The AF_RXRPC procfs helpers format local and remote socket addresses into
fixed 50-byte stack buffers with "%pISpc".
That is too small for the longest current-tree IPv6-with-port form the
formatter can produce. In lib/vsprintf.c, the compressed IPv6 path uses a
dotted-quad tail not only for v4mapped addresses, but also for ISATAP
addresses via ipv6_addr_is_isatap().
As a result, a case such as
[ffff:ffff:ffff:ffff:0:5efe:255.255.255.255]:65535
is possible with the current formatter. That is 50 visible characters, so
51 bytes including the trailing NUL, which does not fit in the existing
char[50] buffers used by net/rxrpc/proc.c.
Size the buffers from the formatter's maximum textual form and switch the
call sites to scnprintf().
Changes since v1:
- correct the changelog to cite the actual maximum current-tree case
explicitly
- frame the proof around the ISATAP formatting path instead of the earlier
mapped-v4 example
Fixes: 75b54cb57c ("rxrpc: Add IPv6 support")
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Anderson Nascimento <anderson@allelesecurity.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Link: https://patch.msgid.link/20260408121252.2249051-22-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
An AF_RXRPC socket can be both client and server at the same time. When
sending new calls (ie. it's acting as a client), it uses rx->key to set the
security, and when accepting incoming calls (ie. it's acting as a server),
it uses rx->securities.
setsockopt(RXRPC_SECURITY_KEY) sets rx->key to point to an rxrpc-type key
and setsockopt(RXRPC_SECURITY_KEYRING) sets rx->securities to point to a
keyring of rxrpc_s-type keys.
Now, it should be possible to use both rx->key and rx->securities on the
same socket - but for userspace AF_RXRPC sockets rxrpc_setsockopt()
prevents that.
Fix this by:
(1) Remove the incorrect check rxrpc_setsockopt(RXRPC_SECURITY_KEYRING)
makes on rx->key.
(2) Move the check that rxrpc_setsockopt(RXRPC_SECURITY_KEY) makes on
rx->key down into rxrpc_request_key().
(3) Remove rxrpc_request_key()'s check on rx->securities.
This (in combination with a previous patch) pushes the checks down into the
functions that set those pointers and removes the cross-checks that prevent
both key and keyring being set.
Fixes: 17926a7932 ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both")
Closes: https://sashiko.dev/#/patchset/20260401105614.1696001-10-dhowells@redhat.com
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Anderson Nascimento <anderson@allelesecurity.com>
cc: Luxiao Xu <rakukuip@gmail.com>
cc: Yuan Tan <yuantan098@gmail.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Link: https://patch.msgid.link/20260408121252.2249051-16-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rxgk_verify_response() decodes auth_len from the packet and is supposed
to verify that it fits in the remaining bytes. The existing check is
inverted, so oversized RESPONSE authenticators are accepted and passed
to rxgk_decrypt_skb(), which can later reach skb_to_sgvec() with an
impossible length and hit BUG_ON(len).
Decoded from the original latest-net reproduction logs with
scripts/decode_stacktrace.sh:
RIP: __skb_to_sgvec()
[net/core/skbuff.c:5285 (discriminator 1)]
Call Trace:
skb_to_sgvec() [net/core/skbuff.c:5305]
rxgk_decrypt_skb() [net/rxrpc/rxgk_common.h:81]
rxgk_verify_response() [net/rxrpc/rxgk.c:1268]
rxrpc_process_connection()
[net/rxrpc/conn_event.c:266 net/rxrpc/conn_event.c:364
net/rxrpc/conn_event.c:386]
process_one_work() [kernel/workqueue.c:3281]
worker_thread()
[kernel/workqueue.c:3353 kernel/workqueue.c:3440]
kthread() [kernel/kthread.c:436]
ret_from_fork() [arch/x86/kernel/process.c:164]
Reject authenticator lengths that exceed the remaining packet payload.
Fixes: 9d1d2b5934 ("rxrpc: rxgk: Implement the yfs-rxgk security class (GSSAPI)")
Signed-off-by: Keenan Dong <keenanat2000@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: Willy Tarreau <w@1wt.eu>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Link: https://patch.msgid.link/20260408121252.2249051-14-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rxgk_verify_authenticator() copies auth_len bytes into a temporary
buffer and then passes p + auth_len as the parser limit to
rxgk_do_verify_authenticator(). Since p is a __be32 *, that inflates the
parser end pointer by a factor of four and lets malformed RESPONSE
authenticators read past the kmalloc() buffer.
Decoded from the original latest-net reproduction logs with
scripts/decode_stacktrace.sh:
BUG: KASAN: slab-out-of-bounds in rxgk_verify_response()
Call Trace:
dump_stack_lvl() [lib/dump_stack.c:123]
print_report() [mm/kasan/report.c:379 mm/kasan/report.c:482]
kasan_report() [mm/kasan/report.c:597]
rxgk_verify_response()
[net/rxrpc/rxgk.c:1103 net/rxrpc/rxgk.c:1167
net/rxrpc/rxgk.c:1274]
rxrpc_process_connection()
[net/rxrpc/conn_event.c:266 net/rxrpc/conn_event.c:364
net/rxrpc/conn_event.c:386]
process_one_work() [kernel/workqueue.c:3281]
worker_thread()
[kernel/workqueue.c:3353 kernel/workqueue.c:3440]
kthread() [kernel/kthread.c:436]
ret_from_fork() [arch/x86/kernel/process.c:164]
Allocated by task 54:
rxgk_verify_response()
[include/linux/slab.h:954 net/rxrpc/rxgk.c:1155
net/rxrpc/rxgk.c:1274]
rxrpc_process_connection()
[net/rxrpc/conn_event.c:266 net/rxrpc/conn_event.c:364
net/rxrpc/conn_event.c:386]
Convert the byte count to __be32 units before constructing the parser
limit.
Fixes: 9d1d2b5934 ("rxrpc: rxgk: Implement the yfs-rxgk security class (GSSAPI)")
Signed-off-by: Keenan Dong <keenanat2000@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: Willy Tarreau <w@1wt.eu>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Link: https://patch.msgid.link/20260408121252.2249051-13-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rxrpc_preparse_xdr_yfs_rxgk() reads the raw key length and ticket length
from the XDR token as u32 values and passes each through round_up(x, 4)
before using the rounded value for validation and allocation. When the raw
length is >= 0xfffffffd, round_up() wraps to 0, so the bounds check and
kzalloc both use 0 while the subsequent memcpy still copies the original
~4 GiB value, producing a heap buffer overflow reachable from an
unprivileged add_key() call.
Fix this by:
(1) Rejecting raw key lengths above AFSTOKEN_GK_KEY_MAX and raw ticket
lengths above AFSTOKEN_GK_TOKEN_MAX before rounding, consistent with
the caps that the RxKAD path already enforces via AFSTOKEN_RK_TIX_MAX.
(2) Sizing the flexible-array allocation from the validated raw key
length via struct_size_t() instead of the rounded value.
(3) Caching the raw lengths so that the later field assignments and
memcpy calls do not re-read from the token, eliminating a class of
TOCTOU re-parse.
The control path (valid token with lengths within bounds) is unaffected.
Fixes: 0ca100ff4d ("rxrpc: Add YFS RxGK (GSSAPI) security class")
Signed-off-by: Oleh Konko <security@1seal.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jeffrey Altman <jaltman@auristor.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Link: https://patch.msgid.link/20260408121252.2249051-6-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Fix rxrpc call removal from the rxnet->calls list to use list_del_rcu()
rather than list_del_init() to prevent stuffing up reading
/proc/net/rxrpc/calls from potentially getting into an infinite loop.
This, however, means that list_empty() no longer works on an entry that's
been deleted from the list, making it harder to detect prior deletion. Fix
this by:
Firstly, make rxrpc_destroy_all_calls() only dump the first ten calls that
are unexpectedly still on the list. Limiting the number of steps means
there's no need to call cond_resched() or to remove calls from the list
here, thereby eliminating the need for rxrpc_put_call() to check for that.
rxrpc_put_call() can then be fixed to unconditionally delete the call from
the list as it is the only place that the deletion occurs.
Fixes: 2baec2c3f8 ("rxrpc: Support network namespacing")
Closes: https://sashiko.dev/#/patchset/20260319150150.4189381-1-dhowells%40redhat.com
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Jeffrey Altman <jaltman@auristor.com>
cc: Linus Torvalds <torvalds@linux-foundation.org>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Link: https://patch.msgid.link/20260408121252.2249051-5-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In rxrpc_new_client_call_for_sendmsg(), a key with no payload is meant to
be substituted for a NULL key pointer, but the variable this is done with
is subsequently not used.
Fix this by using "key" rather than "rx->key" when filling in the
connection parameters.
Note that this only affects direct use of AF_RXRPC; the kAFS filesystem
doesn't use sendmsg() directly and so bypasses the issue. Further,
AF_RXRPC passes a NULL key in if no key is set, so using an anonymous key
in that manner works. Since this hasn't been noticed to this point, it
might be better just to remove the "key" variable and the code that sets it
- and, arguably, rxrpc_init_client_call_security() would be a better place
to handle it.
Fixes: 19ffa01c9c ("rxrpc: Use structs to hold connection params and protocol info")
Closes: https://sashiko.dev/#/patchset/20260319150150.4189381-1-dhowells%40redhat.com
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Jeffrey Altman <jaltman@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Link: https://patch.msgid.link/20260408121252.2249051-4-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The seg6 lwtunnel uses a single dst_cache per encap route, shared
between seg6_input_core() and seg6_output_core(). These two paths
can perform the post-encap SID lookup in different routing contexts
(e.g., ip rules matching on the ingress interface, or VRF table
separation). Whichever path runs first populates the cache, and the
other reuses it blindly, bypassing its own lookup.
Fix this by splitting the cache into cache_input and cache_output,
so each path maintains its own cached dst independently.
Fixes: 6c8702c60b ("ipv6: sr: add support for SRH encapsulation and injection with lwtunnels")
Cc: stable@vger.kernel.org
Signed-off-by: Andrea Mayer <andrea.mayer@uniroma2.it>
Reviewed-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Reviewed-by: Justin Iurman <justin.iurman@gmail.com>
Link: https://patch.msgid.link/20260404004405.4057-2-andrea.mayer@uniroma2.it
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Sashiko points out that we use qops in __net_mp_open_rxq()
but never validate they are null. This was introduced when
check was moved from netdev_rx_queue_restart().
Look at ops directly instead of the locking config.
qops imply netdev_need_ops_lock(). We used netdev_need_ops_lock()
initially to signify that the real_num_rx_queues check below
is safe without rtnl_lock, but I'm not sure if this is actually
clear to most people, anyway.
Fixes: da7772a2b4 ("net: move mp->rx_page_size validation to __net_mp_open_rxq()")
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/20260404001938.2425670-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The -EBUSY handling in tls_do_encryption(), introduced by commit
8590541473 ("net: tls: handle backlogging of crypto requests"), has
a use-after-free due to double cleanup of encrypt_pending and the
scatterlist entry.
When crypto_aead_encrypt() returns -EBUSY, the request is enqueued to
the cryptd backlog and the async callback tls_encrypt_done() will be
invoked upon completion. That callback unconditionally restores the
scatterlist entry (sge->offset, sge->length) and decrements
ctx->encrypt_pending. However, if tls_encrypt_async_wait() returns an
error, the synchronous error path in tls_do_encryption() performs the
same cleanup again, double-decrementing encrypt_pending and
double-restoring the scatterlist.
The double-decrement corrupts the encrypt_pending sentinel (initialized
to 1), making tls_encrypt_async_wait() permanently skip the wait for
pending async callbacks. A subsequent sendmsg can then free the
tls_rec via bpf_exec_tx_verdict() while a cryptd callback is still
pending, resulting in a use-after-free when the callback fires on the
freed record.
Fix this by skipping the synchronous cleanup when the -EBUSY async
wait returns an error, since the callback has already handled
encrypt_pending and sge restoration.
Fixes: 8590541473 ("net: tls: handle backlogging of crypto requests")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Alifa Ramdhan <ramdhan@starlabs.sg>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://patch.msgid.link/20260403013617.2838875-1-ramdhan@starlabs.sg
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
SKB_SMALL_HEAD_CACHE_SIZE is intentionally set to a non-power-of-2
value (e.g. 704 on x86_64) to avoid collisions with generic kmalloc
bucket sizes. This ensures that skb_kfree_head() can reliably use
skb_end_offset to distinguish skb heads allocated from
skb_small_head_cache vs. generic kmalloc caches.
However, when KFENCE is enabled, kfence_ksize() returns the exact
requested allocation size instead of the slab bucket size. If a caller
(e.g. bpf_test_init) allocates skb head data via kzalloc() and the
requested size happens to equal SKB_SMALL_HEAD_CACHE_SIZE, then
slab_build_skb() -> ksize() returns that exact value. After subtracting
skb_shared_info overhead, skb_end_offset ends up matching
SKB_SMALL_HEAD_HEADROOM, causing skb_kfree_head() to incorrectly free
the object to skb_small_head_cache instead of back to the original
kmalloc cache, resulting in a slab cross-cache free:
kmem_cache_free(skbuff_small_head): Wrong slab cache. Expected
skbuff_small_head but got kmalloc-1k
Fix this by always calling kfree(head) in skb_kfree_head(). This keeps
the free path generic and avoids allocator-specific misclassification
for KFENCE objects.
Fixes: bf9f1baa27 ("net: add dedicated kmem_cache for typical/small skb->head")
Reported-by: Antonius <antonius@bluedragonsec.com>
Closes: https://lore.kernel.org/netdev/CAK8a0jxC5L5N7hq-DT2_NhUyjBxrPocoiDazzsBk4TGgT1r4-A@mail.gmail.com/
Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260403014517.142550-1-jiayuan.chen@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
AF_XDP bind currently accepts zero-copy pool configurations without
verifying that the device MTU fits into the usable frame space provided
by the UMEM chunk.
This becomes a problem since we started to respect tailroom which is
subtracted from chunk_size (among with headroom). 2k chunk size might
not provide enough space for standard 1500 MTU, so let us catch such
settings at bind time. Furthermore, validate whether underlying HW will
be able to satisfy configured MTU wrt XSK's frame size multiplied by
supported Rx buffer chain length (that is exposed via
net_device::xdp_zc_max_segs).
Fixes: 24ea50127e ("xsk: support mbuf on ZC RX")
Reviewed-by: Björn Töpel <bjorn@kernel.org>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://patch.msgid.link/20260402154958.562179-5-maciej.fijalkowski@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently xp_assign_dev_shared() is missing XDP_USE_SG being propagated
to flags so set it in order to preserve mtu check that is supposed to be
done only when no multi-buffer setup is in picture.
Also, this flag has the same value as XDP_UMEM_TX_SW_CSUM so we could
get unexpected SG setups for software Tx checksums. Since csum flag is
UAPI, modify value of XDP_UMEM_SG_FLAG.
Fixes: d609f3d228 ("xsk: add multi-buffer support for sockets sharing umem")
Reviewed-by: Björn Töpel <bjorn@kernel.org>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://patch.msgid.link/20260402154958.562179-4-maciej.fijalkowski@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Multi-buffer XDP stores information about frags in skb_shared_info that
sits at the tailroom of a packet. The storage space is reserved via
xdp_data_hard_end():
((xdp)->data_hard_start + (xdp)->frame_sz - \
SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))
and then we refer to it via macro below:
static inline struct skb_shared_info *
xdp_get_shared_info_from_buff(const struct xdp_buff *xdp)
{
return (struct skb_shared_info *)xdp_data_hard_end(xdp);
}
Currently we do not respect this tailroom space in multi-buffer AF_XDP
ZC scenario. To address this, introduce xsk_pool_get_tailroom() and use
it within xsk_pool_get_rx_frame_size() which is used in ZC drivers to
configure length of HW Rx buffer.
Typically drivers on Rx Hw buffers side work on 128 byte alignment so
let us align the value returned by xsk_pool_get_rx_frame_size() in order
to avoid addressing this on driver's side. This addresses the fact that
idpf uses mentioned function *before* pool->dev being set so we were at
risk that after subtracting tailroom we would not provide 128-byte
aligned value to HW.
Since xsk_pool_get_rx_frame_size() is actively used in xsk_rcv_check()
and __xsk_rcv(), add a variant of this routine that will not include 128
byte alignment and therefore old behavior is preserved.
Reviewed-by: Björn Töpel <bjorn@kernel.org>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Fixes: 24ea50127e ("xsk: support mbuf on ZC RX")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://patch.msgid.link/20260402154958.562179-3-maciej.fijalkowski@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The current headroom validation in xdp_umem_reg() could leave us with
insufficient space dedicated to even receive minimum-sized ethernet
frame. Furthermore if multi-buffer would come to play then
skb_shared_info stored at the end of XSK frame would be corrupted.
HW typically works with 128-aligned sizes so let us provide this value
as bare minimum.
Multi-buffer setting is known later in the configuration process so
besides accounting for 128 bytes, let us also take care of tailroom space
upfront.
Reviewed-by: Björn Töpel <bjorn@kernel.org>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Fixes: 99e3a236dd ("xsk: Add missing check on user supplied headroom size")
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://patch.msgid.link/20260402154958.562179-2-maciej.fijalkowski@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
ipv6_stub->ipv6_dev_find() may return ERR_PTR(-EAFNOSUPPORT) when the
IPv6 stack is not active (CONFIG_IPV6=m and not loaded), and passing
this error pointer to dev_hold() will cause a kernel crash with
null-ptr-deref.
Instead, silently discard the request. RFC 8335 does not appear to
define a specific response for the case where an IPv6 interface
identifier is syntactically valid but the implementation cannot perform
the lookup at runtime, and silently dropping the request may safer than
misreporting "No Such Interface".
Fixes: d329ea5bd8 ("icmp: add response to RFC 8335 PROBE messages")
Signed-off-by: Yiqi Sun <sunyiqixm@gmail.com>
Link: https://patch.msgid.link/20260402070419.2291578-1-sunyiqixm@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When querying a nexthop object via RTM_GETNEXTHOP, the kernel currently
allocates a fixed-size skb using NLMSG_GOODSIZE. While sufficient for
single nexthops and small Equal-Cost Multi-Path groups, this fixed
allocation fails for large nexthop groups like 512 nexthops.
This results in the following warning splat:
WARNING: net/ipv4/nexthop.c:3395 at rtm_get_nexthop+0x176/0x1c0, CPU#20: rep/4608
[...]
RIP: 0010:rtm_get_nexthop (net/ipv4/nexthop.c:3395)
[...]
Call Trace:
<TASK>
rtnetlink_rcv_msg (net/core/rtnetlink.c:6989)
netlink_rcv_skb (net/netlink/af_netlink.c:2550)
netlink_unicast (net/netlink/af_netlink.c:1319 net/netlink/af_netlink.c:1344)
netlink_sendmsg (net/netlink/af_netlink.c:1894)
____sys_sendmsg (net/socket.c:721 net/socket.c:736 net/socket.c:2585)
___sys_sendmsg (net/socket.c:2641)
__sys_sendmsg (net/socket.c:2671)
do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
</TASK>
Fix this by allocating the size dynamically using nh_nlmsg_size() and
using nlmsg_new(), this is consistent with nexthop_notify() behavior. In
addition, adjust nh_nlmsg_size_grp() so it calculates the size needed
based on flags passed. While at it, also add the size of NHA_FDB for
nexthop group size calculation as it was missing too.
This cannot be reproduced via iproute2 as the group size is currently
limited and the command fails as follows:
addattr_l ERROR: message exceeded bound of 1048
Fixes: 430a049190 ("nexthop: Add support for nexthop groups")
Reported-by: Yiming Qian <yimingqian591@gmail.com>
Closes: https://lore.kernel.org/netdev/CAL_bE8Li2h4KO+AQFXW4S6Yb_u5X4oSKnkywW+LPFjuErhqELA@mail.gmail.com/
Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260402072613.25262-2-fmancera@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Currently NHA_HW_STATS_ENABLE is included twice everytime a dump of
nexthop group is performed with NHA_OP_FLAG_DUMP_STATS. As all the stats
querying were moved to nla_put_nh_group_stats(), leave only that
instance of the attribute querying.
Fixes: 5072ae00ae ("net: nexthop: Expose nexthop group HW stats to user space")
Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260402072613.25262-1-fmancera@suse.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The GRP_ACK_MSG handler in tipc_group_proto_rcv() currently decrements
bc_ackers on every inbound group ACK, even when the same member has
already acknowledged the current broadcast round.
Because bc_ackers is a u16, a duplicate ACK received after the last
legitimate ACK wraps the counter to 65535. Once wrapped,
tipc_group_bc_cong() keeps reporting congestion and later group
broadcasts on the affected socket stay blocked until the group is
recreated.
Fix this by ignoring duplicate or stale ACKs before touching bc_acked or
bc_ackers. This makes repeated GRP_ACK_MSG handling idempotent and
prevents the underflow path.
Fixes: 2f487712b8 ("tipc: guarantee that group broadcast doesn't bypass group unicast")
Cc: stable@vger.kernel.org
Signed-off-by: Oleh Konko <security@1seal.org>
Reviewed-by: Tung Nguyen <tung.quang.nguyen@est.tech>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/41a4833f368641218e444fdcff822039.security@1seal.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rtnl_newlink() lacks a CAP_NET_ADMIN capability check on the peer
network namespace when creating paired devices (veth, vxcan,
netkit). This allows an unprivileged user with a user namespace
to create interfaces in arbitrary network namespaces, including
init_net.
Add a netlink_ns_capable() check for CAP_NET_ADMIN in the peer
namespace before allowing device creation to proceed.
Fixes: 81adee47df ("net: Support specifying the network namespace upon device creation.")
Signed-off-by: Nikolaos Gkarlis <nickgarlis@gmail.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20260402181432.4126920-1-nickgarlis@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When CONFIG_BRIDGE_VLAN_FILTERING is not set, br_vlan_group() and
nbp_vlan_group() return NULL (br_private.h stub definitions). The
BR_BOOLOPT_FDB_LOCAL_VLAN_0 toggle code is compiled unconditionally and
reaches br_fdb_delete_locals_per_vlan_port() and
br_fdb_insert_locals_per_vlan_port(), where the NULL vlan group pointer
is dereferenced via list_for_each_entry(v, &vg->vlan_list, vlist).
The observed crash is in the delete path, triggered when creating a
bridge with IFLA_BR_MULTI_BOOLOPT containing BR_BOOLOPT_FDB_LOCAL_VLAN_0
via RTM_NEWLINK. The insert helper has the same bug pattern.
Oops: general protection fault, probably for non-canonical address 0xdffffc0000000056: 0000 [#1] KASAN NOPTI
KASAN: null-ptr-deref in range [0x00000000000002b0-0x00000000000002b7]
RIP: 0010:br_fdb_delete_locals_per_vlan+0x2b9/0x310
Call Trace:
br_fdb_toggle_local_vlan_0+0x452/0x4c0
br_toggle_fdb_local_vlan_0+0x31/0x80 net/bridge/br.c:276
br_boolopt_toggle net/bridge/br.c:313
br_boolopt_multi_toggle net/bridge/br.c:364
br_changelink net/bridge/br_netlink.c:1542
br_dev_newlink net/bridge/br_netlink.c:1575
Add NULL checks for the vlan group pointer in both helpers, returning
early when there are no VLANs to iterate. This matches the existing
pattern used by other bridge FDB functions such as br_fdb_add() and
br_fdb_delete().
Fixes: 21446c06b4 ("net: bridge: Introduce UAPI for BR_BOOLOPT_FDB_LOCAL_VLAN_0")
Signed-off-by: Zijing Yin <yzjaurora@gmail.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Acked-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/20260402140153.3925663-1-yzjaurora@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
tcf_csum_act() walks nested VLAN headers directly from skb->data when an
skb still carries in-payload VLAN tags. The current code reads
vlan->h_vlan_encapsulated_proto and then pulls VLAN_HLEN bytes without
first ensuring that the full VLAN header is present in the linear area.
If only part of an inner VLAN header is linearized, accessing
h_vlan_encapsulated_proto reads past the linear area, and the following
skb_pull(VLAN_HLEN) may violate skb invariants.
Fix this by requiring pskb_may_pull(skb, VLAN_HLEN) before accessing and
pulling each nested VLAN header. If the header still is not fully
available, drop the packet through the existing error path.
Fixes: 2ecba2d1e4 ("net: sched: act_csum: Fix csum calc for tagged packets")
Reported-by: Yifan Wu <yifanwucs@gmail.com>
Reported-by: Juefei Pu <tomapufckgml@gmail.com>
Co-developed-by: Yuan Tan <yuantan098@gmail.com>
Signed-off-by: Yuan Tan <yuantan098@gmail.com>
Suggested-by: Xin Liu <bird@lzu.edu.cn>
Tested-by: Ren Wei <enjou1224z@gmail.com>
Signed-off-by: Ruide Cao <caoruide123@gmail.com>
Signed-off-by: Ren Wei <n05ec@lzu.edu.cn>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/22df2fcb49f410203eafa5d97963dd36089f4ecf.1774892775.git.caoruide123@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Yiming Qian reported :
<quote>
I believe I found a locally triggerable kernel bug in the IPv6 sendmsg
ancillary-data path that can panic the kernel via `skb_under_panic()`
(local DoS).
The core issue is a mismatch between:
- a 16-bit length accumulator (`struct ipv6_txoptions::opt_flen`, type
`__u16`) and
- a pointer to the *last* provided destination-options header (`opt->dst1opt`)
when multiple `IPV6_DSTOPTS` control messages (cmsgs) are provided.
- `include/net/ipv6.h`:
- `struct ipv6_txoptions::opt_flen` is `__u16` (wrap possible).
(lines 291-307, especially 298)
- `net/ipv6/datagram.c:ip6_datagram_send_ctl()`:
- Accepts repeated `IPV6_DSTOPTS` and accumulates into `opt_flen`
without rejecting duplicates. (lines 909-933)
- `net/ipv6/ip6_output.c:__ip6_append_data()`:
- Uses `opt->opt_flen + opt->opt_nflen` to compute header
sizes/headroom decisions. (lines 1448-1466, especially 1463-1465)
- `net/ipv6/ip6_output.c:__ip6_make_skb()`:
- Calls `ipv6_push_frag_opts()` if `opt->opt_flen` is non-zero.
(lines 1930-1934)
- `net/ipv6/exthdrs.c:ipv6_push_frag_opts()` / `ipv6_push_exthdr()`:
- Push size comes from `ipv6_optlen(opt->dst1opt)` (based on the
pointed-to header). (lines 1179-1185 and 1206-1211)
1. `opt_flen` is a 16-bit accumulator:
- `include/net/ipv6.h:298` defines `__u16 opt_flen; /* after fragment hdr */`.
2. `ip6_datagram_send_ctl()` accepts *repeated* `IPV6_DSTOPTS` cmsgs
and increments `opt_flen` each time:
- In `net/ipv6/datagram.c:909-933`, for `IPV6_DSTOPTS`:
- It computes `len = ((hdr->hdrlen + 1) << 3);`
- It checks `CAP_NET_RAW` using `ns_capable(net->user_ns,
CAP_NET_RAW)`. (line 922)
- Then it does:
- `opt->opt_flen += len;` (line 927)
- `opt->dst1opt = hdr;` (line 928)
There is no duplicate rejection here (unlike the legacy
`IPV6_2292DSTOPTS` path which rejects duplicates at
`net/ipv6/datagram.c:901-904`).
If enough large `IPV6_DSTOPTS` cmsgs are provided, `opt_flen` wraps
while `dst1opt` still points to a large (2048-byte)
destination-options header.
In the attached PoC (`poc.c`):
- 32 cmsgs with `hdrlen=255` => `len = (255+1)*8 = 2048`
- 1 cmsg with `hdrlen=0` => `len = 8`
- Total increment: `32*2048 + 8 = 65544`, so `(__u16)opt_flen == 8`
- The last cmsg is 2048 bytes, so `dst1opt` points to a 2048-byte header.
3. The transmit path sizes headers using the wrapped `opt_flen`:
- In `net/ipv6/ip6_output.c:1463-1465`:
- `headersize = sizeof(struct ipv6hdr) + (opt ? opt->opt_flen +
opt->opt_nflen : 0) + ...;`
With wrapped `opt_flen`, `headersize`/headroom decisions underestimate
what will be pushed later.
4. When building the final skb, the actual push length comes from
`dst1opt` and is not limited by wrapped `opt_flen`:
- In `net/ipv6/ip6_output.c:1930-1934`:
- `if (opt->opt_flen) proto = ipv6_push_frag_opts(skb, opt, proto);`
- In `net/ipv6/exthdrs.c:1206-1211`, `ipv6_push_frag_opts()` pushes
`dst1opt` via `ipv6_push_exthdr()`.
- In `net/ipv6/exthdrs.c:1179-1184`, `ipv6_push_exthdr()` does:
- `skb_push(skb, ipv6_optlen(opt));`
- `memcpy(h, opt, ipv6_optlen(opt));`
With insufficient headroom, `skb_push()` underflows and triggers
`skb_under_panic()` -> `BUG()`:
- `net/core/skbuff.c:2669-2675` (`skb_push()` calls `skb_under_panic()`)
- `net/core/skbuff.c:207-214` (`skb_panic()` ends in `BUG()`)
- The `IPV6_DSTOPTS` cmsg path requires `CAP_NET_RAW` in the target
netns user namespace (`ns_capable(net->user_ns, CAP_NET_RAW)`).
- Root (or any task with `CAP_NET_RAW`) can trigger this without user
namespaces.
- An unprivileged `uid=1000` user can trigger this if unprivileged
user namespaces are enabled and it can create a userns+netns to obtain
namespaced `CAP_NET_RAW` (the attached PoC does this).
- Local denial of service: kernel BUG/panic (system crash).
- Reproducible with a small userspace PoC.
</quote>
This patch does not reject duplicated options, as this might break
some user applications.
Instead, it makes sure to adjust opt_flen and opt_nflen to correctly
reflect the size of the current option headers, preventing the overflows
and the potential for panics.
This applies to IPV6_DSTOPTS, IPV6_HOPOPTS, and IPV6_RTHDR.
Specifically:
When a new IPV6_DSTOPTS is processed, the length of the old opt->dst1opt
is subtracted from opt->opt_flen before adding the new length.
When a new IPV6_HOPOPTS is processed, the length of the old opt->dst0opt
is subtracted from opt->opt_nflen.
When a new Routing Header (IPV6_RTHDR or IPV6_2292RTHDR) is processed,
the length of the old opt->srcrt is subtracted from opt->opt_nflen.
In the special case within IPV6_2292RTHDR handling where dst1opt is moved
to dst0opt, the length of the old opt->dst0opt is subtracted from
opt->opt_nflen before the new one is added.
Fixes: 333fad5364 ("[IPV6]: Support several new sockopt / ancillary data in Advanced API (RFC3542).")
Reported-by: Yiming Qian <yimingqian591@gmail.com>
Closes: https://lore.kernel.org/netdev/CAL_bE8JNzawgr5OX5m+3jnQDHry2XxhQT5=jThW1zDPtUikRYA@mail.gmail.com/
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260401154721.3740056-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When vlan_vid_add() fails for a secondary slave, the error path calls
vlan_vid_del() on the failing port instead of the peer slave that had
already succeeded. This results in asymmetric VLAN state across the HSR
pair.
Fix this by switching to a centralized unwind path that removes the VID
from any slave device that was already programmed.
Fixes: 1a8a63a530 ("net: hsr: Add VLAN CTAG filter support")
Signed-off-by: Luka Gejak <luka.gejak@linux.dev>
Link: https://patch.msgid.link/20260401092243.52121-3-luka.gejak@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
During node merging, hsr_handle_sup_frame() walks node_curr->seq_blocks
to update node_real without holding node_curr->seq_out_lock. This
allows concurrent mutations from duplicate registration paths, risking
inconsistent state or XArray/bitmap corruption.
Fix this by locking both nodes' seq_out_lock during the merge.
To prevent ABBA deadlocks, locks are acquired in order of memory
address.
Reviewed-by: Felix Maurer <fmaurer@redhat.com>
Fixes: 415e636751 ("hsr: Implement more robust duplicate discard for PRP")
Signed-off-by: Luka Gejak <luka.gejak@linux.dev>
Link: https://patch.msgid.link/20260401092243.52121-2-luka.gejak@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The `child_ns_mode_locked` field lives in `struct net`, which persists
across vsock module reloads. When the module is unloaded and reloaded,
`vsock_net_init()` resets `mode` and `child_ns_mode` back to their
default values, but does not reset `child_ns_mode_locked`.
The stale lock from the previous module load causes subsequent writes
to `child_ns_mode` to silently fail: `vsock_net_set_child_mode()` sees
the old lock, skips updating the actual value, and returns success
when the requested mode matches the stale lock. The sysctl handler
reports no error, but `child_ns_mode` remains unchanged.
Steps to reproduce:
$ modprobe vsock
$ echo local > /proc/sys/net/vsock/child_ns_mode
$ cat /proc/sys/net/vsock/child_ns_mode
local
$ modprobe -r vsock
$ modprobe vsock
$ echo local > /proc/sys/net/vsock/child_ns_mode
$ cat /proc/sys/net/vsock/child_ns_mode
global <--- expected "local"
Fix this by initializing `child_ns_mode_locked` to 0 (unlocked) in
`vsock_net_init()`, so the write-once mechanism works correctly after
module reload.
Fixes: 102eab95f0 ("vsock: lock down child_ns_mode as write-once")
Reported-by: Jin Liu <jinl@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Bobby Eshleman <bobbyeshleman@meta.com>
Link: https://patch.msgid.link/20260401092153.28462-1-sgarzare@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
flow_change() calls tcf_block_q() and dereferences q->handle to derive
a default baseclass. Shared blocks leave block->q NULL, causing a NULL
deref when a flow filter without a fully qualified baseclass is created
on a shared block.
Check tcf_block_shared() before accessing block->q and return -EINVAL
for shared blocks. This avoids the null-deref shown below:
=======================================================================
KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f]
RIP: 0010:flow_change (net/sched/cls_flow.c:508)
Call Trace:
tc_new_tfilter (net/sched/cls_api.c:2432)
rtnetlink_rcv_msg (net/core/rtnetlink.c:6980)
[...]
=======================================================================
Fixes: 1abf272022 ("net: sched: tcindex, fw, flow: use tcf_block_q helper to get struct Qdisc")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260331050217.504278-2-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The old-method path in fw_classify() calls tcf_block_q() and
dereferences q->handle. Shared blocks leave block->q NULL, causing a
NULL deref when an empty cls_fw filter is attached to a shared block
and a packet with a nonzero major skb mark is classified.
Reject the configuration in fw_change() when the old method (no
TCA_OPTIONS) is used on a shared block, since fw_classify()'s
old-method path needs block->q which is NULL for shared blocks.
The fixed null-ptr-deref calling stack:
KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f]
RIP: 0010:fw_classify (net/sched/cls_fw.c:81)
Call Trace:
tcf_classify (./include/net/tc_wrapper.h:197 net/sched/cls_api.c:1764 net/sched/cls_api.c:1860)
tc_run (net/core/dev.c:4401)
__dev_queue_xmit (net/core/dev.c:4535 net/core/dev.c:4790)
Fixes: 1abf272022 ("net: sched: tcindex, fw, flow: use tcf_block_q helper to get struct Qdisc")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Link: https://patch.msgid.link/20260331050217.504278-1-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
When alloc_skb fails in x25_queue_rx_frame it calls kfree_skb(skb) at
line 48 and returns 1 (error).
This error propagates back through the call chain:
x25_queue_rx_frame returns 1
|
v
x25_state3_machine receives the return value 1 and takes the else
branch at line 278, setting queued=0 and returning 0
|
v
x25_process_rx_frame returns queued=0
|
v
x25_backlog_rcv at line 452 sees queued=0 and calls kfree_skb(skb)
again
This would free the same skb twice. Looking at x25_backlog_rcv:
net/x25/x25_in.c:x25_backlog_rcv() {
...
queued = x25_process_rx_frame(sk, skb);
...
if (!queued)
kfree_skb(skb);
}
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Link: https://patch.msgid.link/20260331-x25_fraglen-v4-1-3e69f18464b4@dev.tdt.de
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
In netem_enqueue(), the packet corruption logic uses
get_random_u32_below(skb_headlen(skb)) to select an index for
modifying skb->data. When an AF_PACKET TX_RING sends fully non-linear
packets over an IPIP tunnel, skb_headlen(skb) evaluates to 0.
Passing 0 to get_random_u32_below() takes the variable-ceil slow path
which returns an unconstrained 32-bit random integer. Using this
unconstrained value as an offset into skb->data results in an
out-of-bounds memory access.
Fix this by verifying skb_headlen(skb) is non-zero before attempting
to corrupt the linear data area. Fully non-linear packets will silently
bypass the corruption logic.
Fixes: c865e5d99e ("[PKT_SCHED] netem: packet corruption option")
Reported-by: Yifan Wu <yifanwucs@gmail.com>
Reported-by: Juefei Pu <tomapufckgml@gmail.com>
Signed-off-by: Yuan Tan <tanyuan98@outlook.com>
Signed-off-by: Xin Liu <bird@lzu.edu.cn>
Signed-off-by: Yuhang Zheng <z1652074432@gmail.com>
Signed-off-by: Yucheng Lu <kanolyc@gmail.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Link: https://patch.msgid.link/45435c0935df877853a81e6d06205ac738ec65fa.1774941614.git.kanolyc@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains Netfilter fixes for net. Note that most
of the bugs fixed here are >5 years old. The large PR is not due to an
increase in regressions.
1) Flowtable hardware offload support in IPv6 can lead to out-of-bounds
when populating the rule action array when combined with double-tagged
vlan. Bump the maximum number of actions from 16 to 24 and check that
such limit is never reached, otherwise bail out. This bugs stems from
the original flowtable hardware offload support.
2) nfnetlink_log does not include the netlink header size of the trailing
NLMSG_DONE message when calculating the skb size. From Florian Westphal.
3) Reject names in xt_cgroup and xt_rateest extensions which are not
nul-terminated. Also from Florian.
4) Use nla_strcmp in ipset lookup by set name, since IPSET_ATTR_NAME and
IPSET_ATTR_NAMEREF are of NLA_STRING type. From Florian Westphal.
5) When unregistering conntrack helpers, pass the helper that is going
away so the expectation cleanup is done accordingly, otherwise UaF is
possible when accessing expectation that refer to the helper that is
gone. From Qi Tang.
6) Zero expectation NAT fields to address leaking kernel memory through
the expectation netlink dump when unset. Also from Qi Tang.
7) Use the master conntrack helper when creating expectations via
ctnetlink, ignore the suggested helper through CTA_EXPECT_HELP_NAME.
This allows to address a possible read of kernel memory off the
expectation object boundary.
8) Fix incorrect release of the hash bucket logic in ipset when the
bucket is empty, leading to shrinking the hash bucket to size 0
which deals to out-of-bound write in next element additions.
From Yifan Wu.
9) Allow the use of x_tables extensions that explicitly declare
NFPROTO_ARP support only. This is to avoid an incorrect hook number
validation due to non-overlapping arp and inet hook number
definitions.
10) Reject immediate NF_QUEUE verdict in nf_tables. The userspace
nft tool always uses the nft_queue expression for queueing.
This ensures this verdict cannot be used for the arp family,
which does supported this.
* tag 'nf-26-04-01' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
netfilter: nf_tables: reject immediate NF_QUEUE verdict
netfilter: x_tables: restrict xt_check_match/xt_check_target extensions for NFPROTO_ARP
netfilter: ipset: drop logically empty buckets in mtype_del
netfilter: ctnetlink: ignore explicit helper on new expectations
netfilter: ctnetlink: zero expect NAT fields when CTA_EXPECT_NAT absent
netfilter: nf_conntrack_helper: pass helper to expect cleanup
netfilter: ipset: use nla_strcmp for IPSET_ATTR_NAME attr
netfilter: x_tables: ensure names are nul-terminated
netfilter: nfnetlink_log: account for netlink header size
netfilter: flowtable: strictly check for maximum number of actions
====================
Link: https://patch.msgid.link/20260401103646.1015423-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rds_ib_get_mr() extracts the rds_ib_connection from conn->c_transport_data
and passes it to rds_ib_reg_frmr() for FRWR memory registration. On a
fresh outgoing connection, ic is allocated in rds_ib_conn_alloc() with
i_cm_id = NULL because the connection worker has not yet called
rds_ib_conn_path_connect() to create the rdma_cm_id. When sendmsg() with
RDS_CMSG_RDMA_MAP is called on such a connection, the sendmsg path parses
the control message before any connection establishment, allowing
rds_ib_post_reg_frmr() to dereference ic->i_cm_id->qp and crash the
kernel.
The existing guard in rds_ib_reg_frmr() only checks for !ic (added in
commit 9e630bcb77), which does not catch this case since ic is allocated
early and is always non-NULL once the connection object exists.
KASAN: null-ptr-deref in range [0x0000000000000010-0x0000000000000017]
RIP: 0010:rds_ib_post_reg_frmr+0x50e/0x920
Call Trace:
rds_ib_post_reg_frmr (net/rds/ib_frmr.c:167)
rds_ib_map_frmr (net/rds/ib_frmr.c:252)
rds_ib_reg_frmr (net/rds/ib_frmr.c:430)
rds_ib_get_mr (net/rds/ib_rdma.c:615)
__rds_rdma_map (net/rds/rdma.c:295)
rds_cmsg_rdma_map (net/rds/rdma.c:860)
rds_sendmsg (net/rds/send.c:1363)
____sys_sendmsg
do_syscall_64
Add a check in rds_ib_get_mr() that verifies ic, i_cm_id, and qp are all
non-NULL before proceeding with FRMR registration, mirroring the guard
already present in rds_ib_post_inv(). Return -ENODEV when the connection
is not ready, which the existing error handling in rds_cmsg_send() converts
to -EAGAIN for userspace retry and triggers rds_conn_connect_if_down() to
start the connection worker.
Fixes: 1659185fb4 ("RDS: IB: Support Fastreg MR (FRMR) memory registration mode")
Reported-by: Xiang Mei <xmei5@asu.edu>
Signed-off-by: Weiming Shi <bestswngs@gmail.com>
Reviewed-by: Allison Henderson <achender@kernel.org>
Link: https://patch.msgid.link/20260330163237.2752440-2-bestswngs@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
fib6_metric_set() may be called concurrently from softirq context without
holding the FIB table lock. A typical path is:
ndisc_router_discovery()
spin_unlock_bh(&table->tb6_lock) <- lock released
fib6_metric_set(rt, RTAX_HOPLIMIT, ...) <- lockless call
When two CPUs process Router Advertisement packets for the same router
simultaneously, they can both arrive at fib6_metric_set() with the same
fib6_info pointer whose fib6_metrics still points to dst_default_metrics.
if (f6i->fib6_metrics == &dst_default_metrics) { /* both CPUs: true */
struct dst_metrics *p = kzalloc_obj(*p, GFP_ATOMIC);
refcount_set(&p->refcnt, 1);
f6i->fib6_metrics = p; /* CPU1 overwrites CPU0's p -> p0 leaked */
}
The dst_metrics allocated by the losing CPU has refcnt=1 but no pointer
to it anywhere in memory, producing a kmemleak report:
unreferenced object 0xff1100025aca1400 (size 96):
comm "softirq", pid 0, jiffies 4299271239
backtrace:
kmalloc_trace+0x28a/0x380
fib6_metric_set+0xcd/0x180
ndisc_router_discovery+0x12dc/0x24b0
icmpv6_rcv+0xc16/0x1360
Fix this by:
- Set val for p->metrics before published via cmpxchg() so the metrics
value is ready before the pointer becomes visible to other CPUs.
- Replace the plain pointer store with cmpxchg() and free the allocation
safely when competition failed.
- Add READ_ONCE()/WRITE_ONCE() for metrics[] setting in the non-default
metrics path to prevent compiler-based data races.
Fixes: d4ead6b34b ("net/ipv6: move metrics from dst to rt6_info")
Reported-by: Fei Liu <feliu@redhat.com>
Reviewed-by: Jiayuan Chen <jiayuan.chen@linux.dev>
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20260331-b4-fib6_metric_set-kmemleak-v3-1-88d27f4d8825@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
