linux_dsm_epyc7002/net/sctp/output.c

842 lines
23 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/* SCTP kernel implementation
* (C) Copyright IBM Corp. 2001, 2004
* Copyright (c) 1999-2000 Cisco, Inc.
* Copyright (c) 1999-2001 Motorola, Inc.
*
* This file is part of the SCTP kernel implementation
*
* These functions handle output processing.
*
* Please send any bug reports or fixes you make to the
* email address(es):
* lksctp developers <linux-sctp@vger.kernel.org>
*
* Written or modified by:
* La Monte H.P. Yarroll <piggy@acm.org>
* Karl Knutson <karl@athena.chicago.il.us>
* Jon Grimm <jgrimm@austin.ibm.com>
* Sridhar Samudrala <sri@us.ibm.com>
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/wait.h>
#include <linux/time.h>
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/init.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 15:04:11 +07:00
#include <linux/slab.h>
#include <net/inet_ecn.h>
#include <net/ip.h>
#include <net/icmp.h>
#include <net/net_namespace.h>
#include <linux/socket.h> /* for sa_family_t */
#include <net/sock.h>
#include <net/sctp/sctp.h>
#include <net/sctp/sm.h>
#include <net/sctp/checksum.h>
/* Forward declarations for private helpers. */
static enum sctp_xmit __sctp_packet_append_chunk(struct sctp_packet *packet,
struct sctp_chunk *chunk);
static enum sctp_xmit sctp_packet_can_append_data(struct sctp_packet *packet,
struct sctp_chunk *chunk);
static void sctp_packet_append_data(struct sctp_packet *packet,
struct sctp_chunk *chunk);
static enum sctp_xmit sctp_packet_will_fit(struct sctp_packet *packet,
struct sctp_chunk *chunk,
u16 chunk_len);
static void sctp_packet_reset(struct sctp_packet *packet)
{
/* sctp_packet_transmit() relies on this to reset size to the
* current overhead after sending packets.
*/
packet->size = packet->overhead;
packet->has_cookie_echo = 0;
packet->has_sack = 0;
packet->has_data = 0;
packet->has_auth = 0;
packet->ipfragok = 0;
packet->auth = NULL;
}
/* Config a packet.
* This appears to be a followup set of initializations.
*/
void sctp_packet_config(struct sctp_packet *packet, __u32 vtag,
int ecn_capable)
{
struct sctp_transport *tp = packet->transport;
struct sctp_association *asoc = tp->asoc;
struct sctp_sock *sp = NULL;
struct sock *sk;
net: sctp: rework debugging framework to use pr_debug and friends We should get rid of all own SCTP debug printk macros and use the ones that the kernel offers anyway instead. This makes the code more readable and conform to the kernel code, and offers all the features of dynamic debbuging that pr_debug() et al has, such as only turning on/off portions of debug messages at runtime through debugfs. The runtime cost of having CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing, is negligible [1]. If kernel debugging is completly turned off, then these statements will also compile into "empty" functions. While we're at it, we also need to change the Kconfig option as it /now/ only refers to the ifdef'ed code portions in outqueue.c that enable further debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code was enabled with this Kconfig option and has now been removed, we transform those code parts into WARNs resp. where appropriate BUG_ONs so that those bugs can be more easily detected as probably not many people have SCTP debugging permanently turned on. To turn on all SCTP debugging, the following steps are needed: # mount -t debugfs none /sys/kernel/debug # echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control This can be done more fine-grained on a per file, per line basis and others as described in [2]. [1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf [2] Documentation/dynamic-debug-howto.txt Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-29 00:49:40 +07:00
pr_debug("%s: packet:%p vtag:0x%x\n", __func__, packet, vtag);
packet->vtag = vtag;
/* do the following jobs only once for a flush schedule */
if (!sctp_packet_empty(packet))
return;
/* set packet max_size with pathmtu, then calculate overhead */
packet->max_size = tp->pathmtu;
if (asoc) {
sk = asoc->base.sk;
sp = sctp_sk(sk);
}
packet->overhead = sctp_mtu_payload(sp, 0, 0);
packet->size = packet->overhead;
if (!asoc)
return;
/* update dst or transport pathmtu if in need */
if (!sctp_transport_dst_check(tp)) {
sctp_transport_route(tp, NULL, sp);
if (asoc->param_flags & SPP_PMTUD_ENABLE)
sctp_assoc_sync_pmtu(asoc);
} else if (!sctp_transport_pmtu_check(tp)) {
if (asoc->param_flags & SPP_PMTUD_ENABLE)
sctp_assoc_sync_pmtu(asoc);
}
if (asoc->pmtu_pending) {
if (asoc->param_flags & SPP_PMTUD_ENABLE)
sctp_assoc_sync_pmtu(asoc);
asoc->pmtu_pending = 0;
}
/* If there a is a prepend chunk stick it on the list before
* any other chunks get appended.
*/
if (ecn_capable) {
struct sctp_chunk *chunk = sctp_get_ecne_prepend(asoc);
if (chunk)
sctp_packet_append_chunk(packet, chunk);
}
if (!tp->dst)
return;
/* set packet max_size with gso_max_size if gso is enabled*/
rcu_read_lock();
if (__sk_dst_get(sk) != tp->dst) {
dst_hold(tp->dst);
sk_setup_caps(sk, tp->dst);
}
packet->max_size = sk_can_gso(sk) ? tp->dst->dev->gso_max_size
: asoc->pathmtu;
rcu_read_unlock();
}
/* Initialize the packet structure. */
void sctp_packet_init(struct sctp_packet *packet,
struct sctp_transport *transport,
__u16 sport, __u16 dport)
{
net: sctp: rework debugging framework to use pr_debug and friends We should get rid of all own SCTP debug printk macros and use the ones that the kernel offers anyway instead. This makes the code more readable and conform to the kernel code, and offers all the features of dynamic debbuging that pr_debug() et al has, such as only turning on/off portions of debug messages at runtime through debugfs. The runtime cost of having CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing, is negligible [1]. If kernel debugging is completly turned off, then these statements will also compile into "empty" functions. While we're at it, we also need to change the Kconfig option as it /now/ only refers to the ifdef'ed code portions in outqueue.c that enable further debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code was enabled with this Kconfig option and has now been removed, we transform those code parts into WARNs resp. where appropriate BUG_ONs so that those bugs can be more easily detected as probably not many people have SCTP debugging permanently turned on. To turn on all SCTP debugging, the following steps are needed: # mount -t debugfs none /sys/kernel/debug # echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control This can be done more fine-grained on a per file, per line basis and others as described in [2]. [1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf [2] Documentation/dynamic-debug-howto.txt Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-29 00:49:40 +07:00
pr_debug("%s: packet:%p transport:%p\n", __func__, packet, transport);
packet->transport = transport;
packet->source_port = sport;
packet->destination_port = dport;
INIT_LIST_HEAD(&packet->chunk_list);
/* The overhead will be calculated by sctp_packet_config() */
packet->overhead = 0;
sctp_packet_reset(packet);
packet->vtag = 0;
}
/* Free a packet. */
void sctp_packet_free(struct sctp_packet *packet)
{
struct sctp_chunk *chunk, *tmp;
net: sctp: rework debugging framework to use pr_debug and friends We should get rid of all own SCTP debug printk macros and use the ones that the kernel offers anyway instead. This makes the code more readable and conform to the kernel code, and offers all the features of dynamic debbuging that pr_debug() et al has, such as only turning on/off portions of debug messages at runtime through debugfs. The runtime cost of having CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing, is negligible [1]. If kernel debugging is completly turned off, then these statements will also compile into "empty" functions. While we're at it, we also need to change the Kconfig option as it /now/ only refers to the ifdef'ed code portions in outqueue.c that enable further debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code was enabled with this Kconfig option and has now been removed, we transform those code parts into WARNs resp. where appropriate BUG_ONs so that those bugs can be more easily detected as probably not many people have SCTP debugging permanently turned on. To turn on all SCTP debugging, the following steps are needed: # mount -t debugfs none /sys/kernel/debug # echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control This can be done more fine-grained on a per file, per line basis and others as described in [2]. [1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf [2] Documentation/dynamic-debug-howto.txt Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-29 00:49:40 +07:00
pr_debug("%s: packet:%p\n", __func__, packet);
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) {
list_del_init(&chunk->list);
sctp_chunk_free(chunk);
}
}
/* This routine tries to append the chunk to the offered packet. If adding
* the chunk causes the packet to exceed the path MTU and COOKIE_ECHO chunk
* is not present in the packet, it transmits the input packet.
* Data can be bundled with a packet containing a COOKIE_ECHO chunk as long
* as it can fit in the packet, but any more data that does not fit in this
* packet can be sent only after receiving the COOKIE_ACK.
*/
enum sctp_xmit sctp_packet_transmit_chunk(struct sctp_packet *packet,
struct sctp_chunk *chunk,
int one_packet, gfp_t gfp)
{
enum sctp_xmit retval;
pr_debug("%s: packet:%p size:%zu chunk:%p size:%d\n", __func__,
packet, packet->size, chunk, chunk->skb ? chunk->skb->len : -1);
switch ((retval = (sctp_packet_append_chunk(packet, chunk)))) {
case SCTP_XMIT_PMTU_FULL:
if (!packet->has_cookie_echo) {
int error = 0;
error = sctp_packet_transmit(packet, gfp);
if (error < 0)
chunk->skb->sk->sk_err = -error;
/* If we have an empty packet, then we can NOT ever
* return PMTU_FULL.
*/
if (!one_packet)
retval = sctp_packet_append_chunk(packet,
chunk);
}
break;
case SCTP_XMIT_RWND_FULL:
case SCTP_XMIT_OK:
case SCTP_XMIT_DELAY:
break;
}
return retval;
}
/* Try to bundle an auth chunk into the packet. */
static enum sctp_xmit sctp_packet_bundle_auth(struct sctp_packet *pkt,
struct sctp_chunk *chunk)
{
struct sctp_association *asoc = pkt->transport->asoc;
enum sctp_xmit retval = SCTP_XMIT_OK;
struct sctp_chunk *auth;
/* if we don't have an association, we can't do authentication */
if (!asoc)
return retval;
/* See if this is an auth chunk we are bundling or if
* auth is already bundled.
*/
if (chunk->chunk_hdr->type == SCTP_CID_AUTH || pkt->has_auth)
return retval;
/* if the peer did not request this chunk to be authenticated,
* don't do it
*/
if (!chunk->auth)
return retval;
auth = sctp_make_auth(asoc, chunk->shkey->key_id);
if (!auth)
return retval;
auth->shkey = chunk->shkey;
sctp_auth_shkey_hold(auth->shkey);
retval = __sctp_packet_append_chunk(pkt, auth);
if (retval != SCTP_XMIT_OK)
sctp_chunk_free(auth);
return retval;
}
/* Try to bundle a SACK with the packet. */
static enum sctp_xmit sctp_packet_bundle_sack(struct sctp_packet *pkt,
struct sctp_chunk *chunk)
{
enum sctp_xmit retval = SCTP_XMIT_OK;
/* If sending DATA and haven't aleady bundled a SACK, try to
* bundle one in to the packet.
*/
if (sctp_chunk_is_data(chunk) && !pkt->has_sack &&
!pkt->has_cookie_echo) {
struct sctp_association *asoc;
2009-07-29 23:05:57 +07:00
struct timer_list *timer;
asoc = pkt->transport->asoc;
2009-07-29 23:05:57 +07:00
timer = &asoc->timers[SCTP_EVENT_TIMEOUT_SACK];
2009-07-29 23:05:57 +07:00
/* If the SACK timer is running, we have a pending SACK */
if (timer_pending(timer)) {
struct sctp_chunk *sack;
sctp: be more restrictive in transport selection on bundled sacks It was noticed recently that when we send data on a transport, its possible that we might bundle a sack that arrived on a different transport. While this isn't a major problem, it does go against the SHOULD requirement in section 6.4 of RFC 2960: An endpoint SHOULD transmit reply chunks (e.g., SACK, HEARTBEAT ACK, etc.) to the same destination transport address from which it received the DATA or control chunk to which it is replying. This rule should also be followed if the endpoint is bundling DATA chunks together with the reply chunk. This patch seeks to correct that. It restricts the bundling of sack operations to only those transports which have moved the ctsn of the association forward since the last sack. By doing this we guarantee that we only bundle outbound saks on a transport that has received a chunk since the last sack. This brings us into stricter compliance with the RFC. Vlad had initially suggested that we strictly allow only sack bundling on the transport that last moved the ctsn forward. While this makes sense, I was concerned that doing so prevented us from bundling in the case where we had received chunks that moved the ctsn on multiple transports. In those cases, the RFC allows us to select any of the transports having received chunks to bundle the sack on. so I've modified the approach to allow for that, by adding a state variable to each transport that tracks weather it has moved the ctsn since the last sack. This I think keeps our behavior (and performance), close enough to our current profile that I think we can do this without a sysctl knob to enable/disable it. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> CC: Vlad Yaseivch <vyasevich@gmail.com> CC: David S. Miller <davem@davemloft.net> CC: linux-sctp@vger.kernel.org Reported-by: Michele Baldessari <michele@redhat.com> Reported-by: sorin serban <sserban@redhat.com> Acked-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-30 10:04:26 +07:00
if (pkt->transport->sack_generation !=
pkt->transport->asoc->peer.sack_generation)
return retval;
asoc->a_rwnd = asoc->rwnd;
sack = sctp_make_sack(asoc);
if (sack) {
retval = __sctp_packet_append_chunk(pkt, sack);
if (retval != SCTP_XMIT_OK) {
sctp_chunk_free(sack);
goto out;
}
SCTP_INC_STATS(asoc->base.net,
SCTP_MIB_OUTCTRLCHUNKS);
asoc->stats.octrlchunks++;
asoc->peer.sack_needed = 0;
2009-07-29 23:05:57 +07:00
if (del_timer(timer))
sctp_association_put(asoc);
}
}
}
out:
return retval;
}
/* Append a chunk to the offered packet reporting back any inability to do
* so.
*/
static enum sctp_xmit __sctp_packet_append_chunk(struct sctp_packet *packet,
struct sctp_chunk *chunk)
{
__u16 chunk_len = SCTP_PAD4(ntohs(chunk->chunk_hdr->length));
enum sctp_xmit retval = SCTP_XMIT_OK;
/* Check to see if this chunk will fit into the packet */
retval = sctp_packet_will_fit(packet, chunk, chunk_len);
if (retval != SCTP_XMIT_OK)
goto finish;
/* We believe that this chunk is OK to add to the packet */
switch (chunk->chunk_hdr->type) {
case SCTP_CID_DATA:
case SCTP_CID_I_DATA:
/* Account for the data being in the packet */
sctp_packet_append_data(packet, chunk);
/* Disallow SACK bundling after DATA. */
packet->has_sack = 1;
/* Disallow AUTH bundling after DATA */
packet->has_auth = 1;
/* Let it be knows that packet has DATA in it */
packet->has_data = 1;
/* timestamp the chunk for rtx purposes */
chunk->sent_at = jiffies;
/* Mainly used for prsctp RTX policy */
chunk->sent_count++;
break;
case SCTP_CID_COOKIE_ECHO:
packet->has_cookie_echo = 1;
break;
case SCTP_CID_SACK:
packet->has_sack = 1;
sctp: Add support to per-association statistics via a new SCTP_GET_ASSOC_STATS call The current SCTP stack is lacking a mechanism to have per association statistics. This is an implementation modeled after OpenSolaris' SCTP_GET_ASSOC_STATS. Userspace part will follow on lksctp if/when there is a general ACK on this. V4: - Move ipackets++ before q->immediate.func() for consistency reasons - Move sctp_max_rto() at the end of sctp_transport_update_rto() to avoid returning bogus RTO values - return asoc->rto_min when max_obs_rto value has not changed V3: - Increase ictrlchunks in sctp_assoc_bh_rcv() as well - Move ipackets++ to sctp_inq_push() - return 0 when no rto updates took place since the last call V2: - Implement partial retrieval of stat struct to cope for future expansion - Kill the rtxpackets counter as it cannot be precise anyway - Rename outseqtsns to outofseqtsns to make it clearer that these are out of sequence unexpected TSNs - Move asoc->ipackets++ under a lock to avoid potential miscounts - Fold asoc->opackets++ into the already existing asoc check - Kill unneeded (q->asoc) test when increasing rtxchunks - Do not count octrlchunks if sending failed (SCTP_XMIT_OK != 0) - Don't count SHUTDOWNs as SACKs - Move SCTP_GET_ASSOC_STATS to the private space API - Adjust the len check in sctp_getsockopt_assoc_stats() to allow for future struct growth - Move association statistics in their own struct - Update idupchunks when we send a SACK with dup TSNs - return min_rto in max_rto when RTO has not changed. Also return the transport when max_rto last changed. Signed-off: Michele Baldessari <michele@acksyn.org> Acked-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-12-01 11:49:42 +07:00
if (chunk->asoc)
chunk->asoc->stats.osacks++;
break;
case SCTP_CID_AUTH:
packet->has_auth = 1;
packet->auth = chunk;
break;
}
/* It is OK to send this chunk. */
list_add_tail(&chunk->list, &packet->chunk_list);
packet->size += chunk_len;
chunk->transport = packet->transport;
finish:
return retval;
}
/* Append a chunk to the offered packet reporting back any inability to do
* so.
*/
enum sctp_xmit sctp_packet_append_chunk(struct sctp_packet *packet,
struct sctp_chunk *chunk)
{
enum sctp_xmit retval = SCTP_XMIT_OK;
net: sctp: rework debugging framework to use pr_debug and friends We should get rid of all own SCTP debug printk macros and use the ones that the kernel offers anyway instead. This makes the code more readable and conform to the kernel code, and offers all the features of dynamic debbuging that pr_debug() et al has, such as only turning on/off portions of debug messages at runtime through debugfs. The runtime cost of having CONFIG_DYNAMIC_DEBUG enabled, but none of the debug statements printing, is negligible [1]. If kernel debugging is completly turned off, then these statements will also compile into "empty" functions. While we're at it, we also need to change the Kconfig option as it /now/ only refers to the ifdef'ed code portions in outqueue.c that enable further debugging/tracing of SCTP transaction fields. Also, since SCTP_ASSERT code was enabled with this Kconfig option and has now been removed, we transform those code parts into WARNs resp. where appropriate BUG_ONs so that those bugs can be more easily detected as probably not many people have SCTP debugging permanently turned on. To turn on all SCTP debugging, the following steps are needed: # mount -t debugfs none /sys/kernel/debug # echo -n 'module sctp +p' > /sys/kernel/debug/dynamic_debug/control This can be done more fine-grained on a per file, per line basis and others as described in [2]. [1] https://www.kernel.org/doc/ols/2009/ols2009-pages-39-46.pdf [2] Documentation/dynamic-debug-howto.txt Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-29 00:49:40 +07:00
pr_debug("%s: packet:%p chunk:%p\n", __func__, packet, chunk);
/* Data chunks are special. Before seeing what else we can
* bundle into this packet, check to see if we are allowed to
* send this DATA.
*/
if (sctp_chunk_is_data(chunk)) {
retval = sctp_packet_can_append_data(packet, chunk);
if (retval != SCTP_XMIT_OK)
goto finish;
}
/* Try to bundle AUTH chunk */
retval = sctp_packet_bundle_auth(packet, chunk);
if (retval != SCTP_XMIT_OK)
goto finish;
/* Try to bundle SACK chunk */
retval = sctp_packet_bundle_sack(packet, chunk);
if (retval != SCTP_XMIT_OK)
goto finish;
retval = __sctp_packet_append_chunk(packet, chunk);
finish:
return retval;
}
static void sctp_packet_gso_append(struct sk_buff *head, struct sk_buff *skb)
{
if (SCTP_OUTPUT_CB(head)->last == head)
skb_shinfo(head)->frag_list = skb;
else
SCTP_OUTPUT_CB(head)->last->next = skb;
SCTP_OUTPUT_CB(head)->last = skb;
head->truesize += skb->truesize;
head->data_len += skb->len;
head->len += skb->len;
refcount_add(skb->truesize, &head->sk->sk_wmem_alloc);
__skb_header_release(skb);
}
static int sctp_packet_pack(struct sctp_packet *packet,
struct sk_buff *head, int gso, gfp_t gfp)
{
struct sctp_transport *tp = packet->transport;
struct sctp_auth_chunk *auth = NULL;
struct sctp_chunk *chunk, *tmp;
int pkt_count = 0, pkt_size;
struct sock *sk = head->sk;
struct sk_buff *nskb;
int auth_len = 0;
if (gso) {
skb_shinfo(head)->gso_type = sk->sk_gso_type;
SCTP_OUTPUT_CB(head)->last = head;
} else {
nskb = head;
pkt_size = packet->size;
goto merge;
}
do {
/* calculate the pkt_size and alloc nskb */
pkt_size = packet->overhead;
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list,
list) {
int padded = SCTP_PAD4(chunk->skb->len);
if (chunk == packet->auth)
auth_len = padded;
else if (auth_len + padded + packet->overhead >
tp->pathmtu)
return 0;
else if (pkt_size + padded > tp->pathmtu)
break;
pkt_size += padded;
}
nskb = alloc_skb(pkt_size + MAX_HEADER, gfp);
if (!nskb)
return 0;
skb_reserve(nskb, packet->overhead + MAX_HEADER);
merge:
/* merge chunks into nskb and append nskb into head list */
pkt_size -= packet->overhead;
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) {
int padding;
list_del_init(&chunk->list);
if (sctp_chunk_is_data(chunk)) {
if (!sctp_chunk_retransmitted(chunk) &&
!tp->rto_pending) {
chunk->rtt_in_progress = 1;
tp->rto_pending = 1;
}
}
padding = SCTP_PAD4(chunk->skb->len) - chunk->skb->len;
if (padding)
skb_put_zero(chunk->skb, padding);
if (chunk == packet->auth)
auth = (struct sctp_auth_chunk *)
skb_tail_pointer(nskb);
skb_put_data(nskb, chunk->skb->data, chunk->skb->len);
pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, rtt_in_progress:%d\n",
chunk,
sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)),
chunk->has_tsn ? "TSN" : "No TSN",
chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0,
ntohs(chunk->chunk_hdr->length), chunk->skb->len,
chunk->rtt_in_progress);
pkt_size -= SCTP_PAD4(chunk->skb->len);
if (!sctp_chunk_is_data(chunk) && chunk != packet->auth)
sctp_chunk_free(chunk);
if (!pkt_size)
break;
}
if (auth) {
sctp_auth_calculate_hmac(tp->asoc, nskb, auth,
packet->auth->shkey, gfp);
/* free auth if no more chunks, or add it back */
if (list_empty(&packet->chunk_list))
sctp_chunk_free(packet->auth);
else
list_add(&packet->auth->list,
&packet->chunk_list);
}
if (gso)
sctp_packet_gso_append(head, nskb);
pkt_count++;
} while (!list_empty(&packet->chunk_list));
if (gso) {
memset(head->cb, 0, max(sizeof(struct inet_skb_parm),
sizeof(struct inet6_skb_parm)));
skb_shinfo(head)->gso_segs = pkt_count;
skb_shinfo(head)->gso_size = GSO_BY_FRAGS;
rcu_read_lock();
if (skb_dst(head) != tp->dst) {
dst_hold(tp->dst);
sk_setup_caps(sk, tp->dst);
}
rcu_read_unlock();
goto chksum;
}
if (sctp_checksum_disable)
return 1;
if (!(skb_dst(head)->dev->features & NETIF_F_SCTP_CRC) ||
dst_xfrm(skb_dst(head)) || packet->ipfragok) {
struct sctphdr *sh =
(struct sctphdr *)skb_transport_header(head);
sh->checksum = sctp_compute_cksum(head, 0);
} else {
chksum:
head->ip_summed = CHECKSUM_PARTIAL;
head->csum_not_inet = 1;
head->csum_start = skb_transport_header(head) - head->head;
head->csum_offset = offsetof(struct sctphdr, checksum);
}
return pkt_count;
}
/* All packets are sent to the network through this function from
* sctp_outq_tail().
*
* The return value is always 0 for now.
*/
int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp)
{
struct sctp_transport *tp = packet->transport;
struct sctp_association *asoc = tp->asoc;
struct sctp_chunk *chunk, *tmp;
int pkt_count, gso = 0;
struct dst_entry *dst;
struct sk_buff *head;
struct sctphdr *sh;
struct sock *sk;
pr_debug("%s: packet:%p\n", __func__, packet);
if (list_empty(&packet->chunk_list))
return 0;
chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list);
sk = chunk->skb->sk;
/* check gso */
if (packet->size > tp->pathmtu && !packet->ipfragok) {
if (!sk_can_gso(sk)) {
pr_err_once("Trying to GSO but underlying device doesn't support it.");
goto out;
}
gso = 1;
}
/* alloc head skb */
head = alloc_skb((gso ? packet->overhead : packet->size) +
MAX_HEADER, gfp);
if (!head)
goto out;
skb_reserve(head, packet->overhead + MAX_HEADER);
skb_set_owner_w(head, sk);
/* set sctp header */
sh = skb_push(head, sizeof(struct sctphdr));
skb_reset_transport_header(head);
sh->source = htons(packet->source_port);
sh->dest = htons(packet->destination_port);
sh->vtag = htonl(packet->vtag);
sh->checksum = 0;
/* drop packet if no dst */
dst = dst_clone(tp->dst);
if (!dst) {
IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
kfree_skb(head);
goto out;
}
skb_dst_set(head, dst);
/* pack up chunks */
pkt_count = sctp_packet_pack(packet, head, gso, gfp);
if (!pkt_count) {
kfree_skb(head);
goto out;
}
pr_debug("***sctp_transmit_packet*** skb->len:%d\n", head->len);
/* start autoclose timer */
if (packet->has_data && sctp_state(asoc, ESTABLISHED) &&
asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]) {
struct timer_list *timer =
&asoc->timers[SCTP_EVENT_TIMEOUT_AUTOCLOSE];
unsigned long timeout =
asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE];
if (!mod_timer(timer, jiffies + timeout))
sctp_association_hold(asoc);
}
/* sctp xmit */
tp->af_specific->ecn_capable(sk);
if (asoc) {
asoc->stats.opackets += pkt_count;
if (asoc->peer.last_sent_to != tp)
asoc->peer.last_sent_to = tp;
}
head->ignore_df = packet->ipfragok;
if (tp->dst_pending_confirm)
skb_set_dst_pending_confirm(head, 1);
/* neighbour should be confirmed on successful transmission or
* positive error
*/
if (tp->af_specific->sctp_xmit(head, tp) >= 0 &&
tp->dst_pending_confirm)
tp->dst_pending_confirm = 0;
out:
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) {
list_del_init(&chunk->list);
if (!sctp_chunk_is_data(chunk))
sctp_chunk_free(chunk);
}
sctp_packet_reset(packet);
return 0;
}
/********************************************************************
* 2nd Level Abstractions
********************************************************************/
/* This private function check to see if a chunk can be added */
static enum sctp_xmit sctp_packet_can_append_data(struct sctp_packet *packet,
struct sctp_chunk *chunk)
{
size_t datasize, rwnd, inflight, flight_size;
struct sctp_transport *transport = packet->transport;
struct sctp_association *asoc = transport->asoc;
struct sctp_outq *q = &asoc->outqueue;
/* RFC 2960 6.1 Transmission of DATA Chunks
*
* A) At any given time, the data sender MUST NOT transmit new data to
* any destination transport address if its peer's rwnd indicates
* that the peer has no buffer space (i.e. rwnd is 0, see Section
* 6.2.1). However, regardless of the value of rwnd (including if it
* is 0), the data sender can always have one DATA chunk in flight to
* the receiver if allowed by cwnd (see rule B below). This rule
* allows the sender to probe for a change in rwnd that the sender
* missed due to the SACK having been lost in transit from the data
* receiver to the data sender.
*/
rwnd = asoc->peer.rwnd;
inflight = q->outstanding_bytes;
flight_size = transport->flight_size;
datasize = sctp_data_size(chunk);
if (datasize > rwnd && inflight > 0)
/* We have (at least) one data chunk in flight,
* so we can't fall back to rule 6.1 B).
*/
return SCTP_XMIT_RWND_FULL;
/* RFC 2960 6.1 Transmission of DATA Chunks
*
* B) At any given time, the sender MUST NOT transmit new data
* to a given transport address if it has cwnd or more bytes
* of data outstanding to that transport address.
*/
/* RFC 7.2.4 & the Implementers Guide 2.8.
*
* 3) ...
* When a Fast Retransmit is being performed the sender SHOULD
* ignore the value of cwnd and SHOULD NOT delay retransmission.
*/
if (chunk->fast_retransmit != SCTP_NEED_FRTX &&
flight_size >= transport->cwnd)
return SCTP_XMIT_RWND_FULL;
/* Nagle's algorithm to solve small-packet problem:
* Inhibit the sending of new chunks when new outgoing data arrives
* if any previously transmitted data on the connection remains
* unacknowledged.
*/
if ((sctp_sk(asoc->base.sk)->nodelay || inflight == 0) &&
!asoc->force_delay)
/* Nothing unacked */
return SCTP_XMIT_OK;
if (!sctp_packet_empty(packet))
/* Append to packet */
return SCTP_XMIT_OK;
if (!sctp_state(asoc, ESTABLISHED))
return SCTP_XMIT_OK;
/* Check whether this chunk and all the rest of pending data will fit
* or delay in hopes of bundling a full sized packet.
*/
if (chunk->skb->len + q->out_qlen > transport->pathmtu -
packet->overhead - sctp_datachk_len(&chunk->asoc->stream) - 4)
/* Enough data queued to fill a packet */
return SCTP_XMIT_OK;
/* Don't delay large message writes that may have been fragmented */
if (!chunk->msg->can_delay)
return SCTP_XMIT_OK;
/* Defer until all data acked or packet full */
return SCTP_XMIT_DELAY;
}
/* This private function does management things when adding DATA chunk */
static void sctp_packet_append_data(struct sctp_packet *packet,
struct sctp_chunk *chunk)
{
struct sctp_transport *transport = packet->transport;
size_t datasize = sctp_data_size(chunk);
struct sctp_association *asoc = transport->asoc;
u32 rwnd = asoc->peer.rwnd;
/* Keep track of how many bytes are in flight over this transport. */
transport->flight_size += datasize;
/* Keep track of how many bytes are in flight to the receiver. */
asoc->outqueue.outstanding_bytes += datasize;
sctp: Do not account for sizeof(struct sk_buff) in estimated rwnd When checking whether a DATA chunk fits into the estimated rwnd a full sizeof(struct sk_buff) is added to the needed chunk size. This quickly exhausts the available rwnd space and leads to packets being sent which are much below the PMTU limit. This can lead to much worse performance. The reason for this behaviour was to avoid putting too much memory pressure on the receiver. The concept is not completely irational because a Linux receiver does in fact clone an skb for each DATA chunk delivered. However, Linux also reserves half the available socket buffer space for data structures therefore usage of it is already accounted for. When proposing to change this the last time it was noted that this behaviour was introduced to solve a performance issue caused by rwnd overusage in combination with small DATA chunks. Trying to reproduce this I found that with the sk_buff overhead removed, the performance would improve significantly unless socket buffer limits are increased. The following numbers have been gathered using a patched iperf supporting SCTP over a live 1 Gbit ethernet network. The -l option was used to limit DATA chunk sizes. The numbers listed are based on the average of 3 test runs each. Default values have been used for sk_(r|w)mem. Chunk Size Unpatched No Overhead ------------------------------------- 4 15.2 Kbit [!] 12.2 Mbit [!] 8 35.8 Kbit [!] 26.0 Mbit [!] 16 95.5 Kbit [!] 54.4 Mbit [!] 32 106.7 Mbit 102.3 Mbit 64 189.2 Mbit 188.3 Mbit 128 331.2 Mbit 334.8 Mbit 256 537.7 Mbit 536.0 Mbit 512 766.9 Mbit 766.6 Mbit 1024 810.1 Mbit 808.6 Mbit Signed-off-by: Thomas Graf <tgraf@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:11:40 +07:00
/* Update our view of the receiver's rwnd. */
if (datasize < rwnd)
rwnd -= datasize;
else
rwnd = 0;
asoc->peer.rwnd = rwnd;
sctp: Fix mis-ordering of user space data when multihoming in use Recently had a bug reported to me, in which the user was sending packets with a payload containing a sequence number. The packets were getting delivered in order according the chunk TSN values, but the sequence values in the payload were arriving out of order. At first I thought it must be an application error, but we eventually found it to be a problem on the transmit side in the sctp stack. The conditions for the error are that multihoming must be in use, and it helps if each transport has a different pmtu. The problem occurs in sctp_outq_flush. Basically we dequeue packets from the data queue, and attempt to append them to the orrered packet for a given transport. After we append a data chunk we add the trasport to the end of a list of transports to have their packets sent at the end of sctp_outq_flush. The problem occurs when a data chunks fills up a offered packet on a transport. The function that does the appending (sctp_packet_transmit_chunk), will try to call sctp_packet_transmit on the full packet, and then append the chunk to a new packet. This call to sctp_packet_transmit, sends that packet ahead of the others that may be queued in the transport_list in sctp_outq_flush. The result is that frames that were sent in one order from the user space sending application get re-ordered prior to tsn assignment in sctp_packet_transmit, resulting in mis-sequencing of data payloads, even though tsn ordering is correct. The fix is to change where we assign a tsn. By doing this earlier, we are then free to place chunks in packets, whatever way we see fit and the protocol will make sure to do all the appropriate re-ordering on receive as is needed. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: William Reich <reich@ulticom.com> Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
2009-11-24 03:54:00 +07:00
sctp_chunk_assign_tsn(chunk);
asoc->stream.si->assign_number(chunk);
}
static enum sctp_xmit sctp_packet_will_fit(struct sctp_packet *packet,
struct sctp_chunk *chunk,
u16 chunk_len)
{
enum sctp_xmit retval = SCTP_XMIT_OK;
size_t psize, pmtu, maxsize;
/* Don't bundle in this packet if this chunk's auth key doesn't
* match other chunks already enqueued on this packet. Also,
* don't bundle the chunk with auth key if other chunks in this
* packet don't have auth key.
*/
if ((packet->auth && chunk->shkey != packet->auth->shkey) ||
(!packet->auth && chunk->shkey &&
chunk->chunk_hdr->type != SCTP_CID_AUTH))
return SCTP_XMIT_PMTU_FULL;
psize = packet->size;
if (packet->transport->asoc)
pmtu = packet->transport->asoc->pathmtu;
else
pmtu = packet->transport->pathmtu;
/* Decide if we need to fragment or resubmit later. */
if (psize + chunk_len > pmtu) {
/* It's OK to fragment at IP level if any one of the following
* is true:
* 1. The packet is empty (meaning this chunk is greater
* the MTU)
* 2. The packet doesn't have any data in it yet and data
* requires authentication.
*/
if (sctp_packet_empty(packet) ||
(!packet->has_data && chunk->auth)) {
/* We no longer do re-fragmentation.
* Just fragment at the IP layer, if we
* actually hit this condition
*/
packet->ipfragok = 1;
goto out;
}
/* Similarly, if this chunk was built before a PMTU
* reduction, we have to fragment it at IP level now. So
* if the packet already contains something, we need to
* flush.
*/
maxsize = pmtu - packet->overhead;
if (packet->auth)
maxsize -= SCTP_PAD4(packet->auth->skb->len);
if (chunk_len > maxsize)
retval = SCTP_XMIT_PMTU_FULL;
/* It is also okay to fragment if the chunk we are
* adding is a control chunk, but only if current packet
* is not a GSO one otherwise it causes fragmentation of
* a large frame. So in this case we allow the
* fragmentation by forcing it to be in a new packet.
*/
if (!sctp_chunk_is_data(chunk) && packet->has_data)
retval = SCTP_XMIT_PMTU_FULL;
if (psize + chunk_len > packet->max_size)
/* Hit GSO/PMTU limit, gotta flush */
retval = SCTP_XMIT_PMTU_FULL;
if (!packet->transport->burst_limited &&
psize + chunk_len > (packet->transport->cwnd >> 1))
/* Do not allow a single GSO packet to use more
* than half of cwnd.
*/
retval = SCTP_XMIT_PMTU_FULL;
if (packet->transport->burst_limited &&
psize + chunk_len > (packet->transport->burst_limited >> 1))
/* Do not allow a single GSO packet to use more
* than half of original cwnd.
*/
retval = SCTP_XMIT_PMTU_FULL;
/* Otherwise it will fit in the GSO packet */
}
out:
return retval;
}