linux/net/sctp/protocol.c

1621 lines
43 KiB
C
Raw Normal View History

/* 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
*
* This file is part of the SCTP kernel implementation
*
* Initialization/cleanup for SCTP protocol support.
*
* This SCTP implementation is free software;
* you can redistribute it and/or modify it under the terms of
* the GNU General Public License as published by
* the Free Software Foundation; either version 2, or (at your option)
* any later version.
*
* This SCTP implementation is distributed in the hope that it
* will be useful, but WITHOUT ANY WARRANTY; without even the implied
* ************************
* warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
* See the GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with GNU CC; see the file COPYING. If not, see
* <http://www.gnu.org/licenses/>.
*
* 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@us.ibm.com>
* Sridhar Samudrala <sri@us.ibm.com>
* Daisy Chang <daisyc@us.ibm.com>
* Ardelle Fan <ardelle.fan@intel.com>
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/module.h>
#include <linux/init.h>
#include <linux/netdevice.h>
#include <linux/inetdevice.h>
#include <linux/seq_file.h>
mm: remove include/linux/bootmem.h Move remaining definitions and declarations from include/linux/bootmem.h into include/linux/memblock.h and remove the redundant header. The includes were replaced with the semantic patch below and then semi-automated removal of duplicated '#include <linux/memblock.h> @@ @@ - #include <linux/bootmem.h> + #include <linux/memblock.h> [sfr@canb.auug.org.au: dma-direct: fix up for the removal of linux/bootmem.h] Link: http://lkml.kernel.org/r/20181002185342.133d1680@canb.auug.org.au [sfr@canb.auug.org.au: powerpc: fix up for removal of linux/bootmem.h] Link: http://lkml.kernel.org/r/20181005161406.73ef8727@canb.auug.org.au [sfr@canb.auug.org.au: x86/kaslr, ACPI/NUMA: fix for linux/bootmem.h removal] Link: http://lkml.kernel.org/r/20181008190341.5e396491@canb.auug.org.au Link: http://lkml.kernel.org/r/1536927045-23536-30-git-send-email-rppt@linux.vnet.ibm.com Signed-off-by: Mike Rapoport <rppt@linux.vnet.ibm.com> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Chris Zankel <chris@zankel.net> Cc: "David S. Miller" <davem@davemloft.net> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Greentime Hu <green.hu@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Guan Xuetao <gxt@pku.edu.cn> Cc: Ingo Molnar <mingo@redhat.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Jonas Bonn <jonas@southpole.se> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Ley Foon Tan <lftan@altera.com> Cc: Mark Salter <msalter@redhat.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Michal Simek <monstr@monstr.eu> Cc: Palmer Dabbelt <palmer@sifive.com> Cc: Paul Burton <paul.burton@mips.com> Cc: Richard Kuo <rkuo@codeaurora.org> Cc: Richard Weinberger <richard@nod.at> Cc: Rich Felker <dalias@libc.org> Cc: Russell King <linux@armlinux.org.uk> Cc: Serge Semin <fancer.lancer@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tony Luck <tony.luck@intel.com> Cc: Vineet Gupta <vgupta@synopsys.com> Cc: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-10-30 22:09:49 +00:00
#include <linux/memblock.h>
#include <linux/highmem.h>
#include <linux/swap.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <net/net_namespace.h>
#include <net/protocol.h>
#include <net/ip.h>
#include <net/ipv6.h>
#include <net/route.h>
#include <net/sctp/sctp.h>
#include <net/addrconf.h>
#include <net/inet_common.h>
#include <net/inet_ecn.h>
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
#define MAX_SCTP_PORT_HASH_ENTRIES (64 * 1024)
/* Global data structures. */
struct sctp_globals sctp_globals __read_mostly;
struct idr sctp_assocs_id;
DEFINE_SPINLOCK(sctp_assocs_id_lock);
static struct sctp_pf *sctp_pf_inet6_specific;
static struct sctp_pf *sctp_pf_inet_specific;
static struct sctp_af *sctp_af_v4_specific;
static struct sctp_af *sctp_af_v6_specific;
struct kmem_cache *sctp_chunk_cachep __read_mostly;
struct kmem_cache *sctp_bucket_cachep __read_mostly;
long sysctl_sctp_mem[3];
int sysctl_sctp_rmem[3];
int sysctl_sctp_wmem[3];
/* Private helper to extract ipv4 address and stash them in
* the protocol structure.
*/
static void sctp_v4_copy_addrlist(struct list_head *addrlist,
struct net_device *dev)
{
struct in_device *in_dev;
struct in_ifaddr *ifa;
struct sctp_sockaddr_entry *addr;
rcu_read_lock();
if ((in_dev = __in_dev_get_rcu(dev)) == NULL) {
rcu_read_unlock();
return;
}
for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next) {
/* Add the address to the local list. */
addr = kzalloc(sizeof(*addr), GFP_ATOMIC);
if (addr) {
addr->a.v4.sin_family = AF_INET;
addr->a.v4.sin_port = 0;
addr->a.v4.sin_addr.s_addr = ifa->ifa_local;
addr->valid = 1;
INIT_LIST_HEAD(&addr->list);
list_add_tail(&addr->list, addrlist);
}
}
rcu_read_unlock();
}
/* Extract our IP addresses from the system and stash them in the
* protocol structure.
*/
static void sctp_get_local_addr_list(struct net *net)
{
struct net_device *dev;
struct list_head *pos;
struct sctp_af *af;
rcu_read_lock();
for_each_netdev_rcu(net, dev) {
list_for_each(pos, &sctp_address_families) {
af = list_entry(pos, struct sctp_af, list);
af->copy_addrlist(&net->sctp.local_addr_list, dev);
}
}
rcu_read_unlock();
}
/* Free the existing local addresses. */
static void sctp_free_local_addr_list(struct net *net)
{
struct sctp_sockaddr_entry *addr;
struct list_head *pos, *temp;
list_for_each_safe(pos, temp, &net->sctp.local_addr_list) {
addr = list_entry(pos, struct sctp_sockaddr_entry, list);
list_del(pos);
kfree(addr);
}
}
/* Copy the local addresses which are valid for 'scope' into 'bp'. */
int sctp_copy_local_addr_list(struct net *net, struct sctp_bind_addr *bp,
enum sctp_scope scope, gfp_t gfp, int copy_flags)
{
struct sctp_sockaddr_entry *addr;
union sctp_addr laddr;
int error = 0;
rcu_read_lock();
list_for_each_entry_rcu(addr, &net->sctp.local_addr_list, list) {
if (!addr->valid)
continue;
if (!sctp_in_scope(net, &addr->a, scope))
continue;
/* Now that the address is in scope, check to see if
* the address type is really supported by the local
* sock as well as the remote peer.
*/
if (addr->a.sa.sa_family == AF_INET &&
!(copy_flags & SCTP_ADDR4_PEERSUPP))
continue;
if (addr->a.sa.sa_family == AF_INET6 &&
(!(copy_flags & SCTP_ADDR6_ALLOWED) ||
!(copy_flags & SCTP_ADDR6_PEERSUPP)))
continue;
laddr = addr->a;
/* also works for setting ipv6 address port */
laddr.v4.sin_port = htons(bp->port);
if (sctp_bind_addr_state(bp, &laddr) != -1)
continue;
error = sctp_add_bind_addr(bp, &addr->a, sizeof(addr->a),
SCTP_ADDR_SRC, GFP_ATOMIC);
if (error)
break;
}
rcu_read_unlock();
return error;
}
/* Copy over any ip options */
static void sctp_v4_copy_ip_options(struct sock *sk, struct sock *newsk)
{
struct inet_sock *newinet, *inet = inet_sk(sk);
struct ip_options_rcu *inet_opt, *newopt = NULL;
newinet = inet_sk(newsk);
rcu_read_lock();
inet_opt = rcu_dereference(inet->inet_opt);
if (inet_opt) {
newopt = sock_kmalloc(newsk, sizeof(*inet_opt) +
inet_opt->opt.optlen, GFP_ATOMIC);
if (newopt)
memcpy(newopt, inet_opt, sizeof(*inet_opt) +
inet_opt->opt.optlen);
else
pr_err("%s: Failed to copy ip options\n", __func__);
}
RCU_INIT_POINTER(newinet->inet_opt, newopt);
rcu_read_unlock();
}
/* Account for the IP options */
static int sctp_v4_ip_options_len(struct sock *sk)
{
struct inet_sock *inet = inet_sk(sk);
struct ip_options_rcu *inet_opt;
int len = 0;
rcu_read_lock();
inet_opt = rcu_dereference(inet->inet_opt);
if (inet_opt)
len = inet_opt->opt.optlen;
rcu_read_unlock();
return len;
}
/* Initialize a sctp_addr from in incoming skb. */
static void sctp_v4_from_skb(union sctp_addr *addr, struct sk_buff *skb,
int is_saddr)
{
/* Always called on head skb, so this is safe */
struct sctphdr *sh = sctp_hdr(skb);
struct sockaddr_in *sa = &addr->v4;
addr->v4.sin_family = AF_INET;
if (is_saddr) {
sa->sin_port = sh->source;
sa->sin_addr.s_addr = ip_hdr(skb)->saddr;
} else {
sa->sin_port = sh->dest;
sa->sin_addr.s_addr = ip_hdr(skb)->daddr;
}
}
/* Initialize an sctp_addr from a socket. */
static void sctp_v4_from_sk(union sctp_addr *addr, struct sock *sk)
{
addr->v4.sin_family = AF_INET;
addr->v4.sin_port = 0;
addr->v4.sin_addr.s_addr = inet_sk(sk)->inet_rcv_saddr;
}
/* Initialize sk->sk_rcv_saddr from sctp_addr. */
static void sctp_v4_to_sk_saddr(union sctp_addr *addr, struct sock *sk)
{
inet_sk(sk)->inet_rcv_saddr = addr->v4.sin_addr.s_addr;
}
/* Initialize sk->sk_daddr from sctp_addr. */
static void sctp_v4_to_sk_daddr(union sctp_addr *addr, struct sock *sk)
{
inet_sk(sk)->inet_daddr = addr->v4.sin_addr.s_addr;
}
/* Initialize a sctp_addr from an address parameter. */
static void sctp_v4_from_addr_param(union sctp_addr *addr,
union sctp_addr_param *param,
__be16 port, int iif)
{
addr->v4.sin_family = AF_INET;
addr->v4.sin_port = port;
addr->v4.sin_addr.s_addr = param->v4.addr.s_addr;
}
/* Initialize an address parameter from a sctp_addr and return the length
* of the address parameter.
*/
static int sctp_v4_to_addr_param(const union sctp_addr *addr,
union sctp_addr_param *param)
{
int length = sizeof(struct sctp_ipv4addr_param);
param->v4.param_hdr.type = SCTP_PARAM_IPV4_ADDRESS;
param->v4.param_hdr.length = htons(length);
param->v4.addr.s_addr = addr->v4.sin_addr.s_addr;
return length;
}
/* Initialize a sctp_addr from a dst_entry. */
static void sctp_v4_dst_saddr(union sctp_addr *saddr, struct flowi4 *fl4,
__be16 port)
{
saddr->v4.sin_family = AF_INET;
saddr->v4.sin_port = port;
saddr->v4.sin_addr.s_addr = fl4->saddr;
}
/* Compare two addresses exactly. */
static int sctp_v4_cmp_addr(const union sctp_addr *addr1,
const union sctp_addr *addr2)
{
if (addr1->sa.sa_family != addr2->sa.sa_family)
return 0;
if (addr1->v4.sin_port != addr2->v4.sin_port)
return 0;
if (addr1->v4.sin_addr.s_addr != addr2->v4.sin_addr.s_addr)
return 0;
return 1;
}
/* Initialize addr struct to INADDR_ANY. */
static void sctp_v4_inaddr_any(union sctp_addr *addr, __be16 port)
{
addr->v4.sin_family = AF_INET;
addr->v4.sin_addr.s_addr = htonl(INADDR_ANY);
addr->v4.sin_port = port;
}
/* Is this a wildcard address? */
static int sctp_v4_is_any(const union sctp_addr *addr)
{
return htonl(INADDR_ANY) == addr->v4.sin_addr.s_addr;
}
/* This function checks if the address is a valid address to be used for
* SCTP binding.
*
* Output:
* Return 0 - If the address is a non-unicast or an illegal address.
* Return 1 - If the address is a unicast.
*/
static int sctp_v4_addr_valid(union sctp_addr *addr,
struct sctp_sock *sp,
const struct sk_buff *skb)
{
/* IPv4 addresses not allowed */
if (sp && ipv6_only_sock(sctp_opt2sk(sp)))
return 0;
/* Is this a non-unicast address or a unusable SCTP address? */
if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr))
return 0;
/* Is this a broadcast address? */
if (skb && skb_rtable(skb)->rt_flags & RTCF_BROADCAST)
return 0;
return 1;
}
/* Should this be available for binding? */
static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp)
{
struct net *net = sock_net(&sp->inet.sk);
int ret = inet_addr_type(net, addr->v4.sin_addr.s_addr);
if (addr->v4.sin_addr.s_addr != htonl(INADDR_ANY) &&
ret != RTN_LOCAL &&
!sp->inet.freebind &&
!net->ipv4.sysctl_ip_nonlocal_bind)
return 0;
if (ipv6_only_sock(sctp_opt2sk(sp)))
return 0;
return 1;
}
/* Checking the loopback, private and other address scopes as defined in
* RFC 1918. The IPv4 scoping is based on the draft for SCTP IPv4
* scoping <draft-stewart-tsvwg-sctp-ipv4-00.txt>.
*
* Level 0 - unusable SCTP addresses
* Level 1 - loopback address
* Level 2 - link-local addresses
* Level 3 - private addresses.
* Level 4 - global addresses
* For INIT and INIT-ACK address list, let L be the level of
* of requested destination address, sender and receiver
* SHOULD include all of its addresses with level greater
* than or equal to L.
*
* IPv4 scoping can be controlled through sysctl option
* net.sctp.addr_scope_policy
*/
static enum sctp_scope sctp_v4_scope(union sctp_addr *addr)
{
enum sctp_scope retval;
/* Check for unusable SCTP addresses. */
if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr)) {
retval = SCTP_SCOPE_UNUSABLE;
} else if (ipv4_is_loopback(addr->v4.sin_addr.s_addr)) {
retval = SCTP_SCOPE_LOOPBACK;
} else if (ipv4_is_linklocal_169(addr->v4.sin_addr.s_addr)) {
retval = SCTP_SCOPE_LINK;
} else if (ipv4_is_private_10(addr->v4.sin_addr.s_addr) ||
ipv4_is_private_172(addr->v4.sin_addr.s_addr) ||
ipv4_is_private_192(addr->v4.sin_addr.s_addr)) {
retval = SCTP_SCOPE_PRIVATE;
} else {
retval = SCTP_SCOPE_GLOBAL;
}
return retval;
}
/* Returns a valid dst cache entry for the given source and destination ip
* addresses. If an association is passed, trys to get a dst entry with a
* source address that matches an address in the bind address list.
*/
static void sctp_v4_get_dst(struct sctp_transport *t, union sctp_addr *saddr,
struct flowi *fl, struct sock *sk)
{
struct sctp_association *asoc = t->asoc;
struct rtable *rt;
struct flowi4 *fl4 = &fl->u.ip4;
struct sctp_bind_addr *bp;
struct sctp_sockaddr_entry *laddr;
struct dst_entry *dst = NULL;
union sctp_addr *daddr = &t->ipaddr;
union sctp_addr dst_saddr;
__u8 tos = inet_sk(sk)->tos;
if (t->dscp & SCTP_DSCP_SET_MASK)
tos = t->dscp & SCTP_DSCP_VAL_MASK;
memset(fl4, 0x0, sizeof(struct flowi4));
fl4->daddr = daddr->v4.sin_addr.s_addr;
fl4->fl4_dport = daddr->v4.sin_port;
fl4->flowi4_proto = IPPROTO_SCTP;
if (asoc) {
fl4->flowi4_tos = RT_CONN_FLAGS_TOS(asoc->base.sk, tos);
fl4->flowi4_oif = asoc->base.sk->sk_bound_dev_if;
fl4->fl4_sport = htons(asoc->base.bind_addr.port);
}
if (saddr) {
fl4->saddr = saddr->v4.sin_addr.s_addr;
fl4->fl4_sport = saddr->v4.sin_port;
}
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>
2013-06-28 17:49:40 +00:00
pr_debug("%s: dst:%pI4, src:%pI4 - ", __func__, &fl4->daddr,
&fl4->saddr);
rt = ip_route_output_key(sock_net(sk), fl4);
if (!IS_ERR(rt))
dst = &rt->dst;
/* If there is no association or if a source address is passed, no
* more validation is required.
*/
if (!asoc || saddr)
goto out;
bp = &asoc->base.bind_addr;
if (dst) {
/* Walk through the bind address list and look for a bind
* address that matches the source address of the returned dst.
*/
sctp_v4_dst_saddr(&dst_saddr, fl4, htons(bp->port));
rcu_read_lock();
list_for_each_entry_rcu(laddr, &bp->address_list, list) {
if (!laddr->valid || (laddr->state == SCTP_ADDR_DEL) ||
(laddr->state != SCTP_ADDR_SRC &&
!asoc->src_out_of_asoc_ok))
continue;
if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a))
goto out_unlock;
}
rcu_read_unlock();
/* None of the bound addresses match the source address of the
* dst. So release it.
*/
dst_release(dst);
dst = NULL;
}
/* Walk through the bind address list and try to get a dst that
* matches a bind address as the source address.
*/
rcu_read_lock();
list_for_each_entry_rcu(laddr, &bp->address_list, list) {
sctp: fix src address selection if using secondary addresses In short, sctp is likely to incorrectly choose src address if socket is bound to secondary addresses. This patch fixes it by adding a new check that checks if such src address belongs to the interface that routing identified as output. This is enough to avoid rp_filter drops on remote peer. Details: Currently, sctp will do a routing attempt without specifying the src address and compare the returned value (preferred source) with the addresses that the socket is bound to. When using secondary addresses, this will not match. Then it will try specifying each of the addresses that the socket is bound to and re-routing, checking if that address is valid as src for that dst. Thing is, this check alone is weak: # ip r l 192.168.100.0/24 dev eth1 proto kernel scope link src 192.168.100.149 192.168.122.0/24 dev eth0 proto kernel scope link src 192.168.122.147 # ip a l 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:15:18:6a brd ff:ff:ff:ff:ff:ff inet 192.168.122.147/24 brd 192.168.122.255 scope global dynamic eth0 valid_lft 2160sec preferred_lft 2160sec inet 192.168.122.148/24 scope global secondary eth0 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:fe15:186a/64 scope link valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:b3:91:46 brd ff:ff:ff:ff:ff:ff inet 192.168.100.149/24 brd 192.168.100.255 scope global dynamic eth1 valid_lft 2162sec preferred_lft 2162sec inet 192.168.100.148/24 scope global secondary eth1 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:feb3:9146/64 scope link valid_lft forever preferred_lft forever 4: ens9: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:05:47:ee brd ff:ff:ff:ff:ff:ff inet6 fe80::5054:ff:fe05:47ee/64 scope link valid_lft forever preferred_lft forever # ip r g 192.168.100.193 from 192.168.122.148 192.168.100.193 from 192.168.122.148 dev eth1 cache Even if you specify an interface: # ip r g 192.168.100.193 from 192.168.122.148 oif eth1 192.168.100.193 from 192.168.122.148 dev eth1 cache Although this would be valid, peers using rp_filter will drop such packets as their src doesn't match the routes for that interface. Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-17 15:34:18 +00:00
struct net_device *odev;
if (!laddr->valid)
continue;
if (laddr->state != SCTP_ADDR_SRC ||
AF_INET != laddr->a.sa.sa_family)
continue;
fl4->fl4_sport = laddr->a.v4.sin_port;
flowi4_update_output(fl4,
asoc->base.sk->sk_bound_dev_if,
RT_CONN_FLAGS_TOS(asoc->base.sk, tos),
daddr->v4.sin_addr.s_addr,
laddr->a.v4.sin_addr.s_addr);
rt = ip_route_output_key(sock_net(sk), fl4);
if (IS_ERR(rt))
continue;
sctp: fix src address selection if using secondary addresses In short, sctp is likely to incorrectly choose src address if socket is bound to secondary addresses. This patch fixes it by adding a new check that checks if such src address belongs to the interface that routing identified as output. This is enough to avoid rp_filter drops on remote peer. Details: Currently, sctp will do a routing attempt without specifying the src address and compare the returned value (preferred source) with the addresses that the socket is bound to. When using secondary addresses, this will not match. Then it will try specifying each of the addresses that the socket is bound to and re-routing, checking if that address is valid as src for that dst. Thing is, this check alone is weak: # ip r l 192.168.100.0/24 dev eth1 proto kernel scope link src 192.168.100.149 192.168.122.0/24 dev eth0 proto kernel scope link src 192.168.122.147 # ip a l 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:15:18:6a brd ff:ff:ff:ff:ff:ff inet 192.168.122.147/24 brd 192.168.122.255 scope global dynamic eth0 valid_lft 2160sec preferred_lft 2160sec inet 192.168.122.148/24 scope global secondary eth0 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:fe15:186a/64 scope link valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:b3:91:46 brd ff:ff:ff:ff:ff:ff inet 192.168.100.149/24 brd 192.168.100.255 scope global dynamic eth1 valid_lft 2162sec preferred_lft 2162sec inet 192.168.100.148/24 scope global secondary eth1 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:feb3:9146/64 scope link valid_lft forever preferred_lft forever 4: ens9: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:05:47:ee brd ff:ff:ff:ff:ff:ff inet6 fe80::5054:ff:fe05:47ee/64 scope link valid_lft forever preferred_lft forever # ip r g 192.168.100.193 from 192.168.122.148 192.168.100.193 from 192.168.122.148 dev eth1 cache Even if you specify an interface: # ip r g 192.168.100.193 from 192.168.122.148 oif eth1 192.168.100.193 from 192.168.122.148 dev eth1 cache Although this would be valid, peers using rp_filter will drop such packets as their src doesn't match the routes for that interface. Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-17 15:34:18 +00:00
/* Ensure the src address belongs to the output
* interface.
*/
odev = __ip_dev_find(sock_net(sk), laddr->a.v4.sin_addr.s_addr,
false);
if (!odev || odev->ifindex != fl4->flowi4_oif) {
sctp: fix dst refcnt leak in sctp_v4_get_dst Fix dst reference count leak in sctp_v4_get_dst() introduced in commit 410f03831 ("sctp: add routing output fallback"): When walking the address_list, successive ip_route_output_key() calls may return the same rt->dst with the reference incremented on each call. The code would not decrement the dst refcount when the dst pointer was identical from the previous iteration, causing the dst refcnt leak. Testcase: ip netns add TEST ip netns exec TEST ip link set lo up ip link add dummy0 type dummy ip link add dummy1 type dummy ip link add dummy2 type dummy ip link set dev dummy0 netns TEST ip link set dev dummy1 netns TEST ip link set dev dummy2 netns TEST ip netns exec TEST ip addr add 192.168.1.1/24 dev dummy0 ip netns exec TEST ip link set dummy0 up ip netns exec TEST ip addr add 192.168.1.2/24 dev dummy1 ip netns exec TEST ip link set dummy1 up ip netns exec TEST ip addr add 192.168.1.3/24 dev dummy2 ip netns exec TEST ip link set dummy2 up ip netns exec TEST sctp_test -H 192.168.1.2 -P 20002 -h 192.168.1.1 -p 20000 -s -B 192.168.1.3 ip netns del TEST In 4.4 and 4.9 kernels this results to: [ 354.179591] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 364.419674] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 374.663664] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 384.903717] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 395.143724] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 405.383645] unregister_netdevice: waiting for lo to become free. Usage count = 1 ... Fixes: 410f03831 ("sctp: add routing output fallback") Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses") Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com> Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-05 19:48:14 +00:00
if (!dst)
dst = &rt->dst;
else
dst_release(&rt->dst);
sctp: fix src address selection if using secondary addresses In short, sctp is likely to incorrectly choose src address if socket is bound to secondary addresses. This patch fixes it by adding a new check that checks if such src address belongs to the interface that routing identified as output. This is enough to avoid rp_filter drops on remote peer. Details: Currently, sctp will do a routing attempt without specifying the src address and compare the returned value (preferred source) with the addresses that the socket is bound to. When using secondary addresses, this will not match. Then it will try specifying each of the addresses that the socket is bound to and re-routing, checking if that address is valid as src for that dst. Thing is, this check alone is weak: # ip r l 192.168.100.0/24 dev eth1 proto kernel scope link src 192.168.100.149 192.168.122.0/24 dev eth0 proto kernel scope link src 192.168.122.147 # ip a l 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:15:18:6a brd ff:ff:ff:ff:ff:ff inet 192.168.122.147/24 brd 192.168.122.255 scope global dynamic eth0 valid_lft 2160sec preferred_lft 2160sec inet 192.168.122.148/24 scope global secondary eth0 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:fe15:186a/64 scope link valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:b3:91:46 brd ff:ff:ff:ff:ff:ff inet 192.168.100.149/24 brd 192.168.100.255 scope global dynamic eth1 valid_lft 2162sec preferred_lft 2162sec inet 192.168.100.148/24 scope global secondary eth1 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:feb3:9146/64 scope link valid_lft forever preferred_lft forever 4: ens9: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:05:47:ee brd ff:ff:ff:ff:ff:ff inet6 fe80::5054:ff:fe05:47ee/64 scope link valid_lft forever preferred_lft forever # ip r g 192.168.100.193 from 192.168.122.148 192.168.100.193 from 192.168.122.148 dev eth1 cache Even if you specify an interface: # ip r g 192.168.100.193 from 192.168.122.148 oif eth1 192.168.100.193 from 192.168.122.148 dev eth1 cache Although this would be valid, peers using rp_filter will drop such packets as their src doesn't match the routes for that interface. Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-17 15:34:18 +00:00
continue;
}
sctp: fix src address selection if using secondary addresses In short, sctp is likely to incorrectly choose src address if socket is bound to secondary addresses. This patch fixes it by adding a new check that checks if such src address belongs to the interface that routing identified as output. This is enough to avoid rp_filter drops on remote peer. Details: Currently, sctp will do a routing attempt without specifying the src address and compare the returned value (preferred source) with the addresses that the socket is bound to. When using secondary addresses, this will not match. Then it will try specifying each of the addresses that the socket is bound to and re-routing, checking if that address is valid as src for that dst. Thing is, this check alone is weak: # ip r l 192.168.100.0/24 dev eth1 proto kernel scope link src 192.168.100.149 192.168.122.0/24 dev eth0 proto kernel scope link src 192.168.122.147 # ip a l 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:15:18:6a brd ff:ff:ff:ff:ff:ff inet 192.168.122.147/24 brd 192.168.122.255 scope global dynamic eth0 valid_lft 2160sec preferred_lft 2160sec inet 192.168.122.148/24 scope global secondary eth0 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:fe15:186a/64 scope link valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:b3:91:46 brd ff:ff:ff:ff:ff:ff inet 192.168.100.149/24 brd 192.168.100.255 scope global dynamic eth1 valid_lft 2162sec preferred_lft 2162sec inet 192.168.100.148/24 scope global secondary eth1 valid_lft forever preferred_lft forever inet6 fe80::5054:ff:feb3:9146/64 scope link valid_lft forever preferred_lft forever 4: ens9: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:05:47:ee brd ff:ff:ff:ff:ff:ff inet6 fe80::5054:ff:fe05:47ee/64 scope link valid_lft forever preferred_lft forever # ip r g 192.168.100.193 from 192.168.122.148 192.168.100.193 from 192.168.122.148 dev eth1 cache Even if you specify an interface: # ip r g 192.168.100.193 from 192.168.122.148 oif eth1 192.168.100.193 from 192.168.122.148 dev eth1 cache Although this would be valid, peers using rp_filter will drop such packets as their src doesn't match the routes for that interface. Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-17 15:34:18 +00:00
sctp: fix dst refcnt leak in sctp_v4_get_dst Fix dst reference count leak in sctp_v4_get_dst() introduced in commit 410f03831 ("sctp: add routing output fallback"): When walking the address_list, successive ip_route_output_key() calls may return the same rt->dst with the reference incremented on each call. The code would not decrement the dst refcount when the dst pointer was identical from the previous iteration, causing the dst refcnt leak. Testcase: ip netns add TEST ip netns exec TEST ip link set lo up ip link add dummy0 type dummy ip link add dummy1 type dummy ip link add dummy2 type dummy ip link set dev dummy0 netns TEST ip link set dev dummy1 netns TEST ip link set dev dummy2 netns TEST ip netns exec TEST ip addr add 192.168.1.1/24 dev dummy0 ip netns exec TEST ip link set dummy0 up ip netns exec TEST ip addr add 192.168.1.2/24 dev dummy1 ip netns exec TEST ip link set dummy1 up ip netns exec TEST ip addr add 192.168.1.3/24 dev dummy2 ip netns exec TEST ip link set dummy2 up ip netns exec TEST sctp_test -H 192.168.1.2 -P 20002 -h 192.168.1.1 -p 20000 -s -B 192.168.1.3 ip netns del TEST In 4.4 and 4.9 kernels this results to: [ 354.179591] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 364.419674] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 374.663664] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 384.903717] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 395.143724] unregister_netdevice: waiting for lo to become free. Usage count = 1 [ 405.383645] unregister_netdevice: waiting for lo to become free. Usage count = 1 ... Fixes: 410f03831 ("sctp: add routing output fallback") Fixes: 0ca50d12f ("sctp: fix src address selection if using secondary addresses") Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com> Acked-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-05 19:48:14 +00:00
dst_release(dst);
dst = &rt->dst;
break;
}
out_unlock:
rcu_read_unlock();
out:
t->dst = dst;
if (dst)
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>
2013-06-28 17:49:40 +00:00
pr_debug("rt_dst:%pI4, rt_src:%pI4\n",
&fl4->daddr, &fl4->saddr);
else
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>
2013-06-28 17:49:40 +00:00
pr_debug("no route\n");
}
/* For v4, the source address is cached in the route entry(dst). So no need
* to cache it separately and hence this is an empty routine.
*/
static void sctp_v4_get_saddr(struct sctp_sock *sk,
struct sctp_transport *t,
struct flowi *fl)
{
union sctp_addr *saddr = &t->saddr;
struct rtable *rt = (struct rtable *)t->dst;
if (rt) {
saddr->v4.sin_family = AF_INET;
saddr->v4.sin_addr.s_addr = fl->u.ip4.saddr;
}
}
/* What interface did this skb arrive on? */
static int sctp_v4_skb_iif(const struct sk_buff *skb)
{
return inet_iif(skb);
}
/* Was this packet marked by Explicit Congestion Notification? */
static int sctp_v4_is_ce(const struct sk_buff *skb)
{
return INET_ECN_is_ce(ip_hdr(skb)->tos);
}
/* Create and initialize a new sk for the socket returned by accept(). */
static struct sock *sctp_v4_create_accept_sk(struct sock *sk,
net: Work around lockdep limitation in sockets that use sockets Lockdep issues a circular dependency warning when AFS issues an operation through AF_RXRPC from a context in which the VFS/VM holds the mmap_sem. The theory lockdep comes up with is as follows: (1) If the pagefault handler decides it needs to read pages from AFS, it calls AFS with mmap_sem held and AFS begins an AF_RXRPC call, but creating a call requires the socket lock: mmap_sem must be taken before sk_lock-AF_RXRPC (2) afs_open_socket() opens an AF_RXRPC socket and binds it. rxrpc_bind() binds the underlying UDP socket whilst holding its socket lock. inet_bind() takes its own socket lock: sk_lock-AF_RXRPC must be taken before sk_lock-AF_INET (3) Reading from a TCP socket into a userspace buffer might cause a fault and thus cause the kernel to take the mmap_sem, but the TCP socket is locked whilst doing this: sk_lock-AF_INET must be taken before mmap_sem However, lockdep's theory is wrong in this instance because it deals only with lock classes and not individual locks. The AF_INET lock in (2) isn't really equivalent to the AF_INET lock in (3) as the former deals with a socket entirely internal to the kernel that never sees userspace. This is a limitation in the design of lockdep. Fix the general case by: (1) Double up all the locking keys used in sockets so that one set are used if the socket is created by userspace and the other set is used if the socket is created by the kernel. (2) Store the kern parameter passed to sk_alloc() in a variable in the sock struct (sk_kern_sock). This informs sock_lock_init(), sock_init_data() and sk_clone_lock() as to the lock keys to be used. Note that the child created by sk_clone_lock() inherits the parent's kern setting. (3) Add a 'kern' parameter to ->accept() that is analogous to the one passed in to ->create() that distinguishes whether kernel_accept() or sys_accept4() was the caller and can be passed to sk_alloc(). Note that a lot of accept functions merely dequeue an already allocated socket. I haven't touched these as the new socket already exists before we get the parameter. Note also that there are a couple of places where I've made the accepted socket unconditionally kernel-based: irda_accept() rds_rcp_accept_one() tcp_accept_from_sock() because they follow a sock_create_kern() and accept off of that. Whilst creating this, I noticed that lustre and ocfs don't create sockets through sock_create_kern() and thus they aren't marked as for-kernel, though they appear to be internal. I wonder if these should do that so that they use the new set of lock keys. Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-03-09 08:09:05 +00:00
struct sctp_association *asoc,
bool kern)
{
struct sock *newsk = sk_alloc(sock_net(sk), PF_INET, GFP_KERNEL,
net: Work around lockdep limitation in sockets that use sockets Lockdep issues a circular dependency warning when AFS issues an operation through AF_RXRPC from a context in which the VFS/VM holds the mmap_sem. The theory lockdep comes up with is as follows: (1) If the pagefault handler decides it needs to read pages from AFS, it calls AFS with mmap_sem held and AFS begins an AF_RXRPC call, but creating a call requires the socket lock: mmap_sem must be taken before sk_lock-AF_RXRPC (2) afs_open_socket() opens an AF_RXRPC socket and binds it. rxrpc_bind() binds the underlying UDP socket whilst holding its socket lock. inet_bind() takes its own socket lock: sk_lock-AF_RXRPC must be taken before sk_lock-AF_INET (3) Reading from a TCP socket into a userspace buffer might cause a fault and thus cause the kernel to take the mmap_sem, but the TCP socket is locked whilst doing this: sk_lock-AF_INET must be taken before mmap_sem However, lockdep's theory is wrong in this instance because it deals only with lock classes and not individual locks. The AF_INET lock in (2) isn't really equivalent to the AF_INET lock in (3) as the former deals with a socket entirely internal to the kernel that never sees userspace. This is a limitation in the design of lockdep. Fix the general case by: (1) Double up all the locking keys used in sockets so that one set are used if the socket is created by userspace and the other set is used if the socket is created by the kernel. (2) Store the kern parameter passed to sk_alloc() in a variable in the sock struct (sk_kern_sock). This informs sock_lock_init(), sock_init_data() and sk_clone_lock() as to the lock keys to be used. Note that the child created by sk_clone_lock() inherits the parent's kern setting. (3) Add a 'kern' parameter to ->accept() that is analogous to the one passed in to ->create() that distinguishes whether kernel_accept() or sys_accept4() was the caller and can be passed to sk_alloc(). Note that a lot of accept functions merely dequeue an already allocated socket. I haven't touched these as the new socket already exists before we get the parameter. Note also that there are a couple of places where I've made the accepted socket unconditionally kernel-based: irda_accept() rds_rcp_accept_one() tcp_accept_from_sock() because they follow a sock_create_kern() and accept off of that. Whilst creating this, I noticed that lustre and ocfs don't create sockets through sock_create_kern() and thus they aren't marked as for-kernel, though they appear to be internal. I wonder if these should do that so that they use the new set of lock keys. Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-03-09 08:09:05 +00:00
sk->sk_prot, kern);
struct inet_sock *newinet;
if (!newsk)
goto out;
sock_init_data(NULL, newsk);
sctp_copy_sock(newsk, sk, asoc);
sock_reset_flag(newsk, SOCK_ZAPPED);
sctp_v4_copy_ip_options(sk, newsk);
newinet = inet_sk(newsk);
newinet->inet_daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr;
sk_refcnt_debug_inc(newsk);
if (newsk->sk_prot->init(newsk)) {
sk_common_release(newsk);
newsk = NULL;
}
out:
return newsk;
}
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
static int sctp_v4_addr_to_user(struct sctp_sock *sp, union sctp_addr *addr)
{
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
/* No address mapping for V4 sockets */
return sizeof(struct sockaddr_in);
}
/* Dump the v4 addr to the seq file. */
static void sctp_v4_seq_dump_addr(struct seq_file *seq, union sctp_addr *addr)
{
seq_printf(seq, "%pI4 ", &addr->v4.sin_addr);
}
static void sctp_v4_ecn_capable(struct sock *sk)
{
INET_ECN_xmit(sk);
}
static void sctp_addr_wq_timeout_handler(struct timer_list *t)
{
struct net *net = from_timer(net, t, sctp.addr_wq_timer);
struct sctp_sockaddr_entry *addrw, *temp;
struct sctp_sock *sp;
spin_lock_bh(&net->sctp.addr_wq_lock);
list_for_each_entry_safe(addrw, temp, &net->sctp.addr_waitq, list) {
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>
2013-06-28 17:49:40 +00:00
pr_debug("%s: the first ent in wq:%p is addr:%pISc for cmd:%d at "
"entry:%p\n", __func__, &net->sctp.addr_waitq, &addrw->a.sa,
addrw->state, addrw);
#if IS_ENABLED(CONFIG_IPV6)
/* Now we send an ASCONF for each association */
/* Note. we currently don't handle link local IPv6 addressees */
if (addrw->a.sa.sa_family == AF_INET6) {
struct in6_addr *in6;
if (ipv6_addr_type(&addrw->a.v6.sin6_addr) &
IPV6_ADDR_LINKLOCAL)
goto free_next;
in6 = (struct in6_addr *)&addrw->a.v6.sin6_addr;
if (ipv6_chk_addr(net, in6, NULL, 0) == 0 &&
addrw->state == SCTP_ADDR_NEW) {
unsigned long timeo_val;
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>
2013-06-28 17:49:40 +00:00
pr_debug("%s: this is on DAD, trying %d sec "
"later\n", __func__,
SCTP_ADDRESS_TICK_DELAY);
timeo_val = jiffies;
timeo_val += msecs_to_jiffies(SCTP_ADDRESS_TICK_DELAY);
mod_timer(&net->sctp.addr_wq_timer, timeo_val);
break;
}
}
#endif
list_for_each_entry(sp, &net->sctp.auto_asconf_splist, auto_asconf_list) {
struct sock *sk;
sk = sctp_opt2sk(sp);
/* ignore bound-specific endpoints */
if (!sctp_is_ep_boundall(sk))
continue;
bh_lock_sock(sk);
if (sctp_asconf_mgmt(sp, addrw) < 0)
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>
2013-06-28 17:49:40 +00:00
pr_debug("%s: sctp_asconf_mgmt failed\n", __func__);
bh_unlock_sock(sk);
}
#if IS_ENABLED(CONFIG_IPV6)
free_next:
#endif
list_del(&addrw->list);
kfree(addrw);
}
spin_unlock_bh(&net->sctp.addr_wq_lock);
}
static void sctp_free_addr_wq(struct net *net)
{
struct sctp_sockaddr_entry *addrw;
struct sctp_sockaddr_entry *temp;
spin_lock_bh(&net->sctp.addr_wq_lock);
del_timer(&net->sctp.addr_wq_timer);
list_for_each_entry_safe(addrw, temp, &net->sctp.addr_waitq, list) {
list_del(&addrw->list);
kfree(addrw);
}
spin_unlock_bh(&net->sctp.addr_wq_lock);
}
/* lookup the entry for the same address in the addr_waitq
* sctp_addr_wq MUST be locked
*/
static struct sctp_sockaddr_entry *sctp_addr_wq_lookup(struct net *net,
struct sctp_sockaddr_entry *addr)
{
struct sctp_sockaddr_entry *addrw;
list_for_each_entry(addrw, &net->sctp.addr_waitq, list) {
if (addrw->a.sa.sa_family != addr->a.sa.sa_family)
continue;
if (addrw->a.sa.sa_family == AF_INET) {
if (addrw->a.v4.sin_addr.s_addr ==
addr->a.v4.sin_addr.s_addr)
return addrw;
} else if (addrw->a.sa.sa_family == AF_INET6) {
if (ipv6_addr_equal(&addrw->a.v6.sin6_addr,
&addr->a.v6.sin6_addr))
return addrw;
}
}
return NULL;
}
void sctp_addr_wq_mgmt(struct net *net, struct sctp_sockaddr_entry *addr, int cmd)
{
struct sctp_sockaddr_entry *addrw;
unsigned long timeo_val;
/* first, we check if an opposite message already exist in the queue.
* If we found such message, it is removed.
* This operation is a bit stupid, but the DHCP client attaches the
* new address after a couple of addition and deletion of that address
*/
spin_lock_bh(&net->sctp.addr_wq_lock);
/* Offsets existing events in addr_wq */
addrw = sctp_addr_wq_lookup(net, addr);
if (addrw) {
if (addrw->state != cmd) {
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>
2013-06-28 17:49:40 +00:00
pr_debug("%s: offsets existing entry for %d, addr:%pISc "
"in wq:%p\n", __func__, addrw->state, &addrw->a.sa,
&net->sctp.addr_waitq);
list_del(&addrw->list);
kfree(addrw);
}
spin_unlock_bh(&net->sctp.addr_wq_lock);
return;
}
/* OK, we have to add the new address to the wait queue */
addrw = kmemdup(addr, sizeof(struct sctp_sockaddr_entry), GFP_ATOMIC);
if (addrw == NULL) {
spin_unlock_bh(&net->sctp.addr_wq_lock);
return;
}
addrw->state = cmd;
list_add_tail(&addrw->list, &net->sctp.addr_waitq);
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>
2013-06-28 17:49:40 +00:00
pr_debug("%s: add new entry for cmd:%d, addr:%pISc in wq:%p\n",
__func__, addrw->state, &addrw->a.sa, &net->sctp.addr_waitq);
if (!timer_pending(&net->sctp.addr_wq_timer)) {
timeo_val = jiffies;
timeo_val += msecs_to_jiffies(SCTP_ADDRESS_TICK_DELAY);
mod_timer(&net->sctp.addr_wq_timer, timeo_val);
}
spin_unlock_bh(&net->sctp.addr_wq_lock);
}
/* Event handler for inet address addition/deletion events.
* The sctp_local_addr_list needs to be protocted by a spin lock since
* multiple notifiers (say IPv4 and IPv6) may be running at the same
* time and thus corrupt the list.
* The reader side is protected with RCU.
*/
static int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev,
void *ptr)
{
struct in_ifaddr *ifa = (struct in_ifaddr *)ptr;
struct sctp_sockaddr_entry *addr = NULL;
struct sctp_sockaddr_entry *temp;
struct net *net = dev_net(ifa->ifa_dev->dev);
int found = 0;
switch (ev) {
case NETDEV_UP:
addr = kmalloc(sizeof(struct sctp_sockaddr_entry), GFP_ATOMIC);
if (addr) {
addr->a.v4.sin_family = AF_INET;
addr->a.v4.sin_port = 0;
addr->a.v4.sin_addr.s_addr = ifa->ifa_local;
addr->valid = 1;
spin_lock_bh(&net->sctp.local_addr_lock);
list_add_tail_rcu(&addr->list, &net->sctp.local_addr_list);
sctp_addr_wq_mgmt(net, addr, SCTP_ADDR_NEW);
spin_unlock_bh(&net->sctp.local_addr_lock);
}
break;
case NETDEV_DOWN:
spin_lock_bh(&net->sctp.local_addr_lock);
list_for_each_entry_safe(addr, temp,
&net->sctp.local_addr_list, list) {
if (addr->a.sa.sa_family == AF_INET &&
addr->a.v4.sin_addr.s_addr ==
ifa->ifa_local) {
sctp_addr_wq_mgmt(net, addr, SCTP_ADDR_DEL);
found = 1;
addr->valid = 0;
list_del_rcu(&addr->list);
break;
}
}
spin_unlock_bh(&net->sctp.local_addr_lock);
if (found)
kfree_rcu(addr, rcu);
break;
}
return NOTIFY_DONE;
}
/*
* Initialize the control inode/socket with a control endpoint data
* structure. This endpoint is reserved exclusively for the OOTB processing.
*/
static int sctp_ctl_sock_init(struct net *net)
{
int err;
sa_family_t family = PF_INET;
if (sctp_get_pf_specific(PF_INET6))
family = PF_INET6;
err = inet_ctl_sock_create(&net->sctp.ctl_sock, family,
SOCK_SEQPACKET, IPPROTO_SCTP, net);
/* If IPv6 socket could not be created, try the IPv4 socket */
if (err < 0 && family == PF_INET6)
err = inet_ctl_sock_create(&net->sctp.ctl_sock, AF_INET,
SOCK_SEQPACKET, IPPROTO_SCTP,
net);
if (err < 0) {
pr_err("Failed to create the SCTP control socket\n");
return err;
}
return 0;
}
/* Register address family specific functions. */
int sctp_register_af(struct sctp_af *af)
{
switch (af->sa_family) {
case AF_INET:
if (sctp_af_v4_specific)
return 0;
sctp_af_v4_specific = af;
break;
case AF_INET6:
if (sctp_af_v6_specific)
return 0;
sctp_af_v6_specific = af;
break;
default:
return 0;
}
INIT_LIST_HEAD(&af->list);
list_add_tail(&af->list, &sctp_address_families);
return 1;
}
/* Get the table of functions for manipulating a particular address
* family.
*/
struct sctp_af *sctp_get_af_specific(sa_family_t family)
{
switch (family) {
case AF_INET:
return sctp_af_v4_specific;
case AF_INET6:
return sctp_af_v6_specific;
default:
return NULL;
}
}
/* Common code to initialize a AF_INET msg_name. */
static void sctp_inet_msgname(char *msgname, int *addr_len)
{
struct sockaddr_in *sin;
sin = (struct sockaddr_in *)msgname;
*addr_len = sizeof(struct sockaddr_in);
sin->sin_family = AF_INET;
memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
}
/* Copy the primary address of the peer primary address as the msg_name. */
static void sctp_inet_event_msgname(struct sctp_ulpevent *event, char *msgname,
int *addr_len)
{
struct sockaddr_in *sin, *sinfrom;
if (msgname) {
struct sctp_association *asoc;
asoc = event->asoc;
sctp_inet_msgname(msgname, addr_len);
sin = (struct sockaddr_in *)msgname;
sinfrom = &asoc->peer.primary_addr.v4;
sin->sin_port = htons(asoc->peer.port);
sin->sin_addr.s_addr = sinfrom->sin_addr.s_addr;
}
}
/* Initialize and copy out a msgname from an inbound skb. */
static void sctp_inet_skb_msgname(struct sk_buff *skb, char *msgname, int *len)
{
if (msgname) {
struct sctphdr *sh = sctp_hdr(skb);
struct sockaddr_in *sin = (struct sockaddr_in *)msgname;
sctp_inet_msgname(msgname, len);
sin->sin_port = sh->source;
sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
}
}
/* Do we support this AF? */
static int sctp_inet_af_supported(sa_family_t family, struct sctp_sock *sp)
{
/* PF_INET only supports AF_INET addresses. */
return AF_INET == family;
}
/* Address matching with wildcards allowed. */
static int sctp_inet_cmp_addr(const union sctp_addr *addr1,
const union sctp_addr *addr2,
struct sctp_sock *opt)
{
/* PF_INET only supports AF_INET addresses. */
if (addr1->sa.sa_family != addr2->sa.sa_family)
return 0;
if (htonl(INADDR_ANY) == addr1->v4.sin_addr.s_addr ||
htonl(INADDR_ANY) == addr2->v4.sin_addr.s_addr)
return 1;
if (addr1->v4.sin_addr.s_addr == addr2->v4.sin_addr.s_addr)
return 1;
return 0;
}
/* Verify that provided sockaddr looks bindable. Common verification has
* already been taken care of.
*/
static int sctp_inet_bind_verify(struct sctp_sock *opt, union sctp_addr *addr)
{
return sctp_v4_available(addr, opt);
}
/* Verify that sockaddr looks sendable. Common verification has already
* been taken care of.
*/
static int sctp_inet_send_verify(struct sctp_sock *opt, union sctp_addr *addr)
{
return 1;
}
/* Fill in Supported Address Type information for INIT and INIT-ACK
* chunks. Returns number of addresses supported.
*/
static int sctp_inet_supported_addrs(const struct sctp_sock *opt,
__be16 *types)
{
types[0] = SCTP_PARAM_IPV4_ADDRESS;
return 1;
}
/* Wrapper routine that calls the ip transmit routine. */
static inline int sctp_v4_xmit(struct sk_buff *skb,
struct sctp_transport *transport)
{
struct inet_sock *inet = inet_sk(skb->sk);
__u8 dscp = inet->tos;
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>
2013-06-28 17:49:40 +00:00
pr_debug("%s: skb:%p, len:%d, src:%pI4, dst:%pI4\n", __func__, skb,
skb->len, &transport->fl.u.ip4.saddr,
&transport->fl.u.ip4.daddr);
if (transport->dscp & SCTP_DSCP_SET_MASK)
dscp = transport->dscp & SCTP_DSCP_VAL_MASK;
inet->pmtudisc = transport->param_flags & SPP_PMTUD_ENABLE ?
IP_PMTUDISC_DO : IP_PMTUDISC_DONT;
SCTP_INC_STATS(sock_net(&inet->sk), SCTP_MIB_OUTSCTPPACKS);
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>
2013-06-28 17:49:40 +00:00
return __ip_queue_xmit(&inet->sk, skb, &transport->fl, dscp);
}
static struct sctp_af sctp_af_inet;
static struct sctp_pf sctp_pf_inet = {
.event_msgname = sctp_inet_event_msgname,
.skb_msgname = sctp_inet_skb_msgname,
.af_supported = sctp_inet_af_supported,
.cmp_addr = sctp_inet_cmp_addr,
.bind_verify = sctp_inet_bind_verify,
.send_verify = sctp_inet_send_verify,
.supported_addrs = sctp_inet_supported_addrs,
.create_accept_sk = sctp_v4_create_accept_sk,
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
.addr_to_user = sctp_v4_addr_to_user,
.to_sk_saddr = sctp_v4_to_sk_saddr,
.to_sk_daddr = sctp_v4_to_sk_daddr,
.copy_ip_options = sctp_v4_copy_ip_options,
.af = &sctp_af_inet
};
/* Notifier for inetaddr addition/deletion events. */
static struct notifier_block sctp_inetaddr_notifier = {
.notifier_call = sctp_inetaddr_event,
};
/* Socket operations. */
static const struct proto_ops inet_seqpacket_ops = {
.family = PF_INET,
.owner = THIS_MODULE,
.release = inet_release, /* Needs to be wrapped... */
.bind = inet_bind,
.connect = sctp_inet_connect,
.socketpair = sock_no_socketpair,
.accept = inet_accept,
.getname = inet_getname, /* Semantics are different. */
.poll = sctp_poll,
.ioctl = inet_ioctl,
.listen = sctp_inet_listen,
.shutdown = inet_shutdown, /* Looks harmless. */
.setsockopt = sock_common_setsockopt, /* IP_SOL IP_OPTION is a problem */
.getsockopt = sock_common_getsockopt,
.sendmsg = inet_sendmsg,
.recvmsg = inet_recvmsg,
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_sock_common_setsockopt,
.compat_getsockopt = compat_sock_common_getsockopt,
#endif
};
/* Registration with AF_INET family. */
static struct inet_protosw sctp_seqpacket_protosw = {
.type = SOCK_SEQPACKET,
.protocol = IPPROTO_SCTP,
.prot = &sctp_prot,
.ops = &inet_seqpacket_ops,
.flags = SCTP_PROTOSW_FLAG
};
static struct inet_protosw sctp_stream_protosw = {
.type = SOCK_STREAM,
.protocol = IPPROTO_SCTP,
.prot = &sctp_prot,
.ops = &inet_seqpacket_ops,
.flags = SCTP_PROTOSW_FLAG
};
/* Register with IP layer. */
static const struct net_protocol sctp_protocol = {
.handler = sctp_rcv,
.err_handler = sctp_v4_err,
.no_policy = 1,
.netns_ok = 1,
.icmp_strict_tag_validation = 1,
};
/* IPv4 address related functions. */
static struct sctp_af sctp_af_inet = {
.sa_family = AF_INET,
.sctp_xmit = sctp_v4_xmit,
.setsockopt = ip_setsockopt,
.getsockopt = ip_getsockopt,
.get_dst = sctp_v4_get_dst,
.get_saddr = sctp_v4_get_saddr,
.copy_addrlist = sctp_v4_copy_addrlist,
.from_skb = sctp_v4_from_skb,
.from_sk = sctp_v4_from_sk,
.from_addr_param = sctp_v4_from_addr_param,
.to_addr_param = sctp_v4_to_addr_param,
.cmp_addr = sctp_v4_cmp_addr,
.addr_valid = sctp_v4_addr_valid,
.inaddr_any = sctp_v4_inaddr_any,
.is_any = sctp_v4_is_any,
.available = sctp_v4_available,
.scope = sctp_v4_scope,
.skb_iif = sctp_v4_skb_iif,
.is_ce = sctp_v4_is_ce,
.seq_dump_addr = sctp_v4_seq_dump_addr,
.ecn_capable = sctp_v4_ecn_capable,
.net_header_len = sizeof(struct iphdr),
.sockaddr_len = sizeof(struct sockaddr_in),
.ip_options_len = sctp_v4_ip_options_len,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_ip_setsockopt,
.compat_getsockopt = compat_ip_getsockopt,
#endif
};
struct sctp_pf *sctp_get_pf_specific(sa_family_t family)
{
switch (family) {
case PF_INET:
return sctp_pf_inet_specific;
case PF_INET6:
return sctp_pf_inet6_specific;
default:
return NULL;
}
}
/* Register the PF specific function table. */
int sctp_register_pf(struct sctp_pf *pf, sa_family_t family)
{
switch (family) {
case PF_INET:
if (sctp_pf_inet_specific)
return 0;
sctp_pf_inet_specific = pf;
break;
case PF_INET6:
if (sctp_pf_inet6_specific)
return 0;
sctp_pf_inet6_specific = pf;
break;
default:
return 0;
}
return 1;
}
static inline int init_sctp_mibs(struct net *net)
{
net->sctp.sctp_statistics = alloc_percpu(struct sctp_mib);
if (!net->sctp.sctp_statistics)
return -ENOMEM;
return 0;
}
static inline void cleanup_sctp_mibs(struct net *net)
{
free_percpu(net->sctp.sctp_statistics);
}
static void sctp_v4_pf_init(void)
{
/* Initialize the SCTP specific PF functions. */
sctp_register_pf(&sctp_pf_inet, PF_INET);
sctp_register_af(&sctp_af_inet);
}
static void sctp_v4_pf_exit(void)
{
list_del(&sctp_af_inet.list);
}
static int sctp_v4_protosw_init(void)
{
int rc;
rc = proto_register(&sctp_prot, 1);
if (rc)
return rc;
/* Register SCTP(UDP and TCP style) with socket layer. */
inet_register_protosw(&sctp_seqpacket_protosw);
inet_register_protosw(&sctp_stream_protosw);
return 0;
}
static void sctp_v4_protosw_exit(void)
{
inet_unregister_protosw(&sctp_stream_protosw);
inet_unregister_protosw(&sctp_seqpacket_protosw);
proto_unregister(&sctp_prot);
}
static int sctp_v4_add_protocol(void)
{
/* Register notifier for inet address additions/deletions. */
register_inetaddr_notifier(&sctp_inetaddr_notifier);
/* Register SCTP with inet layer. */
if (inet_add_protocol(&sctp_protocol, IPPROTO_SCTP) < 0)
return -EAGAIN;
return 0;
}
static void sctp_v4_del_protocol(void)
{
inet_del_protocol(&sctp_protocol, IPPROTO_SCTP);
unregister_inetaddr_notifier(&sctp_inetaddr_notifier);
}
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
static int __net_init sctp_defaults_init(struct net *net)
{
int status;
/*
* 14. Suggested SCTP Protocol Parameter Values
*/
/* The following protocol parameters are RECOMMENDED: */
/* RTO.Initial - 3 seconds */
net->sctp.rto_initial = SCTP_RTO_INITIAL;
/* RTO.Min - 1 second */
net->sctp.rto_min = SCTP_RTO_MIN;
/* RTO.Max - 60 seconds */
net->sctp.rto_max = SCTP_RTO_MAX;
/* RTO.Alpha - 1/8 */
net->sctp.rto_alpha = SCTP_RTO_ALPHA;
/* RTO.Beta - 1/4 */
net->sctp.rto_beta = SCTP_RTO_BETA;
/* Valid.Cookie.Life - 60 seconds */
net->sctp.valid_cookie_life = SCTP_DEFAULT_COOKIE_LIFE;
/* Whether Cookie Preservative is enabled(1) or not(0) */
net->sctp.cookie_preserve_enable = 1;
/* Default sctp sockets to use md5 as their hmac alg */
#if defined (CONFIG_SCTP_DEFAULT_COOKIE_HMAC_MD5)
net->sctp.sctp_hmac_alg = "md5";
#elif defined (CONFIG_SCTP_DEFAULT_COOKIE_HMAC_SHA1)
net->sctp.sctp_hmac_alg = "sha1";
#else
net->sctp.sctp_hmac_alg = NULL;
#endif
/* Max.Burst - 4 */
net->sctp.max_burst = SCTP_DEFAULT_MAX_BURST;
/* Enable pf state by default */
net->sctp.pf_enable = 1;
/* Association.Max.Retrans - 10 attempts
* Path.Max.Retrans - 5 attempts (per destination address)
* Max.Init.Retransmits - 8 attempts
*/
net->sctp.max_retrans_association = 10;
net->sctp.max_retrans_path = 5;
net->sctp.max_retrans_init = 8;
/* Sendbuffer growth - do per-socket accounting */
net->sctp.sndbuf_policy = 0;
/* Rcvbuffer growth - do per-socket accounting */
net->sctp.rcvbuf_policy = 0;
/* HB.interval - 30 seconds */
net->sctp.hb_interval = SCTP_DEFAULT_TIMEOUT_HEARTBEAT;
/* delayed SACK timeout */
net->sctp.sack_timeout = SCTP_DEFAULT_TIMEOUT_SACK;
/* Disable ADDIP by default. */
net->sctp.addip_enable = 0;
net->sctp.addip_noauth = 0;
net->sctp.default_auto_asconf = 0;
/* Enable PR-SCTP by default. */
net->sctp.prsctp_enable = 1;
/* Disable RECONF by default. */
net->sctp.reconf_enable = 0;
/* Disable AUTH by default. */
net->sctp.auth_enable = 0;
/* Set SCOPE policy to enabled */
net->sctp.scope_policy = SCTP_SCOPE_POLICY_ENABLE;
/* Set the default rwnd update threshold */
net->sctp.rwnd_upd_shift = SCTP_DEFAULT_RWND_SHIFT;
/* Initialize maximum autoclose timeout. */
net->sctp.max_autoclose = INT_MAX / HZ;
status = sctp_sysctl_net_register(net);
if (status)
goto err_sysctl_register;
/* Allocate and initialise sctp mibs. */
status = init_sctp_mibs(net);
if (status)
goto err_init_mibs;
#ifdef CONFIG_PROC_FS
/* Initialize proc fs directory. */
status = sctp_proc_init(net);
if (status)
goto err_init_proc;
#endif
sctp_dbg_objcnt_init(net);
/* Initialize the local address list. */
INIT_LIST_HEAD(&net->sctp.local_addr_list);
spin_lock_init(&net->sctp.local_addr_lock);
sctp_get_local_addr_list(net);
/* Initialize the address event list */
INIT_LIST_HEAD(&net->sctp.addr_waitq);
INIT_LIST_HEAD(&net->sctp.auto_asconf_splist);
spin_lock_init(&net->sctp.addr_wq_lock);
net->sctp.addr_wq_timer.expires = 0;
timer_setup(&net->sctp.addr_wq_timer, sctp_addr_wq_timeout_handler, 0);
return 0;
#ifdef CONFIG_PROC_FS
err_init_proc:
cleanup_sctp_mibs(net);
#endif
err_init_mibs:
sctp_sysctl_net_unregister(net);
err_sysctl_register:
return status;
}
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
static void __net_exit sctp_defaults_exit(struct net *net)
{
/* Free the local address list */
sctp_free_addr_wq(net);
sctp_free_local_addr_list(net);
#ifdef CONFIG_PROC_FS
remove_proc_subtree("sctp", net->proc_net);
net->sctp.proc_net_sctp = NULL;
#endif
cleanup_sctp_mibs(net);
sctp_sysctl_net_unregister(net);
}
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
static struct pernet_operations sctp_defaults_ops = {
.init = sctp_defaults_init,
.exit = sctp_defaults_exit,
};
static int __net_init sctp_ctrlsock_init(struct net *net)
{
int status;
/* Initialize the control inode/socket for handling OOTB packets. */
status = sctp_ctl_sock_init(net);
if (status)
pr_err("Failed to initialize the SCTP control sock\n");
return status;
}
static void __net_init sctp_ctrlsock_exit(struct net *net)
{
/* Free the control endpoint. */
inet_ctl_sock_destroy(net->sctp.ctl_sock);
}
static struct pernet_operations sctp_ctrlsock_ops = {
.init = sctp_ctrlsock_init,
.exit = sctp_ctrlsock_exit,
};
/* Initialize the universe into something sensible. */
static __init int sctp_init(void)
{
int i;
int status = -EINVAL;
unsigned long goal;
unsigned long limit;
mm: reference totalram_pages and managed_pages once per function Patch series "mm: convert totalram_pages, totalhigh_pages and managed pages to atomic", v5. This series converts totalram_pages, totalhigh_pages and zone->managed_pages to atomic variables. totalram_pages, zone->managed_pages and totalhigh_pages updates are protected by managed_page_count_lock, but readers never care about it. Convert these variables to atomic to avoid readers potentially seeing a store tear. Main motivation was that managed_page_count_lock handling was complicating things. It was discussed in length here, https://lore.kernel.org/patchwork/patch/995739/#1181785 It seemes better to remove the lock and convert variables to atomic. With the change, preventing poteintial store-to-read tearing comes as a bonus. This patch (of 4): This is in preparation to a later patch which converts totalram_pages and zone->managed_pages to atomic variables. Please note that re-reading the value might lead to a different value and as such it could lead to unexpected behavior. There are no known bugs as a result of the current code but it is better to prevent from them in principle. Link: http://lkml.kernel.org/r/1542090790-21750-2-git-send-email-arunks@codeaurora.org Signed-off-by: Arun KS <arunks@codeaurora.org> Reviewed-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Reviewed-by: David Hildenbrand <david@redhat.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Reviewed-by: Pavel Tatashin <pasha.tatashin@soleen.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 08:34:20 +00:00
unsigned long nr_pages = totalram_pages;
int max_share;
int order;
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
int num_entries;
int max_entry_order;
sock_skb_cb_check_size(sizeof(struct sctp_ulpevent));
/* Allocate bind_bucket and chunk caches. */
status = -ENOBUFS;
sctp_bucket_cachep = kmem_cache_create("sctp_bind_bucket",
sizeof(struct sctp_bind_bucket),
0, SLAB_HWCACHE_ALIGN,
NULL);
if (!sctp_bucket_cachep)
goto out;
sctp_chunk_cachep = kmem_cache_create("sctp_chunk",
sizeof(struct sctp_chunk),
0, SLAB_HWCACHE_ALIGN,
NULL);
if (!sctp_chunk_cachep)
goto err_chunk_cachep;
status = percpu_counter_init(&sctp_sockets_allocated, 0, GFP_KERNEL);
if (status)
goto err_percpu_counter_init;
/* Implementation specific variables. */
/* Initialize default stream count setup information. */
sctp_max_instreams = SCTP_DEFAULT_INSTREAMS;
sctp_max_outstreams = SCTP_DEFAULT_OUTSTREAMS;
/* Initialize handle used for association ids. */
idr_init(&sctp_assocs_id);
limit = nr_free_buffer_pages() / 8;
limit = max(limit, 128UL);
sysctl_sctp_mem[0] = limit / 4 * 3;
sysctl_sctp_mem[1] = limit;
sysctl_sctp_mem[2] = sysctl_sctp_mem[0] * 2;
/* Set per-socket limits to no more than 1/128 the pressure threshold*/
limit = (sysctl_sctp_mem[1]) << (PAGE_SHIFT - 7);
max_share = min(4UL*1024*1024, limit);
sysctl_sctp_rmem[0] = SK_MEM_QUANTUM; /* give each asoc 1 page min */
sysctl_sctp_rmem[1] = 1500 * SKB_TRUESIZE(1);
sysctl_sctp_rmem[2] = max(sysctl_sctp_rmem[1], max_share);
[NET] CORE: Introducing new memory accounting interface. This patch introduces new memory accounting functions for each network protocol. Most of them are renamed from memory accounting functions for stream protocols. At the same time, some stream memory accounting functions are removed since other functions do same thing. Renaming: sk_stream_free_skb() -> sk_wmem_free_skb() __sk_stream_mem_reclaim() -> __sk_mem_reclaim() sk_stream_mem_reclaim() -> sk_mem_reclaim() sk_stream_mem_schedule -> __sk_mem_schedule() sk_stream_pages() -> sk_mem_pages() sk_stream_rmem_schedule() -> sk_rmem_schedule() sk_stream_wmem_schedule() -> sk_wmem_schedule() sk_charge_skb() -> sk_mem_charge() Removeing sk_stream_rfree(): consolidates into sock_rfree() sk_stream_set_owner_r(): consolidates into skb_set_owner_r() sk_stream_mem_schedule() The following functions are added. sk_has_account(): check if the protocol supports accounting sk_mem_uncharge(): do the opposite of sk_mem_charge() In addition, to achieve consolidation, updating sk_wmem_queued is removed from sk_mem_charge(). Next, to consolidate memory accounting functions, this patch adds memory accounting calls to network core functions. Moreover, present memory accounting call is renamed to new accounting call. Finally we replace present memory accounting calls with new interface in TCP and SCTP. Signed-off-by: Takahiro Yasui <tyasui@redhat.com> Signed-off-by: Hideo Aoki <haoki@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-31 08:11:19 +00:00
sysctl_sctp_wmem[0] = SK_MEM_QUANTUM;
sysctl_sctp_wmem[1] = 16*1024;
sysctl_sctp_wmem[2] = max(64*1024, max_share);
/* Size and allocate the association hash table.
* The methodology is similar to that of the tcp hash tables.
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
* Though not identical. Start by getting a goal size
*/
mm: reference totalram_pages and managed_pages once per function Patch series "mm: convert totalram_pages, totalhigh_pages and managed pages to atomic", v5. This series converts totalram_pages, totalhigh_pages and zone->managed_pages to atomic variables. totalram_pages, zone->managed_pages and totalhigh_pages updates are protected by managed_page_count_lock, but readers never care about it. Convert these variables to atomic to avoid readers potentially seeing a store tear. Main motivation was that managed_page_count_lock handling was complicating things. It was discussed in length here, https://lore.kernel.org/patchwork/patch/995739/#1181785 It seemes better to remove the lock and convert variables to atomic. With the change, preventing poteintial store-to-read tearing comes as a bonus. This patch (of 4): This is in preparation to a later patch which converts totalram_pages and zone->managed_pages to atomic variables. Please note that re-reading the value might lead to a different value and as such it could lead to unexpected behavior. There are no known bugs as a result of the current code but it is better to prevent from them in principle. Link: http://lkml.kernel.org/r/1542090790-21750-2-git-send-email-arunks@codeaurora.org Signed-off-by: Arun KS <arunks@codeaurora.org> Reviewed-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Reviewed-by: David Hildenbrand <david@redhat.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Reviewed-by: Pavel Tatashin <pasha.tatashin@soleen.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 08:34:20 +00:00
if (nr_pages >= (128 * 1024))
goal = nr_pages >> (22 - PAGE_SHIFT);
else
mm: reference totalram_pages and managed_pages once per function Patch series "mm: convert totalram_pages, totalhigh_pages and managed pages to atomic", v5. This series converts totalram_pages, totalhigh_pages and zone->managed_pages to atomic variables. totalram_pages, zone->managed_pages and totalhigh_pages updates are protected by managed_page_count_lock, but readers never care about it. Convert these variables to atomic to avoid readers potentially seeing a store tear. Main motivation was that managed_page_count_lock handling was complicating things. It was discussed in length here, https://lore.kernel.org/patchwork/patch/995739/#1181785 It seemes better to remove the lock and convert variables to atomic. With the change, preventing poteintial store-to-read tearing comes as a bonus. This patch (of 4): This is in preparation to a later patch which converts totalram_pages and zone->managed_pages to atomic variables. Please note that re-reading the value might lead to a different value and as such it could lead to unexpected behavior. There are no known bugs as a result of the current code but it is better to prevent from them in principle. Link: http://lkml.kernel.org/r/1542090790-21750-2-git-send-email-arunks@codeaurora.org Signed-off-by: Arun KS <arunks@codeaurora.org> Reviewed-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Reviewed-by: David Hildenbrand <david@redhat.com> Acked-by: Michal Hocko <mhocko@suse.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Reviewed-by: Pavel Tatashin <pasha.tatashin@soleen.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 08:34:20 +00:00
goal = nr_pages >> (24 - PAGE_SHIFT);
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
/* Then compute the page order for said goal */
order = get_order(goal);
/* Now compute the required page order for the maximum sized table we
* want to create
*/
max_entry_order = get_order(MAX_SCTP_PORT_HASH_ENTRIES *
sizeof(struct sctp_bind_hashbucket));
/* Limit the page order by that maximum hash table size */
order = min(order, max_entry_order);
/* Allocate and initialize the endpoint hash table. */
sctp_ep_hashsize = 64;
sctp_ep_hashtable =
treewide: kmalloc() -> kmalloc_array() The kmalloc() function has a 2-factor argument form, kmalloc_array(). This patch replaces cases of: kmalloc(a * b, gfp) with: kmalloc_array(a * b, gfp) as well as handling cases of: kmalloc(a * b * c, gfp) with: kmalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kmalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kmalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The tools/ directory was manually excluded, since it has its own implementation of kmalloc(). The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kmalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kmalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kmalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(char) * COUNT + COUNT , ...) | kmalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kmalloc + kmalloc_array ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kmalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kmalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kmalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kmalloc(C1 * C2 * C3, ...) | kmalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kmalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kmalloc(sizeof(THING) * C2, ...) | kmalloc(sizeof(TYPE) * C2, ...) | kmalloc(C1 * C2 * C3, ...) | kmalloc(C1 * C2, ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - (E1) * E2 + E1, E2 , ...) | - kmalloc + kmalloc_array ( - (E1) * (E2) + E1, E2 , ...) | - kmalloc + kmalloc_array ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 20:55:00 +00:00
kmalloc_array(64, sizeof(struct sctp_hashbucket), GFP_KERNEL);
if (!sctp_ep_hashtable) {
pr_err("Failed endpoint_hash alloc\n");
status = -ENOMEM;
goto err_ehash_alloc;
}
for (i = 0; i < sctp_ep_hashsize; i++) {
rwlock_init(&sctp_ep_hashtable[i].lock);
INIT_HLIST_HEAD(&sctp_ep_hashtable[i].chain);
}
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
/* Allocate and initialize the SCTP port hash table.
* Note that order is initalized to start at the max sized
* table we want to support. If we can't get that many pages
* reduce the order and try again
*/
do {
sctp_port_hashtable = (struct sctp_bind_hashbucket *)
__get_free_pages(GFP_KERNEL | __GFP_NOWARN, order);
} while (!sctp_port_hashtable && --order > 0);
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
if (!sctp_port_hashtable) {
pr_err("Failed bind hash alloc\n");
status = -ENOMEM;
goto err_bhash_alloc;
}
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
/* Now compute the number of entries that will fit in the
* port hash space we allocated
*/
num_entries = (1UL << order) * PAGE_SIZE /
sizeof(struct sctp_bind_hashbucket);
/* And finish by rounding it down to the nearest power of two
* this wastes some memory of course, but its needed because
* the hash function operates based on the assumption that
* that the number of entries is a power of two
*/
sctp_port_hashsize = rounddown_pow_of_two(num_entries);
for (i = 0; i < sctp_port_hashsize; i++) {
spin_lock_init(&sctp_port_hashtable[i].lock);
INIT_HLIST_HEAD(&sctp_port_hashtable[i].chain);
}
status = sctp_transport_hashtable_init();
if (status)
goto err_thash_alloc;
sctp: Fix port hash table size computation Dmitry Vyukov noted recently that the sctp_port_hashtable had an error in its size computation, observing that the current method never guaranteed that the hashsize (measured in number of entries) would be a power of two, which the input hash function for that table requires. The root cause of the problem is that two values need to be computed (one, the allocation order of the storage requries, as passed to __get_free_pages, and two the number of entries for the hash table). Both need to be ^2, but for different reasons, and the existing code is simply computing one order value, and using it as the basis for both, which is wrong (i.e. it assumes that ((1<<order)*PAGE_SIZE)/sizeof(bucket) is still ^2 when its not). To fix this, we change the logic slightly. We start by computing a goal allocation order (which is limited by the maximum size hash table we want to support. Then we attempt to allocate that size table, decreasing the order until a successful allocation is made. Then, with the resultant successful order we compute the number of buckets that hash table supports, which we then round down to the nearest power of two, giving us the number of entries the table actually supports. I've tested this locally here, using non-debug and spinlock-debug kernels, and the number of entries in the hashtable consistently work out to be powers of two in all cases. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> CC: Dmitry Vyukov <dvyukov@google.com> CC: Vladislav Yasevich <vyasevich@gmail.com> CC: "David S. Miller" <davem@davemloft.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-02-18 21:10:57 +00:00
pr_info("Hash tables configured (bind %d/%d)\n", sctp_port_hashsize,
num_entries);
sctp_sysctl_register();
INIT_LIST_HEAD(&sctp_address_families);
sctp_v4_pf_init();
sctp_v6_pf_init();
sctp_sched_ops_init();
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
status = register_pernet_subsys(&sctp_defaults_ops);
if (status)
goto err_register_defaults;
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
status = sctp_v4_protosw_init();
if (status)
goto err_protosw_init;
status = sctp_v6_protosw_init();
if (status)
goto err_v6_protosw_init;
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
status = register_pernet_subsys(&sctp_ctrlsock_ops);
if (status)
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
goto err_register_ctrlsock;
status = sctp_v4_add_protocol();
if (status)
goto err_add_protocol;
/* Register SCTP with inet6 layer. */
status = sctp_v6_add_protocol();
if (status)
goto err_v6_add_protocol;
if (sctp_offload_init() < 0)
pr_crit("%s: Cannot add SCTP protocol offload\n", __func__);
out:
return status;
err_v6_add_protocol:
sctp_v4_del_protocol();
err_add_protocol:
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
unregister_pernet_subsys(&sctp_ctrlsock_ops);
err_register_ctrlsock:
sctp_v6_protosw_exit();
err_v6_protosw_init:
sctp_v4_protosw_exit();
err_protosw_init:
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
unregister_pernet_subsys(&sctp_defaults_ops);
err_register_defaults:
sctp_v4_pf_exit();
sctp_v6_pf_exit();
sctp_sysctl_unregister();
free_pages((unsigned long)sctp_port_hashtable,
get_order(sctp_port_hashsize *
sizeof(struct sctp_bind_hashbucket)));
err_bhash_alloc:
sctp_transport_hashtable_destroy();
err_thash_alloc:
kfree(sctp_ep_hashtable);
err_ehash_alloc:
percpu_counter_destroy(&sctp_sockets_allocated);
err_percpu_counter_init:
kmem_cache_destroy(sctp_chunk_cachep);
err_chunk_cachep:
kmem_cache_destroy(sctp_bucket_cachep);
goto out;
}
/* Exit handler for the SCTP protocol. */
static __exit void sctp_exit(void)
{
/* BUG. This should probably do something useful like clean
* up all the remaining associations and all that memory.
*/
/* Unregister with inet6/inet layers. */
sctp_v6_del_protocol();
sctp_v4_del_protocol();
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
unregister_pernet_subsys(&sctp_ctrlsock_ops);
/* Free protosw registrations */
sctp_v6_protosw_exit();
sctp_v4_protosw_exit();
sctp: fix race on protocol/netns initialization Consider sctp module is unloaded and is being requested because an user is creating a sctp socket. During initialization, sctp will add the new protocol type and then initialize pernet subsys: status = sctp_v4_protosw_init(); if (status) goto err_protosw_init; status = sctp_v6_protosw_init(); if (status) goto err_v6_protosw_init; status = register_pernet_subsys(&sctp_net_ops); The problem is that after those calls to sctp_v{4,6}_protosw_init(), it is possible for userspace to create SCTP sockets like if the module is already fully loaded. If that happens, one of the possible effects is that we will have readers for net->sctp.local_addr_list list earlier than expected and sctp_net_init() does not take precautions while dealing with that list, leading to a potential panic but not limited to that, as sctp_sock_init() will copy a bunch of blank/partially initialized values from net->sctp. The race happens like this: CPU 0 | CPU 1 socket() | __sock_create | socket() inet_create | __sock_create list_for_each_entry_rcu( | answer, &inetsw[sock->type], | list) { | inet_create /* no hits */ | if (unlikely(err)) { | ... | request_module() | /* socket creation is blocked | * the module is fully loaded | */ | sctp_init | sctp_v4_protosw_init | inet_register_protosw | list_add_rcu(&p->list, | last_perm); | | list_for_each_entry_rcu( | answer, &inetsw[sock->type], sctp_v6_protosw_init | list) { | /* hit, so assumes protocol | * is already loaded | */ | /* socket creation continues | * before netns is initialized | */ register_pernet_subsys | Simply inverting the initialization order between register_pernet_subsys() and sctp_v4_protosw_init() is not possible because register_pernet_subsys() will create a control sctp socket, so the protocol must be already visible by then. Deferring the socket creation to a work-queue is not good specially because we loose the ability to handle its errors. So, as suggested by Vlad, the fix is to split netns initialization in two moments: defaults and control socket, so that the defaults are already loaded by when we register the protocol, while control socket initialization is kept at the same moment it is today. Fixes: 4db67e808640 ("sctp: Make the address lists per network namespace") Signed-off-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-09-10 20:31:15 +00:00
unregister_pernet_subsys(&sctp_defaults_ops);
/* Unregister with socket layer. */
sctp_v6_pf_exit();
sctp_v4_pf_exit();
sctp_sysctl_unregister();
free_pages((unsigned long)sctp_port_hashtable,
get_order(sctp_port_hashsize *
sizeof(struct sctp_bind_hashbucket)));
kfree(sctp_ep_hashtable);
sctp_transport_hashtable_destroy();
percpu_counter_destroy(&sctp_sockets_allocated);
rcu_barrier(); /* Wait for completion of call_rcu()'s */
kmem_cache_destroy(sctp_chunk_cachep);
kmem_cache_destroy(sctp_bucket_cachep);
}
module_init(sctp_init);
module_exit(sctp_exit);
/*
* __stringify doesn't likes enums, so use IPPROTO_SCTP value (132) directly.
*/
MODULE_ALIAS("net-pf-" __stringify(PF_INET) "-proto-132");
MODULE_ALIAS("net-pf-" __stringify(PF_INET6) "-proto-132");
MODULE_AUTHOR("Linux Kernel SCTP developers <linux-sctp@vger.kernel.org>");
MODULE_DESCRIPTION("Support for the SCTP protocol (RFC2960)");
module_param_named(no_checksums, sctp_checksum_disable, bool, 0644);
MODULE_PARM_DESC(no_checksums, "Disable checksums computing and verification");
MODULE_LICENSE("GPL");