2005-04-17 06:20:36 +08:00
|
|
|
/*
|
|
|
|
* DECnet An implementation of the DECnet protocol suite for the LINUX
|
|
|
|
* operating system. DECnet is implemented using the BSD Socket
|
|
|
|
* interface as the means of communication with the user level.
|
|
|
|
*
|
|
|
|
* DECnet Network Services Protocol (Output)
|
|
|
|
*
|
|
|
|
* Author: Eduardo Marcelo Serrat <emserrat@geocities.com>
|
|
|
|
*
|
|
|
|
* Changes:
|
|
|
|
*
|
|
|
|
* Steve Whitehouse: Split into dn_nsp_in.c and dn_nsp_out.c from
|
|
|
|
* original dn_nsp.c.
|
|
|
|
* Steve Whitehouse: Updated to work with my new routing architecture.
|
|
|
|
* Steve Whitehouse: Added changes from Eduardo Serrat's patches.
|
|
|
|
* Steve Whitehouse: Now conninits have the "return" bit set.
|
|
|
|
* Steve Whitehouse: Fixes to check alloc'd skbs are non NULL!
|
|
|
|
* Moved output state machine into one function
|
|
|
|
* Steve Whitehouse: New output state machine
|
|
|
|
* Paul Koning: Connect Confirm message fix.
|
|
|
|
* Eduardo Serrat: Fix to stop dn_nsp_do_disc() sending malformed packets.
|
|
|
|
* Steve Whitehouse: dn_nsp_output() and friends needed a spring clean
|
|
|
|
* Steve Whitehouse: Moved dn_nsp_send() in here from route.h
|
|
|
|
*/
|
|
|
|
|
|
|
|
/******************************************************************************
|
|
|
|
(c) 1995-1998 E.M. Serrat emserrat@geocities.com
|
2007-02-09 22:24:40 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
This program is free software; you can redistribute it and/or modify
|
|
|
|
it under the terms of the GNU General Public License as published by
|
|
|
|
the Free Software Foundation; either version 2 of the License, or
|
|
|
|
any later version.
|
|
|
|
|
|
|
|
This program is distributed in the hope that it will be useful,
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
GNU General Public License for more details.
|
|
|
|
*******************************************************************************/
|
|
|
|
|
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/socket.h>
|
|
|
|
#include <linux/in.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/timer.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/sockios.h>
|
|
|
|
#include <linux/net.h>
|
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/inet.h>
|
|
|
|
#include <linux/route.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 16:04:11 +08:00
|
|
|
#include <linux/slab.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
#include <net/sock.h>
|
|
|
|
#include <linux/fcntl.h>
|
|
|
|
#include <linux/mm.h>
|
2007-02-09 22:24:40 +08:00
|
|
|
#include <linux/termios.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
#include <linux/interrupt.h>
|
|
|
|
#include <linux/proc_fs.h>
|
|
|
|
#include <linux/stat.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/poll.h>
|
|
|
|
#include <linux/if_packet.h>
|
|
|
|
#include <net/neighbour.h>
|
|
|
|
#include <net/dst.h>
|
|
|
|
#include <net/flow.h>
|
|
|
|
#include <net/dn.h>
|
|
|
|
#include <net/dn_nsp.h>
|
|
|
|
#include <net/dn_dev.h>
|
|
|
|
#include <net/dn_route.h>
|
|
|
|
|
|
|
|
|
|
|
|
static int nsp_backoff[NSP_MAXRXTSHIFT + 1] = { 1, 2, 4, 8, 16, 32, 64, 64, 64, 64, 64, 64, 64 };
|
|
|
|
|
|
|
|
static void dn_nsp_send(struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
struct sock *sk = skb->sk;
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct dst_entry *dst;
|
2011-03-13 06:17:10 +08:00
|
|
|
struct flowidn fld;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-03-14 00:06:52 +08:00
|
|
|
skb_reset_transport_header(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
scp->stamp = jiffies;
|
|
|
|
|
|
|
|
dst = sk_dst_check(sk, 0);
|
|
|
|
if (dst) {
|
|
|
|
try_again:
|
2009-06-02 13:19:30 +08:00
|
|
|
skb_dst_set(skb, dst);
|
2015-10-08 05:48:35 +08:00
|
|
|
dst_output(&init_net, skb->sk, skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2011-03-13 06:17:10 +08:00
|
|
|
memset(&fld, 0, sizeof(fld));
|
|
|
|
fld.flowidn_oif = sk->sk_bound_dev_if;
|
|
|
|
fld.saddr = dn_saddr2dn(&scp->addr);
|
|
|
|
fld.daddr = dn_saddr2dn(&scp->peer);
|
|
|
|
dn_sk_ports_copy(&fld, scp);
|
|
|
|
fld.flowidn_proto = DNPROTO_NSP;
|
|
|
|
if (dn_route_output_sock(&sk->sk_dst_cache, &fld, sk, 0) == 0) {
|
2005-04-17 06:20:36 +08:00
|
|
|
dst = sk_dst_get(sk);
|
|
|
|
sk->sk_route_caps = dst->dev->features;
|
|
|
|
goto try_again;
|
|
|
|
}
|
|
|
|
|
|
|
|
sk->sk_err = EHOSTUNREACH;
|
|
|
|
if (!sock_flag(sk, SOCK_DEAD))
|
|
|
|
sk->sk_state_change(sk);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If sk == NULL, then we assume that we are supposed to be making
|
|
|
|
* a routing layer skb. If sk != NULL, then we are supposed to be
|
|
|
|
* creating an skb for the NSP layer.
|
|
|
|
*
|
|
|
|
* The eventual aim is for each socket to have a cached header size
|
|
|
|
* for its outgoing packets, and to set hdr from this when sk != NULL.
|
|
|
|
*/
|
2005-10-07 14:46:04 +08:00
|
|
|
struct sk_buff *dn_alloc_skb(struct sock *sk, int size, gfp_t pri)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct sk_buff *skb;
|
|
|
|
int hdr = 64;
|
|
|
|
|
|
|
|
if ((skb = alloc_skb(size + hdr, pri)) == NULL)
|
|
|
|
return NULL;
|
|
|
|
|
2007-12-12 02:51:49 +08:00
|
|
|
skb->protocol = htons(ETH_P_DNA_RT);
|
2005-04-17 06:20:36 +08:00
|
|
|
skb->pkt_type = PACKET_OUTGOING;
|
|
|
|
|
|
|
|
if (sk)
|
|
|
|
skb_set_owner_w(skb, sk);
|
|
|
|
|
|
|
|
skb_reserve(skb, hdr);
|
|
|
|
|
|
|
|
return skb;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Calculate persist timer based upon the smoothed round
|
|
|
|
* trip time and the variance. Backoff according to the
|
|
|
|
* nsp_backoff[] array.
|
|
|
|
*/
|
|
|
|
unsigned long dn_nsp_persist(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
|
|
|
|
unsigned long t = ((scp->nsp_srtt >> 2) + scp->nsp_rttvar) >> 1;
|
|
|
|
|
|
|
|
t *= nsp_backoff[scp->nsp_rxtshift];
|
|
|
|
|
|
|
|
if (t < HZ) t = HZ;
|
|
|
|
if (t > (600*HZ)) t = (600*HZ);
|
|
|
|
|
|
|
|
if (scp->nsp_rxtshift < NSP_MAXRXTSHIFT)
|
|
|
|
scp->nsp_rxtshift++;
|
|
|
|
|
|
|
|
/* printk(KERN_DEBUG "rxtshift %lu, t=%lu\n", scp->nsp_rxtshift, t); */
|
|
|
|
|
|
|
|
return t;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This is called each time we get an estimate for the rtt
|
|
|
|
* on the link.
|
|
|
|
*/
|
|
|
|
static void dn_nsp_rtt(struct sock *sk, long rtt)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
long srtt = (long)scp->nsp_srtt;
|
|
|
|
long rttvar = (long)scp->nsp_rttvar;
|
|
|
|
long delta;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the jiffies clock flips over in the middle of timestamp
|
|
|
|
* gathering this value might turn out negative, so we make sure
|
|
|
|
* that is it always positive here.
|
|
|
|
*/
|
2007-02-09 22:24:40 +08:00
|
|
|
if (rtt < 0)
|
2005-04-17 06:20:36 +08:00
|
|
|
rtt = -rtt;
|
|
|
|
/*
|
|
|
|
* Add new rtt to smoothed average
|
|
|
|
*/
|
|
|
|
delta = ((rtt << 3) - srtt);
|
|
|
|
srtt += (delta >> 3);
|
2007-02-09 22:24:40 +08:00
|
|
|
if (srtt >= 1)
|
2005-04-17 06:20:36 +08:00
|
|
|
scp->nsp_srtt = (unsigned long)srtt;
|
|
|
|
else
|
|
|
|
scp->nsp_srtt = 1;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Add new rtt varience to smoothed varience
|
|
|
|
*/
|
|
|
|
delta >>= 1;
|
|
|
|
rttvar += ((((delta>0)?(delta):(-delta)) - rttvar) >> 2);
|
2007-02-09 22:24:40 +08:00
|
|
|
if (rttvar >= 1)
|
2005-04-17 06:20:36 +08:00
|
|
|
scp->nsp_rttvar = (unsigned long)rttvar;
|
|
|
|
else
|
|
|
|
scp->nsp_rttvar = 1;
|
|
|
|
|
|
|
|
/* printk(KERN_DEBUG "srtt=%lu rttvar=%lu\n", scp->nsp_srtt, scp->nsp_rttvar); */
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* dn_nsp_clone_and_send - Send a data packet by cloning it
|
|
|
|
* @skb: The packet to clone and transmit
|
|
|
|
* @gfp: memory allocation flag
|
|
|
|
*
|
|
|
|
* Clone a queued data or other data packet and transmit it.
|
|
|
|
*
|
|
|
|
* Returns: The number of times the packet has been sent previously
|
|
|
|
*/
|
2012-04-15 13:58:06 +08:00
|
|
|
static inline unsigned int dn_nsp_clone_and_send(struct sk_buff *skb,
|
2005-10-07 14:46:04 +08:00
|
|
|
gfp_t gfp)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct dn_skb_cb *cb = DN_SKB_CB(skb);
|
|
|
|
struct sk_buff *skb2;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
if ((skb2 = skb_clone(skb, gfp)) != NULL) {
|
|
|
|
ret = cb->xmit_count;
|
|
|
|
cb->xmit_count++;
|
|
|
|
cb->stamp = jiffies;
|
|
|
|
skb2->sk = skb->sk;
|
|
|
|
dn_nsp_send(skb2);
|
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* dn_nsp_output - Try and send something from socket queues
|
|
|
|
* @sk: The socket whose queues are to be investigated
|
|
|
|
*
|
|
|
|
* Try and send the packet on the end of the data and other data queues.
|
|
|
|
* Other data gets priority over data, and if we retransmit a packet we
|
|
|
|
* reduce the window by dividing it in two.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
void dn_nsp_output(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct sk_buff *skb;
|
2012-04-15 13:58:06 +08:00
|
|
|
unsigned int reduce_win = 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* First we check for otherdata/linkservice messages
|
|
|
|
*/
|
|
|
|
if ((skb = skb_peek(&scp->other_xmit_queue)) != NULL)
|
|
|
|
reduce_win = dn_nsp_clone_and_send(skb, GFP_ATOMIC);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If we may not send any data, we don't.
|
|
|
|
* If we are still trying to get some other data down the
|
|
|
|
* channel, we don't try and send any data.
|
|
|
|
*/
|
|
|
|
if (reduce_win || (scp->flowrem_sw != DN_SEND))
|
|
|
|
goto recalc_window;
|
|
|
|
|
|
|
|
if ((skb = skb_peek(&scp->data_xmit_queue)) != NULL)
|
|
|
|
reduce_win = dn_nsp_clone_and_send(skb, GFP_ATOMIC);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If we've sent any frame more than once, we cut the
|
|
|
|
* send window size in half. There is always a minimum
|
|
|
|
* window size of one available.
|
|
|
|
*/
|
|
|
|
recalc_window:
|
|
|
|
if (reduce_win) {
|
|
|
|
scp->snd_window >>= 1;
|
|
|
|
if (scp->snd_window < NSP_MIN_WINDOW)
|
|
|
|
scp->snd_window = NSP_MIN_WINDOW;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
int dn_nsp_xmit_timeout(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
|
|
|
|
dn_nsp_output(sk);
|
|
|
|
|
2005-07-09 05:57:23 +08:00
|
|
|
if (!skb_queue_empty(&scp->data_xmit_queue) ||
|
|
|
|
!skb_queue_empty(&scp->other_xmit_queue))
|
2005-04-17 06:20:36 +08:00
|
|
|
scp->persist = dn_nsp_persist(sk);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2006-03-21 14:42:39 +08:00
|
|
|
static inline __le16 *dn_mk_common_header(struct dn_scp *scp, struct sk_buff *skb, unsigned char msgflag, int len)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
unsigned char *ptr = skb_push(skb, len);
|
|
|
|
|
|
|
|
BUG_ON(len < 5);
|
|
|
|
|
|
|
|
*ptr++ = msgflag;
|
2006-03-21 14:42:39 +08:00
|
|
|
*((__le16 *)ptr) = scp->addrrem;
|
2005-04-17 06:20:36 +08:00
|
|
|
ptr += 2;
|
2006-03-21 14:42:39 +08:00
|
|
|
*((__le16 *)ptr) = scp->addrloc;
|
2005-04-17 06:20:36 +08:00
|
|
|
ptr += 2;
|
2006-03-21 14:42:39 +08:00
|
|
|
return (__le16 __force *)ptr;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2006-03-21 14:42:39 +08:00
|
|
|
static __le16 *dn_mk_ack_header(struct sock *sk, struct sk_buff *skb, unsigned char msgflag, int hlen, int other)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
unsigned short acknum = scp->numdat_rcv & 0x0FFF;
|
|
|
|
unsigned short ackcrs = scp->numoth_rcv & 0x0FFF;
|
2006-03-21 14:42:39 +08:00
|
|
|
__le16 *ptr;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
BUG_ON(hlen < 9);
|
|
|
|
|
|
|
|
scp->ackxmt_dat = acknum;
|
|
|
|
scp->ackxmt_oth = ackcrs;
|
|
|
|
acknum |= 0x8000;
|
|
|
|
ackcrs |= 0x8000;
|
|
|
|
|
|
|
|
/* If this is an "other data/ack" message, swap acknum and ackcrs */
|
2017-10-29 04:39:48 +08:00
|
|
|
if (other)
|
|
|
|
swap(acknum, ackcrs);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
/* Set "cross subchannel" bit in ackcrs */
|
|
|
|
ackcrs |= 0x2000;
|
|
|
|
|
2012-06-04 01:41:40 +08:00
|
|
|
ptr = dn_mk_common_header(scp, skb, msgflag, hlen);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-11-27 16:12:47 +08:00
|
|
|
*ptr++ = cpu_to_le16(acknum);
|
|
|
|
*ptr++ = cpu_to_le16(ackcrs);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
return ptr;
|
|
|
|
}
|
|
|
|
|
2006-03-21 14:42:39 +08:00
|
|
|
static __le16 *dn_nsp_mk_data_header(struct sock *sk, struct sk_buff *skb, int oth)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct dn_skb_cb *cb = DN_SKB_CB(skb);
|
2006-03-21 14:42:39 +08:00
|
|
|
__le16 *ptr = dn_mk_ack_header(sk, skb, cb->nsp_flags, 11, oth);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (unlikely(oth)) {
|
|
|
|
cb->segnum = scp->numoth;
|
|
|
|
seq_add(&scp->numoth, 1);
|
|
|
|
} else {
|
|
|
|
cb->segnum = scp->numdat;
|
|
|
|
seq_add(&scp->numdat, 1);
|
|
|
|
}
|
2008-11-27 16:12:47 +08:00
|
|
|
*(ptr++) = cpu_to_le16(cb->segnum);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
return ptr;
|
|
|
|
}
|
|
|
|
|
2005-10-05 13:41:48 +08:00
|
|
|
void dn_nsp_queue_xmit(struct sock *sk, struct sk_buff *skb,
|
2005-10-07 14:46:04 +08:00
|
|
|
gfp_t gfp, int oth)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct dn_skb_cb *cb = DN_SKB_CB(skb);
|
|
|
|
unsigned long t = ((scp->nsp_srtt >> 2) + scp->nsp_rttvar) >> 1;
|
|
|
|
|
|
|
|
cb->xmit_count = 0;
|
|
|
|
dn_nsp_mk_data_header(sk, skb, oth);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Slow start: If we have been idle for more than
|
|
|
|
* one RTT, then reset window to min size.
|
|
|
|
*/
|
|
|
|
if ((jiffies - scp->stamp) > t)
|
|
|
|
scp->snd_window = NSP_MIN_WINDOW;
|
|
|
|
|
|
|
|
if (oth)
|
|
|
|
skb_queue_tail(&scp->other_xmit_queue, skb);
|
|
|
|
else
|
|
|
|
skb_queue_tail(&scp->data_xmit_queue, skb);
|
|
|
|
|
|
|
|
if (scp->flowrem_sw != DN_SEND)
|
|
|
|
return;
|
|
|
|
|
|
|
|
dn_nsp_clone_and_send(skb, gfp);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
int dn_nsp_check_xmit_queue(struct sock *sk, struct sk_buff *skb, struct sk_buff_head *q, unsigned short acknum)
|
|
|
|
{
|
|
|
|
struct dn_skb_cb *cb = DN_SKB_CB(skb);
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
2009-05-29 07:43:52 +08:00
|
|
|
struct sk_buff *skb2, *n, *ack = NULL;
|
2005-04-17 06:20:36 +08:00
|
|
|
int wakeup = 0;
|
|
|
|
int try_retrans = 0;
|
|
|
|
unsigned long reftime = cb->stamp;
|
|
|
|
unsigned long pkttime;
|
|
|
|
unsigned short xmit_count;
|
|
|
|
unsigned short segnum;
|
|
|
|
|
2009-05-29 07:43:52 +08:00
|
|
|
skb_queue_walk_safe(q, skb2, n) {
|
2005-04-17 06:20:36 +08:00
|
|
|
struct dn_skb_cb *cb2 = DN_SKB_CB(skb2);
|
|
|
|
|
|
|
|
if (dn_before_or_equal(cb2->segnum, acknum))
|
|
|
|
ack = skb2;
|
|
|
|
|
|
|
|
/* printk(KERN_DEBUG "ack: %s %04x %04x\n", ack ? "ACK" : "SKIP", (int)cb2->segnum, (int)acknum); */
|
|
|
|
|
|
|
|
if (ack == NULL)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
/* printk(KERN_DEBUG "check_xmit_queue: %04x, %d\n", acknum, cb2->xmit_count); */
|
|
|
|
|
|
|
|
/* Does _last_ packet acked have xmit_count > 1 */
|
|
|
|
try_retrans = 0;
|
|
|
|
/* Remember to wake up the sending process */
|
|
|
|
wakeup = 1;
|
|
|
|
/* Keep various statistics */
|
|
|
|
pkttime = cb2->stamp;
|
|
|
|
xmit_count = cb2->xmit_count;
|
|
|
|
segnum = cb2->segnum;
|
|
|
|
/* Remove and drop ack'ed packet */
|
2005-08-10 10:25:21 +08:00
|
|
|
skb_unlink(ack, q);
|
2005-04-17 06:20:36 +08:00
|
|
|
kfree_skb(ack);
|
|
|
|
ack = NULL;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We don't expect to see acknowledgements for packets we
|
|
|
|
* haven't sent yet.
|
|
|
|
*/
|
|
|
|
WARN_ON(xmit_count == 0);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the packet has only been sent once, we can use it
|
|
|
|
* to calculate the RTT and also open the window a little
|
|
|
|
* further.
|
|
|
|
*/
|
|
|
|
if (xmit_count == 1) {
|
2007-02-09 22:24:40 +08:00
|
|
|
if (dn_equal(segnum, acknum))
|
2005-04-17 06:20:36 +08:00
|
|
|
dn_nsp_rtt(sk, (long)(pkttime - reftime));
|
|
|
|
|
|
|
|
if (scp->snd_window < scp->max_window)
|
|
|
|
scp->snd_window++;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Packet has been sent more than once. If this is the last
|
|
|
|
* packet to be acknowledged then we want to send the next
|
|
|
|
* packet in the send queue again (assumes the remote host does
|
|
|
|
* go-back-N error control).
|
|
|
|
*/
|
|
|
|
if (xmit_count > 1)
|
|
|
|
try_retrans = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (try_retrans)
|
|
|
|
dn_nsp_output(sk);
|
|
|
|
|
|
|
|
return wakeup;
|
|
|
|
}
|
|
|
|
|
|
|
|
void dn_nsp_send_data_ack(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct sk_buff *skb = NULL;
|
|
|
|
|
|
|
|
if ((skb = dn_alloc_skb(sk, 9, GFP_ATOMIC)) == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
skb_reserve(skb, 9);
|
|
|
|
dn_mk_ack_header(sk, skb, 0x04, 9, 0);
|
|
|
|
dn_nsp_send(skb);
|
|
|
|
}
|
|
|
|
|
|
|
|
void dn_nsp_send_oth_ack(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct sk_buff *skb = NULL;
|
|
|
|
|
|
|
|
if ((skb = dn_alloc_skb(sk, 9, GFP_ATOMIC)) == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
skb_reserve(skb, 9);
|
|
|
|
dn_mk_ack_header(sk, skb, 0x14, 9, 1);
|
|
|
|
dn_nsp_send(skb);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void dn_send_conn_ack (struct sock *sk)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct sk_buff *skb = NULL;
|
2007-02-09 22:24:40 +08:00
|
|
|
struct nsp_conn_ack_msg *msg;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if ((skb = dn_alloc_skb(sk, 3, sk->sk_allocation)) == NULL)
|
|
|
|
return;
|
|
|
|
|
networking: make skb_put & friends return void pointers
It seems like a historic accident that these return unsigned char *,
and in many places that means casts are required, more often than not.
Make these functions (skb_put, __skb_put and pskb_put) return void *
and remove all the casts across the tree, adding a (u8 *) cast only
where the unsigned char pointer was used directly, all done with the
following spatch:
@@
expression SKB, LEN;
typedef u8;
identifier fn = { skb_put, __skb_put };
@@
- *(fn(SKB, LEN))
+ *(u8 *)fn(SKB, LEN)
@@
expression E, SKB, LEN;
identifier fn = { skb_put, __skb_put };
type T;
@@
- E = ((T *)(fn(SKB, LEN)))
+ E = fn(SKB, LEN)
which actually doesn't cover pskb_put since there are only three
users overall.
A handful of stragglers were converted manually, notably a macro in
drivers/isdn/i4l/isdn_bsdcomp.c and, oddly enough, one of the many
instances in net/bluetooth/hci_sock.c. In the former file, I also
had to fix one whitespace problem spatch introduced.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:21 +08:00
|
|
|
msg = skb_put(skb, 3);
|
2007-02-09 22:24:40 +08:00
|
|
|
msg->msgflg = 0x24;
|
2005-04-17 06:20:36 +08:00
|
|
|
msg->dstaddr = scp->addrrem;
|
|
|
|
|
2007-02-09 22:24:40 +08:00
|
|
|
dn_nsp_send(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static int dn_nsp_retrans_conn_conf(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
|
|
|
|
if (scp->state == DN_CC)
|
|
|
|
dn_send_conn_conf(sk, GFP_ATOMIC);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-10-07 14:46:04 +08:00
|
|
|
void dn_send_conn_conf(struct sock *sk, gfp_t gfp)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct sk_buff *skb = NULL;
|
2007-02-09 22:24:40 +08:00
|
|
|
struct nsp_conn_init_msg *msg;
|
2008-11-27 16:12:47 +08:00
|
|
|
__u8 len = (__u8)le16_to_cpu(scp->conndata_out.opt_optl);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2006-11-08 07:09:17 +08:00
|
|
|
if ((skb = dn_alloc_skb(sk, 50 + len, gfp)) == NULL)
|
2005-04-17 06:20:36 +08:00
|
|
|
return;
|
|
|
|
|
networking: make skb_put & friends return void pointers
It seems like a historic accident that these return unsigned char *,
and in many places that means casts are required, more often than not.
Make these functions (skb_put, __skb_put and pskb_put) return void *
and remove all the casts across the tree, adding a (u8 *) cast only
where the unsigned char pointer was used directly, all done with the
following spatch:
@@
expression SKB, LEN;
typedef u8;
identifier fn = { skb_put, __skb_put };
@@
- *(fn(SKB, LEN))
+ *(u8 *)fn(SKB, LEN)
@@
expression E, SKB, LEN;
identifier fn = { skb_put, __skb_put };
type T;
@@
- E = ((T *)(fn(SKB, LEN)))
+ E = fn(SKB, LEN)
which actually doesn't cover pskb_put since there are only three
users overall.
A handful of stragglers were converted manually, notably a macro in
drivers/isdn/i4l/isdn_bsdcomp.c and, oddly enough, one of the many
instances in net/bluetooth/hci_sock.c. In the former file, I also
had to fix one whitespace problem spatch introduced.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:21 +08:00
|
|
|
msg = skb_put(skb, sizeof(*msg));
|
2007-02-09 22:24:40 +08:00
|
|
|
msg->msgflg = 0x28;
|
2005-04-17 06:20:36 +08:00
|
|
|
msg->dstaddr = scp->addrrem;
|
2007-02-09 22:24:40 +08:00
|
|
|
msg->srcaddr = scp->addrloc;
|
|
|
|
msg->services = scp->services_loc;
|
|
|
|
msg->info = scp->info_loc;
|
2008-11-27 16:12:47 +08:00
|
|
|
msg->segsize = cpu_to_le16(scp->segsize_loc);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
networking: add and use skb_put_u8()
Joe and Bjørn suggested that it'd be nicer to not have the
cast in the fairly common case of doing
*(u8 *)skb_put(skb, 1) = c;
Add skb_put_u8() for this case, and use it across the code,
using the following spatch:
@@
expression SKB, C, S;
typedef u8;
identifier fn = {skb_put};
fresh identifier fn2 = fn ## "_u8";
@@
- *(u8 *)fn(SKB, S) = C;
+ fn2(SKB, C);
Note that due to the "S", the spatch isn't perfect, it should
have checked that S is 1, but there's also places that use a
sizeof expression like sizeof(var) or sizeof(u8) etc. Turns
out that nobody ever did something like
*(u8 *)skb_put(skb, 2) = c;
which would be wrong anyway since the second byte wouldn't be
initialized.
Suggested-by: Joe Perches <joe@perches.com>
Suggested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:24 +08:00
|
|
|
skb_put_u8(skb, len);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-02-09 22:24:40 +08:00
|
|
|
if (len > 0)
|
networking: introduce and use skb_put_data()
A common pattern with skb_put() is to just want to memcpy()
some data into the new space, introduce skb_put_data() for
this.
An spatch similar to the one for skb_put_zero() converts many
of the places using it:
@@
identifier p, p2;
expression len, skb, data;
type t, t2;
@@
(
-p = skb_put(skb, len);
+p = skb_put_data(skb, data, len);
|
-p = (t)skb_put(skb, len);
+p = skb_put_data(skb, data, len);
)
(
p2 = (t2)p;
-memcpy(p2, data, len);
|
-memcpy(p, data, len);
)
@@
type t, t2;
identifier p, p2;
expression skb, data;
@@
t *p;
...
(
-p = skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
|
-p = (t *)skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
)
(
p2 = (t2)p;
-memcpy(p2, data, sizeof(*p));
|
-memcpy(p, data, sizeof(*p));
)
@@
expression skb, len, data;
@@
-memcpy(skb_put(skb, len), data, len);
+skb_put_data(skb, data, len);
(again, manually post-processed to retain some comments)
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:20 +08:00
|
|
|
skb_put_data(skb, scp->conndata_out.opt_data, len);
|
2007-02-09 22:24:40 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
dn_nsp_send(skb);
|
|
|
|
|
|
|
|
scp->persist = dn_nsp_persist(sk);
|
|
|
|
scp->persist_fxn = dn_nsp_retrans_conn_conf;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-02-09 22:24:40 +08:00
|
|
|
static __inline__ void dn_nsp_do_disc(struct sock *sk, unsigned char msgflg,
|
2005-10-07 14:46:04 +08:00
|
|
|
unsigned short reason, gfp_t gfp,
|
2005-10-05 13:41:48 +08:00
|
|
|
struct dst_entry *dst,
|
2006-03-21 14:42:39 +08:00
|
|
|
int ddl, unsigned char *dd, __le16 rem, __le16 loc)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct sk_buff *skb = NULL;
|
|
|
|
int size = 7 + ddl + ((msgflg == NSP_DISCINIT) ? 1 : 0);
|
|
|
|
unsigned char *msg;
|
|
|
|
|
|
|
|
if ((dst == NULL) || (rem == 0)) {
|
2012-05-14 05:56:26 +08:00
|
|
|
net_dbg_ratelimited("DECnet: dn_nsp_do_disc: BUG! Please report this to SteveW@ACM.org rem=%u dst=%p\n",
|
|
|
|
le16_to_cpu(rem), dst);
|
2005-04-17 06:20:36 +08:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((skb = dn_alloc_skb(sk, size, gfp)) == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
msg = skb_put(skb, size);
|
|
|
|
*msg++ = msgflg;
|
2006-03-21 14:42:39 +08:00
|
|
|
*(__le16 *)msg = rem;
|
2005-04-17 06:20:36 +08:00
|
|
|
msg += 2;
|
2006-03-21 14:42:39 +08:00
|
|
|
*(__le16 *)msg = loc;
|
2005-04-17 06:20:36 +08:00
|
|
|
msg += 2;
|
2008-11-27 16:12:47 +08:00
|
|
|
*(__le16 *)msg = cpu_to_le16(reason);
|
2005-04-17 06:20:36 +08:00
|
|
|
msg += 2;
|
|
|
|
if (msgflg == NSP_DISCINIT)
|
|
|
|
*msg++ = ddl;
|
|
|
|
|
|
|
|
if (ddl) {
|
|
|
|
memcpy(msg, dd, ddl);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This doesn't go via the dn_nsp_send() function since we need
|
|
|
|
* to be able to send disc packets out which have no socket
|
|
|
|
* associations.
|
|
|
|
*/
|
2009-06-02 13:19:30 +08:00
|
|
|
skb_dst_set(skb, dst_clone(dst));
|
2015-10-08 05:48:35 +08:00
|
|
|
dst_output(&init_net, skb->sk, skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-02-09 22:24:40 +08:00
|
|
|
void dn_nsp_send_disc(struct sock *sk, unsigned char msgflg,
|
2005-10-07 14:46:04 +08:00
|
|
|
unsigned short reason, gfp_t gfp)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
int ddl = 0;
|
|
|
|
|
|
|
|
if (msgflg == NSP_DISCINIT)
|
2008-11-27 16:12:47 +08:00
|
|
|
ddl = le16_to_cpu(scp->discdata_out.opt_optl);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (reason == 0)
|
2008-11-27 16:12:47 +08:00
|
|
|
reason = le16_to_cpu(scp->discdata_out.opt_status);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2013-01-23 05:09:50 +08:00
|
|
|
dn_nsp_do_disc(sk, msgflg, reason, gfp, __sk_dst_get(sk), ddl,
|
2005-04-17 06:20:36 +08:00
|
|
|
scp->discdata_out.opt_data, scp->addrrem, scp->addrloc);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-02-09 22:24:40 +08:00
|
|
|
void dn_nsp_return_disc(struct sk_buff *skb, unsigned char msgflg,
|
2005-04-17 06:20:36 +08:00
|
|
|
unsigned short reason)
|
|
|
|
{
|
|
|
|
struct dn_skb_cb *cb = DN_SKB_CB(skb);
|
|
|
|
int ddl = 0;
|
2005-10-07 14:46:04 +08:00
|
|
|
gfp_t gfp = GFP_ATOMIC;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-06-02 13:19:30 +08:00
|
|
|
dn_nsp_do_disc(NULL, msgflg, reason, gfp, skb_dst(skb), ddl,
|
2005-04-17 06:20:36 +08:00
|
|
|
NULL, cb->src_port, cb->dst_port);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void dn_nsp_send_link(struct sock *sk, unsigned char lsflags, char fcval)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct sk_buff *skb;
|
|
|
|
unsigned char *ptr;
|
2005-10-07 14:46:04 +08:00
|
|
|
gfp_t gfp = GFP_ATOMIC;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if ((skb = dn_alloc_skb(sk, DN_MAX_NSP_DATA_HEADER + 2, gfp)) == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
skb_reserve(skb, DN_MAX_NSP_DATA_HEADER);
|
|
|
|
ptr = skb_put(skb, 2);
|
|
|
|
DN_SKB_CB(skb)->nsp_flags = 0x10;
|
|
|
|
*ptr++ = lsflags;
|
|
|
|
*ptr = fcval;
|
|
|
|
|
|
|
|
dn_nsp_queue_xmit(sk, skb, gfp, 1);
|
|
|
|
|
|
|
|
scp->persist = dn_nsp_persist(sk);
|
|
|
|
scp->persist_fxn = dn_nsp_xmit_timeout;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int dn_nsp_retrans_conninit(struct sock *sk)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
|
|
|
|
if (scp->state == DN_CI)
|
|
|
|
dn_nsp_send_conninit(sk, NSP_RCI);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void dn_nsp_send_conninit(struct sock *sk, unsigned char msgflg)
|
|
|
|
{
|
|
|
|
struct dn_scp *scp = DN_SK(sk);
|
|
|
|
struct nsp_conn_init_msg *msg;
|
|
|
|
unsigned char aux;
|
|
|
|
unsigned char menuver;
|
|
|
|
struct dn_skb_cb *cb;
|
|
|
|
unsigned char type = 1;
|
2005-10-07 14:46:04 +08:00
|
|
|
gfp_t allocation = (msgflg == NSP_CI) ? sk->sk_allocation : GFP_ATOMIC;
|
2005-04-17 06:20:36 +08:00
|
|
|
struct sk_buff *skb = dn_alloc_skb(sk, 200, allocation);
|
|
|
|
|
|
|
|
if (!skb)
|
|
|
|
return;
|
|
|
|
|
|
|
|
cb = DN_SKB_CB(skb);
|
networking: make skb_put & friends return void pointers
It seems like a historic accident that these return unsigned char *,
and in many places that means casts are required, more often than not.
Make these functions (skb_put, __skb_put and pskb_put) return void *
and remove all the casts across the tree, adding a (u8 *) cast only
where the unsigned char pointer was used directly, all done with the
following spatch:
@@
expression SKB, LEN;
typedef u8;
identifier fn = { skb_put, __skb_put };
@@
- *(fn(SKB, LEN))
+ *(u8 *)fn(SKB, LEN)
@@
expression E, SKB, LEN;
identifier fn = { skb_put, __skb_put };
type T;
@@
- E = ((T *)(fn(SKB, LEN)))
+ E = fn(SKB, LEN)
which actually doesn't cover pskb_put since there are only three
users overall.
A handful of stragglers were converted manually, notably a macro in
drivers/isdn/i4l/isdn_bsdcomp.c and, oddly enough, one of the many
instances in net/bluetooth/hci_sock.c. In the former file, I also
had to fix one whitespace problem spatch introduced.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:21 +08:00
|
|
|
msg = skb_put(skb, sizeof(*msg));
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
msg->msgflg = msgflg;
|
|
|
|
msg->dstaddr = 0x0000; /* Remote Node will assign it*/
|
|
|
|
|
|
|
|
msg->srcaddr = scp->addrloc;
|
|
|
|
msg->services = scp->services_loc; /* Requested flow control */
|
2007-02-09 22:24:40 +08:00
|
|
|
msg->info = scp->info_loc; /* Version Number */
|
2008-11-27 16:12:47 +08:00
|
|
|
msg->segsize = cpu_to_le16(scp->segsize_loc); /* Max segment size */
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (scp->peer.sdn_objnum)
|
|
|
|
type = 0;
|
|
|
|
|
2007-04-20 11:29:13 +08:00
|
|
|
skb_put(skb, dn_sockaddr2username(&scp->peer,
|
|
|
|
skb_tail_pointer(skb), type));
|
|
|
|
skb_put(skb, dn_sockaddr2username(&scp->addr,
|
|
|
|
skb_tail_pointer(skb), 2));
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
menuver = DN_MENUVER_ACC | DN_MENUVER_USR;
|
|
|
|
if (scp->peer.sdn_flags & SDF_PROXY)
|
|
|
|
menuver |= DN_MENUVER_PRX;
|
|
|
|
if (scp->peer.sdn_flags & SDF_UICPROXY)
|
|
|
|
menuver |= DN_MENUVER_UIC;
|
|
|
|
|
networking: add and use skb_put_u8()
Joe and Bjørn suggested that it'd be nicer to not have the
cast in the fairly common case of doing
*(u8 *)skb_put(skb, 1) = c;
Add skb_put_u8() for this case, and use it across the code,
using the following spatch:
@@
expression SKB, C, S;
typedef u8;
identifier fn = {skb_put};
fresh identifier fn2 = fn ## "_u8";
@@
- *(u8 *)fn(SKB, S) = C;
+ fn2(SKB, C);
Note that due to the "S", the spatch isn't perfect, it should
have checked that S is 1, but there's also places that use a
sizeof expression like sizeof(var) or sizeof(u8) etc. Turns
out that nobody ever did something like
*(u8 *)skb_put(skb, 2) = c;
which would be wrong anyway since the second byte wouldn't be
initialized.
Suggested-by: Joe Perches <joe@perches.com>
Suggested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:24 +08:00
|
|
|
skb_put_u8(skb, menuver); /* Menu Version */
|
2007-02-09 22:24:40 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
aux = scp->accessdata.acc_userl;
|
networking: add and use skb_put_u8()
Joe and Bjørn suggested that it'd be nicer to not have the
cast in the fairly common case of doing
*(u8 *)skb_put(skb, 1) = c;
Add skb_put_u8() for this case, and use it across the code,
using the following spatch:
@@
expression SKB, C, S;
typedef u8;
identifier fn = {skb_put};
fresh identifier fn2 = fn ## "_u8";
@@
- *(u8 *)fn(SKB, S) = C;
+ fn2(SKB, C);
Note that due to the "S", the spatch isn't perfect, it should
have checked that S is 1, but there's also places that use a
sizeof expression like sizeof(var) or sizeof(u8) etc. Turns
out that nobody ever did something like
*(u8 *)skb_put(skb, 2) = c;
which would be wrong anyway since the second byte wouldn't be
initialized.
Suggested-by: Joe Perches <joe@perches.com>
Suggested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:24 +08:00
|
|
|
skb_put_u8(skb, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (aux > 0)
|
networking: introduce and use skb_put_data()
A common pattern with skb_put() is to just want to memcpy()
some data into the new space, introduce skb_put_data() for
this.
An spatch similar to the one for skb_put_zero() converts many
of the places using it:
@@
identifier p, p2;
expression len, skb, data;
type t, t2;
@@
(
-p = skb_put(skb, len);
+p = skb_put_data(skb, data, len);
|
-p = (t)skb_put(skb, len);
+p = skb_put_data(skb, data, len);
)
(
p2 = (t2)p;
-memcpy(p2, data, len);
|
-memcpy(p, data, len);
)
@@
type t, t2;
identifier p, p2;
expression skb, data;
@@
t *p;
...
(
-p = skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
|
-p = (t *)skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
)
(
p2 = (t2)p;
-memcpy(p2, data, sizeof(*p));
|
-memcpy(p, data, sizeof(*p));
)
@@
expression skb, len, data;
@@
-memcpy(skb_put(skb, len), data, len);
+skb_put_data(skb, data, len);
(again, manually post-processed to retain some comments)
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:20 +08:00
|
|
|
skb_put_data(skb, scp->accessdata.acc_user, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
aux = scp->accessdata.acc_passl;
|
networking: add and use skb_put_u8()
Joe and Bjørn suggested that it'd be nicer to not have the
cast in the fairly common case of doing
*(u8 *)skb_put(skb, 1) = c;
Add skb_put_u8() for this case, and use it across the code,
using the following spatch:
@@
expression SKB, C, S;
typedef u8;
identifier fn = {skb_put};
fresh identifier fn2 = fn ## "_u8";
@@
- *(u8 *)fn(SKB, S) = C;
+ fn2(SKB, C);
Note that due to the "S", the spatch isn't perfect, it should
have checked that S is 1, but there's also places that use a
sizeof expression like sizeof(var) or sizeof(u8) etc. Turns
out that nobody ever did something like
*(u8 *)skb_put(skb, 2) = c;
which would be wrong anyway since the second byte wouldn't be
initialized.
Suggested-by: Joe Perches <joe@perches.com>
Suggested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:24 +08:00
|
|
|
skb_put_u8(skb, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (aux > 0)
|
networking: introduce and use skb_put_data()
A common pattern with skb_put() is to just want to memcpy()
some data into the new space, introduce skb_put_data() for
this.
An spatch similar to the one for skb_put_zero() converts many
of the places using it:
@@
identifier p, p2;
expression len, skb, data;
type t, t2;
@@
(
-p = skb_put(skb, len);
+p = skb_put_data(skb, data, len);
|
-p = (t)skb_put(skb, len);
+p = skb_put_data(skb, data, len);
)
(
p2 = (t2)p;
-memcpy(p2, data, len);
|
-memcpy(p, data, len);
)
@@
type t, t2;
identifier p, p2;
expression skb, data;
@@
t *p;
...
(
-p = skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
|
-p = (t *)skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
)
(
p2 = (t2)p;
-memcpy(p2, data, sizeof(*p));
|
-memcpy(p, data, sizeof(*p));
)
@@
expression skb, len, data;
@@
-memcpy(skb_put(skb, len), data, len);
+skb_put_data(skb, data, len);
(again, manually post-processed to retain some comments)
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:20 +08:00
|
|
|
skb_put_data(skb, scp->accessdata.acc_pass, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
aux = scp->accessdata.acc_accl;
|
networking: add and use skb_put_u8()
Joe and Bjørn suggested that it'd be nicer to not have the
cast in the fairly common case of doing
*(u8 *)skb_put(skb, 1) = c;
Add skb_put_u8() for this case, and use it across the code,
using the following spatch:
@@
expression SKB, C, S;
typedef u8;
identifier fn = {skb_put};
fresh identifier fn2 = fn ## "_u8";
@@
- *(u8 *)fn(SKB, S) = C;
+ fn2(SKB, C);
Note that due to the "S", the spatch isn't perfect, it should
have checked that S is 1, but there's also places that use a
sizeof expression like sizeof(var) or sizeof(u8) etc. Turns
out that nobody ever did something like
*(u8 *)skb_put(skb, 2) = c;
which would be wrong anyway since the second byte wouldn't be
initialized.
Suggested-by: Joe Perches <joe@perches.com>
Suggested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:24 +08:00
|
|
|
skb_put_u8(skb, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (aux > 0)
|
networking: introduce and use skb_put_data()
A common pattern with skb_put() is to just want to memcpy()
some data into the new space, introduce skb_put_data() for
this.
An spatch similar to the one for skb_put_zero() converts many
of the places using it:
@@
identifier p, p2;
expression len, skb, data;
type t, t2;
@@
(
-p = skb_put(skb, len);
+p = skb_put_data(skb, data, len);
|
-p = (t)skb_put(skb, len);
+p = skb_put_data(skb, data, len);
)
(
p2 = (t2)p;
-memcpy(p2, data, len);
|
-memcpy(p, data, len);
)
@@
type t, t2;
identifier p, p2;
expression skb, data;
@@
t *p;
...
(
-p = skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
|
-p = (t *)skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
)
(
p2 = (t2)p;
-memcpy(p2, data, sizeof(*p));
|
-memcpy(p, data, sizeof(*p));
)
@@
expression skb, len, data;
@@
-memcpy(skb_put(skb, len), data, len);
+skb_put_data(skb, data, len);
(again, manually post-processed to retain some comments)
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:20 +08:00
|
|
|
skb_put_data(skb, scp->accessdata.acc_acc, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-11-27 16:12:47 +08:00
|
|
|
aux = (__u8)le16_to_cpu(scp->conndata_out.opt_optl);
|
networking: add and use skb_put_u8()
Joe and Bjørn suggested that it'd be nicer to not have the
cast in the fairly common case of doing
*(u8 *)skb_put(skb, 1) = c;
Add skb_put_u8() for this case, and use it across the code,
using the following spatch:
@@
expression SKB, C, S;
typedef u8;
identifier fn = {skb_put};
fresh identifier fn2 = fn ## "_u8";
@@
- *(u8 *)fn(SKB, S) = C;
+ fn2(SKB, C);
Note that due to the "S", the spatch isn't perfect, it should
have checked that S is 1, but there's also places that use a
sizeof expression like sizeof(var) or sizeof(u8) etc. Turns
out that nobody ever did something like
*(u8 *)skb_put(skb, 2) = c;
which would be wrong anyway since the second byte wouldn't be
initialized.
Suggested-by: Joe Perches <joe@perches.com>
Suggested-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:24 +08:00
|
|
|
skb_put_u8(skb, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (aux > 0)
|
networking: introduce and use skb_put_data()
A common pattern with skb_put() is to just want to memcpy()
some data into the new space, introduce skb_put_data() for
this.
An spatch similar to the one for skb_put_zero() converts many
of the places using it:
@@
identifier p, p2;
expression len, skb, data;
type t, t2;
@@
(
-p = skb_put(skb, len);
+p = skb_put_data(skb, data, len);
|
-p = (t)skb_put(skb, len);
+p = skb_put_data(skb, data, len);
)
(
p2 = (t2)p;
-memcpy(p2, data, len);
|
-memcpy(p, data, len);
)
@@
type t, t2;
identifier p, p2;
expression skb, data;
@@
t *p;
...
(
-p = skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
|
-p = (t *)skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
)
(
p2 = (t2)p;
-memcpy(p2, data, sizeof(*p));
|
-memcpy(p, data, sizeof(*p));
)
@@
expression skb, len, data;
@@
-memcpy(skb_put(skb, len), data, len);
+skb_put_data(skb, data, len);
(again, manually post-processed to retain some comments)
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-16 20:29:20 +08:00
|
|
|
skb_put_data(skb, scp->conndata_out.opt_data, aux);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
scp->persist = dn_nsp_persist(sk);
|
|
|
|
scp->persist_fxn = dn_nsp_retrans_conninit;
|
|
|
|
|
|
|
|
cb->rt_flags = DN_RT_F_RQR;
|
|
|
|
|
2007-02-09 22:24:40 +08:00
|
|
|
dn_nsp_send(skb);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|