Commit graph

1,385,869 commits

Author SHA1 Message Date
Jonas Gorski
5f93185a75 net: dsa: b53: allow leaky reserved multicast
Allow reserved multicast to ignore VLAN membership so STP and other
management protocols work without a PVID VLAN configured when using a
vlan aware bridge.

Fixes: 967dd82ffc ("net: dsa: b53: Add support for Broadcom RoboSwitch")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Tested-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20250429201710.330937-2-jonas.gorski@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 19:30:34 -07:00
Dave Marquardt
46431fd522 net: ibmveth: Refactored veth_pool_store for better maintainability
Make veth_pool_store detect requested pool changes, close device if
necessary, update pool, and reopen device.

Signed-off-by: Dave Marquardt <davemarq@linux.ibm.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://patch.msgid.link/20250506160004.328347-1-davemarq@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:41:08 -07:00
Eric Biggers
4e0146a94c crypto: arm64/sha256 - fix build when CONFIG_PREEMPT_VOLUNTARY=y
Fix the build of sha256-ce.S when CONFIG_PREEMPT_VOLUNTARY=y by passing
the correct label to the cond_yield macro.  Also adjust the code to
execute only one branch instruction when CONFIG_PREEMPT_VOLUNTARY=n.

Fixes: 6e36be511d ("crypto: arm64/sha256 - implement library instead of shash")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202505071811.yYpLUbav-lkp@intel.com/
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2025-05-08 09:36:45 +08:00
Herbert Xu
3bbc902ece crypto: powerpc/poly1305 - Add missing poly1305_emit_arch
Rename poly1305_emit_64 to poly1305_emit_arch to conform with
the expectation of the poly1305 library.

Reported-by: Thorsten Leemhuis <linux@leemhuis.info>
Fixes: 14d3197914 ("crypto: powerpc/poly1305 - Add block-only interface")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Tested-by: Thorsten Leemhuis <linux@leemhuis.info>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2025-05-08 09:36:45 +08:00
Colin Ian King
8147e75bd5
ASoC: rt712-sdca: remove redundant else path of if statement
There is an if/else check where the else part is executed if
adc_vol_flag is true, this else path checks if adc_vol_flag
is true (which is a redundant second check) and the if path is
always taken. Remove the redundant check and remove the else
path since that can never occur.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://patch.msgid.link/20250507140907.255562-1-colin.i.king@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2025-05-08 10:32:52 +09:00
Mac Chiang
62a5897607
ASoC: Intel: soc-acpi-intel-ptl-match: Add rt713_vb_l3_rt1320_l3 support
This patch adds support for the rt712_vb multi-function codec and
rt1320 amplifier. Both devices are connected via a single Soundwire link 3,
sharing the sdw3-clock, sdw3-data[0] and sdw3-data[1] data lanes
respectively.

Signed-off-by: Mac Chiang <mac.chiang@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://patch.msgid.link/20250507122649.153579-3-yung-chuan.liao@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2025-05-08 10:32:46 +09:00
Mac Chiang
e79cc43b48
ASoC: Intel: soc-acpi-intel-lnl/ptl-match: Simplify variable name
The variable name contains duplications;thus, remove the redundant strings.

Signed-off-by: Mac Chiang <mac.chiang@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://patch.msgid.link/20250507122649.153579-2-yung-chuan.liao@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2025-05-08 10:32:45 +09:00
Aaron Kling
e979a7c79f
spi: tegra114: Use value to check for invalid delays
A delay unit of 0 is a valid entry, thus it is not valid to check for
unused delays. Instead, check the value field; if that is zero, the
given delay is unset.

Fixes: 4426e6b4ec ("spi: tegra114: Don't fail set_cs_timing when delays are zero")
Cc: stable@vger.kernel.org
Signed-off-by: Aaron Kling <webgeek1234@gmail.com>
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://patch.msgid.link/20250506-spi-tegra114-fixup-v1-1-136dc2f732f3@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
2025-05-08 10:32:40 +09:00
Jakub Kicinski
0a055ec03d Merge branch 'netlink-specs-remove-phantom-structs'
Jakub Kicinski says:

====================
netlink: specs: remove phantom structs

rt-netlink and nl80211 have a few structs which may be helpful for Python
decoding of binary attrs, but which don't actually exist in the C uAPI.
This prevents us from using struct pointers for binary types in C.

We could support this situation better in the codegen, or add these
structs to uAPI. That said Johannes suggested we remove the WiFi
structs for now, and the rt-link ones are semi-broken.
Drop the struct definitions, for now, if someone has a need to use
such structs in Python (as opposed to them being defined for completeness)
we can revist.

v1: https://lore.kernel.org/20250505170215.253672-1-kuba@kernel.org
====================

Link: https://patch.msgid.link/20250506194101.696272-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:52 -07:00
Jakub Kicinski
720447bd0b netlink: specs: rt-link: remove implicit structs from devconf
devconf is even odder than SNMP. On input it reports an array of u32s
which seem to be indexed by the enum values - 1. On output kernel
expects a nest where each attr has the enum type as the nla type.

sub-type: u32 is probably best we can do right now.

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250506194101.696272-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:50 -07:00
Jakub Kicinski
ab91c140be netlink: specs: remove implicit structs for SNMP counters
uAPI doesn't define structs for the SNMP counters, just enums to index
them as arrays. Switch to the same representation in the spec. C codegen
will soon need all the struct types to actually exist.

Note that the existing definition was broken, anyway, as the first
member should be the number of counters reported.

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250506194101.696272-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:49 -07:00
Jakub Kicinski
6c2422396d netlink: specs: ovs: correct struct names
C codegen will soon support using struct types for binary attrs.
Correct the struct names in OvS specs.

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20250506194101.696272-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:49 -07:00
Jakub Kicinski
f22e764d77 netlink: specs: nl80211: drop structs which are not uAPI
C codegen will soon use structs for binary types. A handful of structs
in WiFi carry information elements from the wire, defined by the standard.
The structs are not part of uAPI, so we can't use them in C directly.
We could add them to the uAPI or add some annotation to tell the codegen
to output a local version to the user header. The former seems arbitrary
since we don't expose structs for most of the standard. The latter seems
like a lot of work for a rare occurrence. Drop the struct info for now.

Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://lore.kernel.org/004030652d592b379e730be2f0344bebc4a03475.camel@sipsolutions.net
Link: https://patch.msgid.link/20250506194101.696272-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:49 -07:00
Jakub Kicinski
015b5b8ed1 Merge branch 'tools-ynl-gen-split-presence-metadata'
Jakub Kicinski says:

====================
tools: ynl-gen: split presence metadata

The presence metadata indicates whether given attribute was/should be
added to the Netlink message. We have 3 types of such metadata:
 - bit presence for simple values like integers,
 - len presence for variable size attrs, like binary and strings,
 - count for arrays.

Previously this information was spread around with first two types
living in a dedicated sub-struct called _present. The counts resided
directly in the main struct with an n_ prefix.

Reshuffle these an uniformly store them in dedicated sub-structs.
The immediate motivation is that current scheme causes name collisions
for TC.
====================

Link: https://patch.msgid.link/20250505165208.248049-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:28 -07:00
Jakub Kicinski
d307b9feb8 tools: ynl-gen: move the count into a presence struct too
While we reshuffle the presence members, move the counts as well.
Previously array count members would have been place directly in
the struct, so:

  struct family_op_req {
      struct {
            u32 a:1;
            u32 b:1;
      } _present;
      struct {
            u32 bin;
      } _len;

      u32 a;
      u64 b;
      const unsigned char *bin;
      u32 n_multi;                 << count
      u32 *multi;                  << objects
  };

Since len has been moved to its own presence struct move the count
as well:

  struct family_op_req {
      struct {
            u32 a:1;
            u32 b:1;
      } _present;
      struct {
            u32 bin;
      } _len;
      struct {
            u32 multi;             << count
      } _count;

      u32 a;
      u64 b;
      const unsigned char *bin;
      u32 *multi;                  << objects
  };

This improves the consistency and allows us to remove some hacks
in the codegen. Unlike for len there is no known name collision
with the existing scheme.

Link: https://patch.msgid.link/20250505165208.248049-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:26 -07:00
Jakub Kicinski
b8ae9f70aa tools: ynl-gen: split presence metadata
Each YNL struct contains the data and a sub-struct indicating which
fields are valid. Something like:

  struct family_op_req {
      struct {
            u32 a:1;
            u32 b:1;
	    u32 bin_len;
      } _present;

      u32 a;
      u64 b;
      const unsigned char *bin;
  };

Note that the bin object 'bin' has a length stored, and that length
has a _len suffix added to the field name. This breaks if there
is a explicit field called bin_len, which is the case for some
TC actions. Move the length fields out of the _present struct,
create a new struct called _len:

  struct family_op_req {
      struct {
            u32 a:1;
            u32 b:1;
      } _present;
      struct {
	    u32 bin;
      } _len;

      u32 a;
      u64 b;
      const unsigned char *bin;
  };

This should prevent name collisions and help with the packing
of the struct.

Unfortunately this is a breaking change, but hopefully the migration
isn't too painful.

Link: https://patch.msgid.link/20250505165208.248049-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:25 -07:00
Jakub Kicinski
a512be0ecb tools: ynl-gen: rename basic presence from 'bit' to 'present'
Internal change to the code gen. Rename how we indicate a type
has a single bit presence from using a 'bit' string to 'present'.
This is a noop in terms of generated code but will make next
breaking change easier.

Reviewed-by: David Wei <dw@davidwei.uk>
Link: https://patch.msgid.link/20250505165208.248049-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:21:25 -07:00
Jakub Kicinski
ea78f20175 Merge branch 'bug-fixes-from-xdp-patch-series'
Meghana Malladi says:

====================
Bug fixes from XDP patch series

This patch series fixes the bugs introduced while adding
xdp support in the icssg driver, and were reproduced while
running xdp-trafficgen to generate xdp traffic on icssg interfaces.

v1: https://lore.kernel.org/all/20250428120459.244525-1-m-malladi@ti.com/
====================

Link: https://patch.msgid.link/20250506110546.4065715-1-m-malladi@ti.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:19:17 -07:00
Meghana Malladi
1884fc85ae net: ti: icssg-prueth: Report BQL before sending XDP packets
When sending out any kind of traffic, it is essential that the driver
keeps reporting BQL of the number of bytes that have been sent so that
BQL can track the amount of data in the queue and prevents it from
overflowing. If BQL is not reported, the driver may continue sending
packets even when the queue is full, leading to packet loss, congestion
and decreased network performance. Currently this is missing in
emac_xmit_xdp_frame() and this patch fixes it.

Fixes: 62aa3246f4 ("net: ti: icssg-prueth: Add XDP support")
Signed-off-by: Meghana Malladi <m-malladi@ti.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://patch.msgid.link/20250506110546.4065715-4-m-malladi@ti.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:19:12 -07:00
Meghana Malladi
8b3fae3e23 net: ti: icssg-prueth: Fix kernel panic during concurrent Tx queue access
Add __netif_tx_lock() to ensure that only one packet is being
transmitted at a time to avoid race conditions in the netif_txq
struct and prevent packet data corruption. Failing to do so causes
kernel panic with the following error:

[ 2184.746764] ------------[ cut here ]------------
[ 2184.751412] kernel BUG at lib/dynamic_queue_limits.c:99!
[ 2184.756728] Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP

logs: https://gist.github.com/MeghanaMalladiTI/9c7aa5fc3b7fb03f87c74aad487956e9

The lock is acquired before calling emac_xmit_xdp_frame() and released after the
call returns. This ensures that the TX queue is protected from concurrent access
during the transmission of XDP frames.

Fixes: 62aa3246f4 ("net: ti: icssg-prueth: Add XDP support")
Signed-off-by: Meghana Malladi <m-malladi@ti.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://patch.msgid.link/20250506110546.4065715-3-m-malladi@ti.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:19:11 -07:00
Meghana Malladi
e5641daa0e net: ti: icssg-prueth: Set XDP feature flags for ndev
xdp_features demonstrates what all XDP capabilities are supported
on a given network device. The driver needs to set these xdp_features
flag to let the network stack know what XDP features a given driver
is supporting. These flags need to be set for a given ndev irrespective
of any XDP program being loaded or not.

Fixes: 62aa3246f4 ("net: ti: icssg-prueth: Add XDP support")
Signed-off-by: Meghana Malladi <m-malladi@ti.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Link: https://patch.msgid.link/20250506110546.4065715-2-m-malladi@ti.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:19:11 -07:00
Paul Chaignon
f5c79ffdc2 bpf: Clarify handling of mark and tstamp by redirect_peer
When switching network namespaces with the bpf_redirect_peer helper, the
skb->mark and skb->tstamp fields are not zeroed out like they can be on
a typical netns switch. This patch clarifies that in the helper
description.

Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/ccc86af26d43c5c0b776bcba2601b7479c0d46d0.1746460653.git.paul.chaignon@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:16:33 -07:00
Paul Chaignon
c432722994 bpf: Scrub packet on bpf_redirect_peer
When bpf_redirect_peer is used to redirect packets to a device in
another network namespace, the skb isn't scrubbed. That can lead skb
information from one namespace to be "misused" in another namespace.

As one example, this is causing Cilium to drop traffic when using
bpf_redirect_peer to redirect packets that just went through IPsec
decryption to a container namespace. The following pwru trace shows (1)
the packet path from the host's XFRM layer to the container's XFRM
layer where it's dropped and (2) the number of active skb extensions at
each function.

    NETNS       MARK  IFACE  TUPLE                                FUNC
    4026533547  d00   eth0   10.244.3.124:35473->10.244.2.158:53  xfrm_rcv_cb
                             .active_extensions = (__u8)2,
    4026533547  d00   eth0   10.244.3.124:35473->10.244.2.158:53  xfrm4_rcv_cb
                             .active_extensions = (__u8)2,
    4026533547  d00   eth0   10.244.3.124:35473->10.244.2.158:53  gro_cells_receive
                             .active_extensions = (__u8)2,
    [...]
    4026533547  0     eth0   10.244.3.124:35473->10.244.2.158:53  skb_do_redirect
                             .active_extensions = (__u8)2,
    4026534999  0     eth0   10.244.3.124:35473->10.244.2.158:53  ip_rcv
                             .active_extensions = (__u8)2,
    4026534999  0     eth0   10.244.3.124:35473->10.244.2.158:53  ip_rcv_core
                             .active_extensions = (__u8)2,
    [...]
    4026534999  0     eth0   10.244.3.124:35473->10.244.2.158:53  udp_queue_rcv_one_skb
                             .active_extensions = (__u8)2,
    4026534999  0     eth0   10.244.3.124:35473->10.244.2.158:53  __xfrm_policy_check
                             .active_extensions = (__u8)2,
    4026534999  0     eth0   10.244.3.124:35473->10.244.2.158:53  __xfrm_decode_session
                             .active_extensions = (__u8)2,
    4026534999  0     eth0   10.244.3.124:35473->10.244.2.158:53  security_xfrm_decode_session
                             .active_extensions = (__u8)2,
    4026534999  0     eth0   10.244.3.124:35473->10.244.2.158:53  kfree_skb_reason(SKB_DROP_REASON_XFRM_POLICY)
                             .active_extensions = (__u8)2,

In this case, there are no XFRM policies in the container's network
namespace so the drop is unexpected. When we decrypt the IPsec packet,
the XFRM state used for decryption is set in the skb extensions. This
information is preserved across the netns switch. When we reach the
XFRM policy check in the container's netns, __xfrm_policy_check drops
the packet with LINUX_MIB_XFRMINNOPOLS because a (container-side) XFRM
policy can't be found that matches the (host-side) XFRM state used for
decryption.

This patch fixes this by scrubbing the packet when using
bpf_redirect_peer, as is done on typical netns switches via veth
devices except skb->mark and skb->tstamp are not zeroed.

Fixes: 9aa1206e8f ("bpf: Add redirect_peer helper")
Signed-off-by: Paul Chaignon <paul.chaignon@gmail.com>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Martin KaFai Lau <martin.lau@kernel.org>
Link: https://patch.msgid.link/1728ead5e0fe45e7a6542c36bd4e3ca07a73b7d6.1746460653.git.paul.chaignon@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:16:33 -07:00
Lorenzo Bianconi
4a7843cc8a net: airoha: Add missing field to ppe_mbox_data struct
The official Airoha EN7581 firmware requires adding max_packet field in
ppe_mbox_data struct while the unofficial one used to develop the Airoha
EN7581 flowtable support does not require this field.
This patch does not introduce any real backwards compatible issue since
EN7581 fw is not publicly available in linux-firmware or other
repositories (e.g. OpenWrt) yet and the official fw version will use this
new layout. For this reason this change needs to be backported.
Moreover, make explicit the padding added by the compiler introducing
the rsv array in init_info struct.
At the same time use u32 instead of int for init_info and set_info
struct definitions in ppe_mbox_data struct.

Fixes: 23290c7bc1 ("net: airoha: Introduce Airoha NPU support")
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Link: https://patch.msgid.link/20250506-airoha-en7581-fix-ppe_mbox_data-v5-1-29cabed6864d@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 18:08:41 -07:00
Jakub Kicinski
dc75a43c07 netfilter pull request 25-05-08
-----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCgAdFiEEjF9xRqF1emXiQiqU1w0aZmrPKyEFAmgb2D4ACgkQ1w0aZmrP
 KyHB4g//SzExpaI5eTduMIg9OwLmWyUNpsxCX8Qk3E5FX8jyGZDU1CAyTnbYZyuJ
 VAn+5PwhhswEDs85ZgXiHXMjdOXvmPs1l0IKVVlM4DoJqlRDXpAcuKzPH0m0J46Q
 A+nfIIsFfH8roQq+PhLPQ3TEOnIGK1vgNWVICl4oX5sYjkYTIzPnadEaTpNsG26I
 gZlvpqM9EH7KLWqhcX1h2/O+JB/6oNaFQTUK77qc9sGijMVMUlGUBtDN40aZ1IEW
 RP6l6J1Oea3j3wavmrrQxHSDtvGS4rnC0qTQ4Xs3OIeTrQiou4cmsCqp+ZLmMu7Q
 YNBU04o78uEOX9n3ux6Xm4tr2WLizHFNkGBU+m0WQSsEGYx7CWsALXIqnEsZWDLQ
 IK4y/d7M0rWLNB4nYTePIBPjKt+qzBe0gTgVr9jMLFI/X60vz3XYHJDBovaBL3dd
 skiVohmYxjiQj5Y6WPtZaLvubzVEULfjVGHatYLGheLXHsEO/BjXFJeIfYK6bmfO
 eXAx9pPsnqCbtKzh6n+tQzYCeRhlw+Oh9Uw48MDpeIX+cTUNr+eaN1LalQz3C3zX
 KqfnK/ZlWUWpQmyLzhMA6DwSUKPBbJ5nzl/NuWSowUPfnldT++rGAuW4lFOTtMPo
 DsJDnONuE6FPaNU5Ef3rIivVfrAnCAmRK797psL0buHDY9YRGcM=
 =Jwld
 -----END PGP SIGNATURE-----

Merge tag 'nf-25-05-08' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf

Pablo Neira Ayuso says:

====================
Netfilter/IPVS fixes for net

The following patchset contain Netfilter/IPVS fixes for net:

1) Fix KMSAN uninit-value in do_output_route4, reported by syzbot.
   Patch from Julian Anastasov.

2) ipset hashtable set type breaks up the hashtable into regions of
   2^10 buckets. Fix the macro that determines the hashtable lock
   region to protect concurrent updates. From Jozsef Kadlecsik.

* tag 'nf-25-05-08' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
  netfilter: ipset: fix region locking in hash types
  ipvs: fix uninit-value for saddr in do_output_route4
====================

Link: https://patch.msgid.link/20250507221952.86505-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 17:57:04 -07:00
Eelco Chaudron
6beb6835c1 openvswitch: Fix unsafe attribute parsing in output_userspace()
This patch replaces the manual Netlink attribute iteration in
output_userspace() with nla_for_each_nested(), which ensures that only
well-formed attributes are processed.

Fixes: ccb1352e76 ("net: Add Open vSwitch kernel components.")
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
Acked-by: Ilya Maximets <i.maximets@ovn.org>
Acked-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/0bd65949df61591d9171c0dc13e42cea8941da10.1746541734.git.echaudro@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2025-05-07 16:51:02 -07:00
Dan Williams
fba4ceaa24 configfs-tsm-report: Fix NULL dereference of tsm_ops
Unlike sysfs, the lifetime of configfs objects is controlled by
userspace. There is no mechanism for the kernel to find and delete all
created config-items. Instead, the configfs-tsm-report mechanism has an
expectation that tsm_unregister() can happen at any time and cause
established config-item access to start failing.

That expectation is not fully satisfied. While tsm_report_read(),
tsm_report_{is,is_bin}_visible(), and tsm_report_make_item() safely fail
if tsm_ops have been unregistered, tsm_report_privlevel_store()
tsm_report_provider_show() fail to check for ops registration. Add the
missing checks for tsm_ops having been removed.

Now, in supporting the ability for tsm_unregister() to always succeed,
it leaves the problem of what to do with lingering config-items. The
expectation is that the admin that arranges for the ->remove() (unbind)
of the ${tsm_arch}-guest driver is also responsible for deletion of all
open config-items. Until that deletion happens, ->probe() (reload /
bind) of the ${tsm_arch}-guest driver fails.

This allows for emergency shutdown / revocation of attestation
interfaces, and requires coordinated restart.

Fixes: 70e6f7e2b9 ("configfs-tsm: Introduce a shared ABI for attestation reports")
Cc: stable@vger.kernel.org
Cc: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: Steven Price <steven.price@arm.com>
Cc: Sami Mujawar <sami.mujawar@arm.com>
Cc: Borislav Petkov (AMD) <bp@alien8.de>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Reported-by: Cedric Xing <cedric.xing@intel.com>
Reviewed-by: Kai Huang <kai.huang@intel.com>
Link: https://patch.msgid.link/20250430203331.1177062-1-dan.j.williams@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2025-05-07 16:48:37 -07:00
Dan Carpenter
fa332f5dc6 ipmi:msghandler: Fix potential memory corruption in ipmi_create_user()
The "intf" list iterator is an invalid pointer if the correct
"intf->intf_num" is not found.  Calling atomic_dec(&intf->nr_users) on
and invalid pointer will lead to memory corruption.

We don't really need to call atomic_dec() if we haven't called
atomic_add_return() so update the if (intf->in_shutdown) path as well.

Fixes: 8e76741c3d ("ipmi: Add a limit on the number of users that may use IPMI")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Message-ID: <aBjMZ8RYrOt6NOgi@stanley.mountain>
Signed-off-by: Corey Minyard <corey@minyard.net>
2025-05-07 17:25:48 -05:00
Corey Minyard
971a00454d ipmi:watchdog: Use the new interface for panic messages
It's available, remove all the duplicate code.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
6f7f6605c9 ipmi:msghandler: Export and fix panic messaging capability
Don't have the other users that do things at panic time (the watchdog)
do all this themselves, provide a function to do it.

Also, with the new design where most stuff happens at thread context,
a few things needed to be fixed to avoid doing locking in a panic
context.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
be816bc4f8 Documentation:ipmi: Remove comments about interrupt level
Callbacks no longer run at interrupt level or bh, so remove those
comments.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
6bd0eb6d75 ipmi:ssif: Fix a shutdown race
It was possible for the SSIF thread to stop and quit before the
kthread_stop() call because ssif->stopping was set before the
stop.  So only exit the SSIF thread is kthread_should_stop()
returns true.

There is no need to wake the thread, as the wait will be interrupted
by kthread_stop().

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
87105e0780 ipmi:msghandler: Don't deliver messages to deleted users
Check to see if they have been destroyed before trying to deliver a
message.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
ada2abadda ipmi:si: Rework startup of IPMI devices
It is possible in some situations that IPMI devices won't get started up
properly.  This change makes it so all non-duplicate devices will get
started up.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
ed59cd28af ipmi:msghandler: Add a error return from unhandle LAN cmds
If we get a command from a LAN channel, return an error instead of just
throwing it away.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
8871e77ec7 ipmi:msghandler: Shut down lower layer first at unregister
This makes sure any outstanding messages are returned to the user before
the interface is cleaned up.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
60afcc429c ipmi:msghandler: Remove proc_fs.h
It's no longer used.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
f2a31163d6 ipmi:msghandler: Don't check for shutdown when returning responses
The lower level interface shouldn't attempt to unregister if it has a
callback in the pending queue.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
ff2d2bc9f2 ipmi:msghandler: Don't acquire a user refcount for queued messages
Messages already have a refcount for the user, so there's no need to
account for a new one.

As part of this, grab a refcount to the interface when processing
received messages.  The messages can be freed there, cause the user
then the interface to be freed.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
84fe1ebcc9 ipmi:msghandler: Fix locking around users and interfaces
Now that SRCU is gone from IPMI, it can no longer be sloppy about
locking.  Use the users mutex now when sending a message, not the big
ipmi_interfaces mutex, because it can result in a recursive lock.  The
users mutex will work because the interface destroy code claims it after
setting the interface in shutdown mode.

Also, due to the same changes, rework the refcounting on users and
interfaces.  Remove the refcount to an interface when the user is
freed, not when it is destroyed.  If the interface is destroyed
while the user still exists, the user will still point to the
interface to test that it is valid if the user tries to do anything
but delete the user.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
83d19f03f3 ipmi:msghandler: Remove some user level processing in panic mode
When run to completion is set, don't call things that will claim
mutexes or call user callbacks.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:48 -05:00
Corey Minyard
5017b1b026 ipmi: Add a note about the pretimeout callback
You can't do IPMI calls from the callback, it's called with locks
held.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
b2b69ee34b ipmi:watchdog: Change lock to mutex
Now that the msghandler does all callbacks in user threads, there is
no need to have a lock any more, a mutex will work fine.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
9e91f8a6c8 ipmi:msghandler: Remove srcu for the ipmi_interfaces list
With reworks srcu is no longer necessary, this simplifies locking a lot.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
3be997d5a6 ipmi:msghandler: Remove srcu from the ipmi user structure
With the restructures done, srcu is no longer required, and it's fairly
onerous.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
8b85c0f3cb ipmi:msghandler: Use the system_wq, not system_bh_wq
Everything can be run in thread context now, don't use the bh one.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
646e40bbc7 ipmi_msghandler: Change the events lock to a mutex
It can only be called from thread context now.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
557602f233 ipmi:msghandler: Deliver user messages in a work queue
This simplifies the locking and lets us remove some weird event
handling code.  deliver_response() and friends can now be called
from an atomic context.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
742219863e ipmi:msghandler: Move timer handling into a work queue
Get all operations that manipulate the interface list into thread
context.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00
Corey Minyard
305923bd3b ipmi:msghandler: Rename recv_work to smi_work
It handles both receive and transmit functions, make the name generic.

Signed-off-by: Corey Minyard <cminyard@mvista.com>
2025-05-07 17:25:47 -05:00