linux/net/sctp/ulpevent.c

1191 lines
32 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.
* Copyright (c) 2001 Intel Corp.
* Copyright (c) 2001 Nokia, Inc.
* Copyright (c) 2001 La Monte H.P. Yarroll
*
* These functions manipulate an sctp event. The struct ulpevent is used
* to carry notifications and data to the ULP (sockets).
*
* 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:
* Jon Grimm <jgrimm@us.ibm.com>
* La Monte H.P. Yarroll <piggy@acm.org>
* Ardelle Fan <ardelle.fan@intel.com>
* Sridhar Samudrala <sri@us.ibm.com>
*/
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>
#include <linux/types.h>
#include <linux/skbuff.h>
#include <net/sctp/structs.h>
#include <net/sctp/sctp.h>
#include <net/sctp/sm.h>
static void sctp_ulpevent_receive_data(struct sctp_ulpevent *event,
struct sctp_association *asoc);
static void sctp_ulpevent_release_data(struct sctp_ulpevent *event);
static void sctp_ulpevent_release_frag_data(struct sctp_ulpevent *event);
/* Initialize an ULP event from an given skb. */
static void sctp_ulpevent_init(struct sctp_ulpevent *event,
__u16 msg_flags,
unsigned int len)
{
memset(event, 0, sizeof(struct sctp_ulpevent));
event->msg_flags = msg_flags;
event->rmem_len = len;
}
/* Create a new sctp_ulpevent. */
static struct sctp_ulpevent *sctp_ulpevent_new(int size, __u16 msg_flags,
gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sk_buff *skb;
skb = alloc_skb(size, gfp);
if (!skb)
goto fail;
event = sctp_skb2event(skb);
sctp_ulpevent_init(event, msg_flags, skb->truesize);
return event;
fail:
return NULL;
}
/* Is this a MSG_NOTIFICATION? */
int sctp_ulpevent_is_notification(const struct sctp_ulpevent *event)
{
return MSG_NOTIFICATION == (event->msg_flags & MSG_NOTIFICATION);
}
/* Hold the association in case the msg_name needs read out of
* the association.
*/
static inline void sctp_ulpevent_set_owner(struct sctp_ulpevent *event,
const struct sctp_association *asoc)
{
struct sctp_chunk *chunk = event->chunk;
struct sk_buff *skb;
/* Cast away the const, as we are just wanting to
* bump the reference count.
*/
sctp_association_hold((struct sctp_association *)asoc);
skb = sctp_event2skb(event);
event->asoc = (struct sctp_association *)asoc;
atomic_add(event->rmem_len, &event->asoc->rmem_alloc);
sctp_skb_set_owner_r(skb, asoc->base.sk);
if (chunk && chunk->head_skb && !chunk->head_skb->sk)
chunk->head_skb->sk = asoc->base.sk;
}
/* A simple destructor to give up the reference to the association. */
static inline void sctp_ulpevent_release_owner(struct sctp_ulpevent *event)
{
struct sctp_association *asoc = event->asoc;
atomic_sub(event->rmem_len, &asoc->rmem_alloc);
sctp_association_put(asoc);
}
/* Create and initialize an SCTP_ASSOC_CHANGE event.
*
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* Communication notifications inform the ULP that an SCTP association
* has either begun or ended. The identifier for a new association is
* provided by this notification.
*
* Note: There is no field checking here. If a field is unused it will be
* zero'd out.
*/
struct sctp_ulpevent *sctp_ulpevent_make_assoc_change(
const struct sctp_association *asoc,
__u16 flags, __u16 state, __u16 error, __u16 outbound,
__u16 inbound, struct sctp_chunk *chunk, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_assoc_change *sac;
struct sk_buff *skb;
/* If the lower layer passed in the chunk, it will be
* an ABORT, so we need to include it in the sac_info.
*/
if (chunk) {
/* Copy the chunk data to a new skb and reserve enough
* head room to use as notification.
*/
skb = skb_copy_expand(chunk->skb,
sizeof(struct sctp_assoc_change), 0, gfp);
if (!skb)
goto fail;
/* Embed the event fields inside the cloned skb. */
event = sctp_skb2event(skb);
sctp_ulpevent_init(event, MSG_NOTIFICATION, skb->truesize);
/* Include the notification structure */
sac = skb_push(skb, sizeof(struct sctp_assoc_change));
/* Trim the buffer to the right length. */
skb_trim(skb, sizeof(struct sctp_assoc_change) +
ntohs(chunk->chunk_hdr->length) -
sizeof(struct sctp_chunkhdr));
} else {
event = sctp_ulpevent_new(sizeof(struct sctp_assoc_change),
MSG_NOTIFICATION, gfp);
if (!event)
goto fail;
skb = sctp_event2skb(event);
sac = skb_put(skb, sizeof(struct sctp_assoc_change));
}
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* sac_type:
* It should be SCTP_ASSOC_CHANGE.
*/
sac->sac_type = SCTP_ASSOC_CHANGE;
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* sac_state: 32 bits (signed integer)
* This field holds one of a number of values that communicate the
* event that happened to the association.
*/
sac->sac_state = state;
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* sac_flags: 16 bits (unsigned integer)
* Currently unused.
*/
sac->sac_flags = 0;
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* sac_length: sizeof (__u32)
* This field is the total length of the notification data, including
* the notification header.
*/
sac->sac_length = skb->len;
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* sac_error: 32 bits (signed integer)
*
* If the state was reached due to a error condition (e.g.
* COMMUNICATION_LOST) any relevant error information is available in
* this field. This corresponds to the protocol error codes defined in
* [SCTP].
*/
sac->sac_error = error;
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* sac_outbound_streams: 16 bits (unsigned integer)
* sac_inbound_streams: 16 bits (unsigned integer)
*
* The maximum number of streams allowed in each direction are
* available in sac_outbound_streams and sac_inbound streams.
*/
sac->sac_outbound_streams = outbound;
sac->sac_inbound_streams = inbound;
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* sac_assoc_id: sizeof (sctp_assoc_t)
*
* The association id field, holds the identifier for the association.
* All notifications for a given association have the same association
* identifier. For TCP style socket, this field is ignored.
*/
sctp_ulpevent_set_owner(event, asoc);
sac->sac_assoc_id = sctp_assoc2id(asoc);
return event;
fail:
return NULL;
}
/* Create and initialize an SCTP_PEER_ADDR_CHANGE event.
*
* Socket Extensions for SCTP - draft-01
* 5.3.1.2 SCTP_PEER_ADDR_CHANGE
*
* When a destination address on a multi-homed peer encounters a change
* an interface details event is sent.
*/
static struct sctp_ulpevent *sctp_ulpevent_make_peer_addr_change(
const struct sctp_association *asoc,
const struct sockaddr_storage *aaddr,
int flags, int state, int error, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_paddr_change *spc;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_paddr_change),
MSG_NOTIFICATION, gfp);
if (!event)
goto fail;
skb = sctp_event2skb(event);
spc = skb_put(skb, sizeof(struct sctp_paddr_change));
/* Sockets API Extensions for SCTP
* Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE
*
* spc_type:
*
* It should be SCTP_PEER_ADDR_CHANGE.
*/
spc->spc_type = SCTP_PEER_ADDR_CHANGE;
/* Sockets API Extensions for SCTP
* Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE
*
* spc_length: sizeof (__u32)
*
* This field is the total length of the notification data, including
* the notification header.
*/
spc->spc_length = sizeof(struct sctp_paddr_change);
/* Sockets API Extensions for SCTP
* Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE
*
* spc_flags: 16 bits (unsigned integer)
* Currently unused.
*/
spc->spc_flags = 0;
/* Sockets API Extensions for SCTP
* Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE
*
* spc_state: 32 bits (signed integer)
*
* This field holds one of a number of values that communicate the
* event that happened to the address.
*/
spc->spc_state = state;
/* Sockets API Extensions for SCTP
* Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE
*
* spc_error: 32 bits (signed integer)
*
* If the state was reached due to any error condition (e.g.
* ADDRESS_UNREACHABLE) any relevant error information is available in
* this field.
*/
spc->spc_error = error;
/* Socket Extensions for SCTP
* 5.3.1.1 SCTP_ASSOC_CHANGE
*
* spc_assoc_id: sizeof (sctp_assoc_t)
*
* The association id field, holds the identifier for the association.
* All notifications for a given association have the same association
* identifier. For TCP style socket, this field is ignored.
*/
sctp_ulpevent_set_owner(event, asoc);
spc->spc_assoc_id = sctp_assoc2id(asoc);
/* Sockets API Extensions for SCTP
* Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE
*
* spc_aaddr: sizeof (struct sockaddr_storage)
*
* The affected address field, holds the remote peer's address that is
* encountering the change of state.
*/
memcpy(&spc->spc_aaddr, aaddr, sizeof(struct sockaddr_storage));
/* Map ipv4 address into v4-mapped-on-v6 address. */
sctp: Fixup v4mapped behaviour to comply with Sock API The SCTP socket extensions API document describes the v4mapping option as follows: 8.1.15. Set/Clear IPv4 Mapped Addresses (SCTP_I_WANT_MAPPED_V4_ADDR) This socket option is a Boolean flag which turns on or off the mapping of IPv4 addresses. If this option is turned on, then IPv4 addresses will be mapped to V6 representation. If this option is turned off, then no mapping will be done of V4 addresses and a user will receive both PF_INET6 and PF_INET type addresses on the socket. See [RFC3542] for more details on mapped V6 addresses. This description isn't really in line with what the code does though. Introduce addr_to_user (renamed addr_v4map), which should be called before any sockaddr is passed back to user space. The new function places the sockaddr into the correct format depending on the SCTP_I_WANT_MAPPED_V4_ADDR option. Audit all places that touched v4mapped and either sanely construct a v4 or v6 address then call addr_to_user, or drop the unnecessary v4mapped check entirely. Audit all places that call addr_to_user and verify they are on a sycall return path. Add a custom getname that formats the address properly. Several bugs are addressed: - SCTP_I_WANT_MAPPED_V4_ADDR=0 often returned garbage for addresses to user space - The addr_len returned from recvmsg was not correct when returning AF_INET on a v6 socket - flowlabel and scope_id were not zerod when promoting a v4 to v6 - Some syscalls like bind and connect behaved differently depending on v4mapped Tested bind, getpeername, getsockname, connect, and recvmsg for proper behaviour in v4mapped = 1 and 0 cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Tested-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com> Signed-off-by: Jason Gunthorpe <jgunthorpe@obsidianresearch.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-30 18:40:53 +00:00
sctp_get_pf_specific(asoc->base.sk->sk_family)->addr_to_user(
sctp_sk(asoc->base.sk),
(union sctp_addr *)&spc->spc_aaddr);
return event;
fail:
return NULL;
}
void sctp_ulpevent_notify_peer_addr_change(struct sctp_transport *transport,
int state, int error)
{
struct sctp_association *asoc = transport->asoc;
struct sockaddr_storage addr;
struct sctp_ulpevent *event;
if (asoc->state < SCTP_STATE_ESTABLISHED)
return;
memset(&addr, 0, sizeof(struct sockaddr_storage));
memcpy(&addr, &transport->ipaddr, transport->af_specific->sockaddr_len);
event = sctp_ulpevent_make_peer_addr_change(asoc, &addr, 0, state,
error, GFP_ATOMIC);
if (event)
asoc->stream.si->enqueue_event(&asoc->ulpq, event);
}
/* Create and initialize an SCTP_REMOTE_ERROR notification.
*
* Note: This assumes that the chunk->skb->data already points to the
* operation error payload.
*
* Socket Extensions for SCTP - draft-01
* 5.3.1.3 SCTP_REMOTE_ERROR
*
* A remote peer may send an Operational Error message to its peer.
* This message indicates a variety of error conditions on an
* association. The entire error TLV as it appears on the wire is
* included in a SCTP_REMOTE_ERROR event. Please refer to the SCTP
* specification [SCTP] and any extensions for a list of possible
* error formats.
*/
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
struct sctp_ulpevent *
sctp_ulpevent_make_remote_error(const struct sctp_association *asoc,
struct sctp_chunk *chunk, __u16 flags,
gfp_t gfp)
{
struct sctp_remote_error *sre;
struct sctp_ulpevent *event;
struct sctp_errhdr *ch;
struct sk_buff *skb;
__be16 cause;
int elen;
ch = (struct sctp_errhdr *)(chunk->skb->data);
cause = ch->cause;
elen = SCTP_PAD4(ntohs(ch->length)) - sizeof(*ch);
/* Pull off the ERROR header. */
skb_pull(chunk->skb, sizeof(*ch));
/* Copy the skb to a new skb with room for us to prepend
* notification with.
*/
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
skb = skb_copy_expand(chunk->skb, sizeof(*sre), 0, gfp);
/* Pull off the rest of the cause TLV from the chunk. */
skb_pull(chunk->skb, elen);
if (!skb)
goto fail;
/* Embed the event fields inside the cloned skb. */
event = sctp_skb2event(skb);
sctp_ulpevent_init(event, MSG_NOTIFICATION, skb->truesize);
sre = skb_push(skb, sizeof(*sre));
/* Trim the buffer to the right length. */
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
skb_trim(skb, sizeof(*sre) + elen);
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
/* RFC6458, Section 6.1.3. SCTP_REMOTE_ERROR */
memset(sre, 0, sizeof(*sre));
sre->sre_type = SCTP_REMOTE_ERROR;
sre->sre_flags = 0;
sre->sre_length = skb->len;
sre->sre_error = cause;
sctp_ulpevent_set_owner(event, asoc);
sre->sre_assoc_id = sctp_assoc2id(asoc);
return event;
fail:
return NULL;
}
/* Create and initialize a SCTP_SEND_FAILED notification.
*
* Socket Extensions for SCTP - draft-01
* 5.3.1.4 SCTP_SEND_FAILED
*/
struct sctp_ulpevent *sctp_ulpevent_make_send_failed(
const struct sctp_association *asoc, struct sctp_chunk *chunk,
__u16 flags, __u32 error, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_send_failed *ssf;
struct sk_buff *skb;
/* Pull off any padding. */
int len = ntohs(chunk->chunk_hdr->length);
/* Make skb with more room so we can prepend notification. */
skb = skb_copy_expand(chunk->skb,
sizeof(struct sctp_send_failed), /* headroom */
0, /* tailroom */
gfp);
if (!skb)
goto fail;
/* Pull off the common chunk header and DATA header. */
skb_pull(skb, sctp_datachk_len(&asoc->stream));
len -= sctp_datachk_len(&asoc->stream);
/* Embed the event fields inside the cloned skb. */
event = sctp_skb2event(skb);
sctp_ulpevent_init(event, MSG_NOTIFICATION, skb->truesize);
ssf = skb_push(skb, sizeof(struct sctp_send_failed));
/* Socket Extensions for SCTP
* 5.3.1.4 SCTP_SEND_FAILED
*
* ssf_type:
* It should be SCTP_SEND_FAILED.
*/
ssf->ssf_type = SCTP_SEND_FAILED;
/* Socket Extensions for SCTP
* 5.3.1.4 SCTP_SEND_FAILED
*
* ssf_flags: 16 bits (unsigned integer)
* The flag value will take one of the following values
*
* SCTP_DATA_UNSENT - Indicates that the data was never put on
* the wire.
*
* SCTP_DATA_SENT - Indicates that the data was put on the wire.
* Note that this does not necessarily mean that the
* data was (or was not) successfully delivered.
*/
ssf->ssf_flags = flags;
/* Socket Extensions for SCTP
* 5.3.1.4 SCTP_SEND_FAILED
*
* ssf_length: sizeof (__u32)
* This field is the total length of the notification data, including
* the notification header.
*/
ssf->ssf_length = sizeof(struct sctp_send_failed) + len;
skb_trim(skb, ssf->ssf_length);
/* Socket Extensions for SCTP
* 5.3.1.4 SCTP_SEND_FAILED
*
* ssf_error: 16 bits (unsigned integer)
* This value represents the reason why the send failed, and if set,
* will be a SCTP protocol error code as defined in [SCTP] section
* 3.3.10.
*/
ssf->ssf_error = error;
/* Socket Extensions for SCTP
* 5.3.1.4 SCTP_SEND_FAILED
*
* ssf_info: sizeof (struct sctp_sndrcvinfo)
* The original send information associated with the undelivered
* message.
*/
memcpy(&ssf->ssf_info, &chunk->sinfo, sizeof(struct sctp_sndrcvinfo));
/* Per TSVWG discussion with Randy. Allow the application to
* reassemble a fragmented message.
*/
ssf->ssf_info.sinfo_flags = chunk->chunk_hdr->flags;
/* Socket Extensions for SCTP
* 5.3.1.4 SCTP_SEND_FAILED
*
* ssf_assoc_id: sizeof (sctp_assoc_t)
* The association id field, sf_assoc_id, holds the identifier for the
* association. All notifications for a given association have the
* same association identifier. For TCP style socket, this field is
* ignored.
*/
sctp_ulpevent_set_owner(event, asoc);
ssf->ssf_assoc_id = sctp_assoc2id(asoc);
return event;
fail:
return NULL;
}
struct sctp_ulpevent *sctp_ulpevent_make_send_failed_event(
const struct sctp_association *asoc, struct sctp_chunk *chunk,
__u16 flags, __u32 error, gfp_t gfp)
{
struct sctp_send_failed_event *ssf;
struct sctp_ulpevent *event;
struct sk_buff *skb;
int len;
skb = skb_copy_expand(chunk->skb, sizeof(*ssf), 0, gfp);
if (!skb)
return NULL;
len = ntohs(chunk->chunk_hdr->length);
len -= sctp_datachk_len(&asoc->stream);
skb_pull(skb, sctp_datachk_len(&asoc->stream));
event = sctp_skb2event(skb);
sctp_ulpevent_init(event, MSG_NOTIFICATION, skb->truesize);
ssf = skb_push(skb, sizeof(*ssf));
ssf->ssf_type = SCTP_SEND_FAILED_EVENT;
ssf->ssf_flags = flags;
ssf->ssf_length = sizeof(*ssf) + len;
skb_trim(skb, ssf->ssf_length);
ssf->ssf_error = error;
ssf->ssfe_info.snd_sid = chunk->sinfo.sinfo_stream;
ssf->ssfe_info.snd_ppid = chunk->sinfo.sinfo_ppid;
ssf->ssfe_info.snd_context = chunk->sinfo.sinfo_context;
ssf->ssfe_info.snd_assoc_id = chunk->sinfo.sinfo_assoc_id;
ssf->ssfe_info.snd_flags = chunk->chunk_hdr->flags;
sctp_ulpevent_set_owner(event, asoc);
ssf->ssf_assoc_id = sctp_assoc2id(asoc);
return event;
}
/* Create and initialize a SCTP_SHUTDOWN_EVENT notification.
*
* Socket Extensions for SCTP - draft-01
* 5.3.1.5 SCTP_SHUTDOWN_EVENT
*/
struct sctp_ulpevent *sctp_ulpevent_make_shutdown_event(
const struct sctp_association *asoc,
__u16 flags, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_shutdown_event *sse;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_shutdown_event),
MSG_NOTIFICATION, gfp);
if (!event)
goto fail;
skb = sctp_event2skb(event);
sse = skb_put(skb, sizeof(struct sctp_shutdown_event));
/* Socket Extensions for SCTP
* 5.3.1.5 SCTP_SHUTDOWN_EVENT
*
* sse_type
* It should be SCTP_SHUTDOWN_EVENT
*/
sse->sse_type = SCTP_SHUTDOWN_EVENT;
/* Socket Extensions for SCTP
* 5.3.1.5 SCTP_SHUTDOWN_EVENT
*
* sse_flags: 16 bits (unsigned integer)
* Currently unused.
*/
sse->sse_flags = 0;
/* Socket Extensions for SCTP
* 5.3.1.5 SCTP_SHUTDOWN_EVENT
*
* sse_length: sizeof (__u32)
* This field is the total length of the notification data, including
* the notification header.
*/
sse->sse_length = sizeof(struct sctp_shutdown_event);
/* Socket Extensions for SCTP
* 5.3.1.5 SCTP_SHUTDOWN_EVENT
*
* sse_assoc_id: sizeof (sctp_assoc_t)
* The association id field, holds the identifier for the association.
* All notifications for a given association have the same association
* identifier. For TCP style socket, this field is ignored.
*/
sctp_ulpevent_set_owner(event, asoc);
sse->sse_assoc_id = sctp_assoc2id(asoc);
return event;
fail:
return NULL;
}
/* Create and initialize a SCTP_ADAPTATION_INDICATION notification.
*
* Socket Extensions for SCTP
* 5.3.1.6 SCTP_ADAPTATION_INDICATION
*/
struct sctp_ulpevent *sctp_ulpevent_make_adaptation_indication(
const struct sctp_association *asoc, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_adaptation_event *sai;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_adaptation_event),
MSG_NOTIFICATION, gfp);
if (!event)
goto fail;
skb = sctp_event2skb(event);
sai = skb_put(skb, sizeof(struct sctp_adaptation_event));
sai->sai_type = SCTP_ADAPTATION_INDICATION;
sai->sai_flags = 0;
sai->sai_length = sizeof(struct sctp_adaptation_event);
sai->sai_adaptation_ind = asoc->peer.adaptation_ind;
sctp_ulpevent_set_owner(event, asoc);
sai->sai_assoc_id = sctp_assoc2id(asoc);
return event;
fail:
return NULL;
}
/* A message has been received. Package this message as a notification
* to pass it to the upper layers. Go ahead and calculate the sndrcvinfo
* even if filtered out later.
*
* Socket Extensions for SCTP
* 5.2.2 SCTP Header Information Structure (SCTP_SNDRCV)
*/
struct sctp_ulpevent *sctp_ulpevent_make_rcvmsg(struct sctp_association *asoc,
struct sctp_chunk *chunk,
gfp_t gfp)
{
struct sctp_ulpevent *event = NULL;
struct sk_buff *skb = chunk->skb;
struct sock *sk = asoc->base.sk;
size_t padding, datalen;
int rx_count;
/*
* check to see if we need to make space for this
* new skb, expand the rcvbuffer if needed, or drop
* the frame
*/
if (asoc->ep->rcvbuf_policy)
rx_count = atomic_read(&asoc->rmem_alloc);
else
rx_count = atomic_read(&sk->sk_rmem_alloc);
datalen = ntohs(chunk->chunk_hdr->length);
if (rx_count >= sk->sk_rcvbuf || !sk_rmem_schedule(sk, skb, datalen))
goto fail;
/* Clone the original skb, sharing the data. */
skb = skb_clone(chunk->skb, gfp);
if (!skb)
goto fail;
/* Now that all memory allocations for this chunk succeeded, we
* can mark it as received so the tsn_map is updated correctly.
*/
if (sctp_tsnmap_mark(&asoc->peer.tsn_map,
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 03:04:26 +00:00
ntohl(chunk->subh.data_hdr->tsn),
chunk->transport))
goto fail_mark;
/* First calculate the padding, so we don't inadvertently
* pass up the wrong length to the user.
*
* RFC 2960 - Section 3.2 Chunk Field Descriptions
*
* The total length of a chunk(including Type, Length and Value fields)
* MUST be a multiple of 4 bytes. If the length of the chunk is not a
* multiple of 4 bytes, the sender MUST pad the chunk with all zero
* bytes and this padding is not included in the chunk length field.
* The sender should never pad with more than 3 bytes. The receiver
* MUST ignore the padding bytes.
*/
padding = SCTP_PAD4(datalen) - datalen;
/* Fixup cloned skb with just this chunks data. */
skb_trim(skb, chunk->chunk_end - padding - skb->data);
/* Embed the event fields inside the cloned skb. */
event = sctp_skb2event(skb);
/* Initialize event with flags 0 and correct length
* Since this is a clone of the original skb, only account for
* the data of this chunk as other chunks will be accounted separately.
*/
sctp_ulpevent_init(event, 0, skb->len + sizeof(struct sk_buff));
sctp: allow GSO frags to access the chunk too SCTP will try to access original IP headers on sctp_recvmsg in order to copy the addresses used. There are also other places that do similar access to IP or even SCTP headers. But after 90017accff61 ("sctp: Add GSO support") they aren't always there because they are only present in the header skb. SCTP handles the queueing of incoming data by cloning the incoming skb and limiting to only the relevant payload. This clone has its cb updated to something different and it's then queued on socket rx queue. Thus we need to fix this in two moments. For rx path, not related to socket queue yet, this patch uses a partially copied sctp_input_cb to such GSO frags. This restores the ability to access the headers for this part of the code. Regarding the socket rx queue, it removes iif member from sctp_event and also add a chunk pointer on it. With these changes we're always able to reach the headers again. The biggest change here is that now the sctp_chunk struct and the original skb are only freed after the application consumed the buffer. Note however that the original payload was already like this due to the skb cloning. For iif, SCTP's IPv4 code doesn't use it, so no change is necessary. IPv6 now can fetch it directly from original's IPv6 CB as the original skb is still accessible. In the future we probably can simplify sctp_v*_skb_iif() stuff, as sctp_v4_skb_iif() was called but it's return value not used, and now it's not even called, but such cleanup is out of scope for this change. Fixes: 90017accff61 ("sctp: Add GSO support") Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-13 18:08:57 +00:00
/* And hold the chunk as we need it for getting the IP headers
* later in recvmsg
*/
sctp_chunk_hold(chunk);
event->chunk = chunk;
sctp_ulpevent_receive_data(event, asoc);
event->stream = ntohs(chunk->subh.data_hdr->stream);
if (chunk->chunk_hdr->flags & SCTP_DATA_UNORDERED) {
event->flags |= SCTP_UNORDERED;
event->cumtsn = sctp_tsnmap_get_ctsn(&asoc->peer.tsn_map);
}
event->tsn = ntohl(chunk->subh.data_hdr->tsn);
event->msg_flags |= chunk->chunk_hdr->flags;
return event;
fail_mark:
kfree_skb(skb);
fail:
return NULL;
}
/* Create a partial delivery related event.
*
* 5.3.1.7 SCTP_PARTIAL_DELIVERY_EVENT
*
* When a receiver is engaged in a partial delivery of a
* message this notification will be used to indicate
* various events.
*/
struct sctp_ulpevent *sctp_ulpevent_make_pdapi(
const struct sctp_association *asoc,
__u32 indication, __u32 sid, __u32 seq,
__u32 flags, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_pdapi_event *pd;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_pdapi_event),
MSG_NOTIFICATION, gfp);
if (!event)
goto fail;
skb = sctp_event2skb(event);
pd = skb_put(skb, sizeof(struct sctp_pdapi_event));
/* pdapi_type
* It should be SCTP_PARTIAL_DELIVERY_EVENT
*
* pdapi_flags: 16 bits (unsigned integer)
* Currently unused.
*/
pd->pdapi_type = SCTP_PARTIAL_DELIVERY_EVENT;
pd->pdapi_flags = flags;
pd->pdapi_stream = sid;
pd->pdapi_seq = seq;
/* pdapi_length: 32 bits (unsigned integer)
*
* This field is the total length of the notification data, including
* the notification header. It will generally be sizeof (struct
* sctp_pdapi_event).
*/
pd->pdapi_length = sizeof(struct sctp_pdapi_event);
/* pdapi_indication: 32 bits (unsigned integer)
*
* This field holds the indication being sent to the application.
*/
pd->pdapi_indication = indication;
/* pdapi_assoc_id: sizeof (sctp_assoc_t)
*
* The association id field, holds the identifier for the association.
*/
sctp_ulpevent_set_owner(event, asoc);
pd->pdapi_assoc_id = sctp_assoc2id(asoc);
return event;
fail:
return NULL;
}
struct sctp_ulpevent *sctp_ulpevent_make_authkey(
const struct sctp_association *asoc, __u16 key_id,
__u32 indication, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_authkey_event *ak;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_authkey_event),
MSG_NOTIFICATION, gfp);
if (!event)
goto fail;
skb = sctp_event2skb(event);
ak = skb_put(skb, sizeof(struct sctp_authkey_event));
ak->auth_type = SCTP_AUTHENTICATION_EVENT;
ak->auth_flags = 0;
ak->auth_length = sizeof(struct sctp_authkey_event);
ak->auth_keynumber = key_id;
ak->auth_altkeynumber = 0;
ak->auth_indication = indication;
/*
* The association id field, holds the identifier for the association.
*/
sctp_ulpevent_set_owner(event, asoc);
ak->auth_assoc_id = sctp_assoc2id(asoc);
return event;
fail:
return NULL;
}
/*
* Socket Extensions for SCTP
* 6.3.10. SCTP_SENDER_DRY_EVENT
*/
struct sctp_ulpevent *sctp_ulpevent_make_sender_dry_event(
const struct sctp_association *asoc, gfp_t gfp)
{
struct sctp_ulpevent *event;
struct sctp_sender_dry_event *sdry;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_sender_dry_event),
MSG_NOTIFICATION, gfp);
if (!event)
return NULL;
skb = sctp_event2skb(event);
sdry = skb_put(skb, sizeof(struct sctp_sender_dry_event));
sdry->sender_dry_type = SCTP_SENDER_DRY_EVENT;
sdry->sender_dry_flags = 0;
sdry->sender_dry_length = sizeof(struct sctp_sender_dry_event);
sctp_ulpevent_set_owner(event, asoc);
sdry->sender_dry_assoc_id = sctp_assoc2id(asoc);
return event;
}
struct sctp_ulpevent *sctp_ulpevent_make_stream_reset_event(
const struct sctp_association *asoc, __u16 flags, __u16 stream_num,
__be16 *stream_list, gfp_t gfp)
{
struct sctp_stream_reset_event *sreset;
struct sctp_ulpevent *event;
struct sk_buff *skb;
int length, i;
length = sizeof(struct sctp_stream_reset_event) + 2 * stream_num;
event = sctp_ulpevent_new(length, MSG_NOTIFICATION, gfp);
if (!event)
return NULL;
skb = sctp_event2skb(event);
sreset = skb_put(skb, length);
sreset->strreset_type = SCTP_STREAM_RESET_EVENT;
sreset->strreset_flags = flags;
sreset->strreset_length = length;
sctp_ulpevent_set_owner(event, asoc);
sreset->strreset_assoc_id = sctp_assoc2id(asoc);
for (i = 0; i < stream_num; i++)
sreset->strreset_stream_list[i] = ntohs(stream_list[i]);
return event;
}
struct sctp_ulpevent *sctp_ulpevent_make_assoc_reset_event(
const struct sctp_association *asoc, __u16 flags, __u32 local_tsn,
__u32 remote_tsn, gfp_t gfp)
{
struct sctp_assoc_reset_event *areset;
struct sctp_ulpevent *event;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_assoc_reset_event),
MSG_NOTIFICATION, gfp);
if (!event)
return NULL;
skb = sctp_event2skb(event);
areset = skb_put(skb, sizeof(struct sctp_assoc_reset_event));
areset->assocreset_type = SCTP_ASSOC_RESET_EVENT;
areset->assocreset_flags = flags;
areset->assocreset_length = sizeof(struct sctp_assoc_reset_event);
sctp_ulpevent_set_owner(event, asoc);
areset->assocreset_assoc_id = sctp_assoc2id(asoc);
areset->assocreset_local_tsn = local_tsn;
areset->assocreset_remote_tsn = remote_tsn;
return event;
}
struct sctp_ulpevent *sctp_ulpevent_make_stream_change_event(
const struct sctp_association *asoc, __u16 flags,
__u32 strchange_instrms, __u32 strchange_outstrms, gfp_t gfp)
{
struct sctp_stream_change_event *schange;
struct sctp_ulpevent *event;
struct sk_buff *skb;
event = sctp_ulpevent_new(sizeof(struct sctp_stream_change_event),
MSG_NOTIFICATION, gfp);
if (!event)
return NULL;
skb = sctp_event2skb(event);
schange = skb_put(skb, sizeof(struct sctp_stream_change_event));
schange->strchange_type = SCTP_STREAM_CHANGE_EVENT;
schange->strchange_flags = flags;
schange->strchange_length = sizeof(struct sctp_stream_change_event);
sctp_ulpevent_set_owner(event, asoc);
schange->strchange_assoc_id = sctp_assoc2id(asoc);
schange->strchange_instrms = strchange_instrms;
schange->strchange_outstrms = strchange_outstrms;
return event;
}
/* Return the notification type, assuming this is a notification
* event.
*/
__u16 sctp_ulpevent_get_notification_type(const struct sctp_ulpevent *event)
{
union sctp_notification *notification;
struct sk_buff *skb;
skb = sctp_event2skb(event);
notification = (union sctp_notification *) skb->data;
return notification->sn_header.sn_type;
}
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
/* RFC6458, Section 5.3.2. SCTP Header Information Structure
* (SCTP_SNDRCV, DEPRECATED)
*/
void sctp_ulpevent_read_sndrcvinfo(const struct sctp_ulpevent *event,
struct msghdr *msghdr)
{
struct sctp_sndrcvinfo sinfo;
if (sctp_ulpevent_is_notification(event))
return;
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
memset(&sinfo, 0, sizeof(sinfo));
sinfo.sinfo_stream = event->stream;
sinfo.sinfo_ssn = event->ssn;
sinfo.sinfo_ppid = event->ppid;
sinfo.sinfo_flags = event->flags;
sinfo.sinfo_tsn = event->tsn;
sinfo.sinfo_cumtsn = event->cumtsn;
sinfo.sinfo_assoc_id = sctp_assoc2id(event->asoc);
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
/* Context value that is set via SCTP_CONTEXT socket option. */
sinfo.sinfo_context = event->asoc->default_rcv_context;
/* These fields are not used while receiving. */
sinfo.sinfo_timetolive = 0;
put_cmsg(msghdr, IPPROTO_SCTP, SCTP_SNDRCV,
net: sctp: fix information leaks in ulpevent layer While working on some other SCTP code, I noticed that some structures shared with user space are leaking uninitialized stack or heap buffer. In particular, struct sctp_sndrcvinfo has a 2 bytes hole between .sinfo_flags and .sinfo_ppid that remains unfilled by us in sctp_ulpevent_read_sndrcvinfo() when putting this into cmsg. But also struct sctp_remote_error contains a 2 bytes hole that we don't fill but place into a skb through skb_copy_expand() via sctp_ulpevent_make_remote_error(). Both structures are defined by the IETF in RFC6458: * Section 5.3.2. SCTP Header Information Structure: The sctp_sndrcvinfo structure is defined below: struct sctp_sndrcvinfo { uint16_t sinfo_stream; uint16_t sinfo_ssn; uint16_t sinfo_flags; <-- 2 bytes hole --> uint32_t sinfo_ppid; uint32_t sinfo_context; uint32_t sinfo_timetolive; uint32_t sinfo_tsn; uint32_t sinfo_cumtsn; sctp_assoc_t sinfo_assoc_id; }; * 6.1.3. SCTP_REMOTE_ERROR: A remote peer may send an Operation Error message to its peer. This message indicates a variety of error conditions on an association. The entire ERROR chunk as it appears on the wire is included in an SCTP_REMOTE_ERROR event. Please refer to the SCTP specification [RFC4960] and any extensions for a list of possible error formats. An SCTP error notification has the following format: struct sctp_remote_error { uint16_t sre_type; uint16_t sre_flags; uint32_t sre_length; uint16_t sre_error; <-- 2 bytes hole --> sctp_assoc_t sre_assoc_id; uint8_t sre_data[]; }; Fix this by setting both to 0 before filling them out. We also have other structures shared between user and kernel space in SCTP that contains holes (e.g. struct sctp_paddrthlds), but we copy that buffer over from user space first and thus don't need to care about it in that cases. While at it, we can also remove lengthy comments copied from the draft, instead, we update the comment with the correct RFC number where one can look it up. Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-12 18:30:35 +00:00
sizeof(sinfo), &sinfo);
}
/* RFC6458, Section 5.3.5 SCTP Receive Information Structure
* (SCTP_SNDRCV)
*/
void sctp_ulpevent_read_rcvinfo(const struct sctp_ulpevent *event,
struct msghdr *msghdr)
{
struct sctp_rcvinfo rinfo;
if (sctp_ulpevent_is_notification(event))
return;
memset(&rinfo, 0, sizeof(struct sctp_rcvinfo));
rinfo.rcv_sid = event->stream;
rinfo.rcv_ssn = event->ssn;
rinfo.rcv_ppid = event->ppid;
rinfo.rcv_flags = event->flags;
rinfo.rcv_tsn = event->tsn;
rinfo.rcv_cumtsn = event->cumtsn;
rinfo.rcv_assoc_id = sctp_assoc2id(event->asoc);
rinfo.rcv_context = event->asoc->default_rcv_context;
put_cmsg(msghdr, IPPROTO_SCTP, SCTP_RCVINFO,
sizeof(rinfo), &rinfo);
}
/* RFC6458, Section 5.3.6. SCTP Next Receive Information Structure
* (SCTP_NXTINFO)
*/
static void __sctp_ulpevent_read_nxtinfo(const struct sctp_ulpevent *event,
struct msghdr *msghdr,
const struct sk_buff *skb)
{
struct sctp_nxtinfo nxtinfo;
memset(&nxtinfo, 0, sizeof(nxtinfo));
nxtinfo.nxt_sid = event->stream;
nxtinfo.nxt_ppid = event->ppid;
nxtinfo.nxt_flags = event->flags;
if (sctp_ulpevent_is_notification(event))
nxtinfo.nxt_flags |= SCTP_NOTIFICATION;
nxtinfo.nxt_length = skb->len;
nxtinfo.nxt_assoc_id = sctp_assoc2id(event->asoc);
put_cmsg(msghdr, IPPROTO_SCTP, SCTP_NXTINFO,
sizeof(nxtinfo), &nxtinfo);
}
void sctp_ulpevent_read_nxtinfo(const struct sctp_ulpevent *event,
struct msghdr *msghdr,
struct sock *sk)
{
struct sk_buff *skb;
int err;
skb = sctp_skb_recv_datagram(sk, MSG_PEEK | MSG_DONTWAIT, &err);
if (skb != NULL) {
__sctp_ulpevent_read_nxtinfo(sctp_skb2event(skb),
msghdr, skb);
/* Just release refcount here. */
kfree_skb(skb);
}
}
/* Do accounting for bytes received and hold a reference to the association
* for each skb.
*/
static void sctp_ulpevent_receive_data(struct sctp_ulpevent *event,
struct sctp_association *asoc)
{
struct sk_buff *skb, *frag;
skb = sctp_event2skb(event);
/* Set the owner and charge rwnd for bytes received. */
sctp_ulpevent_set_owner(event, asoc);
Revert "net: sctp: Fix a_rwnd/rwnd management to reflect real state of the receiver's buffer" This reverts commit ef2820a735f7 ("net: sctp: Fix a_rwnd/rwnd management to reflect real state of the receiver's buffer") as it introduced a serious performance regression on SCTP over IPv4 and IPv6, though a not as dramatic on the latter. Measurements are on 10Gbit/s with ixgbe NICs. Current state: [root@Lab200slot2 ~]# iperf3 --sctp -4 -c 192.168.241.3 -V -l 1452 -t 60 iperf version 3.0.1 (10 January 2014) Linux Lab200slot2 3.14.0 #1 SMP Thu Apr 3 23:18:29 EDT 2014 x86_64 Time: Fri, 11 Apr 2014 17:56:21 GMT Connecting to host 192.168.241.3, port 5201 Cookie: Lab200slot2.1397238981.812898.548918 [ 4] local 192.168.241.2 port 38616 connected to 192.168.241.3 port 5201 Starting Test: protocol: SCTP, 1 streams, 1452 byte blocks, omitting 0 seconds, 60 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.09 sec 20.8 MBytes 161 Mbits/sec [ 4] 1.09-2.13 sec 10.8 MBytes 86.8 Mbits/sec [ 4] 2.13-3.15 sec 3.57 MBytes 29.5 Mbits/sec [ 4] 3.15-4.16 sec 4.33 MBytes 35.7 Mbits/sec [ 4] 4.16-6.21 sec 10.4 MBytes 42.7 Mbits/sec [ 4] 6.21-6.21 sec 0.00 Bytes 0.00 bits/sec [ 4] 6.21-7.35 sec 34.6 MBytes 253 Mbits/sec [ 4] 7.35-11.45 sec 22.0 MBytes 45.0 Mbits/sec [ 4] 11.45-11.45 sec 0.00 Bytes 0.00 bits/sec [ 4] 11.45-11.45 sec 0.00 Bytes 0.00 bits/sec [ 4] 11.45-11.45 sec 0.00 Bytes 0.00 bits/sec [ 4] 11.45-12.51 sec 16.0 MBytes 126 Mbits/sec [ 4] 12.51-13.59 sec 20.3 MBytes 158 Mbits/sec [ 4] 13.59-14.65 sec 13.4 MBytes 107 Mbits/sec [ 4] 14.65-16.79 sec 33.3 MBytes 130 Mbits/sec [ 4] 16.79-16.79 sec 0.00 Bytes 0.00 bits/sec [ 4] 16.79-17.82 sec 5.94 MBytes 48.7 Mbits/sec (etc) [root@Lab200slot2 ~]# iperf3 --sctp -6 -c 2001:db8:0:f101::1 -V -l 1400 -t 60 iperf version 3.0.1 (10 January 2014) Linux Lab200slot2 3.14.0 #1 SMP Thu Apr 3 23:18:29 EDT 2014 x86_64 Time: Fri, 11 Apr 2014 19:08:41 GMT Connecting to host 2001:db8:0:f101::1, port 5201 Cookie: Lab200slot2.1397243321.714295.2b3f7c [ 4] local 2001:db8:0:f101::2 port 55804 connected to 2001:db8:0:f101::1 port 5201 Starting Test: protocol: SCTP, 1 streams, 1400 byte blocks, omitting 0 seconds, 60 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 169 MBytes 1.42 Gbits/sec [ 4] 1.00-2.00 sec 201 MBytes 1.69 Gbits/sec [ 4] 2.00-3.00 sec 188 MBytes 1.58 Gbits/sec [ 4] 3.00-4.00 sec 174 MBytes 1.46 Gbits/sec [ 4] 4.00-5.00 sec 165 MBytes 1.39 Gbits/sec [ 4] 5.00-6.00 sec 199 MBytes 1.67 Gbits/sec [ 4] 6.00-7.00 sec 163 MBytes 1.36 Gbits/sec [ 4] 7.00-8.00 sec 174 MBytes 1.46 Gbits/sec [ 4] 8.00-9.00 sec 193 MBytes 1.62 Gbits/sec [ 4] 9.00-10.00 sec 196 MBytes 1.65 Gbits/sec [ 4] 10.00-11.00 sec 157 MBytes 1.31 Gbits/sec [ 4] 11.00-12.00 sec 175 MBytes 1.47 Gbits/sec [ 4] 12.00-13.00 sec 192 MBytes 1.61 Gbits/sec [ 4] 13.00-14.00 sec 199 MBytes 1.67 Gbits/sec (etc) After patch: [root@Lab200slot2 ~]# iperf3 --sctp -4 -c 192.168.240.3 -V -l 1452 -t 60 iperf version 3.0.1 (10 January 2014) Linux Lab200slot2 3.14.0+ #1 SMP Mon Apr 14 12:06:40 EDT 2014 x86_64 Time: Mon, 14 Apr 2014 16:40:48 GMT Connecting to host 192.168.240.3, port 5201 Cookie: Lab200slot2.1397493648.413274.65e131 [ 4] local 192.168.240.2 port 50548 connected to 192.168.240.3 port 5201 Starting Test: protocol: SCTP, 1 streams, 1452 byte blocks, omitting 0 seconds, 60 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 240 MBytes 2.02 Gbits/sec [ 4] 1.00-2.00 sec 239 MBytes 2.01 Gbits/sec [ 4] 2.00-3.00 sec 240 MBytes 2.01 Gbits/sec [ 4] 3.00-4.00 sec 239 MBytes 2.00 Gbits/sec [ 4] 4.00-5.00 sec 245 MBytes 2.05 Gbits/sec [ 4] 5.00-6.00 sec 240 MBytes 2.01 Gbits/sec [ 4] 6.00-7.00 sec 240 MBytes 2.02 Gbits/sec [ 4] 7.00-8.00 sec 239 MBytes 2.01 Gbits/sec With the reverted patch applied, the SCTP/IPv4 performance is back to normal on latest upstream for IPv4 and IPv6 and has same throughput as 3.4.2 test kernel, steady and interval reports are smooth again. Fixes: ef2820a735f7 ("net: sctp: Fix a_rwnd/rwnd management to reflect real state of the receiver's buffer") Reported-by: Peter Butler <pbutler@sonusnet.com> Reported-by: Dongsheng Song <dongsheng.song@gmail.com> Reported-by: Fengguang Wu <fengguang.wu@intel.com> Tested-by: Peter Butler <pbutler@sonusnet.com> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Cc: Matija Glavinic Pecotic <matija.glavinic-pecotic.ext@nsn.com> Cc: Alexander Sverdlin <alexander.sverdlin@nsn.com> Cc: Vlad Yasevich <vyasevich@gmail.com> Acked-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 19:45:17 +00:00
sctp_assoc_rwnd_decrease(asoc, skb_headlen(skb));
if (!skb->data_len)
return;
/* Note: Not clearing the entire event struct as this is just a
* fragment of the real event. However, we still need to do rwnd
* accounting.
* In general, the skb passed from IP can have only 1 level of
* fragments. But we allow multiple levels of fragments.
*/
skb_walk_frags(skb, frag)
sctp_ulpevent_receive_data(sctp_skb2event(frag), asoc);
}
/* Do accounting for bytes just read by user and release the references to
* the association.
*/
static void sctp_ulpevent_release_data(struct sctp_ulpevent *event)
{
struct sk_buff *skb, *frag;
unsigned int len;
/* Current stack structures assume that the rcv buffer is
* per socket. For UDP style sockets this is not true as
* multiple associations may be on a single UDP-style socket.
* Use the local private area of the skb to track the owning
* association.
*/
skb = sctp_event2skb(event);
len = skb->len;
if (!skb->data_len)
goto done;
/* Don't forget the fragments. */
skb_walk_frags(skb, frag) {
/* NOTE: skb_shinfos are recursive. Although IP returns
* skb's with only 1 level of fragments, SCTP reassembly can
* increase the levels.
*/
sctp_ulpevent_release_frag_data(sctp_skb2event(frag));
}
done:
Revert "net: sctp: Fix a_rwnd/rwnd management to reflect real state of the receiver's buffer" This reverts commit ef2820a735f7 ("net: sctp: Fix a_rwnd/rwnd management to reflect real state of the receiver's buffer") as it introduced a serious performance regression on SCTP over IPv4 and IPv6, though a not as dramatic on the latter. Measurements are on 10Gbit/s with ixgbe NICs. Current state: [root@Lab200slot2 ~]# iperf3 --sctp -4 -c 192.168.241.3 -V -l 1452 -t 60 iperf version 3.0.1 (10 January 2014) Linux Lab200slot2 3.14.0 #1 SMP Thu Apr 3 23:18:29 EDT 2014 x86_64 Time: Fri, 11 Apr 2014 17:56:21 GMT Connecting to host 192.168.241.3, port 5201 Cookie: Lab200slot2.1397238981.812898.548918 [ 4] local 192.168.241.2 port 38616 connected to 192.168.241.3 port 5201 Starting Test: protocol: SCTP, 1 streams, 1452 byte blocks, omitting 0 seconds, 60 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.09 sec 20.8 MBytes 161 Mbits/sec [ 4] 1.09-2.13 sec 10.8 MBytes 86.8 Mbits/sec [ 4] 2.13-3.15 sec 3.57 MBytes 29.5 Mbits/sec [ 4] 3.15-4.16 sec 4.33 MBytes 35.7 Mbits/sec [ 4] 4.16-6.21 sec 10.4 MBytes 42.7 Mbits/sec [ 4] 6.21-6.21 sec 0.00 Bytes 0.00 bits/sec [ 4] 6.21-7.35 sec 34.6 MBytes 253 Mbits/sec [ 4] 7.35-11.45 sec 22.0 MBytes 45.0 Mbits/sec [ 4] 11.45-11.45 sec 0.00 Bytes 0.00 bits/sec [ 4] 11.45-11.45 sec 0.00 Bytes 0.00 bits/sec [ 4] 11.45-11.45 sec 0.00 Bytes 0.00 bits/sec [ 4] 11.45-12.51 sec 16.0 MBytes 126 Mbits/sec [ 4] 12.51-13.59 sec 20.3 MBytes 158 Mbits/sec [ 4] 13.59-14.65 sec 13.4 MBytes 107 Mbits/sec [ 4] 14.65-16.79 sec 33.3 MBytes 130 Mbits/sec [ 4] 16.79-16.79 sec 0.00 Bytes 0.00 bits/sec [ 4] 16.79-17.82 sec 5.94 MBytes 48.7 Mbits/sec (etc) [root@Lab200slot2 ~]# iperf3 --sctp -6 -c 2001:db8:0:f101::1 -V -l 1400 -t 60 iperf version 3.0.1 (10 January 2014) Linux Lab200slot2 3.14.0 #1 SMP Thu Apr 3 23:18:29 EDT 2014 x86_64 Time: Fri, 11 Apr 2014 19:08:41 GMT Connecting to host 2001:db8:0:f101::1, port 5201 Cookie: Lab200slot2.1397243321.714295.2b3f7c [ 4] local 2001:db8:0:f101::2 port 55804 connected to 2001:db8:0:f101::1 port 5201 Starting Test: protocol: SCTP, 1 streams, 1400 byte blocks, omitting 0 seconds, 60 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 169 MBytes 1.42 Gbits/sec [ 4] 1.00-2.00 sec 201 MBytes 1.69 Gbits/sec [ 4] 2.00-3.00 sec 188 MBytes 1.58 Gbits/sec [ 4] 3.00-4.00 sec 174 MBytes 1.46 Gbits/sec [ 4] 4.00-5.00 sec 165 MBytes 1.39 Gbits/sec [ 4] 5.00-6.00 sec 199 MBytes 1.67 Gbits/sec [ 4] 6.00-7.00 sec 163 MBytes 1.36 Gbits/sec [ 4] 7.00-8.00 sec 174 MBytes 1.46 Gbits/sec [ 4] 8.00-9.00 sec 193 MBytes 1.62 Gbits/sec [ 4] 9.00-10.00 sec 196 MBytes 1.65 Gbits/sec [ 4] 10.00-11.00 sec 157 MBytes 1.31 Gbits/sec [ 4] 11.00-12.00 sec 175 MBytes 1.47 Gbits/sec [ 4] 12.00-13.00 sec 192 MBytes 1.61 Gbits/sec [ 4] 13.00-14.00 sec 199 MBytes 1.67 Gbits/sec (etc) After patch: [root@Lab200slot2 ~]# iperf3 --sctp -4 -c 192.168.240.3 -V -l 1452 -t 60 iperf version 3.0.1 (10 January 2014) Linux Lab200slot2 3.14.0+ #1 SMP Mon Apr 14 12:06:40 EDT 2014 x86_64 Time: Mon, 14 Apr 2014 16:40:48 GMT Connecting to host 192.168.240.3, port 5201 Cookie: Lab200slot2.1397493648.413274.65e131 [ 4] local 192.168.240.2 port 50548 connected to 192.168.240.3 port 5201 Starting Test: protocol: SCTP, 1 streams, 1452 byte blocks, omitting 0 seconds, 60 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 240 MBytes 2.02 Gbits/sec [ 4] 1.00-2.00 sec 239 MBytes 2.01 Gbits/sec [ 4] 2.00-3.00 sec 240 MBytes 2.01 Gbits/sec [ 4] 3.00-4.00 sec 239 MBytes 2.00 Gbits/sec [ 4] 4.00-5.00 sec 245 MBytes 2.05 Gbits/sec [ 4] 5.00-6.00 sec 240 MBytes 2.01 Gbits/sec [ 4] 6.00-7.00 sec 240 MBytes 2.02 Gbits/sec [ 4] 7.00-8.00 sec 239 MBytes 2.01 Gbits/sec With the reverted patch applied, the SCTP/IPv4 performance is back to normal on latest upstream for IPv4 and IPv6 and has same throughput as 3.4.2 test kernel, steady and interval reports are smooth again. Fixes: ef2820a735f7 ("net: sctp: Fix a_rwnd/rwnd management to reflect real state of the receiver's buffer") Reported-by: Peter Butler <pbutler@sonusnet.com> Reported-by: Dongsheng Song <dongsheng.song@gmail.com> Reported-by: Fengguang Wu <fengguang.wu@intel.com> Tested-by: Peter Butler <pbutler@sonusnet.com> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Cc: Matija Glavinic Pecotic <matija.glavinic-pecotic.ext@nsn.com> Cc: Alexander Sverdlin <alexander.sverdlin@nsn.com> Cc: Vlad Yasevich <vyasevich@gmail.com> Acked-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 19:45:17 +00:00
sctp_assoc_rwnd_increase(event->asoc, len);
sctp: allow GSO frags to access the chunk too SCTP will try to access original IP headers on sctp_recvmsg in order to copy the addresses used. There are also other places that do similar access to IP or even SCTP headers. But after 90017accff61 ("sctp: Add GSO support") they aren't always there because they are only present in the header skb. SCTP handles the queueing of incoming data by cloning the incoming skb and limiting to only the relevant payload. This clone has its cb updated to something different and it's then queued on socket rx queue. Thus we need to fix this in two moments. For rx path, not related to socket queue yet, this patch uses a partially copied sctp_input_cb to such GSO frags. This restores the ability to access the headers for this part of the code. Regarding the socket rx queue, it removes iif member from sctp_event and also add a chunk pointer on it. With these changes we're always able to reach the headers again. The biggest change here is that now the sctp_chunk struct and the original skb are only freed after the application consumed the buffer. Note however that the original payload was already like this due to the skb cloning. For iif, SCTP's IPv4 code doesn't use it, so no change is necessary. IPv6 now can fetch it directly from original's IPv6 CB as the original skb is still accessible. In the future we probably can simplify sctp_v*_skb_iif() stuff, as sctp_v4_skb_iif() was called but it's return value not used, and now it's not even called, but such cleanup is out of scope for this change. Fixes: 90017accff61 ("sctp: Add GSO support") Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-13 18:08:57 +00:00
sctp_chunk_put(event->chunk);
sctp_ulpevent_release_owner(event);
}
static void sctp_ulpevent_release_frag_data(struct sctp_ulpevent *event)
{
struct sk_buff *skb, *frag;
skb = sctp_event2skb(event);
if (!skb->data_len)
goto done;
/* Don't forget the fragments. */
skb_walk_frags(skb, frag) {
/* NOTE: skb_shinfos are recursive. Although IP returns
* skb's with only 1 level of fragments, SCTP reassembly can
* increase the levels.
*/
sctp_ulpevent_release_frag_data(sctp_skb2event(frag));
}
done:
sctp: allow GSO frags to access the chunk too SCTP will try to access original IP headers on sctp_recvmsg in order to copy the addresses used. There are also other places that do similar access to IP or even SCTP headers. But after 90017accff61 ("sctp: Add GSO support") they aren't always there because they are only present in the header skb. SCTP handles the queueing of incoming data by cloning the incoming skb and limiting to only the relevant payload. This clone has its cb updated to something different and it's then queued on socket rx queue. Thus we need to fix this in two moments. For rx path, not related to socket queue yet, this patch uses a partially copied sctp_input_cb to such GSO frags. This restores the ability to access the headers for this part of the code. Regarding the socket rx queue, it removes iif member from sctp_event and also add a chunk pointer on it. With these changes we're always able to reach the headers again. The biggest change here is that now the sctp_chunk struct and the original skb are only freed after the application consumed the buffer. Note however that the original payload was already like this due to the skb cloning. For iif, SCTP's IPv4 code doesn't use it, so no change is necessary. IPv6 now can fetch it directly from original's IPv6 CB as the original skb is still accessible. In the future we probably can simplify sctp_v*_skb_iif() stuff, as sctp_v4_skb_iif() was called but it's return value not used, and now it's not even called, but such cleanup is out of scope for this change. Fixes: 90017accff61 ("sctp: Add GSO support") Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-13 18:08:57 +00:00
sctp_chunk_put(event->chunk);
sctp_ulpevent_release_owner(event);
}
/* Free a ulpevent that has an owner. It includes releasing the reference
* to the owner, updating the rwnd in case of a DATA event and freeing the
* skb.
*/
void sctp_ulpevent_free(struct sctp_ulpevent *event)
{
if (sctp_ulpevent_is_notification(event))
sctp_ulpevent_release_owner(event);
else
sctp_ulpevent_release_data(event);
kfree_skb(sctp_event2skb(event));
}
/* Purge the skb lists holding ulpevents. */
unsigned int sctp_queue_purge_ulpevents(struct sk_buff_head *list)
{
struct sk_buff *skb;
unsigned int data_unread = 0;
while ((skb = skb_dequeue(list)) != NULL) {
struct sctp_ulpevent *event = sctp_skb2event(skb);
if (!sctp_ulpevent_is_notification(event))
data_unread += skb->len;
sctp_ulpevent_free(event);
}
return data_unread;
}