Carolina Jubran [Mon, 10 Mar 2025 22:01:44 +0000 (00:01 +0200)]
net/mlx5e: Prevent bridge link show failure for non-eswitch-allowed devices
mlx5_eswitch_get_vepa returns -EPERM if the device lacks
eswitch_manager capability, blocking mlx5e_bridge_getlink from
retrieving VEPA mode. Since mlx5e_bridge_getlink implements
ndo_bridge_getlink, returning -EPERM causes bridge link show to fail
instead of skipping devices without this capability.
To avoid this, return -EOPNOTSUPP from mlx5e_bridge_getlink when
mlx5_eswitch_get_vepa fails, ensuring the command continues processing
other devices while ignoring those without the necessary capability.
Fixes: 4b89251de024 ("net/mlx5: Support ndo bridge_setlink and getlink") Signed-off-by: Carolina Jubran <cjubran@nvidia.com> Reviewed-by: Jianbo Liu <jianbol@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Link: https://patch.msgid.link/1741644104-97767-7-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Jianbo Liu [Mon, 10 Mar 2025 22:01:43 +0000 (00:01 +0200)]
net/mlx5: Bridge, fix the crash caused by LAG state check
When removing LAG device from bridge, NETDEV_CHANGEUPPER event is
triggered. Driver finds the lower devices (PFs) to flush all the
offloaded entries. And mlx5_lag_is_shared_fdb is checked, it returns
false if one of PF is unloaded. In such case,
mlx5_esw_bridge_lag_rep_get() and its caller return NULL, instead of
the alive PF, and the flush is skipped.
Besides, the bridge fdb entry's lastuse is updated in mlx5 bridge
event handler. But this SWITCHDEV_FDB_ADD_TO_BRIDGE event can be
ignored in this case because the upper interface for bond is deleted,
and the entry will never be aged because lastuse is never updated.
To make things worse, as the entry is alive, mlx5 bridge workqueue
keeps sending that event, which is then handled by kernel bridge
notifier. It causes the following crash when accessing the passed bond
netdev which is already destroyed.
To fix this issue, remove such checks. LAG state is already checked in
commit 15f8f168952f ("net/mlx5: Bridge, verify LAG state when adding
bond to bridge"), driver still need to skip offload if LAG becomes
invalid state after initialization.
Shay Drory [Mon, 10 Mar 2025 22:01:41 +0000 (00:01 +0200)]
net/mlx5: Fix incorrect IRQ pool usage when releasing IRQs
mlx5_irq_pool_get() is a getter for completion IRQ pool only.
However, after the cited commit, mlx5_irq_pool_get() is called during
ctrl IRQ release flow to retrieve the pool, resulting in the use of an
incorrect IRQ pool.
Hence, use the newly introduced mlx5_irq_get_pool() getter to retrieve
the correct IRQ pool based on the IRQ itself. While at it, rename
mlx5_irq_pool_get() to mlx5_irq_table_get_comp_irq_pool() which
accurately reflects its purpose and improves code readability.
Vlad Dogaru [Mon, 10 Mar 2025 22:01:40 +0000 (00:01 +0200)]
net/mlx5: HWS, Rightsize bwc matcher priority
The bwc layer was clamping the matcher priority from 32 bits to 16 bits.
This didn't show up until a matcher was resized, since the initial
native matcher was created using the correct 32 bit value.
The fix also reorders fields to avoid some padding.
Fixes: 2111bb970c78 ("net/mlx5: HWS, added backward-compatible API handling") Signed-off-by: Vlad Dogaru <vdogaru@nvidia.com> Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/1741644104-97767-3-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Yevgeny Kliteynik [Mon, 10 Mar 2025 22:01:39 +0000 (00:01 +0200)]
net/mlx5: DR, use the right action structs for STEv3
Some actions in ConnectX-8 (STEv3) have different structure,
and they are handled separately in ste_ctx_v3.
This separate handling was missing two actions: INSERT_HDR
and REMOVE_HDR, which broke SWS for Linux Bridge.
This patch resolves the issue by introducing dedicated
callbacks for the insert and remove header functions,
with version-specific implementations for each STE variant.
Fixes: 4d617b57574f ("net/mlx5: DR, add support for ConnectX-8 steering") Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com> Reviewed-by: Itamar Gozlan <igozlan@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/1741644104-97767-2-git-send-email-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Xin Long [Sat, 8 Mar 2025 18:05:43 +0000 (13:05 -0500)]
Revert "openvswitch: switch to per-action label counting in conntrack"
Currently, ovs_ct_set_labels() is only called for confirmed conntrack
entries (ct) within ovs_ct_commit(). However, if the conntrack entry
does not have the labels_ext extension, attempting to allocate it in
ovs_ct_get_conn_labels() for a confirmed entry triggers a warning in
nf_ct_ext_add():
WARN_ON(nf_ct_is_confirmed(ct));
This happens when the conntrack entry is created externally before OVS
increments net->ct.labels_used. The issue has become more likely since
commit fcb1aa5163b1 ("openvswitch: switch to per-action label counting
in conntrack"), which changed to use per-action label counting and
increment net->ct.labels_used when a flow with ct action is added.
Since there’s no straightforward way to fully resolve this issue at the
moment, this reverts the commit to avoid breaking existing use cases.
The actions length check is unreliable and produces different results
depending on the initial length of the provided netlink attribute and
the composition of the actual actions inside of it. For example, a
user can add 4088 empty clone() actions without triggering -EMSGSIZE,
on attempt to add 4089 such actions the operation will fail with the
-EMSGSIZE verdict. However, if another 16 KB of other actions will
be *appended* to the previous 4089 clone() actions, the check passes
and the flow is successfully installed into the openvswitch datapath.
The reason for a such a weird behavior is the way memory is allocated.
When ovs_flow_cmd_new() is invoked, it calls ovs_nla_copy_actions(),
that in turn calls nla_alloc_flow_actions() with either the actual
length of the user-provided actions or the MAX_ACTIONS_BUFSIZE. The
function adds the size of the sw_flow_actions structure and then the
actually allocated memory is rounded up to the closest power of two.
So, if the user-provided actions are larger than MAX_ACTIONS_BUFSIZE,
then MAX_ACTIONS_BUFSIZE + sizeof(*sfa) rounded up is 32K + 24 -> 64K.
Later, while copying individual actions, we look at ksize(), which is
64K, so this way the MAX_ACTIONS_BUFSIZE check is not actually
triggered and the user can easily allocate almost 64 KB of actions.
However, when the initial size is less than MAX_ACTIONS_BUFSIZE, but
the actions contain ones that require size increase while copying
(such as clone() or sample()), then the limit check will be performed
during the reserve_sfa_size() and the user will not be allowed to
create actions that yield more than 32 KB internally.
This is one part of the problem. The other part is that it's not
actually possible for the userspace application to know beforehand
if the particular set of actions will be rejected or not.
Certain actions require more space in the internal representation,
e.g. an empty clone() takes 4 bytes in the action list passed in by
the user, but it takes 12 bytes in the internal representation due
to an extra nested attribute, and some actions require less space in
the internal representations, e.g. set(tunnel(..)) normally takes
64+ bytes in the action list provided by the user, but only needs to
store a single pointer in the internal implementation, since all the
data is stored in the tunnel_info structure instead.
And the action size limit is applied to the internal representation,
not to the action list passed by the user. So, it's not possible for
the userpsace application to predict if the certain combination of
actions will be rejected or not, because it is not possible for it to
calculate how much space these actions will take in the internal
representation without knowing kernel internals.
All that is causing random failures in ovs-vswitchd in userspace and
inability to handle certain traffic patterns as a result. For example,
it is reported that adding a bit more than a 1100 VMs in an OpenStack
setup breaks the network due to OVS not being able to handle ARP
traffic anymore in some cases (it tries to install a proper datapath
flow, but the kernel rejects it with -EMSGSIZE, even though the action
list isn't actually that large.)
Kernel behavior must be consistent and predictable in order for the
userspace application to use it in a reasonable way. ovs-vswitchd has
a mechanism to re-direct parts of the traffic and partially handle it
in userspace if the required action list is oversized, but that doesn't
work properly if we can't actually tell if the action list is oversized
or not.
Solution for this is to check the size of the user-provided actions
instead of the internal representation. This commit just removes the
check from the internal part because there is already an implicit size
check imposed by the netlink protocol. The attribute can't be larger
than 64 KB. Realistically, we could reduce the limit to 32 KB, but
we'll be risking to break some existing setups that rely on the fact
that it's possible to create nearly 64 KB action lists today.
Vast majority of flows in real setups are below 100-ish bytes. So
removal of the limit will not change real memory consumption on the
system. The absolutely worst case scenario is if someone adds a flow
with 64 KB of empty clone() actions. That will yield a 192 KB in the
internal representation consuming 256 KB block of memory. However,
that list of actions is not meaningful and also a no-op. Real world
very large action lists (that can occur for a rare cases of BUM
traffic handling) are unlikely to contain a large number of clones and
will likely have a lot of tunnel attributes making the internal
representation comparable in size to the original action list.
So, it should be fine to just remove the limit.
Commit in the 'Fixes' tag is the first one that introduced the
difference between internal representation and the user-provided action
lists, but there were many more afterwards that lead to the situation
we have today.
====================
gre: Fix regressions in IPv6 link-local address generation.
IPv6 link-local address generation has some special cases for GRE
devices. This has led to several regressions in the past, and some of
them are still not fixed. This series fixes the remaining problems,
like the ipv6.conf.<dev>.addr_gen_mode sysctl being ignored and the
router discovery process not being started (see details in patch 1).
To avoid any further regressions, patch 2 adds selftests covering
IPv4 and IPv6 gre/gretap devices with all combinations of currently
supported addr_gen_mode values.
====================
Guillaume Nault [Fri, 7 Mar 2025 19:28:58 +0000 (20:28 +0100)]
selftests: Add IPv6 link-local address generation tests for GRE devices.
GRE devices have their special code for IPv6 link-local address
generation that has been the source of several regressions in the past.
Add selftest to check that all gre, ip6gre, gretap and ip6gretap get an
IPv6 link-link local address in accordance with the
net.ipv6.conf.<dev>.addr_gen_mode sysctl.
Guillaume Nault [Fri, 7 Mar 2025 19:28:53 +0000 (20:28 +0100)]
gre: Fix IPv6 link-local address generation.
Use addrconf_addr_gen() to generate IPv6 link-local addresses on GRE
devices in most cases and fall back to using add_v4_addrs() only in
case the GRE configuration is incompatible with addrconf_addr_gen().
GRE used to use addrconf_addr_gen() until commit e5dd729460ca
("ip/ip6_gre: use the same logic as SIT interfaces when computing v6LL
address") restricted this use to gretap and ip6gretap devices, and
created add_v4_addrs() (borrowed from SIT) for non-Ethernet GRE ones.
The original problem came when commit 9af28511be10 ("addrconf: refuse
isatap eui64 for INADDR_ANY") made __ipv6_isatap_ifid() fail when its
addr parameter was 0. The commit says that this would create an invalid
address, however, I couldn't find any RFC saying that the generated
interface identifier would be wrong. Anyway, since gre over IPv4
devices pass their local tunnel address to __ipv6_isatap_ifid(), that
commit broke their IPv6 link-local address generation when the local
address was unspecified.
Then commit e5dd729460ca ("ip/ip6_gre: use the same logic as SIT
interfaces when computing v6LL address") tried to fix that case by
defining add_v4_addrs() and calling it to generate the IPv6 link-local
address instead of using addrconf_addr_gen() (apart for gretap and
ip6gretap devices, which would still use the regular
addrconf_addr_gen(), since they have a MAC address).
That broke several use cases because add_v4_addrs() isn't properly
integrated into the rest of IPv6 Neighbor Discovery code. Several of
these shortcomings have been fixed over time, but add_v4_addrs()
remains broken on several aspects. In particular, it doesn't send any
Router Sollicitations, so the SLAAC process doesn't start until the
interface receives a Router Advertisement. Also, add_v4_addrs() mostly
ignores the address generation mode of the interface
(/proc/sys/net/ipv6/conf/*/addr_gen_mode), thus breaking the
IN6_ADDR_GEN_MODE_RANDOM and IN6_ADDR_GEN_MODE_STABLE_PRIVACY cases.
Fix the situation by using add_v4_addrs() only in the specific scenario
where the normal method would fail. That is, for interfaces that have
all of the following characteristics:
* run over IPv4,
* transport IP packets directly, not Ethernet (that is, not gretap
interfaces),
* tunnel endpoint is INADDR_ANY (that is, 0),
* device address generation mode is EUI64.
In all other cases, revert back to the regular addrconf_addr_gen().
Also, remove the special case for ip6gre interfaces in add_v4_addrs(),
since ip6gre devices now always use addrconf_addr_gen() instead.
Cong Wang [Thu, 6 Mar 2025 23:23:55 +0000 (15:23 -0800)]
selftests/tc-testing: Add a test case for DRR class with TC_H_ROOT
Integrate the reproduer from Mingi to TDC.
All test results:
1..4
ok 1 0385 - Create DRR with default setting
ok 2 2375 - Delete DRR with handle
ok 3 3092 - Show DRR class
ok 4 4009 - Reject creation of DRR class with classid TC_H_ROOT
Cong Wang [Thu, 6 Mar 2025 23:23:54 +0000 (15:23 -0800)]
net_sched: Prevent creation of classes with TC_H_ROOT
The function qdisc_tree_reduce_backlog() uses TC_H_ROOT as a termination
condition when traversing up the qdisc tree to update parent backlog
counters. However, if a class is created with classid TC_H_ROOT, the
traversal terminates prematurely at this class instead of reaching the
actual root qdisc, causing parent statistics to be incorrectly maintained.
In case of DRR, this could lead to a crash as reported by Mingi Cho.
Prevent the creation of any Qdisc class with classid TC_H_ROOT
(0xFFFFFFFF) across all qdisc types, as suggested by Jamal.
Reported-by: Mingi Cho <mincho@theori.io> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Fixes: 066a3b5b2346 ("[NET_SCHED] sch_api: fix qdisc_tree_decrease_qlen() loop") Link: https://patch.msgid.link/20250306232355.93864-2-xiyou.wangcong@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
David S. Miller [Wed, 12 Mar 2025 13:56:58 +0000 (13:56 +0000)]
Merge tag 'wireless-2025-03-12' of https://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless
Johannes berg says:
====================
Few more fixes:
- cfg80211/mac80211
- stop possible runaway wiphy worker
- EHT should not use reserved MPDU size bits
- don't run worker for stopped interfaces
- fix SA Query processing with MLO
- fix lookup of assoc link BSS entries
- correct station flush on unauthorize
- iwlwifi:
- TSO fixes
- fix non-MSI-X platforms
- stop possible runaway restart worker
- rejigger maintainers so I'm not CC'ed on
everything
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Benjamin Berg [Tue, 11 Mar 2025 11:17:04 +0000 (12:17 +0100)]
wifi: mac80211: fix MPDU length parsing for EHT 5/6 GHz
The MPDU length is only configured using the EHT capabilities element on
2.4 GHz. On 5/6 GHz it is configured using the VHT or HE capabilities
respectively.
Haoxiang Li [Fri, 7 Mar 2025 09:49:52 +0000 (17:49 +0800)]
qlcnic: fix memory leak issues in qlcnic_sriov_common.c
Add qlcnic_sriov_free_vlans() in qlcnic_sriov_alloc_vlans() if
any sriov_vlans fails to be allocated.
Add qlcnic_sriov_free_vlans() to free the memory allocated by
qlcnic_sriov_alloc_vlans() if "sriov->allowed_vlans" fails to
be allocated.
Justin Lai [Thu, 6 Mar 2025 07:05:10 +0000 (15:05 +0800)]
rtase: Fix improper release of ring list entries in rtase_sw_reset
Since rtase_init_ring, which is called within rtase_sw_reset, adds ring
entries already present in the ring list back into the list, it causes
the ring list to form a cycle. This results in list_for_each_entry_safe
failing to find an endpoint during traversal, leading to an error.
Therefore, it is necessary to remove the previously added ring_list nodes
before calling rtase_init_ring.
Hangbin Liu [Thu, 6 Mar 2025 02:39:22 +0000 (02:39 +0000)]
bonding: fix incorrect MAC address setting to receive NS messages
When validation on the backup slave is enabled, we need to validate the
Neighbor Solicitation (NS) messages received on the backup slave. To
receive these messages, the correct destination MAC address must be added
to the slave. However, the target in bonding is a unicast address, which
we cannot use directly. Instead, we should first convert it to a
Solicited-Node Multicast Address and then derive the corresponding MAC
address.
Fix the incorrect MAC address setting on both slave_set_ns_maddr() and
slave_set_ns_maddrs(). Since the two function names are similar. Add
some description for the functions. Also only use one mac_addr variable
in slave_set_ns_maddr() to save some code and logic.
Fixes: 8eb36164d1a6 ("bonding: add ns target multicast address to slave device") Acked-by: Jay Vosburgh <jv@jvosburgh.net> Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org> Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20250306023923.38777-2-liuhangbin@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Matt Johnston [Thu, 6 Mar 2025 02:32:45 +0000 (10:32 +0800)]
net: mctp: unshare packets when reassembling
Ensure that the frag_list used for reassembly isn't shared with other
packets. This avoids incorrect reassembly when packets are cloned, and
prevents a memory leak due to circular references between fragments and
their skb_shared_info.
The upcoming MCTP-over-USB driver uses skb_clone which can trigger the
problem - other MCTP drivers don't share SKBs.
Amit Cohen [Wed, 5 Mar 2025 12:15:09 +0000 (14:15 +0200)]
net: switchdev: Convert blocking notification chain to a raw one
A blocking notification chain uses a read-write semaphore to protect the
integrity of the chain. The semaphore is acquired for writing when
adding / removing notifiers to / from the chain and acquired for reading
when traversing the chain and informing notifiers about an event.
In case of the blocking switchdev notification chain, recursive
notifications are possible which leads to the semaphore being acquired
twice for reading and to lockdep warnings being generated [1].
Specifically, this can happen when the bridge driver processes a
SWITCHDEV_BRPORT_UNOFFLOADED event which causes it to emit notifications
about deferred events when calling switchdev_deferred_process().
Fix this by converting the notification chain to a raw notification
chain in a similar fashion to the netdev notification chain. Protect
the chain using the RTNL mutex by acquiring it when modifying the chain.
Events are always informed under the RTNL mutex, but add an assertion in
call_switchdev_blocking_notifiers() to make sure this is not violated in
the future.
Maintain the "blocking" prefix as events are always emitted from process
context and listeners are allowed to block.
[1]:
WARNING: possible recursive locking detected 6.14.0-rc4-custom-g079270089484 #1 Not tainted
--------------------------------------------
ip/52731 is trying to acquire lock: ffffffff850918d8 ((switchdev_blocking_notif_chain).rwsem){++++}-{4:4}, at: blocking_notifier_call_chain+0x58/0xa0
but task is already holding lock: ffffffff850918d8 ((switchdev_blocking_notif_chain).rwsem){++++}-{4:4}, at: blocking_notifier_call_chain+0x58/0xa0
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock((switchdev_blocking_notif_chain).rwsem);
lock((switchdev_blocking_notif_chain).rwsem);
*** DEADLOCK ***
May be due to missing lock nesting notation
3 locks held by ip/52731:
#0: ffffffff84f795b0 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_newlink+0x727/0x1dc0
#1: ffffffff8731f628 (&net->rtnl_mutex){+.+.}-{4:4}, at: rtnl_newlink+0x790/0x1dc0
#2: ffffffff850918d8 ((switchdev_blocking_notif_chain).rwsem){++++}-{4:4}, at: blocking_notifier_call_chain+0x58/0xa0
====================
eth: bnxt: fix several bugs in the bnxt module
The first fixes setting incorrect skb->truesize.
When xdp-mb prog returns XDP_PASS, skb is allocated and initialized.
Currently, The truesize is calculated as BNXT_RX_PAGE_SIZE *
sinfo->nr_frags, but sinfo->nr_frags is flushed by napi_build_skb().
So, it stores sinfo before calling napi_build_skb() and then use it
for calculate truesize.
The second fixes kernel panic in the bnxt_queue_mem_alloc().
The bnxt_queue_mem_alloc() accesses rx ring descriptor.
rx ring descriptors are allocated when the interface is up and it's
freed when the interface is down.
So, if bnxt_queue_mem_alloc() is called when the interface is down,
kernel panic occurs.
This patch makes the bnxt_queue_mem_alloc() return -ENETDOWN if rx ring
descriptors are not allocated.
The third patch fixes kernel panic in the bnxt_queue_{start | stop}().
When a queue is restarted bnxt_queue_{start | stop}() are called.
These functions set MRU to 0 to stop packet flow and then to set up the
remaining things.
MRU variable is a member of vnic_info[] the first vnic_info is for
default and the second is for ntuple.
The first vnic_info is always allocated when interface is up, but the
second is allocated only when ntuple is enabled.
(ethtool -K eth0 ntuple <on | off>).
Currently, the bnxt_queue_{start | stop}() access
vnic_info[BNXT_VNIC_NTUPLE] regardless of whether ntuple is enabled or
not.
So kernel panic occurs.
This patch make the bnxt_queue_{start | stop}() use bp->nr_vnics instead
of BNXT_VNIC_NTUPLE.
The fourth patch fixes a warning due to checksum state.
The bnxt_rx_pkt() checks whether skb->ip_summed is not CHECKSUM_NONE
before updating ip_summed. if ip_summed is not CHECKSUM_NONE, it WARNS
about it. However, the bnxt_xdp_build_skb() is called in XDP-MB-PASS
path and it updates ip_summed earlier than bnxt_rx_pkt().
So, in the XDP-MB-PASS path, the bnxt_rx_pkt() always warns about
checksum.
Updating ip_summed at the bnxt_xdp_build_skb() is unnecessary and
duplicate, so it is removed.
The fifth patch fixes a kernel panic in the
bnxt_get_queue_stats{rx | tx}().
The bnxt_get_queue_stats{rx | tx}() callback functions are called when
a queue is resetting.
These internally access rx and tx rings without null check, but rings
are allocated and initialized when the interface is up.
So, these functions are called when the interface is down, it
occurs a kernel panic.
The sixth patch fixes memory leak in queue reset logic.
When a queue is resetting, tpa_info is allocated for the new queue and
tpa_info for an old queue is not used anymore.
So it should be freed, but not.
The seventh patch makes net_devmem_unbind_dmabuf() ignore -ENETDOWN.
When devmem socket is closed, net_devmem_unbind_dmabuf() is called to
unbind/release resources.
If interface is down, the driver returns -ENETDOWN.
The -ENETDOWN return value is not an actual error, because the interface
will release resources when the interface is down.
So, net_devmem_unbind_dmabuf() needs to ignore -ENETDOWN.
The last patch adds XDP testcases to
tools/testing/selftests/drivers/net/ping.py.
====================
Taehee Yoo [Sun, 9 Mar 2025 13:42:19 +0000 (13:42 +0000)]
selftests: drv-net: add xdp cases for ping.py
ping.py has 3 cases, test_v4, test_v6 and test_tcp.
But these cases are not executed on the XDP environment.
So, it adds XDP environment, existing tests(test_v4, test_v6, and
test_tcp) are executed too on the below XDP environment.
So, it adds XDP cases.
1. xdp-generic + single-buffer
2. xdp-generic + multi-buffer
3. xdp-native + single-buffer
4. xdp-native + multi-buffer
5. xdp-offload
It also makes test_{v4 | v6 | tcp} sending large size packets. this may
help to check whether multi-buffer is working or not.
Note that the physical interface may be down and then up when xdp is
attached or detached.
This takes some period to activate traffic. So sleep(10) is
added if the test interface is the physical interface.
netdevsim and veth type interfaces skip sleep.
Taehee Yoo [Sun, 9 Mar 2025 13:42:18 +0000 (13:42 +0000)]
net: devmem: do not WARN conditionally after netdev_rx_queue_restart()
When devmem socket is closed, netdev_rx_queue_restart() is called to
reset queue by the net_devmem_unbind_dmabuf(). But callback may return
-ENETDOWN if the interface is down because queues are already freed
when the interface is down so queue reset is not needed.
So, it should not warn if the return value is -ENETDOWN.
Taehee Yoo [Sun, 9 Mar 2025 13:42:17 +0000 (13:42 +0000)]
eth: bnxt: fix memory leak in queue reset
When the queue is reset, the bnxt_alloc_one_tpa_info() is called to
allocate tpa_info for the new queue.
And then the old queue's tpa_info should be removed by the
bnxt_free_one_tpa_info(), but it is not called.
So memory leak occurs.
It adds the bnxt_free_one_tpa_info() in the bnxt_queue_mem_free().
Taehee Yoo [Sun, 9 Mar 2025 13:42:16 +0000 (13:42 +0000)]
eth: bnxt: fix kernel panic in the bnxt_get_queue_stats{rx | tx}
When qstats-get operation is executed, callbacks of netdev_stats_ops
are called. The bnxt_get_queue_stats{rx | tx} collect per-queue stats
from sw_stats in the rings.
But {rx | tx | cp}_ring are allocated when the interface is up.
So, these rings are not allocated when the interface is down.
The qstats-get is allowed even if the interface is down. However,
the bnxt_get_queue_stats{rx | tx}() accesses cp_ring and tx_ring
without null check.
So, it needs to avoid accessing rings if the interface is down.
Reproducer:
ip link set $interface down
./cli.py --spec netdev.yaml --dump qstats-get
OR
ip link set $interface down
python ./stats.py
Taehee Yoo [Sun, 9 Mar 2025 13:42:15 +0000 (13:42 +0000)]
eth: bnxt: do not update checksum in bnxt_xdp_build_skb()
The bnxt_rx_pkt() updates ip_summed value at the end if checksum offload
is enabled.
When the XDP-MB program is attached and it returns XDP_PASS, the
bnxt_xdp_build_skb() is called to update skb_shared_info.
The main purpose of bnxt_xdp_build_skb() is to update skb_shared_info,
but it updates ip_summed value too if checksum offload is enabled.
This is actually duplicate work.
When the bnxt_rx_pkt() updates ip_summed value, it checks if ip_summed
is CHECKSUM_NONE or not.
It means that ip_summed should be CHECKSUM_NONE at this moment.
But ip_summed may already be updated to CHECKSUM_UNNECESSARY in the
XDP-MB-PASS path.
So the by skb_checksum_none_assert() WARNS about it.
This is duplicate work and updating ip_summed in the
bnxt_xdp_build_skb() is not needed.
Taehee Yoo [Sun, 9 Mar 2025 13:42:14 +0000 (13:42 +0000)]
eth: bnxt: do not use BNXT_VNIC_NTUPLE unconditionally in queue restart logic
When a queue is restarted, it sets MRU to 0 for stopping packet flow.
MRU variable is a member of vnic_info[], the first vnic_info is default
and the second is ntuple.
Only when ntuple is enabled(ethtool -K eth0 ntuple on), vnic_info for
ntuple is allocated in init logic.
The bp->nr_vnics indicates how many vnic_info are allocated.
However bnxt_queue_{start | stop}() accesses vnic_info[BNXT_VNIC_NTUPLE]
regardless of ntuple state.
Taehee Yoo [Sun, 9 Mar 2025 13:42:13 +0000 (13:42 +0000)]
eth: bnxt: return fail if interface is down in bnxt_queue_mem_alloc()
The bnxt_queue_mem_alloc() is called to allocate new queue memory when
a queue is restarted.
It internally accesses rx buffer descriptor corresponding to the index.
The rx buffer descriptor is allocated and set when the interface is up
and it's freed when the interface is down.
So, if queue is restarted if interface is down, kernel panic occurs.
Taehee Yoo [Sun, 9 Mar 2025 13:42:12 +0000 (13:42 +0000)]
eth: bnxt: fix truesize for mb-xdp-pass case
When mb-xdp is set and return is XDP_PASS, packet is converted from
xdp_buff to sk_buff with xdp_update_skb_shared_info() in
bnxt_xdp_build_skb().
bnxt_xdp_build_skb() passes incorrect truesize argument to
xdp_update_skb_shared_info().
The truesize is calculated as BNXT_RX_PAGE_SIZE * sinfo->nr_frags but
the skb_shared_info was wiped by napi_build_skb() before.
So it stores sinfo->nr_frags before bnxt_xdp_build_skb() and use it
instead of getting skb_shared_info from xdp_get_shared_info_from_buff().
How to reproduce:
<Node A>
ip link set $interface1 xdp obj xdp_pass.o
ip link set $interface1 mtu 9000 up
ip a a 10.0.0.1/24 dev $interface1
<Node B>
ip link set $interfac2 mtu 9000 up
ip a a 10.0.0.2/24 dev $interface2
ping 10.0.0.1 -s 65000
Following ping.py patch adds xdp-mb-pass case. so ping.py is going to be
able to reproduce this issue.
Oleksij Rempel [Fri, 7 Mar 2025 10:12:23 +0000 (11:12 +0100)]
net: usb: lan78xx: Sanitize return values of register read/write functions
usb_control_msg() returns the number of transferred bytes or a negative
error code. The current implementation propagates the transferred byte
count, which is unintended. This affects code paths that assume a
boolean success/failure check, such as the EEPROM detection logic.
Fix this by ensuring lan78xx_read_reg() and lan78xx_write_reg() return
only 0 on success and preserve negative error codes.
This approach is consistent with existing usage, as the transferred byte
count is not explicitly checked elsewhere.
Fixes: 8b1b2ca83b20 ("net: usb: lan78xx: Improve error handling in EEPROM and OTP operations") Reported-by: Mark Brown <broonie@kernel.org> Closes: https://lore.kernel.org/all/ac965de8-f320-430f-80f6-b16f4e1ba06d@sirena.org.uk Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de> Tested-by: Mark Brown <broonie@kernel.org> Link: https://patch.msgid.link/20250307101223.3025632-1-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Kory Maincent [Fri, 7 Mar 2025 09:12:55 +0000 (10:12 +0100)]
net: ethtool: tsinfo: Fix dump command
Fix missing initialization of ts_info->phc_index in the dump command,
which could cause a netdev interface to incorrectly display a PTP provider
at index 0 instead of "none".
Fix it by initializing the phc_index to -1.
In the same time, restore missing initialization of ts_info.cmd for the
IOCTL case, as it was before the transition from ethnl_default_dumpit to
custom ethnl_tsinfo_dumpit.
Also, remove unnecessary zeroing of ts_info, as it is embedded within
reply_data, which is fully zeroed two lines earlier.
Fixes: b9e3f7dc9ed95 ("net: ethtool: tsinfo: Enhance tsinfo to support several hwtstamp by net topology") Signed-off-by: Kory Maincent <kory.maincent@bootlin.com> Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> Link: https://patch.msgid.link/20250307091255.463559-1-kory.maincent@bootlin.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Wentao Liang [Fri, 7 Mar 2025 02:18:20 +0000 (10:18 +0800)]
net/mlx5: handle errors in mlx5_chains_create_table()
In mlx5_chains_create_table(), the return value of mlx5_get_fdb_sub_ns()
and mlx5_get_flow_namespace() must be checked to prevent NULL pointer
dereferences. If either function fails, the function should log error
message with mlx5_core_warn() and return error pointer.
Breno Leitao [Thu, 6 Mar 2025 13:16:18 +0000 (05:16 -0800)]
netpoll: hold rcu read lock in __netpoll_send_skb()
The function __netpoll_send_skb() is being invoked without holding the
RCU read lock. This oversight triggers a warning message when
CONFIG_PROVE_RCU_LIST is enabled:
====================
net: phy: nxp-c45-tja11xx: add errata for TJA112XA/B
This patch series implements two errata for TJA1120 and TJA1121.
The first errata applicable to both RGMII and SGMII version
of TJA1120 and TJA1121 deals with achieving full silicon performance.
The workaround in this case is putting the PHY in managed mode and
applying a series of PHY writes before the link gest established.
The second errata applicable only to SGMII version of TJA1120 and
TJA1121 deals with achieving a stable operation of SGMII after a
startup event.
The workaround puts the SGMII PCS into power down mode and back up
after restart or wakeup from sleep.
====================
The most recent sillicon versions of TJA1120 and TJA1121 can achieve
full silicon performance by putting the PHY in managed mode.
It is necessary to apply these PHY writes before link gets established.
Application of this fix is required after restart of device and wakeup
from sleep.
Cc: stable@vger.kernel.org Fixes: f1fe5dff2b8a ("net: phy: nxp-c45-tja11xx: add TJA1120 support") Signed-off-by: Andrei Botila <andrei.botila@oss.nxp.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://patch.msgid.link/20250304160619.181046-2-andrei.botila@oss.nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Joseph Huang [Thu, 6 Mar 2025 17:23:05 +0000 (12:23 -0500)]
net: dsa: mv88e6xxx: Verify after ATU Load ops
ATU Load operations could fail silently if there's not enough space
on the device to hold the new entry. When this happens, the symptom
depends on the unknown flood settings. If unknown multicast flood is
disabled, the multicast packets are dropped when the ATU table is
full. If unknown multicast flood is enabled, the multicast packets
will be flooded to all ports. Either way, IGMP snooping is broken
when the ATU Load operation fails silently.
Do a Read-After-Write verification after each fdb/mdb add operation
to make sure that the operation was really successful, and return
-ENOSPC otherwise.
Fixes: defb05b9b9b4 ("net: dsa: mv88e6xxx: Add support for fdb_add, fdb_del, and fdb_getnext") Signed-off-by: Joseph Huang <Joseph.Huang@garmin.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Link: https://patch.msgid.link/20250306172306.3859214-1-Joseph.Huang@garmin.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Mina Almasry [Thu, 6 Mar 2025 21:55:20 +0000 (21:55 +0000)]
netmem: prevent TX of unreadable skbs
Currently on stable trees we have support for netmem/devmem RX but not
TX. It is not safe to forward/redirect an RX unreadable netmem packet
into the device's TX path, as the device may call dma-mapping APIs on
dma addrs that should not be passed to it.
Fix this by preventing the xmit of unreadable skbs.
Tested by configuring tc redirect:
sudo tc qdisc add dev eth1 ingress
sudo tc filter add dev eth1 ingress protocol ip prio 1 flower ip_proto \
tcp src_ip 192.168.1.12 action mirred egress redirect dev eth1
Before, I see unreadable skbs in the driver's TX path passed to dma
mapping APIs.
After, I don't see unreadable skbs in the driver's TX path passed to dma
mapping APIs.
Fixes: 65249feb6b3d ("net: add support for skbs with unreadable frags") Suggested-by: Jakub Kicinski <kuba@kernel.org> Cc: stable@vger.kernel.org Signed-off-by: Mina Almasry <almasrymina@google.com> Link: https://patch.msgid.link/20250306215520.1415465-1-almasrymina@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski [Sat, 8 Mar 2025 03:23:51 +0000 (19:23 -0800)]
Merge tag 'for-net-2025-03-07' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth
Luiz Augusto von Dentz says:
====================
bluetooth pull request for net:
- btusb: Configure altsetting for HCI_USER_CHANNEL
- hci_event: Fix enabling passive scanning
- revert: "hci_core: Fix sleeping function called from invalid context"
- SCO: fix sco_conn refcounting on sco_conn_ready
* tag 'for-net-2025-03-07' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth:
Revert "Bluetooth: hci_core: Fix sleeping function called from invalid context"
Bluetooth: hci_event: Fix enabling passive scanning
Bluetooth: SCO: fix sco_conn refcounting on sco_conn_ready
Bluetooth: btusb: Configure altsetting for HCI_USER_CHANNEL
====================
Luiz Augusto von Dentz [Tue, 4 Mar 2025 15:06:10 +0000 (10:06 -0500)]
Revert "Bluetooth: hci_core: Fix sleeping function called from invalid context"
This reverts commit 4d94f05558271654670d18c26c912da0c1c15549 which has
problems (see [1]) and is no longer needed since 581dd2dc168f
("Bluetooth: hci_event: Fix using rcu_read_(un)lock while iterating")
has reworked the code where the original bug has been found.
Passive scanning shall only be enabled when disconnecting LE links,
otherwise it may start result in triggering scanning when e.g. an ISO
link disconnects:
> HCI Event: LE Meta Event (0x3e) plen 29
LE Connected Isochronous Stream Established (0x19)
Status: Success (0x00)
Connection Handle: 257
CIG Synchronization Delay: 0 us (0x000000)
CIS Synchronization Delay: 0 us (0x000000)
Central to Peripheral Latency: 10000 us (0x002710)
Peripheral to Central Latency: 10000 us (0x002710)
Central to Peripheral PHY: LE 2M (0x02)
Peripheral to Central PHY: LE 2M (0x02)
Number of Subevents: 1
Central to Peripheral Burst Number: 1
Peripheral to Central Burst Number: 1
Central to Peripheral Flush Timeout: 2
Peripheral to Central Flush Timeout: 2
Central to Peripheral MTU: 320
Peripheral to Central MTU: 160
ISO Interval: 10.00 msec (0x0008)
...
> HCI Event: Disconnect Complete (0x05) plen 4
Status: Success (0x00)
Handle: 257
Reason: Remote User Terminated Connection (0x13)
< HCI Command: LE Set Extended Scan Enable (0x08|0x0042) plen 6
Extended scan: Enabled (0x01)
Filter duplicates: Enabled (0x01)
Duration: 0 msec (0x0000)
Period: 0.00 sec (0x0000)
Fixes: 9fcb18ef3acb ("Bluetooth: Introduce LE auto connect options") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Pauli Virtanen [Thu, 27 Feb 2025 21:28:15 +0000 (23:28 +0200)]
Bluetooth: SCO: fix sco_conn refcounting on sco_conn_ready
sco_conn refcount shall not be incremented a second time if the sk
already owns the refcount, so hold only when adding new chan.
Add sco_conn_hold() for clarity, as refcnt is never zero here due to the
sco_conn_add().
Fixes SCO socket shutdown not actually closing the SCO connection.
Fixes: ed9588554943 ("Bluetooth: SCO: remove the redundant sco_conn_put") Signed-off-by: Pauli Virtanen <pav@iki.fi> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Hsin-chen Chuang [Thu, 27 Feb 2025 17:14:10 +0000 (01:14 +0800)]
Bluetooth: btusb: Configure altsetting for HCI_USER_CHANNEL
Automatically configure the altsetting for HCI_USER_CHANNEL when a SCO
is connected.
The motivation is to enable the HCI_USER_CHANNEL user to send out SCO
data through USB Bluetooth chips, which is mainly used for bidirectional
audio transfer (voice call). This was not capable because:
- Per Bluetooth Core Spec v5, Vol 4, Part B, 2.1, the corresponding
alternate setting should be set based on the air mode in order to
transfer SCO data, but
- The Linux Bluetooth HCI_USER_CHANNEL exposes the Bluetooth Host
Controller Interface to the user space, which is something above the
USB layer. The user space is not able to configure the USB alt while
keeping the channel open.
This patch intercepts the HCI_EV_SYNC_CONN_COMPLETE packets in btusb,
extracts the air mode, and configures the alt setting in btusb.
This patch is tested on ChromeOS devices. The USB Bluetooth models
(CVSD, TRANS alt3 and alt6) could work without a customized kernel.
Fixes: b16b327edb4d ("Bluetooth: btusb: add sysfs attribute to control USB alt setting") Signed-off-by: Hsin-chen Chuang <chharry@chromium.org> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Miri Korenblit [Thu, 6 Mar 2025 10:37:59 +0000 (12:37 +0200)]
wifi: cfg80211: cancel wiphy_work before freeing wiphy
A wiphy_work can be queued from the moment the wiphy is allocated and
initialized (i.e. wiphy_new_nm). When a wiphy_work is queued, the
rdev::wiphy_work is getting queued.
If wiphy_free is called before the rdev::wiphy_work had a chance to run,
the wiphy memory will be freed, and then when it eventally gets to run
it'll use invalid memory.
Fix this by canceling the work before freeing the wiphy.
Johannes Berg [Thu, 6 Mar 2025 10:37:58 +0000 (12:37 +0200)]
wifi: mac80211: fix SA Query processing in MLO
When MLO is used and SA Query processing isn't done by
userspace (e.g. wpa_supplicant w/o CONFIG_OCV), then
the mac80211 code kicks in but uses the wrong addresses.
Fix them.
Johannes Berg [Thu, 6 Mar 2025 10:37:57 +0000 (12:37 +0200)]
wifi: nl80211: fix assoc link handling
The refactoring of the assoc link handling in order to
support multi-link reconfiguration broke the setting
of the assoc link ID, and thus resulted in the wrong
BSS "use_for" value being selected. Fix that for both
association and ML reconfiguration.
Fixes: 720fa448f5a7 ("wifi: nl80211: Split the links handling of an association request") Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Link: https://patch.msgid.link/20250306123626.7b233d769c32.I62fd04a8667dd55cedb9a1c0414cc92dd098da75@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Miri Korenblit [Thu, 6 Mar 2025 10:37:56 +0000 (12:37 +0200)]
wifi: mac80211: don't queue sdata::work for a non-running sdata
The worker really shouldn't be queued for a non-running interface.
Also, if ieee80211_setup_sdata is called between queueing and executing
the wk, it will be initialized, which will corrupt wiphy_work_list.
Emmanuel Grumbach [Thu, 6 Mar 2025 10:37:55 +0000 (12:37 +0200)]
wifi: mac80211: flush the station before moving it to UN-AUTHORIZED state
We first want to flush the station to make sure we no longer have any
frames being Tx by the station before the station is moved to
un-authorized state. Failing to do that will lead to races: a frame may
be sent after the station's state has been changed.
Since the API clearly states that the driver can't fail the sta_state()
transition down the list of state, we can easily flush the station
first, and only then call the driver's sta_state().
Miri Korenblit [Thu, 6 Mar 2025 10:25:48 +0000 (12:25 +0200)]
wifi: iwlwifi: trans: cancel restart work on op mode leave
If the restart work happens to run after the opmode left
(i.e. called iwl_trans_op_mode_leave), then the opmode memory (including
its mutex) is likely to be freed already, and trans->opmode is NULL.
Although the hw is stopped in that stage, which means that this restart
got aborted (i.e. STATUS_RESET_PENDING will be cleared),
it still can access trans->opmode (NULL pointer dereference)
or the opmodes memory (which is freed).
Fix this by canceling the restart wk in iwl_trans_op_mode_leave.
Also make sure that the restart wk is really aborted.
Emmanuel Grumbach [Thu, 6 Mar 2025 10:25:47 +0000 (12:25 +0200)]
wifi: iwlwifi: mvm: fix PNVM timeout for non-MSI-X platforms
When MSI-X is not enabled, we mask all the interrupts in the interrupt
handler and re-enable them when the interrupt thread runs. If
STATUS_INT_ENABLED is not set, we won't re-enable in the thread.
In order to get the ALIVE interrupt, we allow the ALIVE interrupt
itself, and RX as well in order to receive the ALIVE notification (which
is received as an RX from the firmware.
The problem is that STATUS_INT_ENABLED is clear until the op_mode calls
trans_fw_alive which means that until trans_fw_alive is called, any
notification from the firmware will not be received.
This became a problem when we inserted the pnvm_load exactly between the
ALIVE and trans_fw_alive.
Fix that by calling trans_fw_alive before loading the PNVM. This will
allow to get the notification from the firmware about PNVM load being
complete and continue the flow normally.
This didn't happen on MSI-X because we don't disable the interrupts in
the ISR when MSI-X is available.
The error in the log looks like this:
iwlwifi 0000:00:03.0: Timeout waiting for PNVM load!
iwlwifi 0000:00:03.0: Failed to start RT ucode: -110
iwlwifi 0000:00:03.0: WRT: Collecting data: ini trigger 13 fired (delay=0ms).
Fixes: 70d3ca86b025 ("iwlwifi: mvm: ring the doorbell and wait for PNVM load completion") Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Link: https://patch.msgid.link/20250306122425.0f2cf207aae1.I025d8f724b44f52eadf6c19069352eb9275613a8@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Ilan Peer [Thu, 6 Mar 2025 10:25:46 +0000 (12:25 +0200)]
wifi: iwlwifi: pcie: Fix TSO preparation
The allocation of the scatter gather data structure should be done
based on the number of memory chunks that need to be mapped, and it
is not dependent on the overall payload length. Fix it.
In addition, as the skb_to_sgvec() function returns an 'int' do not
assign it to an 'unsigned int' as otherwise the error check would be
useless.
Johannes Berg [Thu, 6 Mar 2025 08:28:26 +0000 (09:28 +0100)]
wifi: rework MAINTAINERS entries a bit
Since I really don't want to be CC'ed on every patch
add X: entries for all the drivers that are otherwise
covered. In some cases, add a bit more to drivers that
have other entries, mostly for the vendor directories,
but for libertas also add libertas_tf.
While at it, also add all nl80211-related (vendor)
UAPI header files to the nl80211 entry.
Jakub Kicinski [Fri, 7 Mar 2025 01:58:49 +0000 (17:58 -0800)]
Merge tag 'nf-25-03-06' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains Netfilter fixes for net:
1) Fix racy non-atomic read-then-increment operation with
PREEMPT_RT in nft_ct, from Sebastian Andrzej Siewior.
2) GC is not skipped when jiffies wrap around in nf_conncount,
from Nicklas Bo Jensen.
3) flush_work() on nf_tables_destroy_work waits for the last queued
instance, this could be an instance that is different from the one
that we must wait for, then make destruction work queue.
* tag 'nf-25-03-06' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
netfilter: nf_tables: make destruction work queue pernet
netfilter: nf_conncount: garbage collection is not skipped when jiffies wrap around
netfilter: nft_ct: Use __refcount_inc() for per-CPU nft_ct_pcpu_template.
====================
Jun Yang [Wed, 5 Mar 2025 15:44:10 +0000 (23:44 +0800)]
sched: address a potential NULL pointer dereference in the GRED scheduler.
If kzalloc in gred_init returns a NULL pointer, the code follows the
error handling path, invoking gred_destroy. This, in turn, calls
gred_offload, where memset could receive a NULL pointer as input,
potentially leading to a kernel crash.
When table->opt is NULL in gred_init(), gred_change_table_def()
is not called yet, so it is not necessary to call ->ndo_setup_tc()
in gred_offload().
Signed-off-by: Jun Yang <juny24602@gmail.com> Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com> Fixes: f25c0515c521 ("net: sched: gred: dynamically allocate tc_gred_qopt_offload") Link: https://patch.msgid.link/20250305154410.3505642-1-juny24602@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub Kicinski [Fri, 7 Mar 2025 00:32:11 +0000 (16:32 -0800)]
Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2025-03-05 (ice)
This series contains updates to ice driver.
Larysa removes modification of destination override that caused LLDP
packets to be blocked.
Grzegorz fixes a memory leak in aRFS.
Marcin resolves an issue with operation of switchdev and LAG.
Przemek adjusts order of calls for registering devlink in relation to
health reporters.
* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue:
ice: register devlink prior to creating health reporters
ice: Fix switchdev slow-path in LAG
ice: fix memory leak in aRFS after reset
ice: do not configure destination override for switchdev
====================
- wifi: iwlwifi:
- fix A-MSDU TSO preparation
- free pages allocated when failing to build A-MSDU
- ipv6: fix dst ref loop in ila lwtunnel
- mptcp: fix 'scheduling while atomic' in
mptcp_pm_nl_append_new_local_addr
- bluetooth: add check for mgmt_alloc_skb() in
mgmt_device_connected()
- ethtool: allow NULL nlattrs when getting a phy_device
- eth: be2net: fix sleeping while atomic bugs in
be_ndo_bridge_getlink
Previous releases - always broken:
- core: support TCP GSO case for a few missing flags
- wifi: mac80211:
- fix vendor-specific inheritance
- cleanup sta TXQs on flush
- llc: do not use skb_get() before dev_queue_xmit()
- eth: ipa: nable checksum for IPA_ENDPOINT_AP_MODEM_{RX,TX}
for v4.7"
* tag 'net-6.14-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (41 commits)
net: ipv6: fix missing dst ref drop in ila lwtunnel
net: ipv6: fix dst ref loop in ila lwtunnel
mctp i3c: handle NULL header address
net: dsa: mt7530: Fix traffic flooding for MMIO devices
net-timestamp: support TCP GSO case for a few missing flags
vlan: enforce underlying device type
mptcp: fix 'scheduling while atomic' in mptcp_pm_nl_append_new_local_addr
net: ethtool: netlink: Allow NULL nlattrs when getting a phy_device
ppp: Fix KMSAN uninit-value warning with bpf
net: ipa: Enable checksum for IPA_ENDPOINT_AP_MODEM_{RX,TX} for v4.7
net: ipa: Fix QSB data for v4.7
net: ipa: Fix v4.7 resource group names
net: hns3: make sure ptp clock is unregister and freed if hclge_ptp_get_cycle returns an error
wifi: nl80211: disable multi-link reconfiguration
net: dsa: rtl8366rb: don't prompt users for LED control
be2net: fix sleeping while atomic bugs in be_ndo_bridge_getlink
llc: do not use skb_get() before dev_queue_xmit()
wifi: cfg80211: regulatory: improve invalid hints checking
caif_virtio: fix wrong pointer check in cfv_probe()
net: gso: fix ownership in __udp_gso_segment
...
Linus Torvalds [Thu, 6 Mar 2025 19:19:15 +0000 (09:19 -1000)]
Merge tag 'v6.14-rc5-smb3-fixes' of git://git.samba.org/ksmbd
Pull smb fixes from Steve French:
"Five SMB server fixes, two related client fixes, and minor MAINTAINERS
update:
- Two SMB3 lock fixes fixes (including use after free and bug on fix)
- Fix to race condition that can happen in processing IPC responses
- Four ACL related fixes: one related to endianness of num_aces, and
two related fixes to the checks for num_aces (for both client and
server), and one fixing missing check for num_subauths which can
cause memory corruption
- And minor update to email addresses in MAINTAINERS file"
* tag 'v6.14-rc5-smb3-fixes' of git://git.samba.org/ksmbd:
cifs: fix incorrect validation for num_aces field of smb_acl
ksmbd: fix incorrect validation for num_aces field of smb_acl
smb: common: change the data type of num_aces to le16
ksmbd: fix bug on trap in smb2_lock
ksmbd: fix use-after-free in smb2_lock
ksmbd: fix type confusion via race condition when using ipc_msg_send_request
ksmbd: fix out-of-bounds in parse_sec_desc()
MAINTAINERS: update email address in cifs and ksmbd entry
Linus Torvalds [Thu, 6 Mar 2025 18:18:48 +0000 (08:18 -1000)]
Merge tag 'exfat-for-6.14-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/linkinjeon/exfat
Pull exfat fixes from Namjae Jeon:
- Optimize new cluster allocation by correctly find empty entry slot
- Add a check to prevent excessive bitmap clearing due to invalid
data size of file/dir entry
- Fix incorrect error return for zero-byte writes
* tag 'exfat-for-6.14-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/linkinjeon/exfat:
exfat: add a check for invalid data size
exfat: short-circuit zero-byte writes in exfat_file_write_iter
exfat: fix soft lockup in exfat_clear_bitmap
exfat: fix just enough dentries but allocate a new cluster to dir
Linus Torvalds [Thu, 6 Mar 2025 18:04:49 +0000 (08:04 -1000)]
Merge tag 'vfs-6.14-rc6.fixes' of gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs
Pull vfs fixes from Christian Brauner:
- Fix spelling mistakes in idmappings.rst
- Fix RCU warnings in override_creds()/revert_creds()
- Create new pid namespaces with default limit now that pid_max is
namespaced
* tag 'vfs-6.14-rc6.fixes' of gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs:
pid: Do not set pid_max in new pid namespaces
doc: correcting two prefix errors in idmappings.rst
cred: Fix RCU warnings in override/revert_creds
Linus Torvalds [Thu, 6 Mar 2025 17:53:25 +0000 (07:53 -1000)]
fs/pipe: fix pipe buffer index use in FUSE
This was another case that Rasmus pointed out where the direct access to
the pipe head and tail pointers broke on 32-bit configurations due to
the type changes.
As with the pipe FIONREAD case, fix it by using the appropriate helper
functions that deal with the right pipe index sizing.
Linus Torvalds [Thu, 6 Mar 2025 17:33:58 +0000 (07:33 -1000)]
fs/pipe: do not open-code pipe head/tail logic in FIONREAD
Rasmus points out that we do indeed have other cases of breakage from
the type changes that were introduced on 32-bit targets in order to read
the pipe head and tail values atomically (commit 3d252160b818: "fs/pipe:
Read pipe->{head,tail} atomically outside pipe->mutex").
Fix it up by using the proper helper functions that now deal with the
pipe buffer index types properly. This makes the code simpler and more
obvious.
The compiler does the CSE and loop hoisting of the pipe ring size
masking that we used to do manually, so open-coding this was never a
good idea.
Linus Torvalds [Thu, 6 Mar 2025 17:30:42 +0000 (07:30 -1000)]
fs/pipe: express 'pipe_empty()' in terms of 'pipe_occupancy()'
That's what 'pipe_full()' does, so it's more consistent. But more
importantly it gets the type limits right when the pipe head and tail
are no longer necessarily 'unsigned int'.
Florian Westphal [Thu, 6 Mar 2025 03:05:26 +0000 (04:05 +0100)]
netfilter: nf_tables: make destruction work queue pernet
The call to flush_work before tearing down a table from the netlink
notifier was supposed to make sure that all earlier updates (e.g. rule
add) that might reference that table have been processed.
Unfortunately, flush_work() waits for the last queued instance.
This could be an instance that is different from the one that we must
wait for.
This is because transactions are protected with a pernet mutex, but the
work item is global, so holding the transaction mutex doesn't prevent
another netns from queueing more work.
Make the work item pernet so that flush_work() will wait for all
transactions queued from this netns.
A welcome side effect is that we no longer need to wait for transaction
objects from foreign netns.
The gc work queue is still global. This seems to be ok because nft_set
structures are reference counted and each container structure owns a
reference on the net namespace.
The destroy_list is still protected by a global spinlock rather than
pernet one but the hold time is very short anyway.
v2: call cancel_work_sync before reaping the remaining tables (Pablo).
Fixes: 9f6958ba2e90 ("netfilter: nf_tables: unconditionally flush pending work before notifier") Reported-by: syzbot+5d8c5789c8cb076b2c25@syzkaller.appspotmail.com Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Justin Iurman [Tue, 4 Mar 2025 18:10:39 +0000 (19:10 +0100)]
net: ipv6: fix dst ref loop in ila lwtunnel
This patch follows commit 92191dd10730 ("net: ipv6: fix dst ref loops in
rpl, seg6 and ioam6 lwtunnels") and, on a second thought, the same patch
is also needed for ila (even though the config that triggered the issue
was pathological, but still, we don't want that to happen).
Michal Koutný [Wed, 5 Mar 2025 14:58:49 +0000 (15:58 +0100)]
pid: Do not set pid_max in new pid namespaces
It is already difficult for users to troubleshoot which of multiple pid
limits restricts their workload. The per-(hierarchical-)NS pid_max would
contribute to the confusion.
Also, the implementation copies the limit upon creation from
parent, this pattern showed cumbersome with some attributes in legacy
cgroup controllers -- it's subject to race condition between parent's
limit modification and children creation and once copied it must be
changed in the descendant.
Let's do what other places do (ucounts or cgroup limits) -- create new
pid namespaces without any limit at all. The global limit (actually any
ancestor's limit) is still effectively in place, we avoid the
set/unshare race and bumps of global (ancestral) limit have the desired
effect on pid namespace that do not care.
Lorenzo Bianconi [Tue, 4 Mar 2025 08:50:23 +0000 (09:50 +0100)]
net: dsa: mt7530: Fix traffic flooding for MMIO devices
On MMIO devices (e.g. MT7988 or EN7581) unicast traffic received on lanX
port is flooded on all other user ports if the DSA switch is configured
without VLAN support since PORT_MATRIX in PCR regs contains all user
ports. Similar to MDIO devices (e.g. MT7530 and MT7531) fix the issue
defining default VLAN-ID 0 for MT7530 MMIO devices.
Fixes: 110c18bfed414 ("net: dsa: mt7530: introduce driver for MT7988 built-in switch") Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Reviewed-by: Chester A. Unal <chester.a.unal@arinc9.com> Link: https://patch.msgid.link/20250304-mt7988-flooding-fix-v1-1-905523ae83e9@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Nicklas Bo Jensen [Thu, 27 Feb 2025 13:32:34 +0000 (13:32 +0000)]
netfilter: nf_conncount: garbage collection is not skipped when jiffies wrap around
nf_conncount is supposed to skip garbage collection if it has already
run garbage collection in the same jiffy. Unfortunately, this is broken
when jiffies wrap around which this patch fixes.
The problem is that last_gc in the nf_conncount_list struct is an u32,
but jiffies is an unsigned long which is 8 bytes on my systems. When
those two are compared it only works until last_gc wraps around.
See bug report: https://bugzilla.netfilter.org/show_bug.cgi?id=1778
for more details.
Linus Torvalds [Wed, 5 Mar 2025 17:35:40 +0000 (07:35 -1000)]
fs/pipe: remove buggy and unused 'helper' function
While looking for incorrect users of the pipe head/tail fields (see
commit c27c66afc449: "fs/pipe: Fix pipe_occupancy() with 16-bit
indexes"), I found a bug in pipe_discard_from() that looked entirely
broken.
However, the fix is trivial: this buggy function isn't actually called
by anything, so let's just remove it ASAP.
Linus Torvalds [Wed, 5 Mar 2025 17:08:09 +0000 (07:08 -1000)]
fs/pipe: Fix pipe_occupancy() with 16-bit indexes
The pipe_occupancy() logic implicitly relied on the natural unsigned
modulo arithmetic in C, but that doesn't work for the new 'pipe_index_t'
case, since any arithmetic will be done in 'int' (and here we had also
made it 'unsigned int' due to the function call boundary).
So make the modulo arithmetic explicit by casting the result to the
proper type.
Przemek Kitszel [Wed, 19 Feb 2025 13:30:39 +0000 (14:30 +0100)]
ice: register devlink prior to creating health reporters
ice_health_init() was introduced in the commit 2a82874a3b7b ("ice: add
Tx hang devlink health reporter"). The call to it should have been put
after ice_devlink_register(). It went unnoticed until next reporter by
Konrad, which receives events from FW. FW is reporting all events, also
from prior driver load, and thus it is not unlikely to have something
at the very beginning. And that results in a splat:
[ 24.455950] ? devlink_recover_notify.constprop.0+0x198/0x1b0
[ 24.455973] devlink_health_report+0x5d/0x2a0
[ 24.455976] ? __pfx_ice_health_status_lookup_compare+0x10/0x10 [ice]
[ 24.456044] ice_process_health_status_event+0x1b7/0x200 [ice]
Do the analogous thing for deinit patch.
Fixes: 85d6164ec56d ("ice: add fw and port health reporters") Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov@intel.com> Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Reviewed-by: Konrad Knitter <konrad.knitter@intel.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Marcin Szycik [Thu, 2 Jan 2025 19:07:52 +0000 (20:07 +0100)]
ice: Fix switchdev slow-path in LAG
Ever since removing switchdev control VSI and using PF for port
representor Tx/Rx, switchdev slow-path has been working improperly after
failover in SR-IOV LAG. LAG assumes that the first uplink to be added to
the aggregate will own VFs and have switchdev configured. After
failing-over to the other uplink, representors are still configured to
Tx through the uplink they are set up on, which fails because that
uplink is now down.
On failover, update all PRs on primary uplink to use the currently
active uplink for Tx. Call netif_keep_dst(), as the secondary uplink
might not be in switchdev mode. Also make sure to call
ice_eswitch_set_target_vsi() if uplink is in LAG.
On the Rx path, representors are already working properly, because
default Tx from VFs is set to PF owning the eswitch. After failover the
same PF is receiving traffic from VFs, even though link is down.
Fixes: defd52455aee ("ice: do Tx through PF netdev in slow-path") Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Signed-off-by: Marcin Szycik <marcin.szycik@linux.intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Rafal Romanowski <rafal.romanowski@intel.com> Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Grzegorz Nitka [Thu, 23 Jan 2025 08:15:39 +0000 (09:15 +0100)]
ice: fix memory leak in aRFS after reset
Fix aRFS (accelerated Receive Flow Steering) structures memory leak by
adding a checker to verify if aRFS memory is already allocated while
configuring VSI. aRFS objects are allocated in two cases:
- as part of VSI initialization (at probe), and
- as part of reset handling
However, VSI reconfiguration executed during reset involves memory
allocation one more time, without prior releasing already allocated
resources. This led to the memory leak with the following signature:
Fixes: 28bf26724fdb ("ice: Implement aRFS") Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Signed-off-by: Grzegorz Nitka <grzegorz.nitka@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Rinitha S <sx.rinitha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Larysa Zaremba [Mon, 9 Dec 2024 14:08:53 +0000 (15:08 +0100)]
ice: do not configure destination override for switchdev
After switchdev is enabled and disabled later, LLDP packets sending stops,
despite working perfectly fine before and during switchdev state.
To reproduce (creating/destroying VF is what triggers the reconfiguration):
devlink dev eswitch set pci/<address> mode switchdev
echo '2' > /sys/class/net/<ifname>/device/sriov_numvfs
echo '0' > /sys/class/net/<ifname>/device/sriov_numvfs
This happens because LLDP relies on the destination override functionality.
It needs to 1) set a flag in the descriptor, 2) set the VSI permission to
make it valid. The permissions are set when the PF VSI is first configured,
but switchdev then enables it for the uplink VSI (which is always the PF)
once more when configured and disables when deconfigured, which leads to
software-generated LLDP packets being blocked.
Do not modify the destination override permissions when configuring
switchdev, as the enabled state is the default configuration that is never
modified.
Fixes: 1a1c40df2e80 ("ice: set and release switchdev environment") Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux.intel.com> Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com> Reviewed-by: Simon Horman <horms@kernel.org> Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Jason Xing [Tue, 4 Mar 2025 00:44:29 +0000 (08:44 +0800)]
net-timestamp: support TCP GSO case for a few missing flags
When I read through the TSO codes, I found out that we probably
miss initializing the tx_flags of last seg when TSO is turned
off, which means at the following points no more timestamp
(for this last one) will be generated. There are three flags
to be handled in this patch:
1. SKBTX_HW_TSTAMP
2. SKBTX_BPF
3. SKBTX_SCHED_TSTAMP
Note that SKBTX_BPF[1] was added in 6.14.0-rc2 by commit 6b98ec7e882af ("bpf: Add BPF_SOCK_OPS_TSTAMP_SCHED_CB callback")
and only belongs to net-next branch material for now. The common
issue of the above three flags can be fixed by this single patch.
This patch initializes the tx_flags to SKBTX_ANY_TSTAMP like what
the UDP GSO does to make the newly segmented last skb inherit the
tx_flags so that requested timestamp will be generated in each
certain layer, or else that last one has zero value of tx_flags
which leads to no timestamp at all.
Fixes: 4ed2d765dfacc ("net-timestamp: TCP timestamping") Signed-off-by: Jason Xing <kerneljasonxing@gmail.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Sandeen [Tue, 11 Feb 2025 20:14:21 +0000 (14:14 -0600)]
exfat: short-circuit zero-byte writes in exfat_file_write_iter
When generic_write_checks() returns zero, it means that
iov_iter_count() is zero, and there is no work to do.
Simply return success like all other filesystems do, rather than
proceeding down the write path, which today yields an -EFAULT in
generic_perform_write() via the
(fault_in_iov_iter_readable(i, bytes) == bytes) check when bytes
== 0.
Fixes: 11a347fb6cef ("exfat: change to get file size from DataLength") Reported-by: Noah <kernel-org-10@maxgrass.eu> Signed-off-by: Eric Sandeen <sandeen@redhat.com> Reviewed-by: Yuezhang Mo <Yuezhang.Mo@sony.com> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Namjae Jeon [Fri, 31 Jan 2025 03:55:55 +0000 (12:55 +0900)]
exfat: fix soft lockup in exfat_clear_bitmap
bitmap clear loop will take long time in __exfat_free_cluster()
if data size of file/dir enty is invalid.
If cluster bit in bitmap is already clear, stop clearing bitmap go to
out of loop.
Fixes: 31023864e67a ("exfat: add fat entry operations") Reported-by: Kun Hu <huk23@m.fudan.edu.cn>, Jiaji Qin <jjtan24@m.fudan.edu.cn> Reviewed-by: Sungjong Seo <sj1557.seo@samsung.com> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Yuezhang Mo [Fri, 22 Nov 2024 02:50:55 +0000 (10:50 +0800)]
exfat: fix just enough dentries but allocate a new cluster to dir
This commit fixes the condition for allocating cluster to parent
directory to avoid allocating new cluster to parent directory when
there are just enough empty directory entries at the end of the
parent directory.
Fixes: af02c72d0b62 ("exfat: convert exfat_find_empty_entry() to use dentry cache") Signed-off-by: Yuezhang Mo <Yuezhang.Mo@sony.com> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Linus Torvalds [Wed, 5 Mar 2025 05:05:53 +0000 (19:05 -1000)]
Merge tag 'x86_microcode_for_v6.14_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull AMD microcode loading fixes from Borislav Petkov:
- Load only sha256-signed microcode patch blobs
- Other good cleanups
* tag 'x86_microcode_for_v6.14_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/microcode/AMD: Load only SHA256-checksummed patches
x86/microcode/AMD: Add get_patch_level()
x86/microcode/AMD: Get rid of the _load_microcode_amd() forward declaration
x86/microcode/AMD: Merge early_apply_microcode() into its single callsite
x86/microcode/AMD: Remove unused save_microcode_in_initrd_amd() declarations
x86/microcode/AMD: Remove ugly linebreak in __verify_patch_section() signature
Krister Johansen [Mon, 3 Mar 2025 17:10:13 +0000 (18:10 +0100)]
mptcp: fix 'scheduling while atomic' in mptcp_pm_nl_append_new_local_addr
If multiple connection requests attempt to create an implicit mptcp
endpoint in parallel, more than one caller may end up in
mptcp_pm_nl_append_new_local_addr because none found the address in
local_addr_list during their call to mptcp_pm_nl_get_local_id. In this
case, the concurrent new_local_addr calls may delete the address entry
created by the previous caller. These deletes use synchronize_rcu, but
this is not permitted in some of the contexts where this function may be
called. During packet recv, the caller may be in a rcu read critical
section and have preemption disabled.
An example stack:
BUG: scheduling while atomic: swapper/2/0/0x00000302
This problem seems particularly prevalent if the user advertises an
endpoint that has a different external vs internal address. In the case
where the external address is advertised and multiple connections
already exist, multiple subflow SYNs arrive in parallel which tends to
trigger the race during creation of the first local_addr_list entries
which have the internal address instead.
Fix by skipping the replacement of an existing implicit local address if
called via mptcp_pm_nl_get_local_id.
So, when tb is NULL (such as in the ethnl notify path), we have a nice
crash.
It turns out that there's only the PLCA command that is in that case, as
the other phydev-specific commands don't have a notification.
This commit fixes the crash by passing the cmd index and the nlattr
array separately, allowing NULL-checking it directly inside the helper.
Fixes: c15e065b46dc ("net: ethtool: Allow passing a phy index for some commands") Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Reviewed-by: Kory Maincent <kory.maincent@bootlin.com> Reported-by: Parthiban Veerasooran <parthiban.veerasooran@microchip.com> Link: https://patch.msgid.link/20250301141114.97204-1-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jiayuan Chen [Fri, 28 Feb 2025 14:14:08 +0000 (22:14 +0800)]
ppp: Fix KMSAN uninit-value warning with bpf
Syzbot caught an "KMSAN: uninit-value" warning [1], which is caused by the
ppp driver not initializing a 2-byte header when using socket filter.
The following code can generate a PPP filter BPF program:
'''
struct bpf_program fp;
pcap_t *handle;
handle = pcap_open_dead(DLT_PPP_PPPD, 65535);
pcap_compile(handle, &fp, "ip and outbound", 0, 0);
bpf_dump(&fp, 1);
'''
Its output is:
'''
(000) ldh [2]
(001) jeq #0x21 jt 2 jf 5
(002) ldb [0]
(003) jeq #0x1 jt 4 jf 5
(004) ret #65535
(005) ret #0
'''
Wen can find similar code at the following link:
https://github.com/ppp-project/ppp/blob/master/pppd/options.c#L1680
The maintainer of this code repository is also the original maintainer
of the ppp driver.
As you can see the BPF program skips 2 bytes of data and then reads the
'Protocol' field to determine if it's an IP packet. Then it read the first
byte of the first 2 bytes to determine the direction.
The issue is that only the first byte indicating direction is initialized
in current ppp driver code while the second byte is not initialized.
For normal BPF programs generated by libpcap, uninitialized data won't be
used, so it's not a problem. However, for carefully crafted BPF programs,
such as those generated by syzkaller [2], which start reading from offset
0, the uninitialized data will be used and caught by KMSAN.
Jakub Kicinski [Wed, 5 Mar 2025 00:19:24 +0000 (16:19 -0800)]
Merge branch 'fixes-for-ipa-v4-7'
Luca Weiss says:
====================
Fixes for IPA v4.7
During bringup of IPA v4.7 unfortunately some bits were missed, and it
couldn't be tested much back then due to missing features in tqftpserv
which caused the modem to not enable correctly.
Especially the last commit is important since it makes mobile data
actually functional on SoCs with IPA v4.7 like SM6350 - used on the
Fairphone 4. Before that, you'd get an IP address on the interface but
then e.g. ping never got any response back.
====================
Luca Weiss [Thu, 27 Feb 2025 10:33:42 +0000 (11:33 +0100)]
net: ipa: Enable checksum for IPA_ENDPOINT_AP_MODEM_{RX,TX} for v4.7
Enable the checksum option for these two endpoints in order to allow
mobile data to actually work. Without this, no packets seem to make it
through the IPA.
projects / users / willy / pagecache.git / log