm_r ack_ratio 2, ack_vector0 0x00, elapsed_time 212>
:-)
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-21 01:43:56 +00:00
|
|
|
#include "feat.h"
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-26 19:05:31 +00:00
|
|
|
DEFINE_SNMP_STAT(struct dccp_mib, dccp_statistics) __read_mostly;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_statistics);
|
|
|
|
|
2008-11-26 05:17:14 +00:00
|
|
|
struct percpu_counter dccp_orphan_count;
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_orphan_count);
|
|
|
|
|
2008-11-20 08:40:07 +00:00
|
|
|
struct inet_hashinfo dccp_hashinfo;
|
2006-03-21 05:24:19 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_hashinfo);
|
|
|
|
|
2006-11-20 20:30:17 +00:00
|
|
|
/* the maximum queue length for tx in packets. 0 is no limit */
|
|
|
|
int sysctl_dccp_tx_qlen __read_mostly = 5;
|
|
|
|
|
2006-03-21 05:58:56 +00:00
|
|
|
void dccp_set_state(struct sock *sk, const int state)
|
|
|
|
{
|
|
|
|
const int oldstate = sk->sk_state;
|
|
|
|
|
2007-11-28 13:34:53 +00:00
|
|
|
dccp_pr_debug("%s(%p) %s --> %s\n", dccp_role(sk), sk,
|
2006-03-21 05:58:56 +00:00
|
|
|
dccp_state_name(oldstate), dccp_state_name(state));
|
|
|
|
WARN_ON(state == oldstate);
|
|
|
|
|
|
|
|
switch (state) {
|
|
|
|
case DCCP_OPEN:
|
|
|
|
if (oldstate != DCCP_OPEN)
|
|
|
|
DCCP_INC_STATS(DCCP_MIB_CURRESTAB);
|
2008-12-08 09:15:26 +00:00
|
|
|
/* Client retransmits all Confirm options until entering OPEN */
|
|
|
|
if (oldstate == DCCP_PARTOPEN)
|
|
|
|
dccp_feat_list_purge(&dccp_sk(sk)->dccps_featneg);
|
2006-03-21 05:58:56 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
case DCCP_CLOSED:
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
if (oldstate == DCCP_OPEN || oldstate == DCCP_ACTIVE_CLOSEREQ ||
|
|
|
|
oldstate == DCCP_CLOSING)
|
2006-03-21 05:58:56 +00:00
|
|
|
DCCP_INC_STATS(DCCP_MIB_ESTABRESETS);
|
|
|
|
|
|
|
|
sk->sk_prot->unhash(sk);
|
|
|
|
if (inet_csk(sk)->icsk_bind_hash != NULL &&
|
|
|
|
!(sk->sk_userlocks & SOCK_BINDPORT_LOCK))
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 12:06:04 +00:00
|
|
|
inet_put_port(sk);
|
2006-03-21 05:58:56 +00:00
|
|
|
/* fall through */
|
|
|
|
default:
|
|
|
|
if (oldstate == DCCP_OPEN)
|
|
|
|
DCCP_DEC_STATS(DCCP_MIB_CURRESTAB);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Change state AFTER socket is unhashed to avoid closed
|
|
|
|
* socket sitting in hash tables.
|
|
|
|
*/
|
|
|
|
sk->sk_state = state;
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_set_state);
|
|
|
|
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
static void dccp_finish_passive_close(struct sock *sk)
|
|
|
|
{
|
|
|
|
switch (sk->sk_state) {
|
|
|
|
case DCCP_PASSIVE_CLOSE:
|
|
|
|
/* Node (client or server) has received Close packet. */
|
|
|
|
dccp_send_reset(sk, DCCP_RESET_CODE_CLOSED);
|
|
|
|
dccp_set_state(sk, DCCP_CLOSED);
|
|
|
|
break;
|
|
|
|
case DCCP_PASSIVE_CLOSEREQ:
|
|
|
|
/*
|
|
|
|
* Client received CloseReq. We set the `active' flag so that
|
|
|
|
* dccp_send_close() retransmits the Close as per RFC 4340, 8.3.
|
|
|
|
*/
|
|
|
|
dccp_send_close(sk, 1);
|
|
|
|
dccp_set_state(sk, DCCP_CLOSING);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-03-21 05:58:56 +00:00
|
|
|
void dccp_done(struct sock *sk)
|
|
|
|
{
|
|
|
|
dccp_set_state(sk, DCCP_CLOSED);
|
|
|
|
dccp_clear_xmit_timers(sk);
|
|
|
|
|
|
|
|
sk->sk_shutdown = SHUTDOWN_MASK;
|
|
|
|
|
|
|
|
if (!sock_flag(sk, SOCK_DEAD))
|
|
|
|
sk->sk_state_change(sk);
|
|
|
|
else
|
|
|
|
inet_csk_destroy_sock(sk);
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_done);
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
const char *dccp_packet_name(const int type)
|
|
|
|
{
|
2009-08-05 17:42:58 +00:00
|
|
|
static const char *const dccp_packet_names[] = {
|
2005-08-10 03:14:34 +00:00
|
|
|
[DCCP_PKT_REQUEST] = "REQUEST",
|
|
|
|
[DCCP_PKT_RESPONSE] = "RESPONSE",
|
|
|
|
[DCCP_PKT_DATA] = "DATA",
|
|
|
|
[DCCP_PKT_ACK] = "ACK",
|
|
|
|
[DCCP_PKT_DATAACK] = "DATAACK",
|
|
|
|
[DCCP_PKT_CLOSEREQ] = "CLOSEREQ",
|
|
|
|
[DCCP_PKT_CLOSE] = "CLOSE",
|
|
|
|
[DCCP_PKT_RESET] = "RESET",
|
|
|
|
[DCCP_PKT_SYNC] = "SYNC",
|
|
|
|
[DCCP_PKT_SYNCACK] = "SYNCACK",
|
|
|
|
};
|
|
|
|
|
|
|
|
if (type >= DCCP_NR_PKT_TYPES)
|
|
|
|
return "INVALID";
|
|
|
|
else
|
|
|
|
return dccp_packet_names[type];
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_packet_name);
|
|
|
|
|
|
|
|
const char *dccp_state_name(const int state)
|
|
|
|
{
|
2009-08-05 17:42:58 +00:00
|
|
|
static const char *const dccp_state_names[] = {
|
2007-11-28 13:34:53 +00:00
|
|
|
[DCCP_OPEN] = "OPEN",
|
|
|
|
[DCCP_REQUESTING] = "REQUESTING",
|
|
|
|
[DCCP_PARTOPEN] = "PARTOPEN",
|
|
|
|
[DCCP_LISTEN] = "LISTEN",
|
|
|
|
[DCCP_RESPOND] = "RESPOND",
|
|
|
|
[DCCP_CLOSING] = "CLOSING",
|
|
|
|
[DCCP_ACTIVE_CLOSEREQ] = "CLOSEREQ",
|
|
|
|
[DCCP_PASSIVE_CLOSE] = "PASSIVE_CLOSE",
|
|
|
|
[DCCP_PASSIVE_CLOSEREQ] = "PASSIVE_CLOSEREQ",
|
|
|
|
[DCCP_TIME_WAIT] = "TIME_WAIT",
|
|
|
|
[DCCP_CLOSED] = "CLOSED",
|
2005-08-10 03:14:34 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
if (state >= DCCP_MAX_STATES)
|
|
|
|
return "INVALID STATE!";
|
|
|
|
else
|
|
|
|
return dccp_state_names[state];
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_state_name);
|
|
|
|
|
2006-03-21 06:00:37 +00:00
|
|
|
int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized)
|
2006-03-21 05:23:15 +00:00
|
|
|
{
|
|
|
|
struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
struct inet_connection_sock *icsk = inet_csk(sk);
|
|
|
|
|
2007-11-24 23:42:53 +00:00
|
|
|
icsk->icsk_rto = DCCP_TIMEOUT_INIT;
|
|
|
|
icsk->icsk_syn_retries = sysctl_dccp_request_retries;
|
|
|
|
sk->sk_state = DCCP_CLOSED;
|
|
|
|
sk->sk_write_space = dccp_write_space;
|
|
|
|
icsk->icsk_sync_mss = dccp_sync_mss;
|
2008-09-09 11:27:22 +00:00
|
|
|
dp->dccps_mss_cache = 536;
|
2007-11-24 23:42:53 +00:00
|
|
|
dp->dccps_rate_last = jiffies;
|
|
|
|
dp->dccps_role = DCCP_ROLE_UNDEFINED;
|
|
|
|
dp->dccps_service = DCCP_SERVICE_CODE_IS_ABSENT;
|
2008-09-09 11:27:22 +00:00
|
|
|
dp->dccps_l_ack_ratio = dp->dccps_r_ack_ratio = 1;
|
2007-11-24 23:42:53 +00:00
|
|
|
|
|
|
|
dccp_init_xmit_timers(sk);
|
|
|
|
|
2008-11-05 07:55:49 +00:00
|
|
|
INIT_LIST_HEAD(&dp->dccps_featneg);
|
2008-12-08 09:15:26 +00:00
|
|
|
/* control socket doesn't need feat nego */
|
|
|
|
if (likely(ctl_sock_initialized))
|
|
|
|
return dccp_feat_init(sk);
|
2006-03-21 05:23:15 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_init_sock);
|
|
|
|
|
2008-06-15 00:04:49 +00:00
|
|
|
void dccp_destroy_sock(struct sock *sk)
|
2006-03-21 05:23:15 +00:00
|
|
|
{
|
|
|
|
struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* DCCP doesn't use sk_write_queue, just sk_send_head
|
|
|
|
* for retransmissions
|
|
|
|
*/
|
|
|
|
if (sk->sk_send_head != NULL) {
|
|
|
|
kfree_skb(sk->sk_send_head);
|
|
|
|
sk->sk_send_head = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Clean up a referenced DCCP bind bucket. */
|
|
|
|
if (inet_csk(sk)->icsk_bind_hash != NULL)
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 12:06:04 +00:00
|
|
|
inet_put_port(sk);
|
2006-03-21 05:23:15 +00:00
|
|
|
|
|
|
|
kfree(dp->dccps_service_list);
|
|
|
|
dp->dccps_service_list = NULL;
|
|
|
|
|
dccp ccid-2: Phase out the use of boolean Ack Vector sysctl
This removes the use of the sysctl and the minisock variable for the Send Ack
Vector feature, as it now is handled fully dynamically via feature negotiation
(i.e. when CCID-2 is enabled, Ack Vectors are automatically enabled as per
RFC 4341, 4.).
Using a sysctl in parallel to this implementation would open the door to
crashes, since much of the code relies on tests of the boolean minisock /
sysctl variable. Thus, this patch replaces all tests of type
if (dccp_msk(sk)->dccpms_send_ack_vector)
/* ... */
with
if (dp->dccps_hc_rx_ackvec != NULL)
/* ... */
The dccps_hc_rx_ackvec is allocated by the dccp_hdlr_ackvec() when feature
negotiation concluded that Ack Vectors are to be used on the half-connection.
Otherwise, it is NULL (due to dccp_init_sock/dccp_create_openreq_child),
so that the test is a valid one.
The activation handler for Ack Vectors is called as soon as the feature
negotiation has concluded at the
* server when the Ack marking the transition RESPOND => OPEN arrives;
* client after it has sent its ACK, marking the transition REQUEST => PARTOPEN.
Adding the sequence number of the Response packet to the Ack Vector has been
removed, since
(a) connection establishment implies that the Response has been received;
(b) the CCIDs only look at packets received in the (PART)OPEN state, i.e.
this entry will always be ignored;
(c) it can not be used for anything useful - to detect loss for instance, only
packets received after the loss can serve as pseudo-dupacks.
There was a FIXME to change the error code when dccp_ackvec_add() fails.
I removed this after finding out that:
* the check whether ackno < ISN is already made earlier,
* this Response is likely the 1st packet with an Ackno that the client gets,
* so when dccp_ackvec_add() fails, the reason is likely not a packet error.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-12-08 09:19:06 +00:00
|
|
|
if (dp->dccps_hc_rx_ackvec != NULL) {
|
2006-03-21 05:23:15 +00:00
|
|
|
dccp_ackvec_free(dp->dccps_hc_rx_ackvec);
|
|
|
|
dp->dccps_hc_rx_ackvec = NULL;
|
|
|
|
}
|
|
|
|
ccid_hc_rx_delete(dp->dccps_hc_rx_ccid, sk);
|
|
|
|
ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk);
|
|
|
|
dp->dccps_hc_rx_ccid = dp->dccps_hc_tx_ccid = NULL;
|
|
|
|
|
|
|
|
/* clean up feature negotiation state */
|
2008-11-05 07:56:30 +00:00
|
|
|
dccp_feat_list_purge(&dp->dccps_featneg);
|
2006-03-21 05:23:15 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_destroy_sock);
|
|
|
|
|
2006-11-16 10:30:37 +00:00
|
|
|
static inline int dccp_listen_start(struct sock *sk, int backlog)
|
2005-08-10 03:14:34 +00:00
|
|
|
{
|
2005-09-16 23:58:40 +00:00
|
|
|
struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
|
|
|
|
dp->dccps_role = DCCP_ROLE_LISTEN;
|
dccp: Resolve dependencies of features on choice of CCID
This provides a missing link in the code chain, as several features implicitly
depend and/or rely on the choice of CCID. Most notably, this is the Send Ack Vector
feature, but also Ack Ratio and Send Loss Event Rate (also taken care of).
For Send Ack Vector, the situation is as follows:
* since CCID2 mandates the use of Ack Vectors, there is no point in allowing
endpoints which use CCID2 to disable Ack Vector features such a connection;
* a peer with a TX CCID of CCID2 will always expect Ack Vectors, and a peer
with a RX CCID of CCID2 must always send Ack Vectors (RFC 4341, sec. 4);
* for all other CCIDs, the use of (Send) Ack Vector is optional and thus
negotiable. However, this implies that the code negotiating the use of Ack
Vectors also supports it (i.e. is able to supply and to either parse or
ignore received Ack Vectors). Since this is not the case (CCID-3 has no Ack
Vector support), the use of Ack Vectors is here disabled, with a comment
in the source code.
An analogous consideration arises for the Send Loss Event Rate feature,
since the CCID-3 implementation does not support the loss interval options
of RFC 4342. To make such use explicit, corresponding feature-negotiation
options are inserted which signal the use of the loss event rate option,
as it is used by the CCID3 code.
Lastly, the values of the Ack Ratio feature are matched to the choice of CCID.
The patch implements this as a function which is called after the user has
made all other registrations for changing default values of features.
The table is variable-length, the reserved (and hence for feature-negotiation
invalid, confirmed by considering section 19.4 of RFC 4340) feature number `0'
is used to mark the end of the table.
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Acked-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-11-12 08:48:44 +00:00
|
|
|
/* do not start to listen if feature negotiation setup fails */
|
|
|
|
if (dccp_feat_finalise_settings(dp))
|
|
|
|
return -EPROTO;
|
2006-11-16 10:30:37 +00:00
|
|
|
return inet_csk_listen_start(sk, backlog);
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2007-11-25 00:14:15 +00:00
|
|
|
static inline int dccp_need_reset(int state)
|
|
|
|
{
|
|
|
|
return state != DCCP_CLOSED && state != DCCP_LISTEN &&
|
|
|
|
state != DCCP_REQUESTING;
|
|
|
|
}
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
int dccp_disconnect(struct sock *sk, int flags)
|
|
|
|
{
|
|
|
|
struct inet_connection_sock *icsk = inet_csk(sk);
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
|
|
|
int err = 0;
|
|
|
|
const int old_state = sk->sk_state;
|
|
|
|
|
|
|
|
if (old_state != DCCP_CLOSED)
|
|
|
|
dccp_set_state(sk, DCCP_CLOSED);
|
|
|
|
|
2007-11-25 00:14:15 +00:00
|
|
|
/*
|
|
|
|
* This corresponds to the ABORT function of RFC793, sec. 3.8
|
|
|
|
* TCP uses a RST segment, DCCP a Reset packet with Code 2, "Aborted".
|
|
|
|
*/
|
2005-08-10 03:14:34 +00:00
|
|
|
if (old_state == DCCP_LISTEN) {
|
|
|
|
inet_csk_listen_stop(sk);
|
2007-11-25 00:14:15 +00:00
|
|
|
} else if (dccp_need_reset(old_state)) {
|
|
|
|
dccp_send_reset(sk, DCCP_RESET_CODE_ABORTED);
|
|
|
|
sk->sk_err = ECONNRESET;
|
2005-08-10 03:14:34 +00:00
|
|
|
} else if (old_state == DCCP_REQUESTING)
|
|
|
|
sk->sk_err = ECONNRESET;
|
|
|
|
|
|
|
|
dccp_clear_xmit_timers(sk);
|
2008-08-23 11:28:27 +00:00
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
__skb_queue_purge(&sk->sk_receive_queue);
|
2008-08-23 11:28:27 +00:00
|
|
|
__skb_queue_purge(&sk->sk_write_queue);
|
2005-08-10 03:14:34 +00:00
|
|
|
if (sk->sk_send_head != NULL) {
|
|
|
|
__kfree_skb(sk->sk_send_head);
|
|
|
|
sk->sk_send_head = NULL;
|
|
|
|
}
|
|
|
|
|
2009-10-15 06:30:45 +00:00
|
|
|
inet->inet_dport = 0;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
|
|
|
|
inet_reset_saddr(sk);
|
|
|
|
|
|
|
|
sk->sk_shutdown = 0;
|
|
|
|
sock_reset_flag(sk, SOCK_DONE);
|
|
|
|
|
|
|
|
icsk->icsk_backoff = 0;
|
|
|
|
inet_csk_delack_init(sk);
|
|
|
|
__sk_dst_reset(sk);
|
|
|
|
|
2009-10-15 06:30:45 +00:00
|
|
|
WARN_ON(inet->inet_num && !icsk->icsk_bind_hash);
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
sk->sk_error_report(sk);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_disconnect);
|
|
|
|
|
2005-08-24 04:54:23 +00:00
|
|
|
/*
|
|
|
|
* Wait for a DCCP event.
|
|
|
|
*
|
|
|
|
* Note that we don't need to lock the socket, as the upper poll layers
|
|
|
|
* take care of normal races (between the test and the event) and we don't
|
|
|
|
* go look at any of the socket buffers directly.
|
|
|
|
*/
|
2005-12-14 07:24:16 +00:00
|
|
|
unsigned int dccp_poll(struct file *file, struct socket *sock,
|
|
|
|
poll_table *wait)
|
2005-08-24 04:54:23 +00:00
|
|
|
{
|
|
|
|
unsigned int mask;
|
|
|
|
struct sock *sk = sock->sk;
|
|
|
|
|
2010-04-20 13:03:51 +00:00
|
|
|
sock_poll_wait(file, sk_sleep(sk), wait);
|
2005-08-24 04:54:23 +00:00
|
|
|
if (sk->sk_state == DCCP_LISTEN)
|
|
|
|
return inet_csk_listen_poll(sk);
|
|
|
|
|
|
|
|
/* Socket is not locked. We are protected from async events
|
|
|
|
by poll logic and correct handling of state changes
|
|
|
|
made by another threads is impossible in any case.
|
|
|
|
*/
|
|
|
|
|
|
|
|
mask = 0;
|
|
|
|
if (sk->sk_err)
|
|
|
|
mask = POLLERR;
|
|
|
|
|
|
|
|
if (sk->sk_shutdown == SHUTDOWN_MASK || sk->sk_state == DCCP_CLOSED)
|
|
|
|
mask |= POLLHUP;
|
|
|
|
if (sk->sk_shutdown & RCV_SHUTDOWN)
|
2006-03-25 11:07:39 +00:00
|
|
|
mask |= POLLIN | POLLRDNORM | POLLRDHUP;
|
2005-08-24 04:54:23 +00:00
|
|
|
|
|
|
|
/* Connected? */
|
|
|
|
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_RESPOND)) {
|
|
|
|
if (atomic_read(&sk->sk_rmem_alloc) > 0)
|
|
|
|
mask |= POLLIN | POLLRDNORM;
|
|
|
|
|
|
|
|
if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
|
|
|
|
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk)) {
|
|
|
|
mask |= POLLOUT | POLLWRNORM;
|
|
|
|
} else { /* send SIGIO later */
|
|
|
|
set_bit(SOCK_ASYNC_NOSPACE,
|
|
|
|
&sk->sk_socket->flags);
|
|
|
|
set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
|
|
|
|
|
|
|
|
/* Race breaker. If space is freed after
|
|
|
|
* wspace test but before the flags are set,
|
|
|
|
* IO signal will be lost.
|
|
|
|
*/
|
|
|
|
if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk))
|
|
|
|
mask |= POLLOUT | POLLWRNORM;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return mask;
|
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_poll);
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
int dccp_ioctl(struct sock *sk, int cmd, unsigned long arg)
|
|
|
|
{
|
2007-10-24 03:23:30 +00:00
|
|
|
int rc = -ENOTCONN;
|
|
|
|
|
|
|
|
lock_sock(sk);
|
|
|
|
|
|
|
|
if (sk->sk_state == DCCP_LISTEN)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
switch (cmd) {
|
|
|
|
case SIOCINQ: {
|
|
|
|
struct sk_buff *skb;
|
|
|
|
unsigned long amount = 0;
|
|
|
|
|
|
|
|
skb = skb_peek(&sk->sk_receive_queue);
|
|
|
|
if (skb != NULL) {
|
|
|
|
/*
|
|
|
|
* We will only return the amount of this packet since
|
|
|
|
* that is all that will be read.
|
|
|
|
*/
|
|
|
|
amount = skb->len;
|
|
|
|
}
|
|
|
|
rc = put_user(amount, (int __user *)arg);
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
rc = -ENOIOCTLCMD;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
release_sock(sk);
|
|
|
|
return rc;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_ioctl);
|
|
|
|
|
2006-03-21 03:23:32 +00:00
|
|
|
static int dccp_setsockopt_service(struct sock *sk, const __be32 service,
|
2009-09-30 23:12:20 +00:00
|
|
|
char __user *optval, unsigned int optlen)
|
2005-09-16 23:58:40 +00:00
|
|
|
{
|
|
|
|
struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
struct dccp_service_list *sl = NULL;
|
|
|
|
|
2006-12-10 18:01:18 +00:00
|
|
|
if (service == DCCP_SERVICE_INVALID_VALUE ||
|
2005-09-16 23:58:40 +00:00
|
|
|
optlen > DCCP_SERVICE_LIST_MAX_LEN * sizeof(u32))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (optlen > sizeof(service)) {
|
|
|
|
sl = kmalloc(optlen, GFP_KERNEL);
|
|
|
|
if (sl == NULL)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
sl->dccpsl_nr = optlen / sizeof(u32) - 1;
|
|
|
|
if (copy_from_user(sl->dccpsl_list,
|
|
|
|
optval + sizeof(service),
|
|
|
|
optlen - sizeof(service)) ||
|
|
|
|
dccp_list_has_service(sl, DCCP_SERVICE_INVALID_VALUE)) {
|
|
|
|
kfree(sl);
|
|
|
|
return -EFAULT;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
lock_sock(sk);
|
|
|
|
dp->dccps_service = service;
|
|
|
|
|
2005-11-08 17:41:34 +00:00
|
|
|
kfree(dp->dccps_service_list);
|
2005-09-16 23:58:40 +00:00
|
|
|
|
|
|
|
dp->dccps_service_list = sl;
|
|
|
|
release_sock(sk);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-11-17 06:53:48 +00:00
|
|
|
static int dccp_setsockopt_cscov(struct sock *sk, int cscov, bool rx)
|
|
|
|
{
|
|
|
|
u8 *list, len;
|
|
|
|
int i, rc;
|
|
|
|
|
|
|
|
if (cscov < 0 || cscov > 15)
|
|
|
|
return -EINVAL;
|
|
|
|
/*
|
|
|
|
* Populate a list of permissible values, in the range cscov...15. This
|
|
|
|
* is necessary since feature negotiation of single values only works if
|
|
|
|
* both sides incidentally choose the same value. Since the list starts
|
|
|
|
* lowest-value first, negotiation will pick the smallest shared value.
|
|
|
|
*/
|
|
|
|
if (cscov == 0)
|
|
|
|
return 0;
|
|
|
|
len = 16 - cscov;
|
|
|
|
|
|
|
|
list = kmalloc(len, GFP_KERNEL);
|
|
|
|
if (list == NULL)
|
|
|
|
return -ENOBUFS;
|
|
|
|
|
|
|
|
for (i = 0; i < len; i++)
|
|
|
|
list[i] = cscov++;
|
|
|
|
|
|
|
|
rc = dccp_feat_register_sp(sk, DCCPF_MIN_CSUM_COVER, rx, list, len);
|
|
|
|
|
|
|
|
if (rc == 0) {
|
|
|
|
if (rx)
|
|
|
|
dccp_sk(sk)->dccps_pcrlen = cscov;
|
|
|
|
else
|
|
|
|
dccp_sk(sk)->dccps_pcslen = cscov;
|
|
|
|
}
|
|
|
|
kfree(list);
|
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
2008-11-24 00:02:31 +00:00
|
|
|
static int dccp_setsockopt_ccid(struct sock *sk, int type,
|
2009-09-30 23:12:20 +00:00
|
|
|
char __user *optval, unsigned int optlen)
|
2008-11-24 00:02:31 +00:00
|
|
|
{
|
|
|
|
u8 *val;
|
|
|
|
int rc = 0;
|
|
|
|
|
|
|
|
if (optlen < 1 || optlen > DCCP_FEAT_MAX_SP_VALS)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2010-05-21 22:25:19 +00:00
|
|
|
val = memdup_user(optval, optlen);
|
|
|
|
if (IS_ERR(val))
|
|
|
|
return PTR_ERR(val);
|
2008-11-24 00:02:31 +00:00
|
|
|
|
|
|
|
lock_sock(sk);
|
|
|
|
if (type == DCCP_SOCKOPT_TX_CCID || type == DCCP_SOCKOPT_CCID)
|
|
|
|
rc = dccp_feat_register_sp(sk, DCCPF_CCID, 1, val, optlen);
|
|
|
|
|
|
|
|
if (!rc && (type == DCCP_SOCKOPT_RX_CCID || type == DCCP_SOCKOPT_CCID))
|
|
|
|
rc = dccp_feat_register_sp(sk, DCCPF_CCID, 0, val, optlen);
|
|
|
|
release_sock(sk);
|
|
|
|
|
|
|
|
kfree(val);
|
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
static int do_dccp_setsockopt(struct sock *sk, int level, int optname,
|
2009-09-30 23:12:20 +00:00
|
|
|
char __user *optval, unsigned int optlen)
|
2005-08-10 03:14:34 +00:00
|
|
|
{
|
2006-11-14 14:57:34 +00:00
|
|
|
struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
int val, err = 0;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2008-11-17 06:56:55 +00:00
|
|
|
switch (optname) {
|
|
|
|
case DCCP_SOCKOPT_PACKET_SIZE:
|
|
|
|
DCCP_WARN("sockopt(PACKET_SIZE) is deprecated: fix your app\n");
|
|
|
|
return 0;
|
|
|
|
case DCCP_SOCKOPT_CHANGE_L:
|
|
|
|
case DCCP_SOCKOPT_CHANGE_R:
|
|
|
|
DCCP_WARN("sockopt(CHANGE_L/R) is deprecated: fix your app\n");
|
|
|
|
return 0;
|
2008-11-24 00:02:31 +00:00
|
|
|
case DCCP_SOCKOPT_CCID:
|
|
|
|
case DCCP_SOCKOPT_RX_CCID:
|
|
|
|
case DCCP_SOCKOPT_TX_CCID:
|
|
|
|
return dccp_setsockopt_ccid(sk, optname, optval, optlen);
|
2008-11-17 06:56:55 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
if (optlen < (int)sizeof(int))
|
2005-08-28 07:51:32 +00:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (get_user(val, (int __user *)optval))
|
|
|
|
return -EFAULT;
|
|
|
|
|
2005-09-16 23:58:40 +00:00
|
|
|
if (optname == DCCP_SOCKOPT_SERVICE)
|
|
|
|
return dccp_setsockopt_service(sk, val, optval, optlen);
|
2005-08-28 07:51:32 +00:00
|
|
|
|
2005-09-16 23:58:40 +00:00
|
|
|
lock_sock(sk);
|
2005-08-28 07:51:32 +00:00
|
|
|
switch (optname) {
|
2007-12-13 14:25:01 +00:00
|
|
|
case DCCP_SOCKOPT_SERVER_TIMEWAIT:
|
|
|
|
if (dp->dccps_role != DCCP_ROLE_SERVER)
|
|
|
|
err = -EOPNOTSUPP;
|
|
|
|
else
|
|
|
|
dp->dccps_server_timewait = (val != 0);
|
|
|
|
break;
|
2008-11-17 06:53:48 +00:00
|
|
|
case DCCP_SOCKOPT_SEND_CSCOV:
|
|
|
|
err = dccp_setsockopt_cscov(sk, val, false);
|
2008-09-04 05:30:19 +00:00
|
|
|
break;
|
2008-11-17 06:53:48 +00:00
|
|
|
case DCCP_SOCKOPT_RECV_CSCOV:
|
|
|
|
err = dccp_setsockopt_cscov(sk, val, true);
|
2008-09-04 05:30:19 +00:00
|
|
|
break;
|
2005-08-28 07:51:32 +00:00
|
|
|
default:
|
|
|
|
err = -ENOPROTOOPT;
|
|
|
|
break;
|
|
|
|
}
|
2008-09-09 11:27:22 +00:00
|
|
|
release_sock(sk);
|
2008-11-17 06:56:55 +00:00
|
|
|
|
2005-08-28 07:51:32 +00:00
|
|
|
return err;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
int dccp_setsockopt(struct sock *sk, int level, int optname,
|
2009-09-30 23:12:20 +00:00
|
|
|
char __user *optval, unsigned int optlen)
|
2006-03-21 06:45:21 +00:00
|
|
|
{
|
|
|
|
if (level != SOL_DCCP)
|
|
|
|
return inet_csk(sk)->icsk_af_ops->setsockopt(sk, level,
|
|
|
|
optname, optval,
|
|
|
|
optlen);
|
|
|
|
return do_dccp_setsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
2006-03-21 06:48:35 +00:00
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_setsockopt);
|
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
#ifdef CONFIG_COMPAT
|
|
|
|
int compat_dccp_setsockopt(struct sock *sk, int level, int optname,
|
2009-09-30 23:12:20 +00:00
|
|
|
char __user *optval, unsigned int optlen)
|
2006-03-21 06:45:21 +00:00
|
|
|
{
|
2006-03-21 06:46:16 +00:00
|
|
|
if (level != SOL_DCCP)
|
|
|
|
return inet_csk_compat_setsockopt(sk, level, optname,
|
|
|
|
optval, optlen);
|
2006-03-21 06:45:21 +00:00
|
|
|
return do_dccp_setsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
2006-03-21 06:48:35 +00:00
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
EXPORT_SYMBOL_GPL(compat_dccp_setsockopt);
|
|
|
|
#endif
|
|
|
|
|
2005-09-16 23:58:40 +00:00
|
|
|
static int dccp_getsockopt_service(struct sock *sk, int len,
|
2006-03-21 03:23:32 +00:00
|
|
|
__be32 __user *optval,
|
2005-09-16 23:58:40 +00:00
|
|
|
int __user *optlen)
|
|
|
|
{
|
|
|
|
const struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
const struct dccp_service_list *sl;
|
|
|
|
int err = -ENOENT, slen = 0, total_len = sizeof(u32);
|
|
|
|
|
|
|
|
lock_sock(sk);
|
|
|
|
if ((sl = dp->dccps_service_list) != NULL) {
|
|
|
|
slen = sl->dccpsl_nr * sizeof(u32);
|
|
|
|
total_len += slen;
|
|
|
|
}
|
|
|
|
|
|
|
|
err = -EINVAL;
|
|
|
|
if (total_len > len)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
err = 0;
|
|
|
|
if (put_user(total_len, optlen) ||
|
|
|
|
put_user(dp->dccps_service, optval) ||
|
|
|
|
(sl != NULL && copy_to_user(optval + 1, sl->dccpsl_list, slen)))
|
|
|
|
err = -EFAULT;
|
|
|
|
out:
|
|
|
|
release_sock(sk);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
static int do_dccp_getsockopt(struct sock *sk, int level, int optname,
|
2005-08-14 01:42:25 +00:00
|
|
|
char __user *optval, int __user *optlen)
|
2005-08-10 03:14:34 +00:00
|
|
|
{
|
2005-08-28 07:51:32 +00:00
|
|
|
struct dccp_sock *dp;
|
|
|
|
int val, len;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-28 07:51:32 +00:00
|
|
|
if (get_user(len, optlen))
|
|
|
|
return -EFAULT;
|
|
|
|
|
2007-03-28 18:54:32 +00:00
|
|
|
if (len < (int)sizeof(int))
|
2005-08-28 07:51:32 +00:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
dp = dccp_sk(sk);
|
|
|
|
|
|
|
|
switch (optname) {
|
|
|
|
case DCCP_SOCKOPT_PACKET_SIZE:
|
2006-11-28 21:33:36 +00:00
|
|
|
DCCP_WARN("sockopt(PACKET_SIZE) is deprecated: fix your app\n");
|
2006-11-28 21:42:03 +00:00
|
|
|
return 0;
|
2005-09-18 07:19:32 +00:00
|
|
|
case DCCP_SOCKOPT_SERVICE:
|
|
|
|
return dccp_getsockopt_service(sk, len,
|
2006-03-21 03:23:32 +00:00
|
|
|
(__be32 __user *)optval, optlen);
|
2007-10-04 21:39:22 +00:00
|
|
|
case DCCP_SOCKOPT_GET_CUR_MPS:
|
|
|
|
val = dp->dccps_mss_cache;
|
|
|
|
break;
|
2008-11-12 08:47:26 +00:00
|
|
|
case DCCP_SOCKOPT_AVAILABLE_CCIDS:
|
|
|
|
return ccid_getsockopt_builtin_ccids(sk, len, optval, optlen);
|
2008-11-24 00:04:59 +00:00
|
|
|
case DCCP_SOCKOPT_TX_CCID:
|
|
|
|
val = ccid_get_current_tx_ccid(dp);
|
|
|
|
if (val < 0)
|
|
|
|
return -ENOPROTOOPT;
|
|
|
|
break;
|
|
|
|
case DCCP_SOCKOPT_RX_CCID:
|
|
|
|
val = ccid_get_current_rx_ccid(dp);
|
|
|
|
if (val < 0)
|
|
|
|
return -ENOPROTOOPT;
|
|
|
|
break;
|
2007-12-13 14:25:01 +00:00
|
|
|
case DCCP_SOCKOPT_SERVER_TIMEWAIT:
|
|
|
|
val = dp->dccps_server_timewait;
|
|
|
|
break;
|
2006-11-10 19:43:06 +00:00
|
|
|
case DCCP_SOCKOPT_SEND_CSCOV:
|
|
|
|
val = dp->dccps_pcslen;
|
|
|
|
break;
|
|
|
|
case DCCP_SOCKOPT_RECV_CSCOV:
|
|
|
|
val = dp->dccps_pcrlen;
|
|
|
|
break;
|
2005-09-18 07:19:32 +00:00
|
|
|
case 128 ... 191:
|
|
|
|
return ccid_hc_rx_getsockopt(dp->dccps_hc_rx_ccid, sk, optname,
|
|
|
|
len, (u32 __user *)optval, optlen);
|
|
|
|
case 192 ... 255:
|
|
|
|
return ccid_hc_tx_getsockopt(dp->dccps_hc_tx_ccid, sk, optname,
|
|
|
|
len, (u32 __user *)optval, optlen);
|
2005-08-28 07:51:32 +00:00
|
|
|
default:
|
|
|
|
return -ENOPROTOOPT;
|
|
|
|
}
|
|
|
|
|
2007-12-13 14:27:14 +00:00
|
|
|
len = sizeof(val);
|
2005-08-28 07:51:32 +00:00
|
|
|
if (put_user(len, optlen) || copy_to_user(optval, &val, len))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
return 0;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
int dccp_getsockopt(struct sock *sk, int level, int optname,
|
|
|
|
char __user *optval, int __user *optlen)
|
|
|
|
{
|
|
|
|
if (level != SOL_DCCP)
|
|
|
|
return inet_csk(sk)->icsk_af_ops->getsockopt(sk, level,
|
|
|
|
optname, optval,
|
|
|
|
optlen);
|
|
|
|
return do_dccp_getsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
2006-03-21 06:48:35 +00:00
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_getsockopt);
|
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
#ifdef CONFIG_COMPAT
|
|
|
|
int compat_dccp_getsockopt(struct sock *sk, int level, int optname,
|
2006-03-21 06:48:35 +00:00
|
|
|
char __user *optval, int __user *optlen)
|
2006-03-21 06:45:21 +00:00
|
|
|
{
|
2006-03-21 06:46:16 +00:00
|
|
|
if (level != SOL_DCCP)
|
|
|
|
return inet_csk_compat_getsockopt(sk, level, optname,
|
|
|
|
optval, optlen);
|
2006-03-21 06:45:21 +00:00
|
|
|
return do_dccp_getsockopt(sk, level, optname, optval, optlen);
|
|
|
|
}
|
2006-03-21 06:48:35 +00:00
|
|
|
|
2006-03-21 06:45:21 +00:00
|
|
|
EXPORT_SYMBOL_GPL(compat_dccp_getsockopt);
|
|
|
|
#endif
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
int dccp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
|
|
|
|
size_t len)
|
|
|
|
{
|
|
|
|
const struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
const int flags = msg->msg_flags;
|
|
|
|
const int noblock = flags & MSG_DONTWAIT;
|
|
|
|
struct sk_buff *skb;
|
|
|
|
int rc, size;
|
|
|
|
long timeo;
|
|
|
|
|
|
|
|
if (len > dp->dccps_mss_cache)
|
|
|
|
return -EMSGSIZE;
|
|
|
|
|
|
|
|
lock_sock(sk);
|
2006-11-20 20:30:17 +00:00
|
|
|
|
2008-09-09 11:27:22 +00:00
|
|
|
if (sysctl_dccp_tx_qlen &&
|
|
|
|
(sk->sk_write_queue.qlen >= sysctl_dccp_tx_qlen)) {
|
2006-11-20 20:30:17 +00:00
|
|
|
rc = -EAGAIN;
|
|
|
|
goto out_release;
|
|
|
|
}
|
|
|
|
|
2005-08-10 03:30:56 +00:00
|
|
|
timeo = sock_sndtimeo(sk, noblock);
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* We have to use sk_stream_wait_connect here to set sk_write_pending,
|
|
|
|
* so that the trick in dccp_rcv_request_sent_state_process.
|
|
|
|
*/
|
|
|
|
/* Wait for a connection to finish. */
|
2007-09-26 22:36:08 +00:00
|
|
|
if ((1 << sk->sk_state) & ~(DCCPF_OPEN | DCCPF_PARTOPEN))
|
2005-08-10 03:14:34 +00:00
|
|
|
if ((rc = sk_stream_wait_connect(sk, &timeo)) != 0)
|
2005-08-10 03:30:56 +00:00
|
|
|
goto out_release;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
size = sk->sk_prot->max_header + len;
|
|
|
|
release_sock(sk);
|
|
|
|
skb = sock_alloc_send_skb(sk, size, noblock, &rc);
|
|
|
|
lock_sock(sk);
|
|
|
|
if (skb == NULL)
|
|
|
|
goto out_release;
|
|
|
|
|
|
|
|
skb_reserve(skb, sk->sk_prot->max_header);
|
|
|
|
rc = memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len);
|
2005-08-10 03:30:56 +00:00
|
|
|
if (rc != 0)
|
|
|
|
goto out_discard;
|
|
|
|
|
2008-09-09 11:27:22 +00:00
|
|
|
skb_queue_tail(&sk->sk_write_queue, skb);
|
|
|
|
dccp_write_xmit(sk,0);
|
2005-08-10 03:14:34 +00:00
|
|
|
out_release:
|
|
|
|
release_sock(sk);
|
|
|
|
return rc ? : len;
|
2005-08-10 03:30:56 +00:00
|
|
|
out_discard:
|
|
|
|
kfree_skb(skb);
|
2005-08-10 03:14:34 +00:00
|
|
|
goto out_release;
|
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_sendmsg);
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
int dccp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
|
|
|
|
size_t len, int nonblock, int flags, int *addr_len)
|
|
|
|
{
|
|
|
|
const struct dccp_hdr *dh;
|
|
|
|
long timeo;
|
|
|
|
|
|
|
|
lock_sock(sk);
|
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
if (sk->sk_state == DCCP_LISTEN) {
|
|
|
|
len = -ENOTCONN;
|
2005-08-10 03:14:34 +00:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
timeo = sock_rcvtimeo(sk, nonblock);
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
do {
|
2005-08-13 23:35:17 +00:00
|
|
|
struct sk_buff *skb = skb_peek(&sk->sk_receive_queue);
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
if (skb == NULL)
|
|
|
|
goto verify_sock_status;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
dh = dccp_hdr(skb);
|
2005-08-10 03:14:34 +00:00
|
|
|
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
switch (dh->dccph_type) {
|
|
|
|
case DCCP_PKT_DATA:
|
|
|
|
case DCCP_PKT_DATAACK:
|
2005-08-13 23:35:17 +00:00
|
|
|
goto found_ok_skb;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
case DCCP_PKT_CLOSE:
|
|
|
|
case DCCP_PKT_CLOSEREQ:
|
|
|
|
if (!(flags & MSG_PEEK))
|
|
|
|
dccp_finish_passive_close(sk);
|
|
|
|
/* fall through */
|
|
|
|
case DCCP_PKT_RESET:
|
|
|
|
dccp_pr_debug("found fin (%s) ok!\n",
|
|
|
|
dccp_packet_name(dh->dccph_type));
|
2005-08-13 23:35:17 +00:00
|
|
|
len = 0;
|
|
|
|
goto found_fin_ok;
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
default:
|
|
|
|
dccp_pr_debug("packet_type=%s\n",
|
|
|
|
dccp_packet_name(dh->dccph_type));
|
|
|
|
sk_eat_skb(sk, skb, 0);
|
2005-08-13 23:35:17 +00:00
|
|
|
}
|
|
|
|
verify_sock_status:
|
|
|
|
if (sock_flag(sk, SOCK_DONE)) {
|
|
|
|
len = 0;
|
2005-08-10 03:14:34 +00:00
|
|
|
break;
|
2005-08-13 23:35:17 +00:00
|
|
|
}
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
if (sk->sk_err) {
|
|
|
|
len = sock_error(sk);
|
|
|
|
break;
|
|
|
|
}
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
if (sk->sk_shutdown & RCV_SHUTDOWN) {
|
|
|
|
len = 0;
|
|
|
|
break;
|
|
|
|
}
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
if (sk->sk_state == DCCP_CLOSED) {
|
|
|
|
if (!sock_flag(sk, SOCK_DONE)) {
|
|
|
|
/* This occurs when user tries to read
|
|
|
|
* from never connected socket.
|
|
|
|
*/
|
|
|
|
len = -ENOTCONN;
|
2005-08-10 03:14:34 +00:00
|
|
|
break;
|
|
|
|
}
|
2005-08-13 23:35:17 +00:00
|
|
|
len = 0;
|
|
|
|
break;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
if (!timeo) {
|
|
|
|
len = -EAGAIN;
|
|
|
|
break;
|
|
|
|
}
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
if (signal_pending(current)) {
|
|
|
|
len = sock_intr_errno(timeo);
|
|
|
|
break;
|
|
|
|
}
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2005-08-13 23:35:17 +00:00
|
|
|
sk_wait_data(sk, &timeo);
|
2005-08-10 03:14:34 +00:00
|
|
|
continue;
|
|
|
|
found_ok_skb:
|
2005-08-13 23:35:17 +00:00
|
|
|
if (len > skb->len)
|
|
|
|
len = skb->len;
|
|
|
|
else if (len < skb->len)
|
|
|
|
msg->msg_flags |= MSG_TRUNC;
|
|
|
|
|
|
|
|
if (skb_copy_datagram_iovec(skb, 0, msg->msg_iov, len)) {
|
|
|
|
/* Exception. Bailout! */
|
|
|
|
len = -EFAULT;
|
|
|
|
break;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
2010-02-10 20:26:18 +00:00
|
|
|
if (flags & MSG_TRUNC)
|
|
|
|
len = skb->len;
|
2005-08-10 03:14:34 +00:00
|
|
|
found_fin_ok:
|
|
|
|
if (!(flags & MSG_PEEK))
|
2006-05-24 01:01:28 +00:00
|
|
|
sk_eat_skb(sk, skb, 0);
|
2005-08-10 03:14:34 +00:00
|
|
|
break;
|
2005-08-13 23:35:17 +00:00
|
|
|
} while (1);
|
2005-08-10 03:14:34 +00:00
|
|
|
out:
|
|
|
|
release_sock(sk);
|
2005-08-13 23:35:17 +00:00
|
|
|
return len;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_recvmsg);
|
|
|
|
|
|
|
|
int inet_dccp_listen(struct socket *sock, int backlog)
|
2005-08-10 03:14:34 +00:00
|
|
|
{
|
|
|
|
struct sock *sk = sock->sk;
|
|
|
|
unsigned char old_state;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
lock_sock(sk);
|
|
|
|
|
|
|
|
err = -EINVAL;
|
|
|
|
if (sock->state != SS_UNCONNECTED || sock->type != SOCK_DCCP)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
old_state = sk->sk_state;
|
|
|
|
if (!((1 << old_state) & (DCCPF_CLOSED | DCCPF_LISTEN)))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* Really, if the socket is already in listen state
|
|
|
|
* we can only allow the backlog to be adjusted.
|
|
|
|
*/
|
|
|
|
if (old_state != DCCP_LISTEN) {
|
|
|
|
/*
|
|
|
|
* FIXME: here it probably should be sk->sk_prot->listen_start
|
|
|
|
* see tcp_listen_start
|
|
|
|
*/
|
2006-11-16 10:30:37 +00:00
|
|
|
err = dccp_listen_start(sk, backlog);
|
2005-08-10 03:14:34 +00:00
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
sk->sk_max_ack_backlog = backlog;
|
|
|
|
err = 0;
|
|
|
|
|
|
|
|
out:
|
|
|
|
release_sock(sk);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(inet_dccp_listen);
|
|
|
|
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
static void dccp_terminate_connection(struct sock *sk)
|
2005-08-10 03:14:34 +00:00
|
|
|
{
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
u8 next_state = DCCP_CLOSED;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
switch (sk->sk_state) {
|
|
|
|
case DCCP_PASSIVE_CLOSE:
|
|
|
|
case DCCP_PASSIVE_CLOSEREQ:
|
|
|
|
dccp_finish_passive_close(sk);
|
|
|
|
break;
|
|
|
|
case DCCP_PARTOPEN:
|
|
|
|
dccp_pr_debug("Stop PARTOPEN timer (%p)\n", sk);
|
|
|
|
inet_csk_clear_xmit_timer(sk, ICSK_TIME_DACK);
|
|
|
|
/* fall through */
|
|
|
|
case DCCP_OPEN:
|
|
|
|
dccp_send_close(sk, 1);
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2007-12-13 14:25:01 +00:00
|
|
|
if (dccp_sk(sk)->dccps_role == DCCP_ROLE_SERVER &&
|
|
|
|
!dccp_sk(sk)->dccps_server_timewait)
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
next_state = DCCP_ACTIVE_CLOSEREQ;
|
|
|
|
else
|
|
|
|
next_state = DCCP_CLOSING;
|
|
|
|
/* fall through */
|
|
|
|
default:
|
|
|
|
dccp_set_state(sk, next_state);
|
|
|
|
}
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void dccp_close(struct sock *sk, long timeout)
|
|
|
|
{
|
2006-08-27 02:16:45 +00:00
|
|
|
struct dccp_sock *dp = dccp_sk(sk);
|
2005-08-10 03:14:34 +00:00
|
|
|
struct sk_buff *skb;
|
2007-12-17 00:06:03 +00:00
|
|
|
u32 data_was_unread = 0;
|
2006-05-06 00:09:13 +00:00
|
|
|
int state;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
lock_sock(sk);
|
|
|
|
|
|
|
|
sk->sk_shutdown = SHUTDOWN_MASK;
|
|
|
|
|
|
|
|
if (sk->sk_state == DCCP_LISTEN) {
|
|
|
|
dccp_set_state(sk, DCCP_CLOSED);
|
|
|
|
|
|
|
|
/* Special case. */
|
|
|
|
inet_csk_listen_stop(sk);
|
|
|
|
|
|
|
|
goto adjudge_to_death;
|
|
|
|
}
|
|
|
|
|
2006-08-27 02:16:45 +00:00
|
|
|
sk_stop_timer(sk, &dp->dccps_xmit_timer);
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
/*
|
|
|
|
* We need to flush the recv. buffs. We do this only on the
|
|
|
|
* descriptor close, not protocol-sourced closes, because the
|
|
|
|
*reader process may not have drained the data yet!
|
|
|
|
*/
|
|
|
|
while ((skb = __skb_dequeue(&sk->sk_receive_queue)) != NULL) {
|
2007-12-17 00:06:03 +00:00
|
|
|
data_was_unread += skb->len;
|
2005-08-10 03:14:34 +00:00
|
|
|
__kfree_skb(skb);
|
|
|
|
}
|
|
|
|
|
2007-12-17 00:06:03 +00:00
|
|
|
if (data_was_unread) {
|
|
|
|
/* Unread data was tossed, send an appropriate Reset Code */
|
|
|
|
DCCP_WARN("DCCP: ABORT -- %u bytes unread\n", data_was_unread);
|
|
|
|
dccp_send_reset(sk, DCCP_RESET_CODE_ABORTED);
|
|
|
|
dccp_set_state(sk, DCCP_CLOSED);
|
|
|
|
} else if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) {
|
2005-08-10 03:14:34 +00:00
|
|
|
/* Check zero linger _after_ checking for unread data. */
|
|
|
|
sk->sk_prot->disconnect(sk, 0);
|
[DCCP]: Integrate state transitions for passive-close
This adds the necessary state transitions for the two forms of passive-close
* PASSIVE_CLOSE - which is entered when a host receives a Close;
* PASSIVE_CLOSEREQ - which is entered when a client receives a CloseReq.
Here is a detailed account of what the patch does in each state.
1) Receiving CloseReq
The pseudo-code in 8.5 says:
Step 13: Process CloseReq
If P.type == CloseReq and S.state < CLOSEREQ,
Generate Close
S.state := CLOSING
Set CLOSING timer.
This means we need to address what to do in CLOSED, LISTEN, REQUEST, RESPOND, PARTOPEN, and OPEN.
* CLOSED: silently ignore - it may be a late or duplicate CloseReq;
* LISTEN/RESPOND: will not appear, since Step 7 is performed first (we know we are the client);
* REQUEST: perform Step 13 directly (no need to enqueue packet);
* OPEN/PARTOPEN: enter PASSIVE_CLOSEREQ so that the application has a chance to process unread data.
When already in PASSIVE_CLOSEREQ, no second CloseReq is enqueued. In any other state, the CloseReq is ignored.
I think that this offers some robustness against rare and pathological cases: e.g. a simultaneous close where
the client sends a Close and the server a CloseReq. The client will then be retransmitting its Close until it
gets the Reset, so ignoring the CloseReq while in state CLOSING is sane.
2) Receiving Close
The code below from 8.5 is unconditional.
Step 14: Process Close
If P.type == Close,
Generate Reset(Closed)
Tear down connection
Drop packet and return
Thus we need to consider all states:
* CLOSED: silently ignore, since this can happen when a retransmitted or late Close arrives;
* LISTEN: dccp_rcv_state_process() will generate a Reset ("No Connection");
* REQUEST: perform Step 14 directly (no need to enqueue packet);
* RESPOND: dccp_check_req() will generate a Reset ("Packet Error") -- left it at that;
* OPEN/PARTOPEN: enter PASSIVE_CLOSE so that application has a chance to process unread data;
* CLOSEREQ: server performed active-close -- perform Step 14;
* CLOSING: simultaneous-close: use a tie-breaker to avoid message ping-pong (see comment);
* PASSIVE_CLOSEREQ: ignore - the peer has a bug (sending first a CloseReq and now a Close);
* TIMEWAIT: packet is ignored.
Note that the condition of receiving a packet in state CLOSED here is different from the condition "there
is no socket for such a connection": the socket still exists, but its state indicates it is unusable.
Last, dccp_finish_passive_close sets either DCCP_CLOSED or DCCP_CLOSING = TCP_CLOSING, so that
sk_stream_wait_close() will wait for the final Reset (which will trigger CLOSING => CLOSED).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-28 13:59:48 +00:00
|
|
|
} else if (sk->sk_state != DCCP_CLOSED) {
|
|
|
|
dccp_terminate_connection(sk);
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
sk_stream_wait_close(sk, timeout);
|
|
|
|
|
|
|
|
adjudge_to_death:
|
2006-05-06 00:09:13 +00:00
|
|
|
state = sk->sk_state;
|
|
|
|
sock_hold(sk);
|
|
|
|
sock_orphan(sk);
|
|
|
|
|
2005-08-24 04:50:06 +00:00
|
|
|
/*
|
|
|
|
* It is the last release_sock in its life. It will remove backlog.
|
|
|
|
*/
|
2005-08-10 03:14:34 +00:00
|
|
|
release_sock(sk);
|
|
|
|
/*
|
|
|
|
* Now socket is owned by kernel and we acquire BH lock
|
|
|
|
* to finish close. No need to check for user refs.
|
|
|
|
*/
|
|
|
|
local_bh_disable();
|
|
|
|
bh_lock_sock(sk);
|
2008-07-26 04:43:18 +00:00
|
|
|
WARN_ON(sock_owned_by_user(sk));
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2008-12-30 07:04:08 +00:00
|
|
|
percpu_counter_inc(sk->sk_prot->orphan_count);
|
|
|
|
|
2006-05-06 00:09:13 +00:00
|
|
|
/* Have we already been destroyed by a softirq or backlog? */
|
|
|
|
if (state != DCCP_CLOSED && sk->sk_state == DCCP_CLOSED)
|
|
|
|
goto out;
|
2005-08-24 04:50:06 +00:00
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
if (sk->sk_state == DCCP_CLOSED)
|
|
|
|
inet_csk_destroy_sock(sk);
|
|
|
|
|
|
|
|
/* Otherwise, socket is reprieved until protocol close. */
|
|
|
|
|
2006-05-06 00:09:13 +00:00
|
|
|
out:
|
2005-08-10 03:14:34 +00:00
|
|
|
bh_unlock_sock(sk);
|
|
|
|
local_bh_enable();
|
|
|
|
sock_put(sk);
|
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_close);
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
void dccp_shutdown(struct sock *sk, int how)
|
|
|
|
{
|
[DCCP]: Honour and make use of shutdown option set by user
This extends the DCCP socket API by honouring any shutdown(2) option set by the user.
The behaviour is, as much as possible, made consistent with the API for TCP's shutdown.
This patch exploits the information provided by the user via the socket API to reduce
processing costs:
* if the read end is closed (SHUT_RD), it is not necessary to deliver to input CCID;
* if the write end is closed (SHUT_WR), the same idea applies, but with a difference -
as long as the TX queue has not been drained, we need to receive feedback to keep
congestion-control rates up to date. Hence SHUT_WR is honoured only after the last
packet (under congestion control) has been sent;
* although SHUT_RDWR seems nonsensical, it is nevertheless supported in the same manner
as for TCP (and agrees with test for SHUTDOWN_MASK in dccp_poll() in net/dccp/proto.c).
Furthermore, most of the code already honours the sk_shutdown flags (dccp_recvmsg() for
instance sets the read length to 0 if SHUT_RD had been called); CCID handling is now added
to this by the present patch.
There will also no longer be any delivery when the socket is in the final stages, i.e. when
one of dccp_close(), dccp_fin(), or dccp_done() has been called - which is fine since at
that stage the connection is its final stages.
Motivation and background are on http://www.erg.abdn.ac.uk/users/gerrit/dccp/notes/shutdown
A FIXME has been added to notify the other end if SHUT_RD has been set (RFC 4340, 11.7).
Note: There is a comment in inet_shutdown() in net/ipv4/af_inet.c which asks to "make
sure the socket is a TCP socket". This should probably be extended to mean
`TCP or DCCP socket' (the code is also used by UDP and raw sockets).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-11-21 11:56:48 +00:00
|
|
|
dccp_pr_debug("called shutdown(%x)\n", how);
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2005-12-14 07:24:16 +00:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_shutdown);
|
|
|
|
|
2008-04-10 10:48:43 +00:00
|
|
|
static inline int dccp_mib_init(void)
|
2005-08-10 03:14:34 +00:00
|
|
|
{
|
2010-02-16 15:20:26 +00:00
|
|
|
return snmp_mib_init((void __percpu **)dccp_statistics,
|
2010-06-22 20:58:41 +00:00
|
|
|
sizeof(struct dccp_mib),
|
|
|
|
__alignof__(struct dccp_mib));
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2008-04-10 10:48:43 +00:00
|
|
|
static inline void dccp_mib_exit(void)
|
2006-03-21 05:24:42 +00:00
|
|
|
{
|
2010-02-16 15:20:26 +00:00
|
|
|
snmp_mib_free((void __percpu **)dccp_statistics);
|
2006-03-21 05:24:42 +00:00
|
|
|
}
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
static int thash_entries;
|
|
|
|
module_param(thash_entries, int, 0444);
|
|
|
|
MODULE_PARM_DESC(thash_entries, "Number of ehash buckets");
|
|
|
|
|
2005-08-14 01:42:25 +00:00
|
|
|
#ifdef CONFIG_IP_DCCP_DEBUG
|
2005-08-10 03:14:34 +00:00
|
|
|
int dccp_debug;
|
2008-08-23 11:28:27 +00:00
|
|
|
module_param(dccp_debug, bool, 0644);
|
2005-08-10 03:14:34 +00:00
|
|
|
MODULE_PARM_DESC(dccp_debug, "Enable debug messages");
|
2005-12-14 07:24:16 +00:00
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_debug);
|
2005-08-14 01:42:25 +00:00
|
|
|
#endif
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
static int __init dccp_init(void)
|
|
|
|
{
|
|
|
|
unsigned long goal;
|
|
|
|
int ehash_order, bhash_order, i;
|
2008-11-26 05:17:14 +00:00
|
|
|
int rc;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
2008-04-13 01:35:41 +00:00
|
|
|
BUILD_BUG_ON(sizeof(struct dccp_skb_cb) >
|
|
|
|
FIELD_SIZEOF(struct sk_buff, cb));
|
2008-11-26 05:17:14 +00:00
|
|
|
rc = percpu_counter_init(&dccp_orphan_count, 0);
|
|
|
|
if (rc)
|
2010-03-14 20:13:19 +00:00
|
|
|
goto out_fail;
|
2008-11-26 05:17:14 +00:00
|
|
|
rc = -ENOBUFS;
|
2008-11-20 08:40:07 +00:00
|
|
|
inet_hashinfo_init(&dccp_hashinfo);
|
2005-08-13 23:34:54 +00:00
|
|
|
dccp_hashinfo.bind_bucket_cachep =
|
|
|
|
kmem_cache_create("dccp_bind_bucket",
|
|
|
|
sizeof(struct inet_bind_bucket), 0,
|
2007-07-20 01:11:58 +00:00
|
|
|
SLAB_HWCACHE_ALIGN, NULL);
|
2005-08-10 03:14:34 +00:00
|
|
|
if (!dccp_hashinfo.bind_bucket_cachep)
|
2008-11-26 05:17:14 +00:00
|
|
|
goto out_free_percpu;
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Size and allocate the main established and bind bucket
|
|
|
|
* hash tables.
|
|
|
|
*
|
|
|
|
* The methodology is similar to that of the buffer cache.
|
|
|
|
*/
|
2009-09-22 00:03:05 +00:00
|
|
|
if (totalram_pages >= (128 * 1024))
|
|
|
|
goal = totalram_pages >> (21 - PAGE_SHIFT);
|
2005-08-10 03:14:34 +00:00
|
|
|
else
|
2009-09-22 00:03:05 +00:00
|
|
|
goal = totalram_pages >> (23 - PAGE_SHIFT);
|
2005-08-10 03:14:34 +00:00
|
|
|
|
|
|
|
if (thash_entries)
|
2005-08-13 23:34:54 +00:00
|
|
|
goal = (thash_entries *
|
|
|
|
sizeof(struct inet_ehash_bucket)) >> PAGE_SHIFT;
|
2005-08-10 03:14:34 +00:00
|
|
|
for (ehash_order = 0; (1UL << ehash_order) < goal; ehash_order++)
|
|
|
|
;
|
|
|
|
do {
|
2009-10-09 00:16:19 +00:00
|
|
|
unsigned long hash_size = (1UL << ehash_order) * PAGE_SIZE /
|
2005-08-10 03:14:34 +00:00
|
|
|
sizeof(struct inet_ehash_bucket);
|
2009-10-09 00:16:19 +00:00
|
|
|
|
|
|
|
while (hash_size & (hash_size - 1))
|
|
|
|
hash_size--;
|
|
|
|
dccp_hashinfo.ehash_mask = hash_size - 1;
|
2005-08-10 03:14:34 +00:00
|
|
|
dccp_hashinfo.ehash = (struct inet_ehash_bucket *)
|
2009-07-29 22:04:10 +00:00
|
|
|
__get_free_pages(GFP_ATOMIC|__GFP_NOWARN, ehash_order);
|
2005-08-10 03:14:34 +00:00
|
|
|
} while (!dccp_hashinfo.ehash && --ehash_order > 0);
|
|
|
|
|
|
|
|
if (!dccp_hashinfo.ehash) {
|
2006-11-20 20:39:23 +00:00
|
|
|
DCCP_CRIT("Failed to allocate DCCP established hash table");
|
2005-08-10 03:14:34 +00:00
|
|
|
goto out_free_bind_bucket_cachep;
|
|
|
|
}
|
|
|
|
|
2009-10-09 00:16:19 +00:00
|
|
|
for (i = 0; i <= dccp_hashinfo.ehash_mask; i++) {
|
2008-11-17 03:40:17 +00:00
|
|
|
INIT_HLIST_NULLS_HEAD(&dccp_hashinfo.ehash[i].chain, i);
|
|
|
|
INIT_HLIST_NULLS_HEAD(&dccp_hashinfo.ehash[i].twchain, i);
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
2007-11-07 10:40:20 +00:00
|
|
|
if (inet_ehash_locks_alloc(&dccp_hashinfo))
|
|
|
|
goto out_free_dccp_ehash;
|
|
|
|
|
2005-08-10 03:14:34 +00:00
|
|
|
bhash_order = ehash_order;
|
|
|
|
|
|
|
|
do {
|
|
|
|
dccp_hashinfo.bhash_size = (1UL << bhash_order) * PAGE_SIZE /
|
|
|
|
sizeof(struct inet_bind_hashbucket);
|
2005-08-13 23:34:54 +00:00
|
|
|
if ((dccp_hashinfo.bhash_size > (64 * 1024)) &&
|
|
|
|
bhash_order > 0)
|
2005-08-10 03:14:34 +00:00
|
|
|
continue;
|
|
|
|
dccp_hashinfo.bhash = (struct inet_bind_hashbucket *)
|
2009-07-29 22:04:10 +00:00
|
|
|
__get_free_pages(GFP_ATOMIC|__GFP_NOWARN, bhash_order);
|
2005-08-10 03:14:34 +00:00
|
|
|
} while (!dccp_hashinfo.bhash && --bhash_order >= 0);
|
|
|
|
|
|
|
|
if (!dccp_hashinfo.bhash) {
|
2006-11-20 20:39:23 +00:00
|
|
|
DCCP_CRIT("Failed to allocate DCCP bind hash table");
|
2007-11-07 10:40:20 +00:00
|
|
|
goto out_free_dccp_locks;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
for (i = 0; i < dccp_hashinfo.bhash_size; i++) {
|
|
|
|
spin_lock_init(&dccp_hashinfo.bhash[i].lock);
|
|
|
|
INIT_HLIST_HEAD(&dccp_hashinfo.bhash[i].chain);
|
|
|
|
}
|
|
|
|
|
2006-03-21 05:24:42 +00:00
|
|
|
rc = dccp_mib_init();
|
2006-03-21 01:16:01 +00:00
|
|
|
if (rc)
|
2005-08-10 03:14:34 +00:00
|
|
|
goto out_free_dccp_bhash;
|
|
|
|
|
2006-03-21 01:16:17 +00:00
|
|
|
rc = dccp_ackvec_init();
|
2005-08-10 03:14:34 +00:00
|
|
|
if (rc)
|
2006-03-21 05:25:11 +00:00
|
|
|
goto out_free_dccp_mib;
|
2006-03-21 01:16:17 +00:00
|
|
|
|
2006-03-21 03:25:02 +00:00
|
|
|
rc = dccp_sysctl_init();
|
2006-03-21 01:16:17 +00:00
|
|
|
if (rc)
|
|
|
|
goto out_ackvec_exit;
|
2007-09-26 05:40:13 +00:00
|
|
|
|
2009-01-05 05:42:53 +00:00
|
|
|
rc = ccid_initialize_builtins();
|
|
|
|
if (rc)
|
|
|
|
goto out_sysctl_exit;
|
|
|
|
|
2007-09-26 05:40:13 +00:00
|
|
|
dccp_timestamping_init();
|
2010-03-14 20:13:19 +00:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
2009-01-05 05:42:53 +00:00
|
|
|
out_sysctl_exit:
|
|
|
|
dccp_sysctl_exit();
|
2006-03-21 01:16:17 +00:00
|
|
|
out_ackvec_exit:
|
|
|
|
dccp_ackvec_exit();
|
2006-03-21 05:25:11 +00:00
|
|
|
out_free_dccp_mib:
|
2006-03-21 05:24:42 +00:00
|
|
|
dccp_mib_exit();
|
2005-08-10 03:14:34 +00:00
|
|
|
out_free_dccp_bhash:
|
|
|
|
free_pages((unsigned long)dccp_hashinfo.bhash, bhash_order);
|
2007-11-07 10:40:20 +00:00
|
|
|
out_free_dccp_locks:
|
|
|
|
inet_ehash_locks_free(&dccp_hashinfo);
|
2005-08-10 03:14:34 +00:00
|
|
|
out_free_dccp_ehash:
|
|
|
|
free_pages((unsigned long)dccp_hashinfo.ehash, ehash_order);
|
|
|
|
out_free_bind_bucket_cachep:
|
|
|
|
kmem_cache_destroy(dccp_hashinfo.bind_bucket_cachep);
|
2008-11-26 05:17:14 +00:00
|
|
|
out_free_percpu:
|
|
|
|
percpu_counter_destroy(&dccp_orphan_count);
|
2010-03-14 20:13:19 +00:00
|
|
|
out_fail:
|
|
|
|
dccp_hashinfo.bhash = NULL;
|
|
|
|
dccp_hashinfo.ehash = NULL;
|
|
|
|
dccp_hashinfo.bind_bucket_cachep = NULL;
|
|
|
|
return rc;
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit dccp_fini(void)
|
|
|
|
{
|
2009-01-05 05:42:53 +00:00
|
|
|
ccid_cleanup_builtins();
|
2006-03-21 05:24:42 +00:00
|
|
|
dccp_mib_exit();
|
2005-08-13 23:35:39 +00:00
|
|
|
free_pages((unsigned long)dccp_hashinfo.bhash,
|
|
|
|
get_order(dccp_hashinfo.bhash_size *
|
|
|
|
sizeof(struct inet_bind_hashbucket)));
|
|
|
|
free_pages((unsigned long)dccp_hashinfo.ehash,
|
2009-10-09 00:16:19 +00:00
|
|
|
get_order((dccp_hashinfo.ehash_mask + 1) *
|
2005-08-13 23:35:39 +00:00
|
|
|
sizeof(struct inet_ehash_bucket)));
|
2007-11-07 10:40:20 +00:00
|
|
|
inet_ehash_locks_free(&dccp_hashinfo);
|
2005-08-10 03:14:34 +00:00
|
|
|
kmem_cache_destroy(dccp_hashinfo.bind_bucket_cachep);
|
2006-03-21 01:16:17 +00:00
|
|
|
dccp_ackvec_exit();
|
2006-03-21 03:25:02 +00:00
|
|
|
dccp_sysctl_exit();
|
2009-08-04 21:44:39 +00:00
|
|
|
percpu_counter_destroy(&dccp_orphan_count);
|
2005-08-10 03:14:34 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
module_init(dccp_init);
|
|
|
|
module_exit(dccp_fini);
|
|
|
|
|
|
|
|
MODULE_LICENSE("GPL");
|
|
|
|
MODULE_AUTHOR("Arnaldo Carvalho de Melo <acme@conectiva.com.br>");
|
|
|
|
MODULE_DESCRIPTION("DCCP - Datagram Congestion Controlled Protocol");
|