Yi Chen reports that 'udpclash' loops forever depending on compiler
(and optimization level used); while (x == 1) gets optimized into
for (;;). Add volatile qualifier to avoid that.
While at it, also run it under timeout(1) and fix the resize script
to not ignore the timeout passed as second parameter to insert_flood.
Reported-by: Yi Chen <yiche@redhat.com> Suggested-by: Yi Chen <yiche@redhat.com> Fixes: 78a588363587 ("selftests: netfilter: add conntrack clash resolution test case") Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
In mctp_getsockopt(), unrecognized options currently return -EINVAL.
In contrast, mctp_setsockopt() returns -ENOPROTOOPT for unknown
options.
Update mctp_getsockopt() to also return -ENOPROTOOPT for unknown
options. This aligns the behavior of getsockopt() and setsockopt(),
and matches the standard kernel socket API convention for handling
unsupported options.
Currently SMC code is validating the reserved bits while parsing the incoming
CLC decline message & when this validation fails, its treated as a protocol
error. As a result, the SMC connection is terminated instead of falling back to
TCP. As per RFC7609[1] specs we shouldn't be validating the reserved bits that
is part of CLC message. This patch fixes this issue.
The inetdev_init() function never returns NULL. Check for error
pointers instead.
Fixes: 22600596b675 ("ipv4: give an IPv4 dev to blackhole_netdev") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://patch.msgid.link/aLaQWL9NguWmeM1i@stanley.mountain Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
All paths in probe that call goto defer do so before assigning phydev
and thus it makes sense to cleanup the prior index. It also fixes a bug
where index 0 does not get cleaned up.
Followup of commit c51da3f7a161 ("net: remove sock_i_uid()")
A recent syzbot report was the trigger for this change.
Over the years, we had many problems caused by the
read_lock[_bh](&sk->sk_callback_lock) in sock_i_uid().
We could fix smc_diag_dump_proto() or make a more radical move:
Instead of waiting for new syzbot reports, cache the socket
inode number in sk->sk_ino, so that we no longer
need to acquire sk->sk_callback_lock in sock_i_ino().
This makes socket dumps faster (one less cache line miss,
and two atomic ops avoided).
Prior art:
commit 25a9c8a4431c ("netlink: Add __sock_i_ino() for __netlink_diag_dump().")
commit 4f9bf2a2f5aa ("tcp: Don't acquire inet_listen_hashbucket::lock with disabled BH.")
commit efc3dbc37412 ("rds: Make rds_sock_lock BH rather than IRQ safe.")
Fixes: d2d6422f8bd1 ("x86: Allow to enable PREEMPT_RT.") Reported-by: syzbot+50603c05bbdf4dfdaffa@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/68b73804.050a0220.3db4df.01d8.GAE@google.com/T/#u Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com> Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Link: https://patch.msgid.link/20250902183603.740428-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Difference between sock_i_uid() and sk_uid() is that
after sock_orphan(), sock_i_uid() returns GLOBAL_ROOT_UID
while sk_uid() returns the last cached sk->sk_uid value.
None of sock_i_uid() callers care about this.
Use sk_uid() which is much faster and inlined.
Note that diag/dump users are calling sock_i_ino() and
can not see the full benefit yet.
Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Lorenzo Colitti <lorenzo@google.com> Reviewed-by: Maciej Żenczykowski <maze@google.com> Link: https://patch.msgid.link/20250620133001.4090592-3-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Stable-dep-of: 5d6b58c932ec ("net: lockless sock_i_ino()") Signed-off-by: Sasha Levin <sashal@kernel.org>
The blamed commit introduced the concept of split attribute
counting, and later allocating an array to hold them, however
TypeArrayNest wasn't updated to use the new counting variable.
Abbreviated example from tools/net/ynl/generated/nl80211-user.c:
nl80211_if_combination_attributes_parse(...):
unsigned int n_limits = 0;
[...]
ynl_attr_for_each(attr, nlh, yarg->ys->family->hdr_len)
if (type == NL80211_IFACE_COMB_LIMITS)
ynl_attr_for_each_nested(attr2, attr)
dst->_count.limits++;
if (n_limits) {
dst->_count.limits = n_limits;
/* allocate and parse attributes */
}
In the above example n_limits is guaranteed to always be 0,
hence the conditional is unsatisfiable and is optimized out.
This patch changes the attribute counting to use n_limits++ in the
attribute counting loop in the above example.
When the "proxy" option is enabled on a VXLAN device, the device will
suppress ARP requests and IPv6 Neighbor Solicitation messages if it is
able to reply on behalf of the remote host. That is, if a matching and
valid neighbor entry is configured on the VXLAN device whose MAC address
is not behind the "any" remote (0.0.0.0 / ::).
The code currently assumes that the FDB entry for the neighbor's MAC
address points to a valid remote destination, but this is incorrect if
the entry is associated with an FDB nexthop group. This can result in a
NPD [1][3] which can be reproduced using [2][4].
Fix by checking that the remote destination exists before dereferencing
it.
VXLAN FDB entries can point to either a remote destination or an FDB
nexthop group. The latter is usually used in EVPN deployments where
learning is disabled.
However, when learning is enabled, an incoming packet might try to
refresh an FDB entry that points to an FDB nexthop group and therefore
does not have a remote. Such packets should be dropped, but they are
only dropped after dereferencing the non-existent remote, resulting in a
NPD [1] which can be reproduced using [2].
Fix by dropping such packets earlier. Remove the misleading comment from
first_remote_rcu().
ip address add 192.0.2.1/32 dev lo
ip address add 192.0.2.2/32 dev lo
ip nexthop add id 1 via 192.0.2.3 fdb
ip nexthop add id 10 group 1 fdb
ip link add name vx0 up type vxlan id 10010 local 192.0.2.1 dstport 12345 localbypass
ip link add name vx1 up type vxlan id 10020 local 192.0.2.2 dstport 54321 learning
bridge fdb add 00:11:22:33:44:55 dev vx0 self static dst 192.0.2.2 port 54321 vni 10020
bridge fdb add 00:aa:bb:cc:dd:ee dev vx1 self static nhid 10
mausezahn vx0 -a 00:aa:bb:cc:dd:ee -b 00:11:22:33:44:55 -c 1 -q
Fixes: 1274e1cc4226 ("vxlan: ecmp support for mac fdb entries") Reported-by: Marlin Cremers <mcremers@cloudbear.nl> Reviewed-by: Petr Machata <petrm@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org> Link: https://patch.msgid.link/20250901065035.159644-2-idosch@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
This is because in tcp_v6_syn_recv_sock (and the IPv4 counterpart), when
exiting upon error, inet_csk_prepare_forced_close() and tcp_done() need
to be called. They make sure the newsk will end up being correctly
free'd.
tcp_v4_syn_recv_sock() makes this very clear by having the put_and_exit
label that takes care of things. So, this patch here makes sure
tcp_v4_syn_recv_sock and tcp_v6_syn_recv_sock have similar
error-handling and thus fixes the leak for TCP-AO.
Fixes: 06b22ef29591 ("net/tcp: Wire TCP-AO to request sockets") Signed-off-by: Christoph Paasch <cpaasch@openai.com> Reviewed-by: Dmitry Safonov <0x7f454c46@gmail.com> Link: https://patch.msgid.link/20250830-tcpao_leak-v1-1-e5878c2c3173@openai.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
During GTK rekey, mac80211 issues a clear key (if the old key exists)
followed by an install key operation in the same context. This causes
ath11k to send two WMI commands in quick succession: one to clear the
old key and another to install the new key in the same slot.
Under certain conditions—especially under high load or time sensitive
scenarios, firmware may process these commands asynchronously in a way
that firmware assumes the key is cleared whereas hardware has a valid key.
This inconsistency between hardware and firmware leads to group addressed
packet drops. Only setting the same key again can restore a valid key in
firmware and allow packets to be transmitted.
This issue remained latent because the host's clear key commands were
not effective in firmware until commit 436a4e886598 ("ath11k: clear the
keys properly via DISABLE_KEY"). That commit enabled the host to
explicitly clear group keys, which inadvertently exposed the race.
To mitigate this, restrict group key clearing across all modes (AP, STA,
MESH). During rekey, the new key can simply be set on top of the previous
one, avoiding the need for a clear followed by a set.
However, in AP mode specifically, permit group key clearing when no
stations are associated. This exception supports transitions from secure
modes (e.g., WPA2/WPA3) to open mode, during which all associated peers
are removed and the group key is cleared as part of the transition.
Add a per-BSS station counter to track the presence of stations during
set key operations. Also add a reset_group_keys flag to track the key
re-installation state and avoid repeated installation of the same key
when the number of connected stations transitions to non-zero within a
rekey period.
Additionally, for AP and Mesh modes, when the first station associates,
reinstall the same group key that was last set. This ensures that the
firmware recovers from any race that may have occurred during a previous
key clear when no stations were associated.
This change ensures that key clearing is permitted only when no clients
are connected, avoiding packet loss while enabling dynamic security mode
transitions.
Currently, when updating EMLSR capabilities of a multi-link (ML) station,
only the EMLSR parameters (e.g., padding delay, transition delay, and
timeout) are sent to firmware. However, firmware also requires the
EMLSR support flag to be set in the MLO flags of the peer assoc WMI
command to properly handle EML operating mode notification frames.
Set the ATH12K_WMI_FLAG_MLO_EMLSR_SUPPORT flag in the peer assoc WMI
command when the ML station is EMLSR-capable, so that the firmware can
respond to EHT EML action frames from associated stations.
incorrectly used ixgbe_lp_map in loops intended to populate the
supported and advertised EEE linkmode bitmaps based on ixgbe_ls_map.
This results in incorrect bit setting and potential out-of-bounds
access, since ixgbe_lp_map and ixgbe_ls_map have different sizes
and purposes.
ixgbe_lp_map[i] -> ixgbe_ls_map[i]
Use ixgbe_ls_map for supported and advertised linkmodes, and keep
ixgbe_lp_map usage only for link partner (lp_advertised) mapping.
Fixes: 9356b6db9d05 ("net: ethernet: ixgbe: Convert EEE to use linkmodes") Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
list_first_entry() never returns NULL - if the list is empty, it still
returns a pointer to an invalid object, leading to potential invalid
memory access when dereferenced.
Fix this by using list_first_entry_or_null instead of list_first_entry.
Fixes: e3219ce6a775 ("i40e: Add support for client interface for IWARP driver") Signed-off-by: Zhen Ni <zhen.ni@easystack.cn> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The 'command' and 'netdev_ops' debugfs files are a legacy debugging
interface supported by the i40e driver since its early days by commit 02e9c290814c ("i40e: debugfs interface").
Both of these debugfs files provide a read handler which is mostly useless,
and which is implemented with questionable logic. They both use a static
256 byte buffer which is initialized to the empty string. In the case of
the 'command' file this buffer is literally never used and simply wastes
space. In the case of the 'netdev_ops' file, the last command written is
saved here.
On read, the files contents are presented as the name of the device
followed by a colon and then the contents of their respective static
buffer. For 'command' this will always be "<device>: ". For 'netdev_ops',
this will be "<device>: <last command written>". But note the buffer is
shared between all devices operated by this module. At best, it is mostly
meaningless information, and at worse it could be accessed simultaneously
as there doesn't appear to be any locking mechanism.
We have also recently received multiple reports for both read functions
about their use of snprintf and potential overflow that could result in
reading arbitrary kernel memory. For the 'command' file, this is definitely
impossible, since the static buffer is always zero and never written to.
For the 'netdev_ops' file, it does appear to be possible, if the user
carefully crafts the command input, it will be copied into the buffer,
which could be large enough to cause snprintf to truncate, which then
causes the copy_to_user to read beyond the length of the buffer allocated
by kzalloc.
A minimal fix would be to replace snprintf() with scnprintf() which would
cap the return to the number of bytes written, preventing an overflow. A
more involved fix would be to drop the mostly useless static buffers,
saving 512 bytes and modifying the read functions to stop needing those as
input.
Instead, lets just completely drop the read access to these files. These
are debug interfaces exposed as part of debugfs, and I don't believe that
dropping read access will break any script, as the provided output is
pretty useless. You can find the netdev name through other more standard
interfaces, and the 'netdev_ops' interface can easily result in garbage if
you issue simultaneous writes to multiple devices at once.
In order to properly remove the i40e_dbg_netdev_ops_buf, we need to
refactor its write function to avoid using the static buffer. Instead, use
the same logic as the i40e_dbg_command_write, with an allocated buffer.
Update the code to use this instead of the static buffer, and ensure we
free the buffer on exit. This fixes simultaneous writes to 'netdev_ops' on
multiple devices, and allows us to remove the now unused static buffer
along with removing the read access.
Fixes: 02e9c290814c ("i40e: debugfs interface") Reported-by: Kunwu Chan <chentao@kylinos.cn> Closes: https://lore.kernel.org/intel-wired-lan/20231208031950.47410-1-chentao@kylinos.cn/ Reported-by: Wang Haoran <haoranwangsec@gmail.com> Closes: https://lore.kernel.org/all/CANZ3JQRRiOdtfQJoP9QM=6LS1Jto8PGBGw6y7-TL=BcnzHQn1Q@mail.gmail.com/ Reported-by: Amir Mohammad Jahangirzad <a.jahangirzad@gmail.com> Closes: https://lore.kernel.org/all/20250722115017.206969-1-a.jahangirzad@gmail.com/ Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Dawid Osuchowski <dawid.osuchowski@linux.intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Kunwu Chan <kunwu.chan@linux.dev> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
These errors occur during driver load or when changing the MAC via:
ip link set <iface> address <mac>
Add logic to set the MAC type when sending ADD/DEL (opcodes 535/536) to
the control plane. Since only one primary MAC is supported per vport, the
driver only needs to send an ADD opcode when setting it. Remove the old
address by calling __idpf_del_mac_filter(), which skips the message and
just clears the entry from the internal list. This avoids an error on DEL
as it attempts to remove an address already cleared by the preceding ADD
opcode.
Fixes: ce1b75d0635c ("idpf: add ptypes and MAC filter support") Reported-by: Jian Liu <jianliu@redhat.com> Signed-off-by: Emil Tantilov <emil.s.tantilov@intel.com> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Samuel Salin <Samuel.salin@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Recent versions of the E810 firmware have support for an extra interrupt to
handle report of the "low latency" Tx timestamps coming from the
specialized low latency firmware interface. Instead of polling the
registers, software can wait until the low latency interrupt is fired.
This logic makes use of the Tx timestamp tracking structure, ice_ptp_tx, as
it uses the same "ready" bitmap to track which Tx timestamps complete.
Unfortunately, the ice_ll_ts_intr() function does not check if the
tracker is initialized before its first access. This results in NULL
dereference or use-after-free bugs similar to the issues fixed in the
ice_ptp_ts_irq() function.
Fix this by only checking the in_use bitmap (and other fields) if the
tracker is marked as initialized. The reset flow will clear the init field
under lock before it tears the tracker down, thus preventing any
use-after-free or NULL access.
Fixes: 82e71b226e0e ("ice: Enable SW interrupt from FW for LL TS") Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The E810 device has support for a "low latency" firmware interface to
access and read the Tx timestamps. This interface does not use the standard
Tx timestamp logic, due to the latency overhead of proxying sideband
command requests over the firmware AdminQ.
The logic still makes use of the Tx timestamp tracking structure,
ice_ptp_tx, as it uses the same "ready" bitmap to track which Tx
timestamps complete.
Unfortunately, the ice_ptp_ts_irq() function does not check if the tracker
is initialized before its first access. This results in NULL dereference or
use-after-free bugs similar to the following:
This can occur if a Tx timestamp interrupt races with the driver reset
logic.
Fix this by only checking the in_use bitmap (and other fields) if the
tracker is marked as initialized. The reset flow will clear the init field
under lock before it tears the tracker down, thus preventing any
use-after-free or NULL access.
Fixes: f9472aaabd1f ("ice: Process TSYN IRQ in a separate function") Signed-off-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In the TX completion packet stage of TI SoCs with CPSW2G instance, which
has single external ethernet port, ndev is accessed without being
initialized if no TX packets have been processed. It results into null
pointer dereference, causing kernel to crash. Fix this by having a check
on the number of TX packets which have been processed.
We're not currently setting skb->mac_header on ingress, and the netdev
core rx path expects it. Without it, we'll hit a warning on DEBUG_NETDEV
from commit 1e4033b53db4 ("net: skb_reset_mac_len() must check if
mac_header was set")
Initialise the mac_header to refer to the USB transport header.
As of commit f5d83cf0eeb9 ("net: mctp: unshare packets when
reassembling"), we skb_unshare() in mctp_frag_queue(). The unshare may
invalidate the original skb pointer, so we need to treat the skb as
entirely owned by the fraq queue, even on failure.
If the software RoCE device is used, ibdev->dma_device is a null pointer.
As a result, the problem occurs. Null pointer detection is added to
prevent problems.
Fixes: 0ef69e788411c ("net/smc: optimize for smc_sndbuf_sync_sg_for_device and smc_rmb_sync_sg_for_cpu") Signed-off-by: Liu Jian <liujian56@huawei.com> Reviewed-by: Guangguan Wang <guangguan.wang@linux.alibaba.com> Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev> Reviewed-by: D. Wythe <alibuda@linux.alibaba.com> Link: https://patch.msgid.link/20250828124117.2622624-1-liujian56@huawei.com Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The code currently reads both U32 attributes and U64 attributes as
U64, so when a U32 attribute is provided by userspace (ie, when not
using XPN), on big endian systems, we'll load that value into the
upper 32bits of the next_pn field instead of the lower 32bits. This
means that the value that userspace provided is ignored (we only care
about the lower 32bits for non-XPN), and we'll start using PNs from 0.
Switch to nla_get_uint, which will read the value correctly on all
arches, whether it's 32b or 64b.
macb_start_xmit and macb_tx_poll can be called with bottom-halves
disabled (e.g. from softirq) as well as with interrupts disabled (with
netpoll). Because of this, all other functions taking tx_ptr_lock must
use spin_lock_irqsave.
Fixes: 138badbc21a0 ("net: macb: use NAPI for TX completion path") Reported-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Sean Anderson <sean.anderson@linux.dev> Link: https://patch.msgid.link/20250829143521.1686062-1-sean.anderson@linux.dev Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Replace NULL check with IS_ERR() check after calling page_pool_create()
since this function returns error pointers (ERR_PTR).
Using NULL check could lead to invalid pointer dereference.
Fixes: 8533b14b3d65 ("eth: mlx4: create a page pool for Rx") Signed-off-by: Miaoqian Lin <linmq006@gmail.com> Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> Link: https://patch.msgid.link/20250828121858.67639-1-linmq006@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The icmp_ndo_send function was originally introduced to ensure proper
rate limiting when icmp_send is called by a network device driver,
where the packet's source address may have already been transformed
by SNAT.
However, the original implementation only considers the
IP_CT_DIR_ORIGINAL direction for SNAT and always replaced the packet's
source address with that of the original-direction tuple. This causes
two problems:
1. For SNAT:
Reply-direction packets were incorrectly translated using the source
address of the CT original direction, even though no translation is
required.
2. For DNAT:
Reply-direction packets were not handled at all. In DNAT, the original
direction's destination is translated. Therefore, in the reply
direction the source address must be set to the reply-direction
source, so rate limiting works as intended.
Fix this by using the connection direction to select the correct tuple
for source address translation, and adjust the pre-checks to handle
reply-direction packets in case of DNAT.
Additionally, wrap the `ct->status` access in READ_ONCE(). This avoids
possible KCSAN reports about concurrent updates to `ct->status`.
Fixes: 0b41713b6066 ("icmp: introduce helper for nat'd source address in network device context") Signed-off-by: Fabian Bläse <fabian@blaese.de> Cc: Jason A. Donenfeld <Jason@zx2c4.com> Reviewed-by: Florian Westphal <fw@strlen.de> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The warning in bnxt_alloc_one_rx_ring_netmem() reports the number
of pages allocated for the RX aggregation ring. However, it
mistakenly used bp->rx_ring_size instead of bp->rx_agg_ring_size,
leading to confusing or misleading log output.
Use the correct bp->rx_agg_ring_size value to fix this.
Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.") Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Reviewed-by: Michael Chan <michael.chan@broadcom.com> Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com> Link: https://patch.msgid.link/20250830062331.783783-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
dsp_hwec_enable() allocates dup pointer by kstrdup(arg),
but then it updates dup variable by strsep(&dup, ",").
As a result when it calls kfree(dup), the dup variable may be
a modified pointer that no longer points to the original allocated
memory, causing a memory leak.
The issue is the same pattern as fixed in commit c6a502c22999
("mISDN: Fix memory leak in dsp_pipeline_build()").
Fixes: 9a4381618262 ("mISDN: Remove VLAs") Signed-off-by: Miaoqian Lin <linmq006@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250828081457.36061-1-linmq006@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The ptp_ocp_detach() only shuts down the watchdog timer if it is
pending. However, if the timer handler is already running, the
timer_delete_sync() is not called. This leads to race conditions
where the devlink that contains the ptp_ocp is deallocated while
the timer handler is still accessing it, resulting in use-after-free
bugs. The following details one of the race scenarios.
The current code incorrectly passes (XIRCREG1_ECR | FullDuplex) as
the register address to GetByte(), instead of fetching the register
value and OR-ing it with FullDuplex. This results in an invalid
register access.
Fix it by reading XIRCREG1_ECR first, then or-ing with FullDuplex
before writing it back.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com> Link: https://patch.msgid.link/20250827192645.658496-1-alok.a.tiwari@oracle.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Jakub says:
nft_flowtable.sh is one of the most flake-atious test for netdev CI currently :(
The root cause is two-fold:
1. the failing part of the test is supposed to make sure that ip
fragments are forwarded for offloaded flows.
(flowtable has to pass them to classic forward path).
path mtu discovery for these subtests is disabled.
2. nft_flowtable.sh has two passes. One with fixed mtus/file size and
one where link mtus and file sizes are random.
The CI failures all have same pattern:
re-run with random mtus and file size: -o 27663 -l 4117 -r 10089 -s 54384840
[..]
PASS: dscp_egress: dscp packet counters match
FAIL: file mismatch for ns1 -> ns2
In some cases this error triggers a bit ealier, sometimes in a later
subtest:
re-run with random mtus and file size: -o 20201 -l 4555 -r 12657 -s 9405856
[..]
PASS: dscp_egress: dscp packet counters match
PASS: dscp_fwd: dscp packet counters match
2025/08/17 20:37:52 socat[18954] E write(7, 0x560716b96000, 8192): Broken pipe
FAIL: file mismatch for ns1 -> ns2
-rw------- 1 root root 9405856 Aug 17 20:36 /tmp/tmp.2n63vlTrQe
But all logs I saw show same scenario:
1. Failing tests have pmtu discovery off (i.e., ip fragmentation)
2. The test file is much larger than first-pass default (2M Byte)
3. peers have much larger MTUs compared to the 'network'.
These errors are very reproducible when re-running the test with
the same commandline arguments.
The timeout became much more prominent with 1d2fbaad7cd8 ("tcp: stronger sk_rcvbuf checks"): reassembled packets
typically have a skb->truesize more than double the skb length.
As that commit is intentional and pmtud-off with
large-tcp-packets-as-fragments is not normal adjust the test to use a
smaller file for the pmtu-off subtests.
While at it, add more information to pass/fail messages and
also run the dscp alteration subtest with pmtu discovery enabled.
Link: https://netdev.bots.linux.dev/contest.html?test=nft-flowtable-sh Fixes: f84ab634904c ("selftests: netfilter: nft_flowtable.sh: re-run with random mtu sizes") Reported-by: Jakub Kicinski <kuba@kernel.org> Closes: https://lore.kernel.org/netdev/20250822071330.4168f0db@kernel.org/ Signed-off-by: Florian Westphal <fw@strlen.de> Link: https://patch.msgid.link/20250828214918.3385-1-fw@strlen.de Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Move the creation of debugfs files into a dedicated function, and ensure
they are explicitly removed during vhci_release(), before associated
data structures are freed.
Previously, debugfs files such as "force_suspend", "force_wakeup", and
others were created under hdev->debugfs but not removed in
vhci_release(). Since vhci_release() frees the backing vhci_data
structure, any access to these files after release would result in
use-after-free errors.
Although hdev->debugfs is later freed in hci_release_dev(), user can
access files after vhci_data is freed but before hdev->debugfs is
released.
Fixes: ab4e4380d4e1 ("Bluetooth: Add vhci devcoredump support") Signed-off-by: Ivan Pravdin <ipravdin.official@gmail.com> Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
In the fixed commit, I inadvertently removed two configurations
while combining the 0x0083/0x0084 device IDs. Replace the fixed
matches for the BG versions by a masked match and add the BGN
version back with a similar masked match.
The DSM func 0 (DSM_FUNC_QUERY) returns a bitmap of which
other functions contain valid data, query and check it
before returning other functions data.
Fixes: 9db93491f29e ("iwlwifi: acpi: support device specific method (DSM)") Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220085 Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Link: https://patch.msgid.link/20250828095500.881e17ff8f6a.Ic6d92997d9d5fad127919d6e1b830cd3fe944468@changeid Signed-off-by: Sasha Levin <sashal@kernel.org>
We want to see bits being set in the scratch register upon resume, but
if all the bits are set, it means that we were kicked out of the PCI bus
and that clearly doesn't mean we can assume the firmware is still alive
after the suspend / resume cycle.
The helper registration return value is passed-through by module_init
callbacks which modprobe confuses with the harmless -EEXIST returned
when trying to load an already loaded module.
Make sure modprobe fails so users notice their helper has not been
registered and won't work.
When send a broadcast packet to a tap device, which was added to a bridge,
br_nf_local_in() is called to confirm the conntrack. If another conntrack
with the same hash value is added to the hash table, which can be
triggered by a normal packet to a non-bridge device, the below warning
may happen.
To solve the hash conflict, nf_ct_resolve_clash() try to merge the
conntracks, and update skb->_nfct. However, br_nf_local_in() still use the
old ct from local variable 'nfct' after confirm(), which leads to this
warning.
If confirm() does not insert the conntrack entry and return NF_DROP, the
warning may also occur. There is no need to reserve the WARN_ON_ONCE, just
remove it.
Link: https://lore.kernel.org/netdev/20250820043329.2902014-1-wangliang74@huawei.com/ Fixes: 62e7151ae3eb ("netfilter: bridge: confirm multicast packets before passing them up the stack") Suggested-by: Florian Westphal <fw@strlen.de> Signed-off-by: Wang Liang <wangliang74@huawei.com> Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
Since stations are recreated from scratch, all lists that wcids are added
to must be cleared before calling ieee80211_restart_hw.
Set wcid->sta = 0 for each wcid entry in order to ensure that they are
not added again before they are ready.
Non-station wcid entries must not be passed to the rx functions.
In case of the global wcid entry, it could even lead to corruption in the wcid
array due to pointer being casted to struct mt7996_sta_link using container_of.
Fixes: 7464b12b7d92 ("wifi: mt76: mt7996: rework mt7996_rx_get_wcid to support MLO") Link: https://patch.msgid.link/20250827085352.51636-3-nbd@nbd.name Signed-off-by: Felix Fietkau <nbd@nbd.name> Signed-off-by: Sasha Levin <sashal@kernel.org>
When station mode, don't disconnect when we get
channel switch from AP to DFS channel. Most APs
send CSA request after pass background CAC. In other
case we should disconnect after detect beacon miss.
Without patch when we get CSA to DFS channel get:
"kernel: wlo1: preparing for channel switch failed, disconnecting"
The brcmf_btcoex_detach() only shuts down the btcoex timer, if the
flag timer_on is false. However, the brcmf_btcoex_timerfunc(), which
runs as timer handler, sets timer_on to false. This creates critical
race conditions:
1.If brcmf_btcoex_detach() is called while brcmf_btcoex_timerfunc()
is executing, it may observe timer_on as false and skip the call to
timer_shutdown_sync().
2.The brcmf_btcoex_timerfunc() may then reschedule the brcmf_btcoex_info
worker after the cancel_work_sync() has been executed, resulting in
use-after-free bugs.
The use-after-free bugs occur in two distinct scenarios, depending on
the timing of when the brcmf_btcoex_info struct is freed relative to
the execution of its worker thread.
Scenario 1: Freed before the worker is scheduled
The brcmf_btcoex_info is deallocated before the worker is scheduled.
A race condition can occur when schedule_work(&bt_local->work) is
called after the target memory has been freed. The sequence of events
is detailed below:
The brcmf_btcoex_info is freed after the worker has been scheduled
but before or during its execution. In this case, statements within
the brcmf_btcoex_handler() — such as the container_of macro and
subsequent dereferences of the brcmf_btcoex_info object will cause
a use-after-free access. The following timeline illustrates this
scenario:
CPU0 | CPU1
brcmf_btcoex_detach | brcmf_btcoex_timerfunc
| bt_local->timer_on = false;
if (cfg->btcoex->timer_on) |
... |
cancel_work_sync(); |
... | schedule_work(); // Reschedule
|
kfree(cfg->btcoex); // FREE | brcmf_btcoex_handler() // Worker
/* | btci = container_of(....); // USE
The kfree() above could | ...
also occur at any point | btci-> // USE
during the worker's execution|
*/ |
To resolve the race conditions, drop the conditional check and call
timer_shutdown_sync() directly. It can deactivate the timer reliably,
regardless of its current state. Once stopped, the timer_on state is
then set to false.
Fixes: 61730d4dfffc ("brcmfmac: support critical protocol API for DHCP") Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Duoming Zhou <duoming@zju.edu.cn> Link: https://patch.msgid.link/20250822050839.4413-1-duoming@zju.edu.cn Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Following bss_free() quirk introduced in commit 776b3580178f
("cfg80211: track hidden SSID networks properly"), adjust
cfg80211_update_known_bss() to free the last beacon frame
elements only if they're not shared via the corresponding
'hidden_beacon_bss' pointer.
Add missing microSD slot vqmmc-supply property, otherwise the kernel
might shut down LDO5 regulator and that would power off the microSD
card slot, possibly while it is in use. Add the property to make sure
the kernel is aware of the LDO5 regulator which supplies the microSD
slot and keeps the LDO5 enabled.
Fixes: 562d222f23f0 ("arm64: dts: imx8mp: Add support for Data Modul i.MX8M Plus eDM SBC") Signed-off-by: Marek Vasut <marek.vasut@mailbox.org> Signed-off-by: Shawn Guo <shawnguo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Add missing microSD slot vqmmc-supply property, otherwise the kernel
might shut down LDO5 regulator and that would power off the microSD
card slot, possibly while it is in use. Add the property to make sure
the kernel is aware of the LDO5 regulator which supplies the microSD
slot and keeps the LDO5 enabled.
Fixes: 8d6712695bc8 ("arm64: dts: imx8mp: Add support for DH electronics i.MX8M Plus DHCOM and PDK2") Signed-off-by: Marek Vasut <marek.vasut@mailbox.org> Signed-off-by: Shawn Guo <shawnguo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Fix SD card removal caused by automatic LDO5 power off after boot:
LDO5: disabling
mmc1: card 59b4 removed
EXT4-fs (mmcblk1p2): shut down requested (2)
Aborting journal on device mmcblk1p2-8.
JBD2: I/O error when updating journal superblock for mmcblk1p2-8.
To prevent this, add vqmmc regulator for USDHC, using a GPIO-controlled
regulator that is supplied by LDO5. Since this is implemented on SoM but
used on baseboards with SD-card interface, implement the functionality
on SoM part and optionally enable it on baseboards if needed.
Fixes: 418d1d840e42 ("arm64: dts: freescale: add initial device tree for TQMa8MPQL with i.MX8MP") Signed-off-by: Markus Niebel <Markus.Niebel@ew.tq-group.com> Signed-off-by: Alexander Stein <alexander.stein@ew.tq-group.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Chris reported that even on a BIOS that has a new enough SMU F/W
version there is still a spurious IRQ1. Although the solution was
added to SMU F/W 64.66.0 it turns out there needs to be a matching
SBIOS change to activate it. Thus Linux shouldn't be avoiding the
IRQ1 workaround on newer SMU F/W because there is no indication the
BIOS change is in place.
Drop the match for 64.66.0+ and instead match all RN/CZN/BRC (they
all share same SMU F/W). Adjust the quirk infrastructure to allow
quirking the workaround on or off and also adjust existing quirks
to match properly.
Unfortunately this may cause some systems that did have the SBIOS
change in place to regress in keyboard wakeup but we don't have a
way to know. If a user reports a keyboard wakeup regression they can
run with amd_pmc.disable_workarounds=1 to deactivate the workaround
and share DMI data so that their system can be quirked not to use
the workaround in the upstream kernel.
Reported-by: Chris Bainbridge <chris.bainbridge@gmail.com> Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4449 Tested-by: Chris Bainbridge <chris.bainbridge@gmail.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20250724185156.1827592-1-superm1@kernel.org Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Don't call ZONE FINISH for conventional zones as this will result in I/O
errors. Instead check if the zone that needs finishing is a conventional
zone and if yes skip it.
Also factor out the actual handling of finishing a single zone into a
helper function, as do_zone_finish() is growing ever bigger and the
indentations levels are getting higher.
Reviewed-by: Naohiro Aota <naohiro.aota@wdc.com> Reviewed-by: Anand Jain <anand.jain@oracle.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
[BUG]
If submit_one_sector() failed, the block will be kept dirty, but with
their corresponding range finished in the ordered extent.
This means if a writeback happens later again, we can hit the following
problems:
- ASSERT(block_start != EXTENT_MAP_HOLE) in submit_one_sector()
If the original extent map is a hole, then we can hit this case, as
the new ordered extent failed, we will drop the new extent map and
re-read one from the disk.
- DEBUG_WARN() in btrfs_writepage_cow_fixup()
This is because we no longer have an ordered extent for those dirty
blocks. The original for them is already finished with error.
[CAUSE]
The function submit_one_sector() is not following the regular error
handling of writeback. The common practice is to clear the folio dirty,
start and finish the writeback for the block.
This is normally done by extent_clear_unlock_delalloc() with
PAGE_START_WRITEBACK | PAGE_END_WRITEBACK flags during
run_delalloc_range().
So if we keep those failed blocks dirty, they will stay in the page
cache and wait for the next writeback.
And since the original ordered extent is already finished and removed,
depending on the original extent map, we either hit the ASSERT() inside
submit_one_sector(), or hit the DEBUG_WARN() in
btrfs_writepage_cow_fixup().
[FIX]
Follow the regular error handling to clear the dirty flag for the block,
start and finish writeback for that block instead.
Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Qu Wenruo <wqu@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Make video port registers nonvolatile. As DSP_CTRL register is written
to twice due to gamma LUT enable bit which is set outside of the main
DSP_CTRL initialization within atomic_enable (for rk356x case it is also
necessary to always disable gamma LUT before writing a new LUT) there is
a chance that DSP_CTRL value read-out in gamma LUT init/update code is
not the one which was written by the preceding DSP_CTRL initialization
code within atomic_enable. This might result in misconfigured DSP_CTRL
which leads to no visual output[1]. Since DSP_CTRL write takes effect
after VSYNC[1] the issue is not always present. When tested on Pinetab2
with kernel 6.14 it happenes only when DRM is compiled as a module[1].
In order to confirm that it is a timing issue I inserted 18ms udelay
before vop2_crtc_atomic_try_set_gamma in atomic enable and compiled DRM
as module - this has also fixed the issue.
Link: https://github.com/bluez/bluez/issues/1442 Signed-off-by: Yang Li <yang.li@amlogic.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The set subcommand's -t option is documented as being available for boost
configuration, but it was not actually functioning due to a bug
in the option handling.
When the CPU is offline, the timer of LoongArch is not correctly closed.
This is harmless for real machines, but resulting in an excessively high
cpu usage rate of the offline vCPU thread in the virtual machines.
To correctly close the timer, we have made the following modifications:
Register the cpu hotplug event (CPUHP_AP_LOONGARCH_ARCH_TIMER_STARTING)
for LoongArch. This event's hooks will be called to close the timer when
the CPU is offline.
Clear the timer interrupt when the timer is turned off. Since before the
timer is turned off, there may be a timer interrupt that has already been
in the pending state due to the interruption of the disabled, which also
affects the halt state of the offline vCPU.
Now if preemption happens between protected_save_fpu_context() and
protected_save_lbt_context(), FTOP context is lost. Because FTOP is
saved by protected_save_lbt_context() but protected_save_fpu_context()
disables TM before that. So save LBT before FPU in setup_sigcontext()
to avoid this potential risk.
Commit 05f254a6369ac020fc0382a7cbd3ef64ad997c92 ("ALSA: usb-audio:
Improve filtering of sample rates on Focusrite devices") changed the
check for max_rate in a way which was overly restrictive, forcing
devices to use very high samplerates if they support them, despite
support existing for lower rates as well.
This maintains the intended outcome (ensuring samplerates selected are
supported) while allowing devices with higher maximum samplerates to be
opened at all supported samplerates.
This patch was tested with a Clarett+ 8Pre USB
Fixes: 05f254a6369a ("ALSA: usb-audio: Improve filtering of sample rates on Focusrite devices") Signed-off-by: Tina Wuest <tina@wuest.me> Link: https://patch.msgid.link/20250901092024.140993-1-tina@wuest.me Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
commit 2c6b6a3e8b93 ("ASoC: rsnd: use snd_pcm_direction_name()") uses
snd_pcm_direction_name() instead of original method to get string
"Playback" or "Capture". But io->substream might be NULL in this timing.
Let's re-use original method.
Fixes: 2c6b6a3e8b93 ("ASoC: rsnd: use snd_pcm_direction_name()") Reported-by: Thuan Nguyen <thuan.nguyen-hong@banvien.com.vn> Tested-by: Thuan Nguyen <thuan.nguyen-hong@banvien.com.vn> Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-ID: <87zfbmwq6v.wl-kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
soc-generic-dmaengine-pcm.c uses same dev for both CPU and Platform.
In such case, CPU component driver might not have driver->name, then
snd_soc_lookup_component_nolocked() will be NULL pointer access error.
Care NULL driver name.
Call trace:
strcmp from snd_soc_lookup_component_nolocked+0x64/0xa4
snd_soc_lookup_component_nolocked from snd_soc_unregister_component_by_driver+0x2c/0x44
snd_soc_unregister_component_by_driver from snd_dmaengine_pcm_unregister+0x28/0x64
snd_dmaengine_pcm_unregister from devres_release_all+0x98/0xfc
devres_release_all from device_unbind_cleanup+0xc/0x60
device_unbind_cleanup from really_probe+0x220/0x2c8
really_probe from __driver_probe_device+0x88/0x1a0
__driver_probe_device from driver_probe_device+0x30/0x110
driver_probe_device from __driver_attach+0x90/0x178
__driver_attach from bus_for_each_dev+0x7c/0xcc
bus_for_each_dev from bus_add_driver+0xcc/0x1ec
bus_add_driver from driver_register+0x80/0x11c
driver_register from do_one_initcall+0x58/0x23c
do_one_initcall from kernel_init_freeable+0x198/0x1f4
kernel_init_freeable from kernel_init+0x1c/0x12c
kernel_init from ret_from_fork+0x14/0x28
Fixes: 144d6dfc7482 ("ASoC: soc-core: merge snd_soc_unregister_component() and snd_soc_unregister_component_by_driver()") Reported-by: J. Neuschäfer <j.ne@posteo.net> Closes: https://lore.kernel.org/r/aJb311bMDc9x-dpW@probook Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Reported-by: Ondřej Jirman <megi@xff.cz> Closes: https://lore.kernel.org/r/arxpwzu6nzgjxvsndct65ww2wz4aezb5gjdzlgr24gfx7xvyih@natjg6dg2pj6 Tested-by: J. Neuschäfer <j.ne@posteo.net>
Message-ID: <87ect8ysv8.wl-kuninori.morimoto.gx@renesas.com> Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
At inode_logged() we do a couple lockless checks for ->logged_trans, and
these are generally safe except the second one in case we get a load or
store tearing due to a concurrent call updating ->logged_trans (either at
btrfs_log_inode() or later at inode_logged()).
In the first case it's safe to compare to the current transaction ID since
once ->logged_trans is set the current transaction, we never set it to a
lower value.
In the second case, where we check if it's greater than zero, we are prone
to load/store tearing races, since we can have a concurrent task updating
to the current transaction ID with store tearing for example, instead of
updating with a single 64 bits write, to update with two 32 bits writes or
four 16 bits writes. In that case the reading side at inode_logged() could
see a positive value that does not match the current transaction and then
return a false negative.
Fix this by doing the second check while holding the inode's spinlock, add
some comments about it too. Also add the data_race() annotation to the
first check to avoid any reports from KCSAN (or similar tools) and comment
about it.
Fixes: 0f8ce49821de ("btrfs: avoid inode logging during rename and link when possible") Reviewed-by: Boris Burkov <boris@bur.io> Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
At inode_logged() if we find that the inode was not logged before we
update its ->last_dir_index_offset to (u64)-1 with the goal that the
next directory log operation will see the (u64)-1 and then figure out
it must check what was the index of the last logged dir index key and
update ->last_dir_index_offset to that key's offset (this is done in
update_last_dir_index_offset()).
This however has a possibility for a time window where a race can happen
and lead to directory logging skipping dir index keys that should be
logged. The race happens like this:
1) Task A calls inode_logged(), sees ->logged_trans as 0 and then checks
that the inode item was logged before, but before it sets the inode's
->last_dir_index_offset to (u64)-1...
2) Task B is at btrfs_log_inode() which calls inode_logged() early, and
that has set ->last_dir_index_offset to (u64)-1;
3) Task B then enters log_directory_changes() which calls
update_last_dir_index_offset(). There it sees ->last_dir_index_offset
is (u64)-1 and that the inode was logged before (ctx->logged_before is
true), and so it searches for the last logged dir index key in the log
tree and it finds that it has an offset (index) value of N, so it sets
->last_dir_index_offset to N, so that we can skip index keys that are
less than or equal to N (later at process_dir_items_leaf());
4) Task A now sets ->last_dir_index_offset to (u64)-1, undoing the update
that task B just did;
5) Task B will now skip every index key when it enters
process_dir_items_leaf(), since ->last_dir_index_offset is (u64)-1.
Fix this by making inode_logged() not touch ->last_dir_index_offset and
initializing it to 0 when an inode is loaded (at btrfs_alloc_inode()) and
then having update_last_dir_index_offset() treat a value of 0 as meaning
we must check the log tree and update with the index of the last logged
index key. This is fine since the minimum possible value for
->last_dir_index_offset is 1 (BTRFS_DIR_START_INDEX - 1 = 2 - 1 = 1).
This also simplifies the management of ->last_dir_index_offset and now
all accesses to it are done under the inode's log_mutex.
Fixes: 0f8ce49821de ("btrfs: avoid inode logging during rename and link when possible") Reviewed-by: Boris Burkov <boris@bur.io> Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
There's a race between checking if an inode was logged before and logging
an inode that can cause us to mark an inode as not logged just after it
was logged by a concurrent task:
1) We have inode X which was not logged before neither in the current
transaction not in past transaction since the inode was loaded into
memory, so it's ->logged_trans value is 0;
2) We are at transaction N;
3) Task A calls inode_logged() against inode X, sees that ->logged_trans
is 0 and there is a log tree and so it proceeds to search in the log
tree for an inode item for inode X. It doesn't see any, but before
it sets ->logged_trans to N - 1...
3) Task B calls btrfs_log_inode() against inode X, logs the inode and
sets ->logged_trans to N;
4) Task A now sets ->logged_trans to N - 1;
5) At this point anyone calling inode_logged() gets 0 (inode not logged)
since ->logged_trans is greater than 0 and less than N, but our inode
was really logged. As a consequence operations like rename, unlink and
link that happen afterwards in the current transaction end up not
updating the log when they should.
Fix this by ensuring inode_logged() only updates ->logged_trans in case
the inode item is not found in the log tree if after tacking the inode's
lock (spinlock struct btrfs_inode::lock) the ->logged_trans value is still
zero, since the inode lock is what protects setting ->logged_trans at
btrfs_log_inode().
Fixes: 0f8ce49821de ("btrfs: avoid inode logging during rename and link when possible") Reviewed-by: Boris Burkov <boris@bur.io> Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The upstream commit a40c5d727b8111b5db424a1e43e14a1dcce1e77f ("drm/dp:
Change AUX DPCD probe address from DPCD_REV to LANE0_1_STATUS") the
reverted commit backported causes a regression, on one eDP panel at
least resulting in display flickering, described in detail at the Link:
below. The issue fixed by the upstream commit will need a different
solution, revert the backport for now.
There's a subtle race in the SCM driver: we assign the __scm pointer
before requesting the waitqueue interrupt. Assigning __scm marks the SCM
API as ready to accept calls. It's possible that a user makes a call
right after we set __scm and the firmware raises an interrupt before the
driver's ready to service it. Move the __scm assignment after we request
the interrupt.
This has the added benefit of allowing us to drop the goto label.
Now that qcom_scm_shm_bridge_enable() uses the struct device passed to
it as argument to make the QCOM_SCM_MP_SHM_BRIDGE_ENABLE SCM call, we
can move the TZMem initialization before the assignment of the __scm
pointer in the SCM driver (which marks SCM as ready to users) thus
fixing the potential race between consumer calls and the memory pool
initialization.
qcom_scm_shm_bridge_enable() is used early in the SCM initialization
routine. It makes an SCM call and so expects the internal __scm pointer
in the SCM driver to be assigned. For this reason the tzmem memory pool
is allocated *after* this pointer is assigned. However, this can lead to
a crash if another consumer of the SCM API makes a call using the memory
pool between the assignment of the __scm pointer and the initialization
of the tzmem memory pool.
As qcom_scm_shm_bridge_enable() is a special case, not meant to be
called by ordinary users, pull it into the local SCM header. Make it
take struct device as argument. This is the device that will be used to
make the SCM call as opposed to the global __scm pointer. This will
allow us to move the tzmem initialization *before* the __scm assignment
in the core SCM driver.
qcom_scm_shm_bridge_create() and qcom_scm_shm_bridge_delete() take
struct device as argument but don't use it. Remove it from these
functions' prototypes.
projects / thirdparty / kernel / stable.git / log
