Pablo Neira Ayuso says:
====================
Netfilter updates for net-next
The following patchset contains Netfilter updates for net-next:
1) Extended netlink error reporting if nfnetlink attribute parser fails,
from Donald Hunter.
2) Incorrect request_module() module, from Simon Horman.
3) A series of patches to reduce memory consumption for set element
transactions.
Florian Westphal says:
"When doing a flush on a set or mass adding/removing elements from a
set, each element needs to allocate 96 bytes to hold the transactional
state.
In such cases, virtually all the information in struct nft_trans_elem
is the same.
Change nft_trans_elem to a flex-array, i.e. a single nft_trans_elem
can hold multiple set element pointers.
The number of elements that can be stored in one nft_trans_elem is limited
by the slab allocator, this series limits the compaction to at most 62
elements as it caps the reallocation to 2048 bytes of memory."
4) A series of patches to prepare the transition to dscp_t in .flowi_tos.
From Guillaume Nault.
5) Support for bitwise operations with two source registers,
from Jeremy Sowden.
* tag 'nf-next-24-11-15' of git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next:
netfilter: bitwise: add support for doing AND, OR and XOR directly
netfilter: bitwise: rename some boolean operation functions
netfilter: nf_dup4: Convert nf_dup_ipv4_route() to dscp_t.
netfilter: nft_fib: Convert nft_fib4_eval() to dscp_t.
netfilter: rpfilter: Convert rpfilter_mt() to dscp_t.
netfilter: flow_offload: Convert nft_flow_route() to dscp_t.
netfilter: ipv4: Convert ip_route_me_harder() to dscp_t.
netfilter: nf_tables: allocate element update information dynamically
netfilter: nf_tables: switch trans_elem to real flex array
netfilter: nf_tables: prepare nft audit for set element compaction
netfilter: nf_tables: prepare for multiple elements in nft_trans_elem structure
netfilter: nf_tables: add nft_trans_commit_list_add_elem helper
netfilter: bpf: Pass string literal as format argument of request_module()
netfilter: nfnetlink: Report extack policy errors for batched ops
====================
Link: https://patch.msgid.link/20241115133207.8907-1-pablo@netfilter.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Hitherto, these operations have been converted in user space to
mask-and-xor operations on one register and two immediate values, and it
is the latter which have been evaluated by the kernel. We add support
for evaluating these operations directly in kernel space on one register
and either an immediate value or a second register.
Pablo made a few changes to the original patch:
- EINVAL if NFTA_BITWISE_SREG2 is used with fast version.
- Allow _AND,_OR,_XOR with _DATA != sizeof(u32)
- Dump _SREG2 or _DATA with _AND,_OR,_XOR
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
In the next patch we add support for doing AND, OR and XOR operations
directly in the kernel, so rename some functions and an enum constant
related to mask-and-xor boolean operations.
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp() instead of reading iph->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp() instead of reading iph->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp() instead of reading iph->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp()instead of reading ip_hdr()->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Also, remove the comment about the net/ip.h include file, since it's
now required for the ip4h_dscp() helper too.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use ip4h_dscp()instead of reading iph->tos directly.
ip4h_dscp() returns a dscp_t value which is temporarily converted back
to __u8 with inet_dscp_to_dsfield(). When converting ->flowi4_tos to
dscp_t in the future, we'll only have to remove that
inet_dscp_to_dsfield() call.
Signed-off-by: Guillaume Nault <gnault@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Edward Cree says:
====================
net: make RSS+RXNFC semantics more explicit
The original semantics of ntuple filters with FLOW_RSS were not
fully understood by all drivers, some ignoring the ring_cookie from
the flow rule. Require this support to be explicitly declared by
the driver for filters relying on it to be inserted, and add self-
test coverage for this functionality.
Also teach ethtool_check_max_channel() about this.
====================
Link: https://patch.msgid.link/cover.1731499021.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In ethtool_check_max_channel(), the new RX count must not only cover the
max queue indices in RSS indirection tables and RXNFC destinations
separately, but must also, for RXNFC rules with FLOW_RSS, cover the sum
of the destination queue and the maximum index in the associated RSS
context's indirection table, since that is the highest queue that the
rule can actually deliver traffic to.
It could be argued that the max queue across all custom RSS contexts
(ethtool_get_max_rss_ctx_channel()) need no longer be considered, since
any context to which packets can actually be delivered will be targeted
by some RXNFC rule and its max will thus be allowed for by
ethtool_get_max_rxnfc_channel(). For simplicity we keep both checks, so
even RSS contexts unused by any RXNFC rule must fit the channel count.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Link: https://patch.msgid.link/43257d375434bef388e36181492aa4c458b88336.1731499022.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Ethtool ntuple filters with FLOW_RSS were originally defined as adding
the base queue ID (ring_cookie) to the value from the indirection table,
so that the same table could distribute over more than one set of queues
when used by different filters.
However, some drivers / hardware ignore the ring_cookie, and simply use
the indirection table entries as queue IDs directly. Thus, for drivers
which have not opted in by setting ethtool_ops.cap_rss_rxnfc_adds to
declare that they support the original (addition) semantics, reject in
ethtool_set_rxnfc any filter which combines FLOW_RSS and a nonzero ring.
(For a ring_cookie of zero, both behaviours are equivalent.)
Set the cap bit in sfc, as it is known to support this feature.
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Martin Habets <habetsm.xilinx@gmail.com>
Link: https://patch.msgid.link/cc3da0844083b0e301a33092a6299e4042b65221.1731499022.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The sparx5 switchdev driver can be built either with or without support
for the Lan969x switch. However, it cannot be built-in when the lan969x
driver is a loadable module because of a link-time dependency:
arm-linux-gnueabi-ld: drivers/net/ethernet/microchip/sparx5/sparx5_main.o:(.rodata+0xd44): undefined reference to `lan969x_desc'
Add a Kconfig dependency to reflect this in Kconfig, allowing all
the valid configurations but forcing sparx5 to be a loadable module
as well if lan969x is.
Fixes: 98a0111960 ("net: sparx5: add compatible string for lan969x")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Daniel Machon <daniel.machon@microchip.com>
Link: https://patch.msgid.link/20241113115513.4132548-1-arnd@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Martin KaFai Lau says:
====================
pull-request: bpf-next 2024-11-14
We've added 9 non-merge commits during the last 4 day(s) which contain
a total of 3 files changed, 226 insertions(+), 84 deletions(-).
The main changes are:
1) Fixes to bpf_msg_push/pop_data and test_sockmap. The changes has
dependency on the other changes in the bpf-next/net branch,
from Zijian Zhang.
2) Drop netns codes from mptcp test. Reuse the common helpers in
test_progs, from Geliang Tang.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next:
bpf, sockmap: Fix sk_msg_reset_curr
bpf, sockmap: Several fixes to bpf_msg_pop_data
bpf, sockmap: Several fixes to bpf_msg_push_data
selftests/bpf: Add more tests for test_txmsg_push_pop in test_sockmap
selftests/bpf: Add push/pop checking for msg_verify_data in test_sockmap
selftests/bpf: Fix total_bytes in msg_loop_rx in test_sockmap
selftests/bpf: Fix SENDPAGE data logic in test_sockmap
selftests/bpf: Add txmsg_pass to pull/push/pop in test_sockmap
selftests/bpf: Drop netns helpers in mptcp
====================
Link: https://patch.msgid.link/20241114202832.3187927-1-martin.lau@linux.dev
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Guillaume Nault says:
====================
ipv4: Prepare bpf helpers to .flowi4_tos conversion.
Continue the process of making a dscp_t variable available when setting
.flowi4_tos. This series focuses on the BPF helpers that initialise a
struct flowi4 manually.
The objective is to eventually convert .flowi4_tos to dscp_t, (to get
type annotation and prevent ECN bits from interfering with DSCP).
====================
Link: https://patch.msgid.link/cover.1731064982.git.gnault@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Donald Hunter says:
====================
tools/net/ynl: rework async notification handling
Revert patch 1bf70e6c3a which modified check_ntf() and instead add a
new poll_ntf() with async notification semantics. See patch 2 for a
detailed description.
====================
Link: https://patch.msgid.link/20241113090843.72917-1-donald.hunter@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The notification handling in ynl is currently very simple, using sleep()
to wait a period of time and then handling all the buffered messages in
a single batch.
This patch adds async notification handling so that messages can be
processed as they are received. This makes it possible to use ynl as a
library that supplies notifications in a timely manner.
- Add poll_ntf() to be a generator that yields 1 notification at a
time and blocks until a notification is available.
- Add a --duration parameter to the CLI, with --sleep as an alias.
./tools/net/ynl/cli.py \
--spec <SPEC> --subscribe <TOPIC> [ --duration <SECS> ]
The cli will report any notifications for duration seconds and then
exit. If duration is not specified, then it will poll forever, until
interrupted.
Here is an example python snippet that shows how to use ynl as a library
for receiving notifications:
ynl = YnlFamily(f"{dir}/rt_route.yaml")
ynl.ntf_subscribe('rtnlgrp-ipv4-route')
for event in ynl.poll_ntf():
handle(event)
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20241113090843.72917-3-donald.hunter@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Heiner Kallweit says:
====================
net: phy: switch eee_broken_modes to linkmode bitmap and add accessor
eee_broken_modes has a eee_cap1 register layout currently. This doesn't
allow to flag e.g. 2.5Gbps or 5Gbps BaseT EEE as broken. To overcome
this limitation switch eee_broken_modes to a linkmode bitmap.
Add an accessor for the bitmap and use it in r8169.
====================
Link: https://patch.msgid.link/405734c5-0ed4-40e4-9ac9-91084b9536d6@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Vendor driver r8125 doesn't advertise 2.5G EEE on RTL8125A, and r8126
doesn't advertise 5G EEE. Likely there are compatibility issues,
therefore do the same in r8169.
With this change we don't have to disable 2.5G EEE advertisement in
rtl8125a_config_eee_phy() any longer.
We use new phylib accessor phy_set_eee_broken() to mark the respective
EEE modes as broken.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Link: https://patch.msgid.link/ce185e10-8a2f-4cf8-a49b-fd8fb3c3c8a1@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pull bcachefs fixes from Kent Overstreet:
"This fixes one minor regression from the btree cache fixes (in the
scan_for_btree_nodes repair path) - and the shutdown path fix is the
big one here, in terms of bugs closed:
- Assorted tiny syzbot fixes
- Shutdown path fix: "bch2_btree_write_buffer_flush_going_ro()"
The shutdown path wasn't flushing the btree write buffer, leading
to shutting down while we still had operations in flight. This
fixes a whole slew of syzbot bugs, and undoubtedly other strange
heisenbugs.
* tag 'bcachefs-2024-11-13' of git://evilpiepirate.org/bcachefs:
bcachefs: Fix assertion pop in bch2_ptr_swab()
bcachefs: Fix journal_entry_dev_usage_to_text() overrun
bcachefs: Allow for unknown key types in backpointers fsck
bcachefs: Fix assertion pop in topology repair
bcachefs: Fix hidden btree errors when reading roots
bcachefs: Fix validate_bset() repair path
bcachefs: Fix missing validation for bch_backpointer.level
bcachefs: Fix bch_member.btree_bitmap_shift validation
bcachefs: bch2_btree_write_buffer_flush_going_ro()
Add support for the 'ethtool -d <dev>' command to retrieve and print
a register dump for fbnic. The dump defaults to version 1 and consists
of two parts: all the register sections that can be dumped linearly, and
an RPC RAM section that is structured in an interleaved fashion and
requires special handling. For each register section, the dump also
contains the start and end boundary information which can simplify parsing.
Signed-off-by: Mohsin Bashir <mohsin.bashr@gmail.com>
Link: https://patch.msgid.link/20241112222605.3303211-1-mohsin.bashr@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Move the timeout/expire/flag members from nft_trans_one_elem struct into
a dybamically allocated structure, only needed when timeout update was
requested.
This halves size of nft_trans_one_elem struct and allows to compact up to
124 elements in one transaction container rather than 62.
This halves memory requirements for a large flush or insert transaction,
where ->update remains NULL.
Care has to be taken to release the extra data in all spots, including
abort path.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
When queueing a set element add or removal operation to the transaction
log, check if the previous operation already asks for a the identical
operation on the same set.
If so, store the element reference in the preceding operation.
This significantlty reduces memory consumption when many set add/delete
operations appear in a single transaction.
Example: 10k elements require 937kb of memory (10k allocations from
kmalloc-96 slab).
Assuming we can compact 4 elements in the same set, 468 kbytes
are needed (64 bytes for base struct, nft_trans_elemn, 32 bytes
for nft_trans_one_elem structure, so 2500 allocations from kmalloc-192
slab).
For large batch updates we can compact up to 62 elements
into one single nft_trans_elem structure (~65% mem reduction):
(64 bytes for base struct, nft_trans_elem, 32 byte for nft_trans_one_elem
struct).
We can halve size of nft_trans_one_elem struct by moving
timeout/expire/update_flags into a dynamically allocated structure,
this allows to store 124 elements in a 2k slab nft_trans_elem struct.
This is done in a followup patch.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
nftables audit log format emits the number of added/deleted rules, sets,
set elements and so on, to userspace:
table=t1 family=2 entries=4 op=nft_register_set
~~~~~~~~~
At this time, the 'entries' key is the number of transactions that will
be applied.
The upcoming set element compression will coalesce subsequent
adds/deletes to the same set requests in the same transaction
request to conseve memory.
Without this patch, we'd under-report the number of altered elements.
Increment the audit counter by the number of elements to keep the reported
entries value the same.
Without this, nft_audit.sh selftest fails because the recorded
(expected) entries key is smaller than the expected one.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add helpers to release the individual elements contained in the
trans_elem container structure.
No functional change intended.
Followup patch will add 'nelems' member and will turn 'priv' into
a flexible array.
These helpers can then loop over all elements.
Care needs to be taken to handle a mix of new elements and existing
elements that are being updated (e.g. timeout refresh).
Before this patch, NEWSETELEM transaction with update is released
early so nft_trans_set_elem_destroy() won't get called, so we need
to skip elements marked as update.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add and use a wrapper to append trans_elem structures to the
transaction log.
Unlike the existing helper, pass a gfp_t to indicate if sleeping
is allowed.
This will be used by a followup patch to realloc nft_trans_elem
structures after they gain a flexible array member to reduce
number of such container structures on the transaction list.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Both gcc-14 and clang-18 report that passing a non-string literal as the
format argument of request_module() is potentially insecure.
E.g. clang-18 says:
.../nf_bpf_link.c:46:24: warning: format string is not a string literal (potentially insecure) [-Wformat-security]
46 | err = request_module(mod);
| ^~~
.../kmod.h:25:55: note: expanded from macro 'request_module'
25 | #define request_module(mod...) __request_module(true, mod)
| ^~~
.../nf_bpf_link.c:46:24: note: treat the string as an argument to avoid this
46 | err = request_module(mod);
| ^
| "%s",
.../kmod.h:25:55: note: expanded from macro 'request_module'
25 | #define request_module(mod...) __request_module(true, mod)
| ^
It is always the case where the contents of mod is safe to pass as the
format argument. That is, in my understanding, it never contains any
format escape sequences.
But, it seems better to be safe than sorry. And, as a bonus, compiler
output becomes less verbose by addressing this issue as suggested by
clang-18.
Compile tested only.
Signed-off-by: Simon Horman <horms@kernel.org>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The nftables batch processing does not currently populate extack with
policy errors. Fix this by passing extack when parsing batch messages.
Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Hangbin Liu says:
====================
bonding: fix ns targets not work on hardware NIC
The first patch fixed ns targets not work on hardware NIC when bonding
set arp_validate.
The second patch add a related selftest for bonding.
v4: Thanks Nikolay for the comments:
use bond_slave_ns_maddrs_{add/del} with clear name
fix comments typos
remove _slave_set_ns_maddrs underscore directly
update bond_option_arp_validate_set() change logic
v3: use ndisc_mc_map to convert the mcast mac address (Jay Vosburgh)
v2: only add/del mcast group on backup slaves when arp_validate is set (Jay Vosburgh)
arp_validate doesn't support 3ad, tlb, alb. So let's only do it on ab mode.
====================
Link: https://patch.msgid.link/20241111101650.27685-1-liuhangbin@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Commit 4598380f9c ("bonding: fix ns validation on backup slaves")
tried to resolve the issue where backup slaves couldn't be brought up when
receiving IPv6 Neighbor Solicitation (NS) messages. However, this fix only
worked for drivers that receive all multicast messages, such as the veth
interface.
For standard drivers, the NS multicast message is silently dropped because
the slave device is not a member of the NS target multicast group.
To address this, we need to make the slave device join the NS target
multicast group, ensuring it can receive these IPv6 NS messages to validate
the slave’s status properly.
There are three policies before joining the multicast group:
1. All settings must be under active-backup mode (alb and tlb do not support
arp_validate), with backup slaves and slaves supporting multicast.
2. We can add or remove multicast groups when arp_validate changes.
3. Other operations, such as enslaving, releasing, or setting NS targets,
need to be guarded by arp_validate.
Fixes: 4e24be018e ("bonding: add new parameter ns_targets")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The first PPS latch time needs to be calculated by the driver
(in rounded off seconds) and configured as the start time
offset for the cycle. After synchronizing two PTP clocks
running as master/slave, missing this would cause master
and slave to start immediately with some milliseconds
drift which causes the PPS signal to never synchronize with
the PTP master.
Fixes: 186734c158 ("net: ti: icssg-prueth: add packet timestamping and ptp support")
Signed-off-by: Meghana Malladi <m-malladi@ti.com>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Reviewed-by: MD Danish Anwar <danishanwar@ti.com>
Link: https://patch.msgid.link/20241111095842.478833-1-m-malladi@ti.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
LAN9646 switch is a 6-port switch with functions like KSZ9897. It has
4 internal PHYs and 1 SGMII port. The chip id read from hardware is
same as KSZ9477, so software driver needs to create a new chip id and
group allowable functions under its chip data structure to
differentiate the product.
Signed-off-by: Tristram Ha <tristram.ha@microchip.com>
Link: https://patch.msgid.link/20241109015705.82685-3-Tristram.Ha@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
