* NoisyTCP — CVE-2026-99288 (Was Re: [oss-security] Backports available - cBPF JIT spray hardening)
[not found] ` <2026073004-spruce-clambake-1785@gregkh>
@ 2026-11-19 23:25 ` Jose R Rodriguez
2026-11-20 0:45 ` [oss-security] " Dominique Martinet
0 siblings, 1 reply; 2+ messages in thread
From: Jose R Rodriguez @ 2026-11-19 23:25 UTC (permalink / raw)
To: Greg KH; +Cc: oss-security, Edward Shishkin, Reiserfs Devel
[-- Attachment #1: Type: text/plain, Size: 13882 bytes --]
On 2026-07-30 07:00, Greg KH wrote:
> On Wed, Jul 29, 2026 at 06:31:11PM -0700, Pawan Gupta wrote:
>> On Wed, Jul 29, 2026 at 05:15:46PM -0700, Jose R Rodriguez wrote:
>> > On 2026-07-29 13:04, Pawan Gupta wrote:
>> > > Hi All,
>> > >
>> > > This is an inform distro maintainers about recently upstreamed hardening
>> > > against speculative execution attacks using BPF JIT spraying.
>> > >
>> > > The backports are available here:
>> > >
>> > > 6.1 - https://lore.kernel.org/all/20260727-cbpf-jit-spray-hardening-6-1-y-v1-0-eb80dcf1eb6e@linux.intel.com
>> > > 6.6 - https://lore.kernel.org/all/20260717-cbpf-jit-spray-hardening-6-6-y-v1-0-e04f1b2893de@linux.intel.com
>> > > 6.12 - https://lore.kernel.org/all/20260715-cbpf-jit-spray-hardening-6-12-y-v1-0-d8585a9aed80@linux.intel.com
>> > > 6.18 - https://lore.kernel.org/all/20260713-cbpf-jit-spray-hardening-6-18-y-v1-0-755f60c55705@linux.intel.com
>> > > 7.1 - https://lore.kernel.org/all/20260709-cbpf-jit-spray-hardening-7-1-y-v1-0-5ac5a2d6797f@linux.intel.com
>> > >
>> > > 6.1 backport is queued. Others are part of LTS kernels.
>> > >
>> > > 5.15 and older do not support pack allocator for BPF on which the
>> > > hardening
>> > > is based on. So the series is not directly applicable to 5.15 and older,
>> > > and may need custom hardening patches.
>> >
>> > Any patches out there for kernel 5.17.15? Thanks in advance!
>>
>> Pack allocator was first introduced in 5.18 by commit:
>>
>> 57631054fae6 ("bpf: Introduce bpf_prog_pack allocator")
>>
>> This means that 5.17 is in the same boat as 5.15, it does not support
>> BPF
>> pack allocator, and these hardening patches dont apply.
>>
>> Pack allocator makes the JIT spraying attacks easier, but this doesn't
>> mean
>> that kernels withouth it are immune to such attacks. This needs a
>> fresh
>> assessment and possibly a different hardening approach.
>>
>> BTW, I am confused by the choice of distro kernel 5.17 which is not an
>> LTS
>> kernel.
>
> Yes, that's a very odd choice for anyone to run these days. According
> to the cve tools:
> Total Vulnerable CVE's in 5.17.15 : 7035
>
> That's a non-trivial amount...
Since the time of your comment I have been keeping up with reiser4
-enhanced 5.15 LTS build iterations -- with the latest being 5.15.224.
Notwithstanding, 'NoisyTCP — CVE-2026-99288' came along and the latest
5.15.225 LTS iteration did not address the critical vulnerability. I
have the 5.15 LTS kernel(s) deployed online and I got spooked.
Accordingly, on a best effort basis, I backported the patch from
upstream:
From 0e5944a7abe777d2b8aad3604f7a7531966c06b1 Mon Sep 17 00:00:00 2001
From: Metztli Information Technology <jose@moiocoiatzin.metztli.it>
Date: Thu, 19 Nov 2026 14:45:54 -0800
Subject: [PATCH]
=?UTF-8?q?Best=20effort=205.15.225=20LTS=20backport=20pat?=
=?UTF-8?q?ch=20against=20NoisyTCP=20=E2=80=94=20CVE-2026-99288?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
include/linux/refcount.h | 55 +++++++++++++++++++++++++++++++++
net/core/filter.c | 2 +-
net/ipv4/inet_connection_sock.c | 16 ++++++----
net/ipv4/inet_hashtables.c | 7 +++--
net/ipv4/inet_timewait_sock.c | 4 +--
net/ipv4/syncookies.c | 4 ++-
net/ipv4/tcp_fastopen.c | 2 +-
net/ipv4/tcp_input.c | 4 ++-
net/ipv6/inet6_hashtables.c | 6 ++--
9 files changed, 83 insertions(+), 17 deletions(-)
diff --git a/include/linux/refcount.h b/include/linux/refcount.h
index b8a6e387f8f9..43a29f15d09c 100644
--- a/include/linux/refcount.h
+++ b/include/linux/refcount.h
@@ -142,6 +142,10 @@ static inline void refcount_set(refcount_t *r, int
n)
*
* Return: the refcount's value
*/
+static inline void refcount_set_release(refcount_t *r, int n)
+{
+ atomic_set_release(&r->refs, n);
+}
static inline unsigned int refcount_read(const refcount_t *r)
{
return atomic_read(&r->refs);
@@ -188,6 +192,48 @@ static inline __must_check bool
refcount_add_not_zero(int i, refcount_t *r)
return __refcount_add_not_zero(i, r, NULL);
}
+static inline __must_check
+bool __refcount_add_not_zero_limited_acquire(int i, refcount_t *r, int
*oldp,
+ int limit)
+{
+ int old = refcount_read(r);
+
+ do {
+ if (!old)
+ break;
+
+ if (i > limit - old) {
+ if (oldp)
+ *oldp = old;
+ return false;
+ }
+ } while (!atomic_try_cmpxchg_acquire(&r->refs, &old, old + i));
+
+ if (oldp)
+ *oldp = old;
+
+ if (unlikely(old < 0 || old + i < 0))
+ refcount_warn_saturate(r, REFCOUNT_ADD_NOT_ZERO_OVF);
+
+ return old;
+}
+
+static inline __must_check bool
+__refcount_inc_not_zero_limited_acquire(refcount_t *r, int *oldp, int
limit)
+{
+ return __refcount_add_not_zero_limited_acquire(1, r, oldp, limit);
+}
+
+static inline __must_check
+bool __refcount_add_not_zero_acquire(int i, refcount_t *r, int *oldp)
+{
+ return __refcount_add_not_zero_limited_acquire(i, r, oldp, INT_MAX);
+}
+static inline __must_check bool refcount_add_not_zero_acquire(int i,
refcount_t *r)
+{
+ return __refcount_add_not_zero_acquire(i, r, NULL);
+}
+
static inline void __refcount_add(int i, refcount_t *r, int *oldp)
{
int old = atomic_fetch_add_relaxed(i, &r->refs);
@@ -245,6 +291,15 @@ static inline __must_check bool
refcount_inc_not_zero(refcount_t *r)
return __refcount_inc_not_zero(r, NULL);
}
+static inline __must_check bool
__refcount_inc_not_zero_acquire(refcount_t *r, int *oldp)
+{
+ return __refcount_add_not_zero_acquire(1, r, oldp);
+}
+static inline __must_check bool
refcount_inc_not_zero_acquire(refcount_t *r)
+{
+ return __refcount_inc_not_zero_acquire(r, NULL);
+}
+
static inline void __refcount_inc(refcount_t *r, int *oldp)
{
__refcount_add(1, r, oldp);
diff --git a/net/core/filter.c b/net/core/filter.c
index f4a6ca23eef5..8de20ec6c5f0 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -7197,7 +7197,7 @@ BPF_CALL_3(bpf_sk_assign, struct sk_buff *, skb,
struct sock *, sk, u64, flags)
if (sk_unhashed(sk))
return -EOPNOTSUPP;
if (sk_is_refcounted(sk) &&
- unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
return -ENOENT;
skb_orphan(skb);
diff --git a/net/ipv4/inet_connection_sock.c
b/net/ipv4/inet_connection_sock.c
index ac9f18861aa7..3681198e43d9 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -863,8 +863,8 @@ static void reqsk_timer_handler(struct timer_list
*t)
* hold another count to prevent use-after-free and
* call reqsk_put() just before return.
*/
- refcount_set(&nreq->rsk_refcnt, 2 + 1);
timer_setup(&nreq->rsk_timer, reqsk_timer_handler, TIMER_PINNED);
+ refcount_set_release(&nreq->rsk_refcnt, 2 + 1);
reqsk_queue_migrated(&inet_csk(nsk)->icsk_accept_queue, req);
req = nreq;
@@ -939,10 +939,14 @@ static void reqsk_timer_handler(struct timer_list
*t)
*/
if (nreq) {
__NET_INC_STATS(net, LINUX_MIB_TCPMIGRATEREQFAILURE);
+ /* The timer was not armed, drop its reference here. */
+ reqsk_put(nreq);
no_ownership:
reqsk_migrate_reset(nreq);
reqsk_queue_removed(queue, nreq);
- __reqsk_free(nreq);
+ /* Drop the temporary reference and the one reserved for ehash. */
+ reqsk_put(nreq);
+ reqsk_put(nreq);
}
drop:
@@ -1216,7 +1220,7 @@ struct sock *inet_csk_complete_hashdance(struct
sock *sk, struct sock *child,
goto child_put;
}
- refcount_set(&nreq->rsk_refcnt, 1);
+ refcount_set_release(&nreq->rsk_refcnt, 1);
if (inet_csk_reqsk_queue_add(sk, nreq, child)) {
__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQSUCCESS);
reqsk_migrate_reset(req);
@@ -1226,7 +1230,7 @@ struct sock *inet_csk_complete_hashdance(struct
sock *sk, struct sock *child,
__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE);
reqsk_migrate_reset(nreq);
- __reqsk_free(nreq);
+ reqsk_put(nreq);
} else if (inet_csk_reqsk_queue_add(sk, req, child)) {
return child;
}
@@ -1270,7 +1274,7 @@ void inet_csk_listen_stop(struct sock *sk)
if (nsk) {
nreq = inet_reqsk_clone(req, nsk);
if (nreq) {
- refcount_set(&nreq->rsk_refcnt, 1);
+ refcount_set_release(&nreq->rsk_refcnt, 1);
rcu_read_lock();
if (inet_csk_reqsk_queue_add(nsk, nreq, child)) {
@@ -1282,7 +1286,7 @@ void inet_csk_listen_stop(struct sock *sk)
__NET_INC_STATS(sock_net(nsk),
LINUX_MIB_TCPMIGRATEREQFAILURE);
reqsk_migrate_reset(nreq);
- __reqsk_free(nreq);
+ reqsk_put(nreq);
}
rcu_read_unlock();
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index a7d11b82fe64..dc7ab92c8978 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -381,10 +381,11 @@ struct sock *__inet_lookup_established(struct net
*net,
if (sk->sk_hash != hash)
continue;
if (likely(INET_MATCH(net, sk, acookie, ports, dif, sdif))) {
- if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ if (unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
goto out;
- if (unlikely(!INET_MATCH(net, sk, acookie,
- ports, dif, sdif))) {
+ if (unlikely(hlist_nulls_unhashed_lockless(&sk->sk_nulls_node) ||
+ !INET_MATCH(net, sk, acookie, ports, dif,
+ sdif))) {
sock_gen_put(sk);
goto begin;
}
diff --git a/net/ipv4/inet_timewait_sock.c
b/net/ipv4/inet_timewait_sock.c
index 25809e06f111..aae056089157 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -137,7 +137,7 @@ void inet_twsk_hashdance(struct inet_timewait_sock
*tw, struct sock *sk,
* Also note that after this point, we lost our implicit reference
* so we are not allowed to use tw anymore.
*/
- refcount_set(&tw->tw_refcnt, 3);
+ refcount_set_release(&tw->tw_refcnt, 3);
}
EXPORT_SYMBOL_GPL(inet_twsk_hashdance);
@@ -278,7 +278,7 @@ void inet_twsk_purge(struct inet_hashinfo *hashinfo,
int family)
refcount_read(&sock_net(sk)->ns.count))
continue;
- if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ if (unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
continue;
if (unlikely(sk->sk_family != family ||
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 5a61f9a364ad..ee39ec6452e3 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -203,7 +203,7 @@ struct sock *tcp_get_cookie_sock(struct sock *sk,
struct sk_buff *skb,
child = icsk->icsk_af_ops->syn_recv_sock(sk, skb, req, dst,
NULL, &own_req, NULL);
if (child) {
- refcount_set(&req->rsk_refcnt, 1);
+ refcount_set_release(&req->rsk_refcnt, 1);
tcp_sk(child)->tsoffset = tsoff;
sock_rps_save_rxhash(child, skb);
@@ -217,6 +217,8 @@ struct sock *tcp_get_cookie_sock(struct sock *sk,
struct sk_buff *skb,
bh_unlock_sock(child);
sock_put(child);
+ reqsk_put(req);
+ return NULL;
}
__reqsk_free(req);
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index f3ed7d5ff7ef..ffa975ceb946 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -281,7 +281,7 @@ static struct sock *tcp_fastopen_create_child(struct
sock *sk,
inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS,
TCP_TIMEOUT_INIT, TCP_RTO_MAX);
- refcount_set(&req->rsk_refcnt, 2);
+ refcount_set_release(&req->rsk_refcnt, 2);
/* Now finish processing the fastopen child socket. */
tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB, skb);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 4a481dd4579b..8b8c892810ee 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -7069,7 +7069,9 @@ int tcp_conn_request(struct request_sock_ops
*rsk_ops,
if (!inet_csk_reqsk_queue_add(sk, req, fastopen_sk)) {
bh_unlock_sock(fastopen_sk);
sock_put(fastopen_sk);
- goto drop_and_free;
+ /* tcp_disconnect() already dropped one reqsk ref. */
+ reqsk_put(req);
+ goto drop;
}
sk->sk_data_ready(sk);
bh_unlock_sock(fastopen_sk);
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 869173f176cc..a23cade5fb83 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -74,10 +74,12 @@ struct sock *__inet6_lookup_established(struct net
*net,
continue;
if (!inet6_match(net, sk, saddr, daddr, ports, dif, sdif))
continue;
- if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ if (unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
goto out;
- if (unlikely(!inet6_match(net, sk, saddr, daddr, ports, dif, sdif)))
{
+ if (unlikely(hlist_nulls_unhashed_lockless(&sk->sk_nulls_node) ||
+ !inet6_match(net, sk, saddr, daddr, ports, dif,
+ sdif))) {
sock_gen_put(sk);
goto begin;
}
--
2.47.3
patching 5.15.225 LTS -- which I am currently running online.
>
> good luck!
>
> greg k-h
--
Best Professional Regards.
--
Jose R R
http://metztli.it
---------------------------------------------------------------------------------------------
Download Metztli Reiser4: Debian Trixie w/ Linux 5.17.15-3 AMD64
---------------------------------------------------------------------------------------------
feats ZSTD compression https://sf.net/projects/metztli-reiser4/
-------------------------------------------------------------------------------------------
Official current Reiser4 resources: https://reiser4.wiki.kernel.org/
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: metztli-5_15_225-LTS-NoisyTCP.patch --]
[-- Type: text/x-diff; name=metztli-5_15_225-LTS-NoisyTCP.patch, Size: 10148 bytes --]
From 0e5944a7abe777d2b8aad3604f7a7531966c06b1 Mon Sep 17 00:00:00 2001
From: Metztli Information Technology <jose@moiocoiatzin.metztli.it>
Date: Thu, 19 Nov 2026 14:45:54 -0800
Subject: [PATCH] =?UTF-8?q?Best=20effort=205.15.225=20LTS=20backport=20pat?=
=?UTF-8?q?ch=20against=20NoisyTCP=20=E2=80=94=20CVE-2026-99288?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
include/linux/refcount.h | 55 +++++++++++++++++++++++++++++++++
net/core/filter.c | 2 +-
net/ipv4/inet_connection_sock.c | 16 ++++++----
net/ipv4/inet_hashtables.c | 7 +++--
net/ipv4/inet_timewait_sock.c | 4 +--
net/ipv4/syncookies.c | 4 ++-
net/ipv4/tcp_fastopen.c | 2 +-
net/ipv4/tcp_input.c | 4 ++-
net/ipv6/inet6_hashtables.c | 6 ++--
9 files changed, 83 insertions(+), 17 deletions(-)
diff --git a/include/linux/refcount.h b/include/linux/refcount.h
index b8a6e387f8f9..43a29f15d09c 100644
--- a/include/linux/refcount.h
+++ b/include/linux/refcount.h
@@ -142,6 +142,10 @@ static inline void refcount_set(refcount_t *r, int n)
*
* Return: the refcount's value
*/
+static inline void refcount_set_release(refcount_t *r, int n)
+{
+ atomic_set_release(&r->refs, n);
+}
static inline unsigned int refcount_read(const refcount_t *r)
{
return atomic_read(&r->refs);
@@ -188,6 +192,48 @@ static inline __must_check bool refcount_add_not_zero(int i, refcount_t *r)
return __refcount_add_not_zero(i, r, NULL);
}
+static inline __must_check
+bool __refcount_add_not_zero_limited_acquire(int i, refcount_t *r, int *oldp,
+ int limit)
+{
+ int old = refcount_read(r);
+
+ do {
+ if (!old)
+ break;
+
+ if (i > limit - old) {
+ if (oldp)
+ *oldp = old;
+ return false;
+ }
+ } while (!atomic_try_cmpxchg_acquire(&r->refs, &old, old + i));
+
+ if (oldp)
+ *oldp = old;
+
+ if (unlikely(old < 0 || old + i < 0))
+ refcount_warn_saturate(r, REFCOUNT_ADD_NOT_ZERO_OVF);
+
+ return old;
+}
+
+static inline __must_check bool
+__refcount_inc_not_zero_limited_acquire(refcount_t *r, int *oldp, int limit)
+{
+ return __refcount_add_not_zero_limited_acquire(1, r, oldp, limit);
+}
+
+static inline __must_check
+bool __refcount_add_not_zero_acquire(int i, refcount_t *r, int *oldp)
+{
+ return __refcount_add_not_zero_limited_acquire(i, r, oldp, INT_MAX);
+}
+static inline __must_check bool refcount_add_not_zero_acquire(int i, refcount_t *r)
+{
+ return __refcount_add_not_zero_acquire(i, r, NULL);
+}
+
static inline void __refcount_add(int i, refcount_t *r, int *oldp)
{
int old = atomic_fetch_add_relaxed(i, &r->refs);
@@ -245,6 +291,15 @@ static inline __must_check bool refcount_inc_not_zero(refcount_t *r)
return __refcount_inc_not_zero(r, NULL);
}
+static inline __must_check bool __refcount_inc_not_zero_acquire(refcount_t *r, int *oldp)
+{
+ return __refcount_add_not_zero_acquire(1, r, oldp);
+}
+static inline __must_check bool refcount_inc_not_zero_acquire(refcount_t *r)
+{
+ return __refcount_inc_not_zero_acquire(r, NULL);
+}
+
static inline void __refcount_inc(refcount_t *r, int *oldp)
{
__refcount_add(1, r, oldp);
diff --git a/net/core/filter.c b/net/core/filter.c
index f4a6ca23eef5..8de20ec6c5f0 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -7197,7 +7197,7 @@ BPF_CALL_3(bpf_sk_assign, struct sk_buff *, skb, struct sock *, sk, u64, flags)
if (sk_unhashed(sk))
return -EOPNOTSUPP;
if (sk_is_refcounted(sk) &&
- unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
return -ENOENT;
skb_orphan(skb);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index ac9f18861aa7..3681198e43d9 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -863,8 +863,8 @@ static void reqsk_timer_handler(struct timer_list *t)
* hold another count to prevent use-after-free and
* call reqsk_put() just before return.
*/
- refcount_set(&nreq->rsk_refcnt, 2 + 1);
timer_setup(&nreq->rsk_timer, reqsk_timer_handler, TIMER_PINNED);
+ refcount_set_release(&nreq->rsk_refcnt, 2 + 1);
reqsk_queue_migrated(&inet_csk(nsk)->icsk_accept_queue, req);
req = nreq;
@@ -939,10 +939,14 @@ static void reqsk_timer_handler(struct timer_list *t)
*/
if (nreq) {
__NET_INC_STATS(net, LINUX_MIB_TCPMIGRATEREQFAILURE);
+ /* The timer was not armed, drop its reference here. */
+ reqsk_put(nreq);
no_ownership:
reqsk_migrate_reset(nreq);
reqsk_queue_removed(queue, nreq);
- __reqsk_free(nreq);
+ /* Drop the temporary reference and the one reserved for ehash. */
+ reqsk_put(nreq);
+ reqsk_put(nreq);
}
drop:
@@ -1216,7 +1220,7 @@ struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
goto child_put;
}
- refcount_set(&nreq->rsk_refcnt, 1);
+ refcount_set_release(&nreq->rsk_refcnt, 1);
if (inet_csk_reqsk_queue_add(sk, nreq, child)) {
__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQSUCCESS);
reqsk_migrate_reset(req);
@@ -1226,7 +1230,7 @@ struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE);
reqsk_migrate_reset(nreq);
- __reqsk_free(nreq);
+ reqsk_put(nreq);
} else if (inet_csk_reqsk_queue_add(sk, req, child)) {
return child;
}
@@ -1270,7 +1274,7 @@ void inet_csk_listen_stop(struct sock *sk)
if (nsk) {
nreq = inet_reqsk_clone(req, nsk);
if (nreq) {
- refcount_set(&nreq->rsk_refcnt, 1);
+ refcount_set_release(&nreq->rsk_refcnt, 1);
rcu_read_lock();
if (inet_csk_reqsk_queue_add(nsk, nreq, child)) {
@@ -1282,7 +1286,7 @@ void inet_csk_listen_stop(struct sock *sk)
__NET_INC_STATS(sock_net(nsk),
LINUX_MIB_TCPMIGRATEREQFAILURE);
reqsk_migrate_reset(nreq);
- __reqsk_free(nreq);
+ reqsk_put(nreq);
}
rcu_read_unlock();
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index a7d11b82fe64..dc7ab92c8978 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -381,10 +381,11 @@ struct sock *__inet_lookup_established(struct net *net,
if (sk->sk_hash != hash)
continue;
if (likely(INET_MATCH(net, sk, acookie, ports, dif, sdif))) {
- if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ if (unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
goto out;
- if (unlikely(!INET_MATCH(net, sk, acookie,
- ports, dif, sdif))) {
+ if (unlikely(hlist_nulls_unhashed_lockless(&sk->sk_nulls_node) ||
+ !INET_MATCH(net, sk, acookie, ports, dif,
+ sdif))) {
sock_gen_put(sk);
goto begin;
}
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 25809e06f111..aae056089157 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -137,7 +137,7 @@ void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
* Also note that after this point, we lost our implicit reference
* so we are not allowed to use tw anymore.
*/
- refcount_set(&tw->tw_refcnt, 3);
+ refcount_set_release(&tw->tw_refcnt, 3);
}
EXPORT_SYMBOL_GPL(inet_twsk_hashdance);
@@ -278,7 +278,7 @@ void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family)
refcount_read(&sock_net(sk)->ns.count))
continue;
- if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ if (unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
continue;
if (unlikely(sk->sk_family != family ||
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 5a61f9a364ad..ee39ec6452e3 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -203,7 +203,7 @@ struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
child = icsk->icsk_af_ops->syn_recv_sock(sk, skb, req, dst,
NULL, &own_req, NULL);
if (child) {
- refcount_set(&req->rsk_refcnt, 1);
+ refcount_set_release(&req->rsk_refcnt, 1);
tcp_sk(child)->tsoffset = tsoff;
sock_rps_save_rxhash(child, skb);
@@ -217,6 +217,8 @@ struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb,
bh_unlock_sock(child);
sock_put(child);
+ reqsk_put(req);
+ return NULL;
}
__reqsk_free(req);
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index f3ed7d5ff7ef..ffa975ceb946 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -281,7 +281,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk,
inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS,
TCP_TIMEOUT_INIT, TCP_RTO_MAX);
- refcount_set(&req->rsk_refcnt, 2);
+ refcount_set_release(&req->rsk_refcnt, 2);
/* Now finish processing the fastopen child socket. */
tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB, skb);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 4a481dd4579b..8b8c892810ee 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -7069,7 +7069,9 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
if (!inet_csk_reqsk_queue_add(sk, req, fastopen_sk)) {
bh_unlock_sock(fastopen_sk);
sock_put(fastopen_sk);
- goto drop_and_free;
+ /* tcp_disconnect() already dropped one reqsk ref. */
+ reqsk_put(req);
+ goto drop;
}
sk->sk_data_ready(sk);
bh_unlock_sock(fastopen_sk);
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 869173f176cc..a23cade5fb83 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -74,10 +74,12 @@ struct sock *__inet6_lookup_established(struct net *net,
continue;
if (!inet6_match(net, sk, saddr, daddr, ports, dif, sdif))
continue;
- if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt)))
+ if (unlikely(!refcount_inc_not_zero_acquire(&sk->sk_refcnt)))
goto out;
- if (unlikely(!inet6_match(net, sk, saddr, daddr, ports, dif, sdif))) {
+ if (unlikely(hlist_nulls_unhashed_lockless(&sk->sk_nulls_node) ||
+ !inet6_match(net, sk, saddr, daddr, ports, dif,
+ sdif))) {
sock_gen_put(sk);
goto begin;
}
--
2.47.3
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [oss-security] NoisyTCP — CVE-2026-99288 (Was Re: [oss-security] Backports available - cBPF JIT spray hardening)
2026-11-19 23:25 ` NoisyTCP — CVE-2026-99288 (Was Re: [oss-security] Backports available - cBPF JIT spray hardening) Jose R Rodriguez
@ 2026-11-20 0:45 ` Dominique Martinet
0 siblings, 0 replies; 2+ messages in thread
From: Dominique Martinet @ 2026-11-20 0:45 UTC (permalink / raw)
To: Jose R Rodriguez, oss-security; +Cc: Greg KH, Edward Shishkin, Reiserfs Devel
Hi Jose,
(not sure why this comes as a reply to the bpf hardening CVE..)
Jose R Rodriguez wrote on Thu, Nov 19, 2026 at 03:25:25PM -0800:
> Since the time of your comment I have been keeping up with reiser4 -enhanced
> 5.15 LTS build iterations -- with the latest being 5.15.224.
> Notwithstanding, 'NoisyTCP — CVE-2026-99288' came along and the latest
> 5.15.225 LTS iteration did not address the critical vulnerability. I have
> the 5.15 LTS kernel(s) deployed online and I got spooked. Accordingly, on a
> best effort basis, I backported the patch from upstream:
The patch "inet: reject unhashed sockets during established lookups"
marked as fixing this CVE already has been backported to older trees,
including 5.15 already and is in the 5.15.226-rc1 (which will likely be
released today (11/20) or tomorrow)
You can already pull the patch from
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
if you want to rebuild your own kernel early
OTOH if you want quick fixes it's best to upgrade to a newer release if
possible - except for embedded or platforms with specific hardware it's
generally not too hard to switch to the latest LTS.
--
Dominique Martinet | Asmadeus
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-11-20 0:45 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <dzp2vbih7jcsxjmcnxnefgi435b7re4kwxvxmvlxrmkra2p5br@vv5ckstkijym>
[not found] ` <14ad9b085cb3fb4b3a9fc091dde74540@metztli.com>
[not found] ` <6mqqzqzujeduybqeojfh7tmblksmqjksocri5ayeudze6l4fqm@ijx64jfxrwbt>
[not found] ` <2026073004-spruce-clambake-1785@gregkh>
2026-11-19 23:25 ` NoisyTCP — CVE-2026-99288 (Was Re: [oss-security] Backports available - cBPF JIT spray hardening) Jose R Rodriguez
2026-11-20 0:45 ` [oss-security] " Dominique Martinet
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.